An Overview and Evaluation of Recent Machine Learning Imputation Methods Using Cardiac Imaging Data.

PubMed

Liu, Yuzhe; Gopalakrishnan, Vanathi

2017-03-01

Many clinical research datasets have a large percentage of missing values that directly impacts their usefulness in yielding high accuracy classifiers when used for training in supervised machine learning. While missing value imputation methods have been shown to work well with smaller percentages of missing values, their ability to impute sparse clinical research data can be problem specific. We previously attempted to learn quantitative guidelines for ordering cardiac magnetic resonance imaging during the evaluation for pediatric cardiomyopathy, but missing data significantly reduced our usable sample size. In this work, we sought to determine if increasing the usable sample size through imputation would allow us to learn better guidelines. We first review several machine learning methods for estimating missing data. Then, we apply four popular methods (mean imputation, decision tree, k-nearest neighbors, and self-organizing maps) to a clinical research dataset of pediatric patients undergoing evaluation for cardiomyopathy. Using Bayesian Rule Learning (BRL) to learn ruleset models, we compared the performance of imputation-augmented models versus unaugmented models. We found that all four imputation-augmented models performed similarly to unaugmented models. While imputation did not improve performance, it did provide evidence for the robustness of our learned models.

SPSS Syntax for Missing Value Imputation in Test and Questionnaire Data

ERIC Educational Resources Information Center

van Ginkel, Joost R.; van der Ark, L. Andries

2005-01-01

A well-known problem in the analysis of test and questionnaire data is that some item scores may be missing. Advanced methods for the imputation of missing data are available, such as multiple imputation under the multivariate normal model and imputation under the saturated logistic model (Schafer, 1997). Accompanying software was made available…

Missing value imputation for microarray data: a comprehensive comparison study and a web tool

PubMed Central

2013-01-01

Background Microarray data are usually peppered with missing values due to various reasons. However, most of the downstream analyses for microarray data require complete datasets. Therefore, accurate algorithms for missing value estimation are needed for improving the performance of microarray data analyses. Although many algorithms have been developed, there are many debates on the selection of the optimal algorithm. The studies about the performance comparison of different algorithms are still incomprehensive, especially in the number of benchmark datasets used, the number of algorithms compared, the rounds of simulation conducted, and the performance measures used. Results In this paper, we performed a comprehensive comparison by using (I) thirteen datasets, (II) nine algorithms, (III) 110 independent runs of simulation, and (IV) three types of measures to evaluate the performance of each imputation algorithm fairly. First, the effects of different types of microarray datasets on the performance of each imputation algorithm were evaluated. Second, we discussed whether the datasets from different species have different impact on the performance of different algorithms. To assess the performance of each algorithm fairly, all evaluations were performed using three types of measures. Our results indicate that the performance of an imputation algorithm mainly depends on the type of a dataset but not on the species where the samples come from. In addition to the statistical measure, two other measures with biological meanings are useful to reflect the impact of missing value imputation on the downstream data analyses. Our study suggests that local-least-squares-based methods are good choices to handle missing values for most of the microarray datasets. Conclusions In this work, we carried out a comprehensive comparison of the algorithms for microarray missing value imputation. Based on such a comprehensive comparison, researchers could choose the optimal algorithm for their datasets easily. Moreover, new imputation algorithms could be compared with the existing algorithms using this comparison strategy as a standard protocol. In addition, to assist researchers in dealing with missing values easily, we built a web-based and easy-to-use imputation tool, MissVIA (http://cosbi.ee.ncku.edu.tw/MissVIA), which supports many imputation algorithms. Once users upload a real microarray dataset and choose the imputation algorithms, MissVIA will determine the optimal algorithm for the users' data through a series of simulations, and then the imputed results can be downloaded for the downstream data analyses. PMID:24565220

The Effects of Methods of Imputation for Missing Values on the Validity and Reliability of Scales

ERIC Educational Resources Information Center

Cokluk, Omay; Kayri, Murat

2011-01-01

The main aim of this study is the comparative examination of the factor structures, corrected item-total correlations, and Cronbach-alpha internal consistency coefficients obtained by different methods used in imputation for missing values in conditions of not having missing values, and having missing values of different rates in terms of testing…

Prediction of regulatory gene pairs using dynamic time warping and gene ontology.

PubMed

Yang, Andy C; Hsu, Hui-Huang; Lu, Ming-Da; Tseng, Vincent S; Shih, Timothy K

2014-01-01

Selecting informative genes is the most important task for data analysis on microarray gene expression data. In this work, we aim at identifying regulatory gene pairs from microarray gene expression data. However, microarray data often contain multiple missing expression values. Missing value imputation is thus needed before further processing for regulatory gene pairs becomes possible. We develop a novel approach to first impute missing values in microarray time series data by combining k-Nearest Neighbour (KNN), Dynamic Time Warping (DTW) and Gene Ontology (GO). After missing values are imputed, we then perform gene regulation prediction based on our proposed DTW-GO distance measurement of gene pairs. Experimental results show that our approach is more accurate when compared with existing missing value imputation methods on real microarray data sets. Furthermore, our approach can also discover more regulatory gene pairs that are known in the literature than other methods.

A Comparison of Joint Model and Fully Conditional Specification Imputation for Multilevel Missing Data

ERIC Educational Resources Information Center

Mistler, Stephen A.; Enders, Craig K.

2017-01-01

Multiple imputation methods can generally be divided into two broad frameworks: joint model (JM) imputation and fully conditional specification (FCS) imputation. JM draws missing values simultaneously for all incomplete variables using a multivariate distribution, whereas FCS imputes variables one at a time from a series of univariate conditional…

Clustering with Missing Values: No Imputation Required

NASA Technical Reports Server (NTRS)

Wagstaff, Kiri

2004-01-01

Clustering algorithms can identify groups in large data sets, such as star catalogs and hyperspectral images. In general, clustering methods cannot analyze items that have missing data values. Common solutions either fill in the missing values (imputation) or ignore the missing data (marginalization). Imputed values are treated as just as reliable as the truly observed data, but they are only as good as the assumptions used to create them. In contrast, we present a method for encoding partially observed features as a set of supplemental soft constraints and introduce the KSC algorithm, which incorporates constraints into the clustering process. In experiments on artificial data and data from the Sloan Digital Sky Survey, we show that soft constraints are an effective way to enable clustering with missing values.

Purposeful Variable Selection and Stratification to Impute Missing FAST Data in Trauma Research

PubMed Central

Fuchs, Paul A.; del Junco, Deborah J.; Fox, Erin E.; Holcomb, John B.; Rahbar, Mohammad H.; Wade, Charles A.; Alarcon, Louis H.; Brasel, Karen J.; Bulger, Eileen M.; Cohen, Mitchell J.; Myers, John G.; Muskat, Peter; Phelan, Herb A.; Schreiber, Martin A.; Cotton, Bryan A.

2013-01-01

Background The Focused Assessment with Sonography for Trauma (FAST) exam is an important variable in many retrospective trauma studies. The purpose of this study was to devise an imputation method to overcome missing data for the FAST exam. Due to variability in patients’ injuries and trauma care, these data are unlikely to be missing completely at random (MCAR), raising concern for validity when analyses exclude patients with missing values. Methods Imputation was conducted under a less restrictive, more plausible missing at random (MAR) assumption. Patients with missing FAST exams had available data on alternate, clinically relevant elements that were strongly associated with FAST results in complete cases, especially when considered jointly. Subjects with missing data (32.7%) were divided into eight mutually exclusive groups based on selected variables that both described the injury and were associated with missing FAST values. Additional variables were selected within each group to classify missing FAST values as positive or negative, and correct FAST exam classification based on these variables was determined for patients with non-missing FAST values. Results Severe head/neck injury (odds ratio, OR=2.04), severe extremity injury (OR=4.03), severe abdominal injury (OR=1.94), no injury (OR=1.94), other abdominal injury (OR=0.47), other head/neck injury (OR=0.57) and other extremity injury (OR=0.45) groups had significant ORs for missing data; the other group odds ratio was not significant (OR=0.84). All 407 missing FAST values were imputed, with 109 classified as positive. Correct classification of non-missing FAST results using the alternate variables was 87.2%. Conclusions Purposeful imputation for missing FAST exams based on interactions among selected variables assessed by simple stratification may be a useful adjunct to sensitivity analysis in the evaluation of imputation strategies under different missing data mechanisms. This approach has the potential for widespread application in clinical and translational research and validation is warranted. Level of Evidence Level II Prognostic or Epidemiological PMID:23778515

MVIAeval: a web tool for comprehensively evaluating the performance of a new missing value imputation algorithm.

PubMed

Wu, Wei-Sheng; Jhou, Meng-Jhun

2017-01-13

Missing value imputation is important for microarray data analyses because microarray data with missing values would significantly degrade the performance of the downstream analyses. Although many microarray missing value imputation algorithms have been developed, an objective and comprehensive performance comparison framework is still lacking. To solve this problem, we previously proposed a framework which can perform a comprehensive performance comparison of different existing algorithms. Also the performance of a new algorithm can be evaluated by our performance comparison framework. However, constructing our framework is not an easy task for the interested researchers. To save researchers' time and efforts, here we present an easy-to-use web tool named MVIAeval (Missing Value Imputation Algorithm evaluator) which implements our performance comparison framework. MVIAeval provides a user-friendly interface allowing users to upload the R code of their new algorithm and select (i) the test datasets among 20 benchmark microarray (time series and non-time series) datasets, (ii) the compared algorithms among 12 existing algorithms, (iii) the performance indices from three existing ones, (iv) the comprehensive performance scores from two possible choices, and (v) the number of simulation runs. The comprehensive performance comparison results are then generated and shown as both figures and tables. MVIAeval is a useful tool for researchers to easily conduct a comprehensive and objective performance evaluation of their newly developed missing value imputation algorithm for microarray data or any data which can be represented as a matrix form (e.g. NGS data or proteomics data). Thus, MVIAeval will greatly expedite the progress in the research of missing value imputation algorithms.

Examining solutions to missing data in longitudinal nursing research.

PubMed

Roberts, Mary B; Sullivan, Mary C; Winchester, Suzy B

2017-04-01

Longitudinal studies are highly valuable in pediatrics because they provide useful data about developmental patterns of child health and behavior over time. When data are missing, the value of the research is impacted. The study's purpose was to (1) introduce a three-step approach to assess and address missing data and (2) illustrate this approach using categorical and continuous-level variables from a longitudinal study of premature infants. A three-step approach with simulations was followed to assess the amount and pattern of missing data and to determine the most appropriate imputation method for the missing data. Patterns of missingness were Missing Completely at Random, Missing at Random, and Not Missing at Random. Missing continuous-level data were imputed using mean replacement, stochastic regression, multiple imputation, and fully conditional specification (FCS). Missing categorical-level data were imputed using last value carried forward, hot-decking, stochastic regression, and FCS. Simulations were used to evaluate these imputation methods under different patterns of missingness at different levels of missing data. The rate of missingness was 16-23% for continuous variables and 1-28% for categorical variables. FCS imputation provided the least difference in mean and standard deviation estimates for continuous measures. FCS imputation was acceptable for categorical measures. Results obtained through simulation reinforced and confirmed these findings. Significant investments are made in the collection of longitudinal data. The prudent handling of missing data can protect these investments and potentially improve the scientific information contained in pediatric longitudinal studies. © 2017 Wiley Periodicals, Inc.

Impact of missing data imputation methods on gene expression clustering and classification.

PubMed

de Souto, Marcilio C P; Jaskowiak, Pablo A; Costa, Ivan G

2015-02-26

Several missing value imputation methods for gene expression data have been proposed in the literature. In the past few years, researchers have been putting a great deal of effort into presenting systematic evaluations of the different imputation algorithms. Initially, most algorithms were assessed with an emphasis on the accuracy of the imputation, using metrics such as the root mean squared error. However, it has become clear that the success of the estimation of the expression value should be evaluated in more practical terms as well. One can consider, for example, the ability of the method to preserve the significant genes in the dataset, or its discriminative/predictive power for classification/clustering purposes. We performed a broad analysis of the impact of five well-known missing value imputation methods on three clustering and four classification methods, in the context of 12 cancer gene expression datasets. We employed a statistical framework, for the first time in this field, to assess whether different imputation methods improve the performance of the clustering/classification methods. Our results suggest that the imputation methods evaluated have a minor impact on the classification and downstream clustering analyses. Simple methods such as replacing the missing values by mean or the median values performed as well as more complex strategies. The datasets analyzed in this study are available at http://costalab.org/Imputation/ .

Two-pass imputation algorithm for missing value estimation in gene expression time series.

PubMed

Tsiporkova, Elena; Boeva, Veselka

2007-10-01

Gene expression microarray experiments frequently generate datasets with multiple values missing. However, most of the analysis, mining, and classification methods for gene expression data require a complete matrix of gene array values. Therefore, the accurate estimation of missing values in such datasets has been recognized as an important issue, and several imputation algorithms have already been proposed to the biological community. Most of these approaches, however, are not particularly suitable for time series expression profiles. In view of this, we propose a novel imputation algorithm, which is specially suited for the estimation of missing values in gene expression time series data. The algorithm utilizes Dynamic Time Warping (DTW) distance in order to measure the similarity between time expression profiles, and subsequently selects for each gene expression profile with missing values a dedicated set of candidate profiles for estimation. Three different DTW-based imputation (DTWimpute) algorithms have been considered: position-wise, neighborhood-wise, and two-pass imputation. These have initially been prototyped in Perl, and their accuracy has been evaluated on yeast expression time series data using several different parameter settings. The experiments have shown that the two-pass algorithm consistently outperforms, in particular for datasets with a higher level of missing entries, the neighborhood-wise and the position-wise algorithms. The performance of the two-pass DTWimpute algorithm has further been benchmarked against the weighted K-Nearest Neighbors algorithm, which is widely used in the biological community; the former algorithm has appeared superior to the latter one. Motivated by these findings, indicating clearly the added value of the DTW techniques for missing value estimation in time series data, we have built an optimized C++ implementation of the two-pass DTWimpute algorithm. The software also provides for a choice between three different initial rough imputation methods.

Combining Fourier and lagged k-nearest neighbor imputation for biomedical time series data.

PubMed

Rahman, Shah Atiqur; Huang, Yuxiao; Claassen, Jan; Heintzman, Nathaniel; Kleinberg, Samantha

2015-12-01

Most clinical and biomedical data contain missing values. A patient's record may be split across multiple institutions, devices may fail, and sensors may not be worn at all times. While these missing values are often ignored, this can lead to bias and error when the data are mined. Further, the data are not simply missing at random. Instead the measurement of a variable such as blood glucose may depend on its prior values as well as that of other variables. These dependencies exist across time as well, but current methods have yet to incorporate these temporal relationships as well as multiple types of missingness. To address this, we propose an imputation method (FLk-NN) that incorporates time lagged correlations both within and across variables by combining two imputation methods, based on an extension to k-NN and the Fourier transform. This enables imputation of missing values even when all data at a time point is missing and when there are different types of missingness both within and across variables. In comparison to other approaches on three biological datasets (simulated and actual Type 1 diabetes datasets, and multi-modality neurological ICU monitoring) the proposed method has the highest imputation accuracy. This was true for up to half the data being missing and when consecutive missing values are a significant fraction of the overall time series length. Copyright © 2015 Elsevier Inc. All rights reserved.

Examining Solutions to Missing Data in Longitudinal Nursing Research

PubMed Central

Roberts, Mary B.; Sullivan, Mary C.; Winchester, Suzy B.

2017-01-01

Purpose Longitudinal studies are highly valuable in pediatrics because they provide useful data about developmental patterns of child health and behavior over time. When data are missing, the value of the research is impacted. The study’s purpose was to: (1) introduce a 3-step approach to assess and address missing data; (2) illustrate this approach using categorical and continuous level variables from a longitudinal study of premature infants. Methods A three-step approach with simulations was followed to assess the amount and pattern of missing data and to determine the most appropriate imputation method for the missing data. Patterns of missingness were Missing Completely at Random, Missing at Random, and Not Missing at Random. Missing continuous-level data were imputed using mean replacement, stochastic regression, multiple imputation, and fully conditional specification. Missing categorical-level data were imputed using last value carried forward, hot-decking, stochastic regression, and fully conditional specification. Simulations were used to evaluate these imputation methods under different patterns of missingness at different levels of missing data. Results The rate of missingness was 16–23% for continuous variables and 1–28% for categorical variables. Fully conditional specification imputation provided the least difference in mean and standard deviation estimates for continuous measures. Fully conditional specification imputation was acceptable for categorical measures. Results obtained through simulation reinforced and confirmed these findings. Practice Implications Significant investments are made in the collection of longitudinal data. The prudent handling of missing data can protect these investments and potentially improve the scientific information contained in pediatric longitudinal studies. PMID:28425202

Missing data exploration: highlighting graphical presentation of missing pattern.

PubMed

Zhang, Zhongheng

2015-12-01

Functions shipped with R base can fulfill many tasks of missing data handling. However, because the data volume of electronic medical record (EMR) system is always very large, more sophisticated methods may be helpful in data management. The article focuses on missing data handling by using advanced techniques. There are three types of missing data, that is, missing completely at random (MCAR), missing at random (MAR) and not missing at random (NMAR). This classification system depends on how missing values are generated. Two packages, Multivariate Imputation by Chained Equations (MICE) and Visualization and Imputation of Missing Values (VIM), provide sophisticated functions to explore missing data pattern. In particular, the VIM package is especially helpful in visual inspection of missing data. Finally, correlation analysis provides information on the dependence of missing data on other variables. Such information is useful in subsequent imputations.

Working with Missing Values

ERIC Educational Resources Information Center

Acock, Alan C.

2005-01-01

Less than optimum strategies for missing values can produce biased estimates, distorted statistical power, and invalid conclusions. After reviewing traditional approaches (listwise, pairwise, and mean substitution), selected alternatives are covered including single imputation, multiple imputation, and full information maximum likelihood…

Nonparametric Multiple Imputation for Questionnaires with Individual Skip Patterns and Constraints: The Case of Income Imputation in The National Educational Panel Study

ERIC Educational Resources Information Center

Aßmann, Christian; Würbach, Ariane; Goßmann, Solange; Geissler, Ferdinand; Bela, Anika

2017-01-01

Large-scale surveys typically exhibit data structures characterized by rich mutual dependencies between surveyed variables and individual-specific skip patterns. Despite high efforts in fieldwork and questionnaire design, missing values inevitably occur. One approach for handling missing values is to provide multiply imputed data sets, thus…

Integrative missing value estimation for microarray data.

PubMed

Hu, Jianjun; Li, Haifeng; Waterman, Michael S; Zhou, Xianghong Jasmine

2006-10-12

Missing value estimation is an important preprocessing step in microarray analysis. Although several methods have been developed to solve this problem, their performance is unsatisfactory for datasets with high rates of missing data, high measurement noise, or limited numbers of samples. In fact, more than 80% of the time-series datasets in Stanford Microarray Database contain less than eight samples. We present the integrative Missing Value Estimation method (iMISS) by incorporating information from multiple reference microarray datasets to improve missing value estimation. For each gene with missing data, we derive a consistent neighbor-gene list by taking reference data sets into consideration. To determine whether the given reference data sets are sufficiently informative for integration, we use a submatrix imputation approach. Our experiments showed that iMISS can significantly and consistently improve the accuracy of the state-of-the-art Local Least Square (LLS) imputation algorithm by up to 15% improvement in our benchmark tests. We demonstrated that the order-statistics-based integrative imputation algorithms can achieve significant improvements over the state-of-the-art missing value estimation approaches such as LLS and is especially good for imputing microarray datasets with a limited number of samples, high rates of missing data, or very noisy measurements. With the rapid accumulation of microarray datasets, the performance of our approach can be further improved by incorporating larger and more appropriate reference datasets.

Treatments of Missing Values in Large National Data Affect Conclusions: The Impact of Multiple Imputation on Arthroplasty Research.

PubMed

Ondeck, Nathaniel T; Fu, Michael C; Skrip, Laura A; McLynn, Ryan P; Su, Edwin P; Grauer, Jonathan N

2018-03-01

Despite the advantages of large, national datasets, one continuing concern is missing data values. Complete case analysis, where only cases with complete data are analyzed, is commonly used rather than more statistically rigorous approaches such as multiple imputation. This study characterizes the potential selection bias introduced using complete case analysis and compares the results of common regressions using both techniques following unicompartmental knee arthroplasty. Patients undergoing unicompartmental knee arthroplasty were extracted from the 2005 to 2015 National Surgical Quality Improvement Program. As examples, the demographics of patients with and without missing preoperative albumin and hematocrit values were compared. Missing data were then treated with both complete case analysis and multiple imputation (an approach that reproduces the variation and associations that would have been present in a full dataset) and the conclusions of common regressions for adverse outcomes were compared. A total of 6117 patients were included, of which 56.7% were missing at least one value. Younger, female, and healthier patients were more likely to have missing preoperative albumin and hematocrit values. The use of complete case analysis removed 3467 patients from the study in comparison with multiple imputation which included all 6117 patients. The 2 methods of handling missing values led to differing associations of low preoperative laboratory values with commonly studied adverse outcomes. The use of complete case analysis can introduce selection bias and may lead to different conclusions in comparison with the statistically rigorous multiple imputation approach. Joint surgeons should consider the methods of handling missing values when interpreting arthroplasty research. Copyright © 2017 Elsevier Inc. All rights reserved.

Missing data exploration: highlighting graphical presentation of missing pattern

PubMed Central

2015-01-01

Functions shipped with R base can fulfill many tasks of missing data handling. However, because the data volume of electronic medical record (EMR) system is always very large, more sophisticated methods may be helpful in data management. The article focuses on missing data handling by using advanced techniques. There are three types of missing data, that is, missing completely at random (MCAR), missing at random (MAR) and not missing at random (NMAR). This classification system depends on how missing values are generated. Two packages, Multivariate Imputation by Chained Equations (MICE) and Visualization and Imputation of Missing Values (VIM), provide sophisticated functions to explore missing data pattern. In particular, the VIM package is especially helpful in visual inspection of missing data. Finally, correlation analysis provides information on the dependence of missing data on other variables. Such information is useful in subsequent imputations. PMID:26807411

Incomplete Data in Smart Grid: Treatment of Values in Electric Vehicle Charging Data

DOE Office of Scientific and Technical Information (OSTI.GOV)

Majipour, Mostafa; Chu, Peter; Gadh, Rajit

2014-11-03

In this paper, five imputation methods namely Constant (zero), Mean, Median, Maximum Likelihood, and Multiple Imputation methods have been applied to compensate for missing values in Electric Vehicle (EV) charging data. The outcome of each of these methods have been used as the input to a prediction algorithm to forecast the EV load in the next 24 hours at each individual outlet. The data is real world data at the outlet level from the UCLA campus parking lots. Given the sparsity of the data, both Median and Constant (=zero) imputations improved the prediction results. Since in most missing value casesmore » in our database, all values of that instance are missing, the multivariate imputation methods did not improve the results significantly compared to univariate approaches.« less

A pattern-mixture model approach for handling missing continuous outcome data in longitudinal cluster randomized trials.

PubMed

Fiero, Mallorie H; Hsu, Chiu-Hsieh; Bell, Melanie L

2017-11-20

We extend the pattern-mixture approach to handle missing continuous outcome data in longitudinal cluster randomized trials, which randomize groups of individuals to treatment arms, rather than the individuals themselves. Individuals who drop out at the same time point are grouped into the same dropout pattern. We approach extrapolation of the pattern-mixture model by applying multilevel multiple imputation, which imputes missing values while appropriately accounting for the hierarchical data structure found in cluster randomized trials. To assess parameters of interest under various missing data assumptions, imputed values are multiplied by a sensitivity parameter, k, which increases or decreases imputed values. Using simulated data, we show that estimates of parameters of interest can vary widely under differing missing data assumptions. We conduct a sensitivity analysis using real data from a cluster randomized trial by increasing k until the treatment effect inference changes. By performing a sensitivity analysis for missing data, researchers can assess whether certain missing data assumptions are reasonable for their cluster randomized trial. Copyright © 2017 John Wiley & Sons, Ltd.

Multiple imputation of missing passenger boarding data in the national census of ferry operators

DOT National Transportation Integrated Search

2008-08-01

This report presents findings from the 2006 National Census of Ferry Operators (NCFO) augmented : with imputed values for passengers and passenger miles. Due to the imputation procedures used to calculate missing data, totals in Table 1 may not corre...

Missing Data and Multiple Imputation in the Context of Multivariate Analysis of Variance

ERIC Educational Resources Information Center

Finch, W. Holmes

2016-01-01

Multivariate analysis of variance (MANOVA) is widely used in educational research to compare means on multiple dependent variables across groups. Researchers faced with the problem of missing data often use multiple imputation of values in place of the missing observations. This study compares the performance of 2 methods for combining p values in…

Order-restricted inference for means with missing values.

PubMed

Wang, Heng; Zhong, Ping-Shou

2017-09-01

Missing values appear very often in many applications, but the problem of missing values has not received much attention in testing order-restricted alternatives. Under the missing at random (MAR) assumption, we impute the missing values nonparametrically using kernel regression. For data with imputation, the classical likelihood ratio test designed for testing the order-restricted means is no longer applicable since the likelihood does not exist. This article proposes a novel method for constructing test statistics for assessing means with an increasing order or a decreasing order based on jackknife empirical likelihood (JEL) ratio. It is shown that the JEL ratio statistic evaluated under the null hypothesis converges to a chi-bar-square distribution, whose weights depend on missing probabilities and nonparametric imputation. Simulation study shows that the proposed test performs well under various missing scenarios and is robust for normally and nonnormally distributed data. The proposed method is applied to an Alzheimer's disease neuroimaging initiative data set for finding a biomarker for the diagnosis of the Alzheimer's disease. © 2017, The International Biometric Society.

Accounting for the Multiple Natures of Missing Values in Label-Free Quantitative Proteomics Data Sets to Compare Imputation Strategies.

PubMed

Lazar, Cosmin; Gatto, Laurent; Ferro, Myriam; Bruley, Christophe; Burger, Thomas

2016-04-01

Missing values are a genuine issue in label-free quantitative proteomics. Recent works have surveyed the different statistical methods to conduct imputation and have compared them on real or simulated data sets and recommended a list of missing value imputation methods for proteomics application. Although insightful, these comparisons do not account for two important facts: (i) depending on the proteomics data set, the missingness mechanism may be of different natures and (ii) each imputation method is devoted to a specific type of missingness mechanism. As a result, we believe that the question at stake is not to find the most accurate imputation method in general but instead the most appropriate one. We describe a series of comparisons that support our views: For instance, we show that a supposedly "under-performing" method (i.e., giving baseline average results), if applied at the "appropriate" time in the data-processing pipeline (before or after peptide aggregation) on a data set with the "appropriate" nature of missing values, can outperform a blindly applied, supposedly "better-performing" method (i.e., the reference method from the state-of-the-art). This leads us to formulate few practical guidelines regarding the choice and the application of an imputation method in a proteomics context.

A nonparametric multiple imputation approach for missing categorical data.

PubMed

Zhou, Muhan; He, Yulei; Yu, Mandi; Hsu, Chiu-Hsieh

2017-06-06

Incomplete categorical variables with more than two categories are common in public health data. However, most of the existing missing-data methods do not use the information from nonresponse (missingness) probabilities. We propose a nearest-neighbour multiple imputation approach to impute a missing at random categorical outcome and to estimate the proportion of each category. The donor set for imputation is formed by measuring distances between each missing value with other non-missing values. The distance function is calculated based on a predictive score, which is derived from two working models: one fits a multinomial logistic regression for predicting the missing categorical outcome (the outcome model) and the other fits a logistic regression for predicting missingness probabilities (the missingness model). A weighting scheme is used to accommodate contributions from two working models when generating the predictive score. A missing value is imputed by randomly selecting one of the non-missing values with the smallest distances. We conduct a simulation to evaluate the performance of the proposed method and compare it with several alternative methods. A real-data application is also presented. The simulation study suggests that the proposed method performs well when missingness probabilities are not extreme under some misspecifications of the working models. However, the calibration estimator, which is also based on two working models, can be highly unstable when missingness probabilities for some observations are extremely high. In this scenario, the proposed method produces more stable and better estimates. In addition, proper weights need to be chosen to balance the contributions from the two working models and achieve optimal results for the proposed method. We conclude that the proposed multiple imputation method is a reasonable approach to dealing with missing categorical outcome data with more than two levels for assessing the distribution of the outcome. In terms of the choices for the working models, we suggest a multinomial logistic regression for predicting the missing outcome and a binary logistic regression for predicting the missingness probability.

The HCUP SID Imputation Project: Improving Statistical Inferences for Health Disparities Research by Imputing Missing Race Data.

PubMed

Ma, Yan; Zhang, Wei; Lyman, Stephen; Huang, Yihe

2018-06-01

To identify the most appropriate imputation method for missing data in the HCUP State Inpatient Databases (SID) and assess the impact of different missing data methods on racial disparities research. HCUP SID. A novel simulation study compared four imputation methods (random draw, hot deck, joint multiple imputation [MI], conditional MI) for missing values for multiple variables, including race, gender, admission source, median household income, and total charges. The simulation was built on real data from the SID to retain their hierarchical data structures and missing data patterns. Additional predictive information from the U.S. Census and American Hospital Association (AHA) database was incorporated into the imputation. Conditional MI prediction was equivalent or superior to the best performing alternatives for all missing data structures and substantially outperformed each of the alternatives in various scenarios. Conditional MI substantially improved statistical inferences for racial health disparities research with the SID. © Health Research and Educational Trust.

Filling the gap in functional trait databases: use of ecological hypotheses to replace missing data.

PubMed

Taugourdeau, Simon; Villerd, Jean; Plantureux, Sylvain; Huguenin-Elie, Olivier; Amiaud, Bernard

2014-04-01

Functional trait databases are powerful tools in ecology, though most of them contain large amounts of missing values. The goal of this study was to test the effect of imputation methods on the evaluation of trait values at species level and on the subsequent calculation of functional diversity indices at community level using functional trait databases. Two simple imputation methods (average and median), two methods based on ecological hypotheses, and one multiple imputation method were tested using a large plant trait database, together with the influence of the percentage of missing data and differences between functional traits. At community level, the complete-case approach and three functional diversity indices calculated from grassland plant communities were included. At the species level, one of the methods based on ecological hypothesis was for all traits more accurate than imputation with average or median values, but the multiple imputation method was superior for most of the traits. The method based on functional proximity between species was the best method for traits with an unbalanced distribution, while the method based on the existence of relationships between traits was the best for traits with a balanced distribution. The ranking of the grassland communities for their functional diversity indices was not robust with the complete-case approach, even for low percentages of missing data. With the imputation methods based on ecological hypotheses, functional diversity indices could be computed with a maximum of 30% of missing data, without affecting the ranking between grassland communities. The multiple imputation method performed well, but not better than single imputation based on ecological hypothesis and adapted to the distribution of the trait values for the functional identity and range of the communities. Ecological studies using functional trait databases have to deal with missing data using imputation methods corresponding to their specific needs and making the most out of the information available in the databases. Within this framework, this study indicates the possibilities and limits of single imputation methods based on ecological hypothesis and concludes that they could be useful when studying the ranking of communities for their functional diversity indices.

Filling the gap in functional trait databases: use of ecological hypotheses to replace missing data

PubMed Central

Taugourdeau, Simon; Villerd, Jean; Plantureux, Sylvain; Huguenin-Elie, Olivier; Amiaud, Bernard

2014-01-01

Functional trait databases are powerful tools in ecology, though most of them contain large amounts of missing values. The goal of this study was to test the effect of imputation methods on the evaluation of trait values at species level and on the subsequent calculation of functional diversity indices at community level using functional trait databases. Two simple imputation methods (average and median), two methods based on ecological hypotheses, and one multiple imputation method were tested using a large plant trait database, together with the influence of the percentage of missing data and differences between functional traits. At community level, the complete-case approach and three functional diversity indices calculated from grassland plant communities were included. At the species level, one of the methods based on ecological hypothesis was for all traits more accurate than imputation with average or median values, but the multiple imputation method was superior for most of the traits. The method based on functional proximity between species was the best method for traits with an unbalanced distribution, while the method based on the existence of relationships between traits was the best for traits with a balanced distribution. The ranking of the grassland communities for their functional diversity indices was not robust with the complete-case approach, even for low percentages of missing data. With the imputation methods based on ecological hypotheses, functional diversity indices could be computed with a maximum of 30% of missing data, without affecting the ranking between grassland communities. The multiple imputation method performed well, but not better than single imputation based on ecological hypothesis and adapted to the distribution of the trait values for the functional identity and range of the communities. Ecological studies using functional trait databases have to deal with missing data using imputation methods corresponding to their specific needs and making the most out of the information available in the databases. Within this framework, this study indicates the possibilities and limits of single imputation methods based on ecological hypothesis and concludes that they could be useful when studying the ranking of communities for their functional diversity indices. PMID:24772273

Comparison of missing value imputation methods in time series: the case of Turkish meteorological data

NASA Astrophysics Data System (ADS)

Yozgatligil, Ceylan; Aslan, Sipan; Iyigun, Cem; Batmaz, Inci

2013-04-01

This study aims to compare several imputation methods to complete the missing values of spatio-temporal meteorological time series. To this end, six imputation methods are assessed with respect to various criteria including accuracy, robustness, precision, and efficiency for artificially created missing data in monthly total precipitation and mean temperature series obtained from the Turkish State Meteorological Service. Of these methods, simple arithmetic average, normal ratio (NR), and NR weighted with correlations comprise the simple ones, whereas multilayer perceptron type neural network and multiple imputation strategy adopted by Monte Carlo Markov Chain based on expectation-maximization (EM-MCMC) are computationally intensive ones. In addition, we propose a modification on the EM-MCMC method. Besides using a conventional accuracy measure based on squared errors, we also suggest the correlation dimension (CD) technique of nonlinear dynamic time series analysis which takes spatio-temporal dependencies into account for evaluating imputation performances. Depending on the detailed graphical and quantitative analysis, it can be said that although computational methods, particularly EM-MCMC method, are computationally inefficient, they seem favorable for imputation of meteorological time series with respect to different missingness periods considering both measures and both series studied. To conclude, using the EM-MCMC algorithm for imputing missing values before conducting any statistical analyses of meteorological data will definitely decrease the amount of uncertainty and give more robust results. Moreover, the CD measure can be suggested for the performance evaluation of missing data imputation particularly with computational methods since it gives more precise results in meteorological time series.

Randomly and Non-Randomly Missing Renal Function Data in the Strong Heart Study: A Comparison of Imputation Methods

PubMed Central

Shara, Nawar; Yassin, Sayf A.; Valaitis, Eduardas; Wang, Hong; Howard, Barbara V.; Wang, Wenyu; Lee, Elisa T.; Umans, Jason G.

2015-01-01

Kidney and cardiovascular disease are widespread among populations with high prevalence of diabetes, such as American Indians participating in the Strong Heart Study (SHS). Studying these conditions simultaneously in longitudinal studies is challenging, because the morbidity and mortality associated with these diseases result in missing data, and these data are likely not missing at random. When such data are merely excluded, study findings may be compromised. In this article, a subset of 2264 participants with complete renal function data from Strong Heart Exams 1 (1989–1991), 2 (1993–1995), and 3 (1998–1999) was used to examine the performance of five methods used to impute missing data: listwise deletion, mean of serial measures, adjacent value, multiple imputation, and pattern-mixture. Three missing at random models and one non-missing at random model were used to compare the performance of the imputation techniques on randomly and non-randomly missing data. The pattern-mixture method was found to perform best for imputing renal function data that were not missing at random. Determining whether data are missing at random or not can help in choosing the imputation method that will provide the most accurate results. PMID:26414328

Randomly and Non-Randomly Missing Renal Function Data in the Strong Heart Study: A Comparison of Imputation Methods.

PubMed

Shara, Nawar; Yassin, Sayf A; Valaitis, Eduardas; Wang, Hong; Howard, Barbara V; Wang, Wenyu; Lee, Elisa T; Umans, Jason G

2015-01-01

Kidney and cardiovascular disease are widespread among populations with high prevalence of diabetes, such as American Indians participating in the Strong Heart Study (SHS). Studying these conditions simultaneously in longitudinal studies is challenging, because the morbidity and mortality associated with these diseases result in missing data, and these data are likely not missing at random. When such data are merely excluded, study findings may be compromised. In this article, a subset of 2264 participants with complete renal function data from Strong Heart Exams 1 (1989-1991), 2 (1993-1995), and 3 (1998-1999) was used to examine the performance of five methods used to impute missing data: listwise deletion, mean of serial measures, adjacent value, multiple imputation, and pattern-mixture. Three missing at random models and one non-missing at random model were used to compare the performance of the imputation techniques on randomly and non-randomly missing data. The pattern-mixture method was found to perform best for imputing renal function data that were not missing at random. Determining whether data are missing at random or not can help in choosing the imputation method that will provide the most accurate results.

Improving record linkage performance in the presence of missing linkage data.

PubMed

Ong, Toan C; Mannino, Michael V; Schilling, Lisa M; Kahn, Michael G

2014-12-01

Existing record linkage methods do not handle missing linking field values in an efficient and effective manner. The objective of this study is to investigate three novel methods for improving the accuracy and efficiency of record linkage when record linkage fields have missing values. By extending the Fellegi-Sunter scoring implementations available in the open-source Fine-grained Record Linkage (FRIL) software system we developed three novel methods to solve the missing data problem in record linkage, which we refer to as: Weight Redistribution, Distance Imputation, and Linkage Expansion. Weight Redistribution removes fields with missing data from the set of quasi-identifiers and redistributes the weight from the missing attribute based on relative proportions across the remaining available linkage fields. Distance Imputation imputes the distance between the missing data fields rather than imputing the missing data value. Linkage Expansion adds previously considered non-linkage fields to the linkage field set to compensate for the missing information in a linkage field. We tested the linkage methods using simulated data sets with varying field value corruption rates. The methods developed had sensitivity ranging from .895 to .992 and positive predictive values (PPV) ranging from .865 to 1 in data sets with low corruption rates. Increased corruption rates lead to decreased sensitivity for all methods. These new record linkage algorithms show promise in terms of accuracy and efficiency and may be valuable for combining large data sets at the patient level to support biomedical and clinical research. Copyright © 2014 Elsevier Inc. All rights reserved.

Multiple imputation to deal with missing EQ-5D-3L data: Should we impute individual domains or the actual index?

PubMed

Simons, Claire L; Rivero-Arias, Oliver; Yu, Ly-Mee; Simon, Judit

2015-04-01

Missing data are a well-known and widely documented problem in cost-effectiveness analyses alongside clinical trials using individual patient-level data. Current methodological research recommends multiple imputation (MI) to deal with missing health outcome data, but there is little guidance on whether MI for multi-attribute questionnaires, such as the EQ-5D-3L, should be carried out at domain or at summary score level. In this paper, we evaluated the impact of imputing individual domains versus imputing index values to deal with missing EQ-5D-3L data using a simulation study and developed recommendations for future practice. We simulated missing data in a patient-level dataset with complete EQ-5D-3L data at one point in time from a large multinational clinical trial (n = 1,814). Different proportions of missing data were generated using a missing at random (MAR) mechanism and three different scenarios were studied. The performance of using each method was evaluated using root mean squared error and mean absolute error of the actual versus predicted EQ-5D-3L indices. In large sample sizes (n > 500) and a missing data pattern that follows mainly unit non-response, imputing domains or the index produced similar results. However, domain imputation became more accurate than index imputation with pattern of missingness following an item non-response. For smaller sample sizes (n < 100), index imputation was more accurate. When MI models were misspecified, both domain and index imputations were inaccurate for any proportion of missing data. The decision between imputing the domains or the EQ-5D-3L index scores depends on the observed missing data pattern and the sample size available for analysis. Analysts conducting this type of exercises should also evaluate the sensitivity of the analysis to the MAR assumption and whether the imputation model is correctly specified.

Cox regression analysis with missing covariates via nonparametric multiple imputation.

PubMed

Hsu, Chiu-Hsieh; Yu, Mandi

2018-01-01

We consider the situation of estimating Cox regression in which some covariates are subject to missing, and there exists additional information (including observed event time, censoring indicator and fully observed covariates) which may be predictive of the missing covariates. We propose to use two working regression models: one for predicting the missing covariates and the other for predicting the missing probabilities. For each missing covariate observation, these two working models are used to define a nearest neighbor imputing set. This set is then used to non-parametrically impute covariate values for the missing observation. Upon the completion of imputation, Cox regression is performed on the multiply imputed datasets to estimate the regression coefficients. In a simulation study, we compare the nonparametric multiple imputation approach with the augmented inverse probability weighted (AIPW) method, which directly incorporates the two working models into estimation of Cox regression, and the predictive mean matching imputation (PMM) method. We show that all approaches can reduce bias due to non-ignorable missing mechanism. The proposed nonparametric imputation method is robust to mis-specification of either one of the two working models and robust to mis-specification of the link function of the two working models. In contrast, the PMM method is sensitive to misspecification of the covariates included in imputation. The AIPW method is sensitive to the selection probability. We apply the approaches to a breast cancer dataset from Surveillance, Epidemiology and End Results (SEER) Program.

Multiple imputation by chained equations for systematically and sporadically missing multilevel data.

PubMed

Resche-Rigon, Matthieu; White, Ian R

2018-06-01

In multilevel settings such as individual participant data meta-analysis, a variable is 'systematically missing' if it is wholly missing in some clusters and 'sporadically missing' if it is partly missing in some clusters. Previously proposed methods to impute incomplete multilevel data handle either systematically or sporadically missing data, but frequently both patterns are observed. We describe a new multiple imputation by chained equations (MICE) algorithm for multilevel data with arbitrary patterns of systematically and sporadically missing variables. The algorithm is described for multilevel normal data but can easily be extended for other variable types. We first propose two methods for imputing a single incomplete variable: an extension of an existing method and a new two-stage method which conveniently allows for heteroscedastic data. We then discuss the difficulties of imputing missing values in several variables in multilevel data using MICE, and show that even the simplest joint multilevel model implies conditional models which involve cluster means and heteroscedasticity. However, a simulation study finds that the proposed methods can be successfully combined in a multilevel MICE procedure, even when cluster means are not included in the imputation models.

Influence of Pattern of Missing Data on Performance of Imputation Methods: An Example Using National Data on Drug Injection in Prisons

PubMed Central

Haji-Maghsoudi, Saiedeh; Haghdoost, Ali-akbar; Rastegari, Azam; Baneshi, Mohammad Reza

2013-01-01

Background: Policy makers need models to be able to detect groups at high risk of HIV infection. Incomplete records and dirty data are frequently seen in national data sets. Presence of missing data challenges the practice of model development. Several studies suggested that performance of imputation methods is acceptable when missing rate is moderate. One of the issues which was of less concern, to be addressed here, is the role of the pattern of missing data. Methods: We used information of 2720 prisoners. Results derived from fitting regression model to whole data were served as gold standard. Missing data were then generated so that 10%, 20% and 50% of data were lost. In scenario 1, we generated missing values, at above rates, in one variable which was significant in gold model (age). In scenario 2, a small proportion of each of independent variable was dropped out. Four imputation methods, under different Event Per Variable (EPV) values, were compared in terms of selection of important variables and parameter estimation. Results: In scenario 2, bias in estimates was low and performances of all methods for handing missing data were similar. All methods at all missing rates were able to detect significance of age. In scenario 1, biases in estimations were increased, in particular at 50% missing rate. Here at EPVs of 10 and 5, imputation methods failed to capture effect of age. Conclusion: In scenario 2, all imputation methods at all missing rates, were able to detect age as being significant. This was not the case in scenario 1. Our results showed that performance of imputation methods depends on the pattern of missing data. PMID:24596839

A comprehensive evaluation of popular proteomics software workflows for label-free proteome quantification and imputation.

PubMed

Välikangas, Tommi; Suomi, Tomi; Elo, Laura L

2017-05-31

Label-free mass spectrometry (MS) has developed into an important tool applied in various fields of biological and life sciences. Several software exist to process the raw MS data into quantified protein abundances, including open source and commercial solutions. Each software includes a set of unique algorithms for different tasks of the MS data processing workflow. While many of these algorithms have been compared separately, a thorough and systematic evaluation of their overall performance is missing. Moreover, systematic information is lacking about the amount of missing values produced by the different proteomics software and the capabilities of different data imputation methods to account for them.In this study, we evaluated the performance of five popular quantitative label-free proteomics software workflows using four different spike-in data sets. Our extensive testing included the number of proteins quantified and the number of missing values produced by each workflow, the accuracy of detecting differential expression and logarithmic fold change and the effect of different imputation and filtering methods on the differential expression results. We found that the Progenesis software performed consistently well in the differential expression analysis and produced few missing values. The missing values produced by the other software decreased their performance, but this difference could be mitigated using proper data filtering or imputation methods. Among the imputation methods, we found that the local least squares (lls) regression imputation consistently increased the performance of the software in the differential expression analysis, and a combination of both data filtering and local least squares imputation increased performance the most in the tested data sets. © The Author 2017. Published by Oxford University Press.

Imputation for multisource data with comparison and assessment techniques

DOE PAGES

Casleton, Emily Michele; Osthus, David Allen; Van Buren, Kendra Lu

2017-12-27

Missing data are prevalent issue in analyses involving data collection. The problem of missing data is exacerbated for multisource analysis, where data from multiple sensors are combined to arrive at a single conclusion. In this scenario, it is more likely to occur and can lead to discarding a large amount of data collected; however, the information from observed sensors can be leveraged to estimate those values not observed. We propose two methods for imputation of multisource data, both of which take advantage of potential correlation between data from different sensors, through ridge regression and a state-space model. These methods, asmore » well as the common median imputation, are applied to data collected from a variety of sensors monitoring an experimental facility. Performance of imputation methods is compared with the mean absolute deviation; however, rather than using this metric to solely rank themethods,we also propose an approach to identify significant differences. Imputation techniqueswill also be assessed by their ability to produce appropriate confidence intervals, through coverage and length, around the imputed values. Finally, performance of imputed datasets is compared with a marginalized dataset through a weighted k-means clustering. In general, we found that imputation through a dynamic linearmodel tended to be the most accurate and to produce the most precise confidence intervals, and that imputing the missing values and down weighting them with respect to observed values in the analysis led to the most accurate performance.« less

Imputation for multisource data with comparison and assessment techniques

DOE Office of Scientific and Technical Information (OSTI.GOV)

Casleton, Emily Michele; Osthus, David Allen; Van Buren, Kendra Lu

Missing data are prevalent issue in analyses involving data collection. The problem of missing data is exacerbated for multisource analysis, where data from multiple sensors are combined to arrive at a single conclusion. In this scenario, it is more likely to occur and can lead to discarding a large amount of data collected; however, the information from observed sensors can be leveraged to estimate those values not observed. We propose two methods for imputation of multisource data, both of which take advantage of potential correlation between data from different sensors, through ridge regression and a state-space model. These methods, asmore » well as the common median imputation, are applied to data collected from a variety of sensors monitoring an experimental facility. Performance of imputation methods is compared with the mean absolute deviation; however, rather than using this metric to solely rank themethods,we also propose an approach to identify significant differences. Imputation techniqueswill also be assessed by their ability to produce appropriate confidence intervals, through coverage and length, around the imputed values. Finally, performance of imputed datasets is compared with a marginalized dataset through a weighted k-means clustering. In general, we found that imputation through a dynamic linearmodel tended to be the most accurate and to produce the most precise confidence intervals, and that imputing the missing values and down weighting them with respect to observed values in the analysis led to the most accurate performance.« less

A Method for Imputing Response Options for Missing Data on Multiple-Choice Assessments

ERIC Educational Resources Information Center

Wolkowitz, Amanda A.; Skorupski, William P.

2013-01-01

When missing values are present in item response data, there are a number of ways one might impute a correct or incorrect response to a multiple-choice item. There are significantly fewer methods for imputing the actual response option an examinee may have provided if he or she had not omitted the item either purposely or accidentally. This…

Saturated linkage map construction in Rubus idaeus using genotyping by sequencing and genome-independent imputation

PubMed Central

2013-01-01

Background Rapid development of highly saturated genetic maps aids molecular breeding, which can accelerate gain per breeding cycle in woody perennial plants such as Rubus idaeus (red raspberry). Recently, robust genotyping methods based on high-throughput sequencing were developed, which provide high marker density, but result in some genotype errors and a large number of missing genotype values. Imputation can reduce the number of missing values and can correct genotyping errors, but current methods of imputation require a reference genome and thus are not an option for most species. Results Genotyping by Sequencing (GBS) was used to produce highly saturated maps for a R. idaeus pseudo-testcross progeny. While low coverage and high variance in sequencing resulted in a large number of missing values for some individuals, a novel method of imputation based on maximum likelihood marker ordering from initial marker segregation overcame the challenge of missing values, and made map construction computationally tractable. The two resulting parental maps contained 4521 and 2391 molecular markers spanning 462.7 and 376.6 cM respectively over seven linkage groups. Detection of precise genomic regions with segregation distortion was possible because of map saturation. Microsatellites (SSRs) linked these results to published maps for cross-validation and map comparison. Conclusions GBS together with genome-independent imputation provides a rapid method for genetic map construction in any pseudo-testcross progeny. Our method of imputation estimates the correct genotype call of missing values and corrects genotyping errors that lead to inflated map size and reduced precision in marker placement. Comparison of SSRs to published R. idaeus maps showed that the linkage maps constructed with GBS and our method of imputation were robust, and marker positioning reliable. The high marker density allowed identification of genomic regions with segregation distortion in R. idaeus, which may help to identify deleterious alleles that are the basis of inbreeding depression in the species. PMID:23324311

Multiple imputation for assessment of exposures to drinking water contaminants: evaluation with the Atrazine Monitoring Program.

PubMed

Jones, Rachael M; Stayner, Leslie T; Demirtas, Hakan

2014-10-01

Drinking water may contain pollutants that harm human health. The frequency of pollutant monitoring may occur quarterly, annually, or less frequently, depending upon the pollutant, the pollutant concentration, and community water system. However, birth and other health outcomes are associated with narrow time-windows of exposure. Infrequent monitoring impedes linkage between water quality and health outcomes for epidemiological analyses. To evaluate the performance of multiple imputation to fill in water quality values between measurements in community water systems (CWSs). The multiple imputation method was implemented in a simulated setting using data from the Atrazine Monitoring Program (AMP, 2006-2009 in five Midwestern states). Values were deleted from the AMP data to leave one measurement per month. Four patterns reflecting drinking water monitoring regulations were used to delete months of data in each CWS: three patterns were missing at random and one pattern was missing not at random. Synthetic health outcome data were created using a linear and a Poisson exposure-response relationship with five levels of hypothesized association, respectively. The multiple imputation method was evaluated by comparing the exposure-response relationships estimated based on multiply imputed data with the hypothesized association. The four patterns deleted 65-92% months of atrazine observations in AMP data. Even with these high rates of missing information, our procedure was able to recover most of the missing information when the synthetic health outcome was included for missing at random patterns and for missing not at random patterns with low-to-moderate exposure-response relationships. Multiple imputation appears to be an effective method for filling in water quality values between measurements. Copyright © 2014 Elsevier Inc. All rights reserved.

Missing data treatments matter: an analysis of multiple imputation for anterior cervical discectomy and fusion procedures.

PubMed

Ondeck, Nathaniel T; Fu, Michael C; Skrip, Laura A; McLynn, Ryan P; Cui, Jonathan J; Basques, Bryce A; Albert, Todd J; Grauer, Jonathan N

2018-04-09

The presence of missing data is a limitation of large datasets, including the National Surgical Quality Improvement Program (NSQIP). In addressing this issue, most studies use complete case analysis, which excludes cases with missing data, thus potentially introducing selection bias. Multiple imputation, a statistically rigorous approach that approximates missing data and preserves sample size, may be an improvement over complete case analysis. The present study aims to evaluate the impact of using multiple imputation in comparison with complete case analysis for assessing the associations between preoperative laboratory values and adverse outcomes following anterior cervical discectomy and fusion (ACDF) procedures. This is a retrospective review of prospectively collected data. Patients undergoing one-level ACDF were identified in NSQIP 2012-2015. Perioperative adverse outcome variables assessed included the occurrence of any adverse event, severe adverse events, and hospital readmission. Missing preoperative albumin and hematocrit values were handled using complete case analysis and multiple imputation. These preoperative laboratory levels were then tested for associations with 30-day postoperative outcomes using logistic regression. A total of 11,999 patients were included. Of this cohort, 63.5% of patients had missing preoperative albumin and 9.9% had missing preoperative hematocrit. When using complete case analysis, only 4,311 patients were studied. The removed patients were significantly younger, healthier, of a common body mass index, and male. Logistic regression analysis failed to identify either preoperative hypoalbuminemia or preoperative anemia as significantly associated with adverse outcomes. When employing multiple imputation, all 11,999 patients were included. Preoperative hypoalbuminemia was significantly associated with the occurrence of any adverse event and severe adverse events. Preoperative anemia was significantly associated with the occurrence of any adverse event, severe adverse events, and hospital readmission. Multiple imputation is a rigorous statistical procedure that is being increasingly used to address missing values in large datasets. Using this technique for ACDF avoided the loss of cases that may have affected the representativeness and power of the study and led to different results than complete case analysis. Multiple imputation should be considered for future spine studies. Copyright © 2018 Elsevier Inc. All rights reserved.

Transitioning to multiple imputation : a new method to impute missing blood alcohol concentration (BAC) values in FARS

DOT National Transportation Integrated Search

2002-01-01

The National Center for Statistics and Analysis (NCSA) of the National Highway Traffic Safety : Administration (NHTSA) has undertaken several approaches to remedy the problem of missing blood alcohol : test results in the Fatality Analysis Reporting ...

Imputation of missing data in time series for air pollutants

NASA Astrophysics Data System (ADS)

Junger, W. L.; Ponce de Leon, A.

2015-02-01

Missing data are major concerns in epidemiological studies of the health effects of environmental air pollutants. This article presents an imputation-based method that is suitable for multivariate time series data, which uses the EM algorithm under the assumption of normal distribution. Different approaches are considered for filtering the temporal component. A simulation study was performed to assess validity and performance of proposed method in comparison with some frequently used methods. Simulations showed that when the amount of missing data was as low as 5%, the complete data analysis yielded satisfactory results regardless of the generating mechanism of the missing data, whereas the validity began to degenerate when the proportion of missing values exceeded 10%. The proposed imputation method exhibited good accuracy and precision in different settings with respect to the patterns of missing observations. Most of the imputations obtained valid results, even under missing not at random. The methods proposed in this study are implemented as a package called mtsdi for the statistical software system R.

A comparison of model-based imputation methods for handling missing predictor values in a linear regression model: A simulation study

NASA Astrophysics Data System (ADS)

Hasan, Haliza; Ahmad, Sanizah; Osman, Balkish Mohd; Sapri, Shamsiah; Othman, Nadirah

2017-08-01

In regression analysis, missing covariate data has been a common problem. Many researchers use ad hoc methods to overcome this problem due to the ease of implementation. However, these methods require assumptions about the data that rarely hold in practice. Model-based methods such as Maximum Likelihood (ML) using the expectation maximization (EM) algorithm and Multiple Imputation (MI) are more promising when dealing with difficulties caused by missing data. Then again, inappropriate methods of missing value imputation can lead to serious bias that severely affects the parameter estimates. The main objective of this study is to provide a better understanding regarding missing data concept that can assist the researcher to select the appropriate missing data imputation methods. A simulation study was performed to assess the effects of different missing data techniques on the performance of a regression model. The covariate data were generated using an underlying multivariate normal distribution and the dependent variable was generated as a combination of explanatory variables. Missing values in covariate were simulated using a mechanism called missing at random (MAR). Four levels of missingness (10%, 20%, 30% and 40%) were imposed. ML and MI techniques available within SAS software were investigated. A linear regression analysis was fitted and the model performance measures; MSE, and R-Squared were obtained. Results of the analysis showed that MI is superior in handling missing data with highest R-Squared and lowest MSE when percent of missingness is less than 30%. Both methods are unable to handle larger than 30% level of missingness.

Microarray missing data imputation based on a set theoretic framework and biological knowledge.

PubMed

Gan, Xiangchao; Liew, Alan Wee-Chung; Yan, Hong

2006-01-01

Gene expressions measured using microarrays usually suffer from the missing value problem. However, in many data analysis methods, a complete data matrix is required. Although existing missing value imputation algorithms have shown good performance to deal with missing values, they also have their limitations. For example, some algorithms have good performance only when strong local correlation exists in data while some provide the best estimate when data is dominated by global structure. In addition, these algorithms do not take into account any biological constraint in their imputation. In this paper, we propose a set theoretic framework based on projection onto convex sets (POCS) for missing data imputation. POCS allows us to incorporate different types of a priori knowledge about missing values into the estimation process. The main idea of POCS is to formulate every piece of prior knowledge into a corresponding convex set and then use a convergence-guaranteed iterative procedure to obtain a solution in the intersection of all these sets. In this work, we design several convex sets, taking into consideration the biological characteristic of the data: the first set mainly exploit the local correlation structure among genes in microarray data, while the second set captures the global correlation structure among arrays. The third set (actually a series of sets) exploits the biological phenomenon of synchronization loss in microarray experiments. In cyclic systems, synchronization loss is a common phenomenon and we construct a series of sets based on this phenomenon for our POCS imputation algorithm. Experiments show that our algorithm can achieve a significant reduction of error compared to the KNNimpute, SVDimpute and LSimpute methods.

Characterizing and Managing Missing Structured Data in Electronic Health Records: Data Analysis.

PubMed

Beaulieu-Jones, Brett K; Lavage, Daniel R; Snyder, John W; Moore, Jason H; Pendergrass, Sarah A; Bauer, Christopher R

2018-02-23

Missing data is a challenge for all studies; however, this is especially true for electronic health record (EHR)-based analyses. Failure to appropriately consider missing data can lead to biased results. While there has been extensive theoretical work on imputation, and many sophisticated methods are now available, it remains quite challenging for researchers to implement these methods appropriately. Here, we provide detailed procedures for when and how to conduct imputation of EHR laboratory results. The objective of this study was to demonstrate how the mechanism of missingness can be assessed, evaluate the performance of a variety of imputation methods, and describe some of the most frequent problems that can be encountered. We analyzed clinical laboratory measures from 602,366 patients in the EHR of Geisinger Health System in Pennsylvania, USA. Using these data, we constructed a representative set of complete cases and assessed the performance of 12 different imputation methods for missing data that was simulated based on 4 mechanisms of missingness (missing completely at random, missing not at random, missing at random, and real data modelling). Our results showed that several methods, including variations of Multivariate Imputation by Chained Equations (MICE) and softImpute, consistently imputed missing values with low error; however, only a subset of the MICE methods was suitable for multiple imputation. The analyses we describe provide an outline of considerations for dealing with missing EHR data, steps that researchers can perform to characterize missingness within their own data, and an evaluation of methods that can be applied to impute clinical data. While the performance of methods may vary between datasets, the process we describe can be generalized to the majority of structured data types that exist in EHRs, and all of our methods and code are publicly available. ©Brett K Beaulieu-Jones, Daniel R Lavage, John W Snyder, Jason H Moore, Sarah A Pendergrass, Christopher R Bauer. Originally published in JMIR Medical Informatics (http://medinform.jmir.org), 23.02.2018.

Missing data and multiple imputation in clinical epidemiological research.

PubMed

Pedersen, Alma B; Mikkelsen, Ellen M; Cronin-Fenton, Deirdre; Kristensen, Nickolaj R; Pham, Tra My; Pedersen, Lars; Petersen, Irene

2017-01-01

Missing data are ubiquitous in clinical epidemiological research. Individuals with missing data may differ from those with no missing data in terms of the outcome of interest and prognosis in general. Missing data are often categorized into the following three types: missing completely at random (MCAR), missing at random (MAR), and missing not at random (MNAR). In clinical epidemiological research, missing data are seldom MCAR. Missing data can constitute considerable challenges in the analyses and interpretation of results and can potentially weaken the validity of results and conclusions. A number of methods have been developed for dealing with missing data. These include complete-case analyses, missing indicator method, single value imputation, and sensitivity analyses incorporating worst-case and best-case scenarios. If applied under the MCAR assumption, some of these methods can provide unbiased but often less precise estimates. Multiple imputation is an alternative method to deal with missing data, which accounts for the uncertainty associated with missing data. Multiple imputation is implemented in most statistical software under the MAR assumption and provides unbiased and valid estimates of associations based on information from the available data. The method affects not only the coefficient estimates for variables with missing data but also the estimates for other variables with no missing data.

Missing data and multiple imputation in clinical epidemiological research

PubMed Central

Pedersen, Alma B; Mikkelsen, Ellen M; Cronin-Fenton, Deirdre; Kristensen, Nickolaj R; Pham, Tra My; Pedersen, Lars; Petersen, Irene

2017-01-01

Missing data are ubiquitous in clinical epidemiological research. Individuals with missing data may differ from those with no missing data in terms of the outcome of interest and prognosis in general. Missing data are often categorized into the following three types: missing completely at random (MCAR), missing at random (MAR), and missing not at random (MNAR). In clinical epidemiological research, missing data are seldom MCAR. Missing data can constitute considerable challenges in the analyses and interpretation of results and can potentially weaken the validity of results and conclusions. A number of methods have been developed for dealing with missing data. These include complete-case analyses, missing indicator method, single value imputation, and sensitivity analyses incorporating worst-case and best-case scenarios. If applied under the MCAR assumption, some of these methods can provide unbiased but often less precise estimates. Multiple imputation is an alternative method to deal with missing data, which accounts for the uncertainty associated with missing data. Multiple imputation is implemented in most statistical software under the MAR assumption and provides unbiased and valid estimates of associations based on information from the available data. The method affects not only the coefficient estimates for variables with missing data but also the estimates for other variables with no missing data. PMID:28352203

Dealing with gene expression missing data.

PubMed

Brás, L P; Menezes, J C

2006-05-01

Compared evaluation of different methods is presented for estimating missing values in microarray data: weighted K-nearest neighbours imputation (KNNimpute), regression-based methods such as local least squares imputation (LLSimpute) and partial least squares imputation (PLSimpute) and Bayesian principal component analysis (BPCA). The influence in prediction accuracy of some factors, such as methods' parameters, type of data relationships used in the estimation process (i.e. row-wise, column-wise or both), missing rate and pattern and type of experiment [time series (TS), non-time series (NTS) or mixed (MIX) experiments] is elucidated. Improvements based on the iterative use of data (iterative LLS and PLS imputation--ILLSimpute and IPLSimpute), the need to perform initial imputations (modified PLS and Helland PLS imputation--MPLSimpute and HPLSimpute) and the type of relationships employed (KNNarray, LLSarray, HPLSarray and alternating PLS--APLSimpute) are proposed. Overall, it is shown that data set properties (type of experiment, missing rate and pattern) affect the data similarity structure, therefore influencing the methods' performance. LLSimpute and ILLSimpute are preferable in the presence of data with a stronger similarity structure (TS and MIX experiments), whereas PLS-based methods (MPLSimpute, IPLSimpute and APLSimpute) are preferable when estimating NTS missing data.

A meta-data based method for DNA microarray imputation.

PubMed

Jörnsten, Rebecka; Ouyang, Ming; Wang, Hui-Yu

2007-03-29

DNA microarray experiments are conducted in logical sets, such as time course profiling after a treatment is applied to the samples, or comparisons of the samples under two or more conditions. Due to cost and design constraints of spotted cDNA microarray experiments, each logical set commonly includes only a small number of replicates per condition. Despite the vast improvement of the microarray technology in recent years, missing values are prevalent. Intuitively, imputation of missing values is best done using many replicates within the same logical set. In practice, there are few replicates and thus reliable imputation within logical sets is difficult. However, it is in the case of few replicates that the presence of missing values, and how they are imputed, can have the most profound impact on the outcome of downstream analyses (e.g. significance analysis and clustering). This study explores the feasibility of imputation across logical sets, using the vast amount of publicly available microarray data to improve imputation reliability in the small sample size setting. We download all cDNA microarray data of Saccharomyces cerevisiae, Arabidopsis thaliana, and Caenorhabditis elegans from the Stanford Microarray Database. Through cross-validation and simulation, we find that, for all three species, our proposed imputation using data from public databases is far superior to imputation within a logical set, sometimes to an astonishing degree. Furthermore, the imputation root mean square error for significant genes is generally a lot less than that of non-significant ones. Since downstream analysis of significant genes, such as clustering and network analysis, can be very sensitive to small perturbations of estimated gene effects, it is highly recommended that researchers apply reliable data imputation prior to further analysis. Our method can also be applied to cDNA microarray experiments from other species, provided good reference data are available.

Effects of imputation on correlation: implications for analysis of mass spectrometry data from multiple biological matrices.

PubMed

Taylor, Sandra L; Ruhaak, L Renee; Kelly, Karen; Weiss, Robert H; Kim, Kyoungmi

2017-03-01

With expanded access to, and decreased costs of, mass spectrometry, investigators are collecting and analyzing multiple biological matrices from the same subject such as serum, plasma, tissue and urine to enhance biomarker discoveries, understanding of disease processes and identification of therapeutic targets. Commonly, each biological matrix is analyzed separately, but multivariate methods such as MANOVAs that combine information from multiple biological matrices are potentially more powerful. However, mass spectrometric data typically contain large amounts of missing values, and imputation is often used to create complete data sets for analysis. The effects of imputation on multiple biological matrix analyses have not been studied. We investigated the effects of seven imputation methods (half minimum substitution, mean substitution, k-nearest neighbors, local least squares regression, Bayesian principal components analysis, singular value decomposition and random forest), on the within-subject correlation of compounds between biological matrices and its consequences on MANOVA results. Through analysis of three real omics data sets and simulation studies, we found the amount of missing data and imputation method to substantially change the between-matrix correlation structure. The magnitude of the correlations was generally reduced in imputed data sets, and this effect increased with the amount of missing data. Significant results from MANOVA testing also were substantially affected. In particular, the number of false positives increased with the level of missing data for all imputation methods. No one imputation method was universally the best, but the simple substitution methods (Half Minimum and Mean) consistently performed poorly. © The Author 2016. Published by Oxford University Press. For Permissions, please email: journals.permissions@oup.com.

Time Series Imputation via L1 Norm-Based Singular Spectrum Analysis

NASA Astrophysics Data System (ADS)

Kalantari, Mahdi; Yarmohammadi, Masoud; Hassani, Hossein; Silva, Emmanuel Sirimal

Missing values in time series data is a well-known and important problem which many researchers have studied extensively in various fields. In this paper, a new nonparametric approach for missing value imputation in time series is proposed. The main novelty of this research is applying the L1 norm-based version of Singular Spectrum Analysis (SSA), namely L1-SSA which is robust against outliers. The performance of the new imputation method has been compared with many other established methods. The comparison is done by applying them to various real and simulated time series. The obtained results confirm that the SSA-based methods, especially L1-SSA can provide better imputation in comparison to other methods.

Use of Multiple Imputation Method to Improve Estimation of Missing Baseline Serum Creatinine in Acute Kidney Injury Research

PubMed Central

Peterson, Josh F.; Eden, Svetlana K.; Moons, Karel G.; Ikizler, T. Alp; Matheny, Michael E.

2013-01-01

Summary Background and objectives Baseline creatinine (BCr) is frequently missing in AKI studies. Common surrogate estimates can misclassify AKI and adversely affect the study of related outcomes. This study examined whether multiple imputation improved accuracy of estimating missing BCr beyond current recommendations to apply assumed estimated GFR (eGFR) of 75 ml/min per 1.73 m2 (eGFR 75). Design, setting, participants, & measurements From 41,114 unique adult admissions (13,003 with and 28,111 without BCr data) at Vanderbilt University Hospital between 2006 and 2008, a propensity score model was developed to predict likelihood of missing BCr. Propensity scoring identified 6502 patients with highest likelihood of missing BCr among 13,003 patients with known BCr to simulate a “missing” data scenario while preserving actual reference BCr. Within this cohort (n=6502), the ability of various multiple-imputation approaches to estimate BCr and classify AKI were compared with that of eGFR 75. Results All multiple-imputation methods except the basic one more closely approximated actual BCr than did eGFR 75. Total AKI misclassification was lower with multiple imputation (full multiple imputation + serum creatinine) (9.0%) than with eGFR 75 (12.3%; P<0.001). Improvements in misclassification were greater in patients with impaired kidney function (full multiple imputation + serum creatinine) (15.3%) versus eGFR 75 (40.5%; P<0.001). Multiple imputation improved specificity and positive predictive value for detecting AKI at the expense of modestly decreasing sensitivity relative to eGFR 75. Conclusions Multiple imputation can improve accuracy in estimating missing BCr and reduce misclassification of AKI beyond currently proposed methods. PMID:23037980

[Comparison of different methods in dealing with HIV viral load data with diversified missing value mechanism on HIV positive MSM].

PubMed

Jiang, Z; Dou, Z; Song, W L; Xu, J; Wu, Z Y

2017-11-10

Objective: To compare results of different methods: in organizing HIV viral load (VL) data with missing values mechanism. Methods We used software SPSS 17.0 to simulate complete and missing data with different missing value mechanism from HIV viral loading data collected from MSM in 16 cities in China in 2013. Maximum Likelihood Methods Using the Expectation and Maximization Algorithm (EM), regressive method, mean imputation, delete method, and Markov Chain Monte Carlo (MCMC) were used to supplement missing data respectively. The results: of different methods were compared according to distribution characteristics, accuracy and precision. Results HIV VL data could not be transferred into a normal distribution. All the methods showed good results in iterating data which is Missing Completely at Random Mechanism (MCAR). For the other types of missing data, regressive and MCMC methods were used to keep the main characteristic of the original data. The means of iterating database with different methods were all close to the original one. The EM, regressive method, mean imputation, and delete method under-estimate VL while MCMC overestimates it. Conclusion: MCMC can be used as the main imputation method for HIV virus loading missing data. The iterated data can be used as a reference for mean HIV VL estimation among the investigated population.

Approach to addressing missing data for electronic medical records and pharmacy claims data research.

PubMed

Bounthavong, Mark; Watanabe, Jonathan H; Sullivan, Kevin M

2015-04-01

The complete capture of all values for each variable of interest in pharmacy research studies remains aspirational. The absence of these possibly influential values is a common problem for pharmacist investigators. Failure to account for missing data may translate to biased study findings and conclusions. Our goal in this analysis was to apply validated statistical methods for missing data to a previously analyzed data set and compare results when missing data methods were implemented versus standard analytics that ignore missing data effects. Using data from a retrospective cohort study, the statistical method of multiple imputation was used to provide regression-based estimates of the missing values to improve available data usable for study outcomes measurement. These findings were then contrasted with a complete-case analysis that restricted estimation to subjects in the cohort that had no missing values. Odds ratios were compared to assess differences in findings of the analyses. A nonadjusted regression analysis ("crude analysis") was also performed as a reference for potential bias. Veterans Integrated Systems Network that includes VA facilities in the Southern California and Nevada regions. New statin users between November 30, 2006, and December 2, 2007, with a diagnosis of dyslipidemia. We compared the odds ratios (ORs) and 95% confidence intervals (CIs) for the crude, complete-case, and multiple imputation analyses for the end points of a 25% or greater reduction in atherogenic lipids. Data were missing for 21.5% of identified patients (1665 subjects of 7739). Regression model results were similar for the crude, complete-case, and multiple imputation analyses with overlap of 95% confidence limits at each end point. The crude, complete-case, and multiple imputation ORs (95% CIs) for a 25% or greater reduction in low-density lipoprotein cholesterol were 3.5 (95% CI 3.1-3.9), 4.3 (95% CI 3.8-4.9), and 4.1 (95% CI 3.7-4.6), respectively. The crude, complete-case, and multiple imputation ORs (95% CIs) for a 25% or greater reduction in non-high-density lipoprotein cholesterol were 3.5 (95% CI 3.1-3.9), 4.5 (95% CI 4.0-5.2), and 4.4 (95% CI 3.9-4.9), respectively. The crude, complete-case, and multiple imputation ORs (95% CIs) for 25% or greater reduction in TGs were 3.1 (95% CI 2.8-3.6), 4.0 (95% CI 3.5-4.6), and 4.1 (95% CI 3.6-4.6), respectively. The use of the multiple imputation method to account for missing data did not alter conclusions based on a complete-case analysis. Given the frequency of missing data in research using electronic health records and pharmacy claims data, multiple imputation may play an important role in the validation of study findings. © 2015 Pharmacotherapy Publications, Inc.

Variable selection under multiple imputation using the bootstrap in a prognostic study

PubMed Central

Heymans, Martijn W; van Buuren, Stef; Knol, Dirk L; van Mechelen, Willem; de Vet, Henrica CW

2007-01-01

Background Missing data is a challenging problem in many prognostic studies. Multiple imputation (MI) accounts for imputation uncertainty that allows for adequate statistical testing. We developed and tested a methodology combining MI with bootstrapping techniques for studying prognostic variable selection. Method In our prospective cohort study we merged data from three different randomized controlled trials (RCTs) to assess prognostic variables for chronicity of low back pain. Among the outcome and prognostic variables data were missing in the range of 0 and 48.1%. We used four methods to investigate the influence of respectively sampling and imputation variation: MI only, bootstrap only, and two methods that combine MI and bootstrapping. Variables were selected based on the inclusion frequency of each prognostic variable, i.e. the proportion of times that the variable appeared in the model. The discriminative and calibrative abilities of prognostic models developed by the four methods were assessed at different inclusion levels. Results We found that the effect of imputation variation on the inclusion frequency was larger than the effect of sampling variation. When MI and bootstrapping were combined at the range of 0% (full model) to 90% of variable selection, bootstrap corrected c-index values of 0.70 to 0.71 and slope values of 0.64 to 0.86 were found. Conclusion We recommend to account for both imputation and sampling variation in sets of missing data. The new procedure of combining MI with bootstrapping for variable selection, results in multivariable prognostic models with good performance and is therefore attractive to apply on data sets with missing values. PMID:17629912

Partitioning error components for accuracy-assessment of near-neighbor methods of imputation

Treesearch

Albert R. Stage; Nicholas L. Crookston

2007-01-01

Imputation is applied for two quite different purposes: to supply missing data to complete a data set for subsequent modeling analyses or to estimate subpopulation totals. Error properties of the imputed values have different effects in these two contexts. We partition errors of imputation derived from similar observation units as arising from three sources:...

Missing data imputation of solar radiation data under different atmospheric conditions.

PubMed

Turrado, Concepción Crespo; López, María Del Carmen Meizoso; Lasheras, Fernando Sánchez; Gómez, Benigno Antonio Rodríguez; Rollé, José Luis Calvo; Juez, Francisco Javier de Cos

2014-10-29

Global solar broadband irradiance on a planar surface is measured at weather stations by pyranometers. In the case of the present research, solar radiation values from nine meteorological stations of the MeteoGalicia real-time observational network, captured and stored every ten minutes, are considered. In this kind of record, the lack of data and/or the presence of wrong values adversely affects any time series study. Consequently, when this occurs, a data imputation process must be performed in order to replace missing data with estimated values. This paper aims to evaluate the multivariate imputation of ten-minute scale data by means of the chained equations method (MICE). This method allows the network itself to impute the missing or wrong data of a solar radiation sensor, by using either all or just a group of the measurements of the remaining sensors. Very good results have been obtained with the MICE method in comparison with other methods employed in this field such as Inverse Distance Weighting (IDW) and Multiple Linear Regression (MLR). The average RMSE value of the predictions for the MICE algorithm was 13.37% while that for the MLR it was 28.19%, and 31.68% for the IDW.

Missing Data Imputation of Solar Radiation Data under Different Atmospheric Conditions

PubMed Central

Turrado, Concepción Crespo; López, María del Carmen Meizoso; Lasheras, Fernando Sánchez; Gómez, Benigno Antonio Rodríguez; Rollé, José Luis Calvo; de Cos Juez, Francisco Javier

2014-01-01

Global solar broadband irradiance on a planar surface is measured at weather stations by pyranometers. In the case of the present research, solar radiation values from nine meteorological stations of the MeteoGalicia real-time observational network, captured and stored every ten minutes, are considered. In this kind of record, the lack of data and/or the presence of wrong values adversely affects any time series study. Consequently, when this occurs, a data imputation process must be performed in order to replace missing data with estimated values. This paper aims to evaluate the multivariate imputation of ten-minute scale data by means of the chained equations method (MICE). This method allows the network itself to impute the missing or wrong data of a solar radiation sensor, by using either all or just a group of the measurements of the remaining sensors. Very good results have been obtained with the MICE method in comparison with other methods employed in this field such as Inverse Distance Weighting (IDW) and Multiple Linear Regression (MLR). The average RMSE value of the predictions for the MICE algorithm was 13.37% while that for the MLR it was 28.19%, and 31.68% for the IDW. PMID:25356644

New Insights into Handling Missing Values in Environmental Epidemiological Studies

PubMed Central

Roda, Célina; Nicolis, Ioannis; Momas, Isabelle; Guihenneuc, Chantal

2014-01-01

Missing data are unavoidable in environmental epidemiologic surveys. The aim of this study was to compare methods for handling large amounts of missing values: omission of missing values, single and multiple imputations (through linear regression or partial least squares regression), and a fully Bayesian approach. These methods were applied to the PARIS birth cohort, where indoor domestic pollutant measurements were performed in a random sample of babies' dwellings. A simulation study was conducted to assess performances of different approaches with a high proportion of missing values (from 50% to 95%). Different simulation scenarios were carried out, controlling the true value of the association (odds ratio of 1.0, 1.2, and 1.4), and varying the health outcome prevalence. When a large amount of data is missing, omitting these missing data reduced statistical power and inflated standard errors, which affected the significance of the association. Single imputation underestimated the variability, and considerably increased risk of type I error. All approaches were conservative, except the Bayesian joint model. In the case of a common health outcome, the fully Bayesian approach is the most efficient approach (low root mean square error, reasonable type I error, and high statistical power). Nevertheless for a less prevalent event, the type I error is increased and the statistical power is reduced. The estimated posterior distribution of the OR is useful to refine the conclusion. Among the methods handling missing values, no approach is absolutely the best but when usual approaches (e.g. single imputation) are not sufficient, joint modelling approach of missing process and health association is more efficient when large amounts of data are missing. PMID:25226278

A comparison of imputation techniques for handling missing predictor values in a risk model with a binary outcome.

PubMed

Ambler, Gareth; Omar, Rumana Z; Royston, Patrick

2007-06-01

Risk models that aim to predict the future course and outcome of disease processes are increasingly used in health research, and it is important that they are accurate and reliable. Most of these risk models are fitted using routinely collected data in hospitals or general practices. Clinical outcomes such as short-term mortality will be near-complete, but many of the predictors may have missing values. A common approach to dealing with this is to perform a complete-case analysis. However, this may lead to overfitted models and biased estimates if entire patient subgroups are excluded. The aim of this paper is to investigate a number of methods for imputing missing data to evaluate their effect on risk model estimation and the reliability of the predictions. Multiple imputation methods, including hotdecking and multiple imputation by chained equations (MICE), were investigated along with several single imputation methods. A large national cardiac surgery database was used to create simulated yet realistic datasets. The results suggest that complete case analysis may produce unreliable risk predictions and should be avoided. Conditional mean imputation performed well in our scenario, but may not be appropriate if using variable selection methods. MICE was amongst the best performing multiple imputation methods with regards to the quality of the predictions. Additionally, it produced the least biased estimates, with good coverage, and hence is recommended for use in practice.

FCMPSO: An Imputation for Missing Data Features in Heart Disease Classification

NASA Astrophysics Data System (ADS)

Salleh, Mohd Najib Mohd; Ashikin Samat, Nurul

2017-08-01

The application of data mining and machine learning in directing clinical research into possible hidden knowledge is becoming greatly influential in medical areas. Heart Disease is a killer disease around the world, and early prevention through efficient methods can help to reduce the mortality number. Medical data may contain many uncertainties, as they are fuzzy and vague in nature. Nonetheless, imprecise features data such as no values and missing values can affect quality of classification results. Nevertheless, the other complete features are still capable to give information in certain features. Therefore, an imputation approach based on Fuzzy C-Means and Particle Swarm Optimization (FCMPSO) is developed in preprocessing stage to help fill in the missing values. Then, the complete dataset is trained in classification algorithm, Decision Tree. The experiment is trained with Heart Disease dataset and the performance is analysed using accuracy, precision, and ROC values. Results show that the performance of Decision Tree is increased after the application of FCMSPO for imputation.

Application of a novel hybrid method for spatiotemporal data imputation: A case study of the Minqin County groundwater level

NASA Astrophysics Data System (ADS)

Zhang, Zhongrong; Yang, Xuan; Li, Hao; Li, Weide; Yan, Haowen; Shi, Fei

2017-10-01

The techniques for data analyses have been widely developed in past years, however, missing data still represent a ubiquitous problem in many scientific fields. In particular, dealing with missing spatiotemporal data presents an enormous challenge. Nonetheless, in recent years, a considerable amount of research has focused on spatiotemporal problems, making spatiotemporal missing data imputation methods increasingly indispensable. In this paper, a novel spatiotemporal hybrid method is proposed to verify and imputed spatiotemporal missing values. This new method, termed SOM-FLSSVM, flexibly combines three advanced techniques: self-organizing feature map (SOM) clustering, the fruit fly optimization algorithm (FOA) and the least squares support vector machine (LSSVM). We employ a cross-validation (CV) procedure and FOA swarm intelligence optimization strategy that can search available parameters and determine the optimal imputation model. The spatiotemporal underground water data for Minqin County, China, were selected to test the reliability and imputation ability of SOM-FLSSVM. We carried out a validation experiment and compared three well-studied models with SOM-FLSSVM using a different missing data ratio from 0.1 to 0.8 in the same data set. The results demonstrate that the new hybrid method performs well in terms of both robustness and accuracy for spatiotemporal missing data.

Identifying Heat Waves in Florida: Considerations of Missing Weather Data

PubMed Central

Leary, Emily; Young, Linda J.; DuClos, Chris; Jordan, Melissa M.

2015-01-01

Background Using current climate models, regional-scale changes for Florida over the next 100 years are predicted to include warming over terrestrial areas and very likely increases in the number of high temperature extremes. No uniform definition of a heat wave exists. Most past research on heat waves has focused on evaluating the aftermath of known heat waves, with minimal consideration of missing exposure information. Objectives To identify and discuss methods of handling and imputing missing weather data and how those methods can affect identified periods of extreme heat in Florida. Methods In addition to ignoring missing data, temporal, spatial, and spatio-temporal models are described and utilized to impute missing historical weather data from 1973 to 2012 from 43 Florida weather monitors. Calculated thresholds are used to define periods of extreme heat across Florida. Results Modeling of missing data and imputing missing values can affect the identified periods of extreme heat, through the missing data itself or through the computed thresholds. The differences observed are related to the amount of missingness during June, July, and August, the warmest months of the warm season (April through September). Conclusions Missing data considerations are important when defining periods of extreme heat. Spatio-temporal methods are recommended for data imputation. A heat wave definition that incorporates information from all monitors is advised. PMID:26619198

Identifying Heat Waves in Florida: Considerations of Missing Weather Data.

PubMed

Leary, Emily; Young, Linda J; DuClos, Chris; Jordan, Melissa M

2015-01-01

Using current climate models, regional-scale changes for Florida over the next 100 years are predicted to include warming over terrestrial areas and very likely increases in the number of high temperature extremes. No uniform definition of a heat wave exists. Most past research on heat waves has focused on evaluating the aftermath of known heat waves, with minimal consideration of missing exposure information. To identify and discuss methods of handling and imputing missing weather data and how those methods can affect identified periods of extreme heat in Florida. In addition to ignoring missing data, temporal, spatial, and spatio-temporal models are described and utilized to impute missing historical weather data from 1973 to 2012 from 43 Florida weather monitors. Calculated thresholds are used to define periods of extreme heat across Florida. Modeling of missing data and imputing missing values can affect the identified periods of extreme heat, through the missing data itself or through the computed thresholds. The differences observed are related to the amount of missingness during June, July, and August, the warmest months of the warm season (April through September). Missing data considerations are important when defining periods of extreme heat. Spatio-temporal methods are recommended for data imputation. A heat wave definition that incorporates information from all monitors is advised.

Comparison of methods for dealing with missing values in the EPV-R.

PubMed

Paniagua, David; Amor, Pedro J; Echeburúa, Enrique; Abad, Francisco J

2017-08-01

The development of an effective instrument to assess the risk of partner violence is a topic of great social relevance. This study evaluates the scale of “Predicción del Riesgo de Violencia Grave Contra la Pareja” –Revisada– (EPV-R - Severe Intimate Partner Violence Risk Prediction Scale-Revised), a tool developed in Spain, which is facing the problem of how to treat the high rate of missing values, as is usual in this type of scale. First, responses to the EPV-R in a sample of 1215 male abusers who were reported to the police were used to analyze the patterns of occurrence of missing values, as well as the factor structure. Second, we analyzed the performance of various imputation methods using simulated data that emulates the missing data mechanism found in the empirical database. The imputation procedure originally proposed by the authors of the scale provides acceptable results, although the application of a method based on the Item Response Theory could provide greater accuracy and offers some additional advantages. Item Response Theory appears to be a useful tool for imputing missing data in this type of questionnaire.

Multiple imputation of missing fMRI data in whole brain analysis

PubMed Central

Vaden, Kenneth I.; Gebregziabher, Mulugeta; Kuchinsky, Stefanie E.; Eckert, Mark A.

2012-01-01

Whole brain fMRI analyses rarely include the entire brain because of missing data that result from data acquisition limits and susceptibility artifact, in particular. This missing data problem is typically addressed by omitting voxels from analysis, which may exclude brain regions that are of theoretical interest and increase the potential for Type II error at cortical boundaries or Type I error when spatial thresholds are used to establish significance. Imputation could significantly expand statistical map coverage, increase power, and enhance interpretations of fMRI results. We examined multiple imputation for group level analyses of missing fMRI data using methods that leverage the spatial information in fMRI datasets for both real and simulated data. Available case analysis, neighbor replacement, and regression based imputation approaches were compared in a general linear model framework to determine the extent to which these methods quantitatively (effect size) and qualitatively (spatial coverage) increased the sensitivity of group analyses. In both real and simulated data analysis, multiple imputation provided 1) variance that was most similar to estimates for voxels with no missing data, 2) fewer false positive errors in comparison to mean replacement, and 3) fewer false negative errors in comparison to available case analysis. Compared to the standard analysis approach of omitting voxels with missing data, imputation methods increased brain coverage in this study by 35% (from 33,323 to 45,071 voxels). In addition, multiple imputation increased the size of significant clusters by 58% and number of significant clusters across statistical thresholds, compared to the standard voxel omission approach. While neighbor replacement produced similar results, we recommend multiple imputation because it uses an informed sampling distribution to deal with missing data across subjects that can include neighbor values and other predictors. Multiple imputation is anticipated to be particularly useful for 1) large fMRI data sets with inconsistent missing voxels across subjects and 2) addressing the problem of increased artifact at ultra-high field, which significantly limit the extent of whole brain coverage and interpretations of results. PMID:22500925

Genetic Diversity Analysis of Highly Incomplete SNP Genotype Data with Imputations: An Empirical Assessment

PubMed Central

Fu, Yong-Bi

2014-01-01

Genotyping by sequencing (GBS) recently has emerged as a promising genomic approach for assessing genetic diversity on a genome-wide scale. However, concerns are not lacking about the uniquely large unbalance in GBS genotype data. Although some genotype imputation has been proposed to infer missing observations, little is known about the reliability of a genetic diversity analysis of GBS data, with up to 90% of observations missing. Here we performed an empirical assessment of accuracy in genetic diversity analysis of highly incomplete single nucleotide polymorphism genotypes with imputations. Three large single-nucleotide polymorphism genotype data sets for corn, wheat, and rice were acquired, and missing data with up to 90% of missing observations were randomly generated and then imputed for missing genotypes with three map-independent imputation methods. Estimating heterozygosity and inbreeding coefficient from original, missing, and imputed data revealed variable patterns of bias from assessed levels of missingness and genotype imputation, but the estimation biases were smaller for missing data without genotype imputation. The estimates of genetic differentiation were rather robust up to 90% of missing observations but became substantially biased when missing genotypes were imputed. The estimates of topology accuracy for four representative samples of interested groups generally were reduced with increased levels of missing genotypes. Probabilistic principal component analysis based imputation performed better in terms of topology accuracy than those analyses of missing data without genotype imputation. These findings are not only significant for understanding the reliability of the genetic diversity analysis with respect to large missing data and genotype imputation but also are instructive for performing a proper genetic diversity analysis of highly incomplete GBS or other genotype data. PMID:24626289

Standard and Robust Methods in Regression Imputation

ERIC Educational Resources Information Center

Moraveji, Behjat; Jafarian, Koorosh

2014-01-01

The aim of this paper is to provide an introduction of new imputation algorithms for estimating missing values from official statistics in larger data sets of data pre-processing, or outliers. The goal is to propose a new algorithm called IRMI (iterative robust model-based imputation). This algorithm is able to deal with all challenges like…

Shrinkage regression-based methods for microarray missing value imputation.

PubMed

Wang, Hsiuying; Chiu, Chia-Chun; Wu, Yi-Ching; Wu, Wei-Sheng

2013-01-01

Missing values commonly occur in the microarray data, which usually contain more than 5% missing values with up to 90% of genes affected. Inaccurate missing value estimation results in reducing the power of downstream microarray data analyses. Many types of methods have been developed to estimate missing values. Among them, the regression-based methods are very popular and have been shown to perform better than the other types of methods in many testing microarray datasets. To further improve the performances of the regression-based methods, we propose shrinkage regression-based methods. Our methods take the advantage of the correlation structure in the microarray data and select similar genes for the target gene by Pearson correlation coefficients. Besides, our methods incorporate the least squares principle, utilize a shrinkage estimation approach to adjust the coefficients of the regression model, and then use the new coefficients to estimate missing values. Simulation results show that the proposed methods provide more accurate missing value estimation in six testing microarray datasets than the existing regression-based methods do. Imputation of missing values is a very important aspect of microarray data analyses because most of the downstream analyses require a complete dataset. Therefore, exploring accurate and efficient methods for estimating missing values has become an essential issue. Since our proposed shrinkage regression-based methods can provide accurate missing value estimation, they are competitive alternatives to the existing regression-based methods.

Should multiple imputation be the method of choice for handling missing data in randomized trials?

PubMed Central

Sullivan, Thomas R; White, Ian R; Salter, Amy B; Ryan, Philip; Lee, Katherine J

2016-01-01

The use of multiple imputation has increased markedly in recent years, and journal reviewers may expect to see multiple imputation used to handle missing data. However in randomized trials, where treatment group is always observed and independent of baseline covariates, other approaches may be preferable. Using data simulation we evaluated multiple imputation, performed both overall and separately by randomized group, across a range of commonly encountered scenarios. We considered both missing outcome and missing baseline data, with missing outcome data induced under missing at random mechanisms. Provided the analysis model was correctly specified, multiple imputation produced unbiased treatment effect estimates, but alternative unbiased approaches were often more efficient. When the analysis model overlooked an interaction effect involving randomized group, multiple imputation produced biased estimates of the average treatment effect when applied to missing outcome data, unless imputation was performed separately by randomized group. Based on these results, we conclude that multiple imputation should not be seen as the only acceptable way to handle missing data in randomized trials. In settings where multiple imputation is adopted, we recommend that imputation is carried out separately by randomized group. PMID:28034175

Should multiple imputation be the method of choice for handling missing data in randomized trials?

PubMed

Sullivan, Thomas R; White, Ian R; Salter, Amy B; Ryan, Philip; Lee, Katherine J

2016-01-01

The use of multiple imputation has increased markedly in recent years, and journal reviewers may expect to see multiple imputation used to handle missing data. However in randomized trials, where treatment group is always observed and independent of baseline covariates, other approaches may be preferable. Using data simulation we evaluated multiple imputation, performed both overall and separately by randomized group, across a range of commonly encountered scenarios. We considered both missing outcome and missing baseline data, with missing outcome data induced under missing at random mechanisms. Provided the analysis model was correctly specified, multiple imputation produced unbiased treatment effect estimates, but alternative unbiased approaches were often more efficient. When the analysis model overlooked an interaction effect involving randomized group, multiple imputation produced biased estimates of the average treatment effect when applied to missing outcome data, unless imputation was performed separately by randomized group. Based on these results, we conclude that multiple imputation should not be seen as the only acceptable way to handle missing data in randomized trials. In settings where multiple imputation is adopted, we recommend that imputation is carried out separately by randomized group.

A comparison of multiple imputation methods for handling missing values in longitudinal data in the presence of a time-varying covariate with a non-linear association with time: a simulation study.

PubMed

De Silva, Anurika Priyanjali; Moreno-Betancur, Margarita; De Livera, Alysha Madhu; Lee, Katherine Jane; Simpson, Julie Anne

2017-07-25

Missing data is a common problem in epidemiological studies, and is particularly prominent in longitudinal data, which involve multiple waves of data collection. Traditional multiple imputation (MI) methods (fully conditional specification (FCS) and multivariate normal imputation (MVNI)) treat repeated measurements of the same time-dependent variable as just another 'distinct' variable for imputation and therefore do not make the most of the longitudinal structure of the data. Only a few studies have explored extensions to the standard approaches to account for the temporal structure of longitudinal data. One suggestion is the two-fold fully conditional specification (two-fold FCS) algorithm, which restricts the imputation of a time-dependent variable to time blocks where the imputation model includes measurements taken at the specified and adjacent times. To date, no study has investigated the performance of two-fold FCS and standard MI methods for handling missing data in a time-varying covariate with a non-linear trajectory over time - a commonly encountered scenario in epidemiological studies. We simulated 1000 datasets of 5000 individuals based on the Longitudinal Study of Australian Children (LSAC). Three missing data mechanisms: missing completely at random (MCAR), and a weak and a strong missing at random (MAR) scenarios were used to impose missingness on body mass index (BMI) for age z-scores; a continuous time-varying exposure variable with a non-linear trajectory over time. We evaluated the performance of FCS, MVNI, and two-fold FCS for handling up to 50% of missing data when assessing the association between childhood obesity and sleep problems. The standard two-fold FCS produced slightly more biased and less precise estimates than FCS and MVNI. We observed slight improvements in bias and precision when using a time window width of two for the two-fold FCS algorithm compared to the standard width of one. We recommend the use of FCS or MVNI in a similar longitudinal setting, and when encountering convergence issues due to a large number of time points or variables with missing values, the two-fold FCS with exploration of a suitable time window.

Multiple imputation in the presence of non-normal data.

PubMed

Lee, Katherine J; Carlin, John B

2017-02-20

Multiple imputation (MI) is becoming increasingly popular for handling missing data. Standard approaches for MI assume normality for continuous variables (conditionally on the other variables in the imputation model). However, it is unclear how to impute non-normally distributed continuous variables. Using simulation and a case study, we compared various transformations applied prior to imputation, including a novel non-parametric transformation, to imputation on the raw scale and using predictive mean matching (PMM) when imputing non-normal data. We generated data from a range of non-normal distributions, and set 50% to missing completely at random or missing at random. We then imputed missing values on the raw scale, following a zero-skewness log, Box-Cox or non-parametric transformation and using PMM with both type 1 and 2 matching. We compared inferences regarding the marginal mean of the incomplete variable and the association with a fully observed outcome. We also compared results from these approaches in the analysis of depression and anxiety symptoms in parents of very preterm compared with term-born infants. The results provide novel empirical evidence that the decision regarding how to impute a non-normal variable should be based on the nature of the relationship between the variables of interest. If the relationship is linear in the untransformed scale, transformation can introduce bias irrespective of the transformation used. However, if the relationship is non-linear, it may be important to transform the variable to accurately capture this relationship. A useful alternative is to impute the variable using PMM with type 1 matching. Copyright © 2016 John Wiley & Sons, Ltd. Copyright © 2016 John Wiley & Sons, Ltd.

Multiple imputation: an application to income nonresponse in the National Survey on Recreation and the Environment

Treesearch

Stanley J. Zarnoch; H. Ken Cordell; Carter J. Betz; John C. Bergstrom

2010-01-01

Multiple imputation is used to create values for missing family income data in the National Survey on Recreation and the Environment. We present an overview of the survey and a description of the missingness pattern for family income and other key variables. We create a logistic model for the multiple imputation process and to impute data sets for family income. We...

Multiple imputation for handling missing outcome data when estimating the relative risk.

PubMed

Sullivan, Thomas R; Lee, Katherine J; Ryan, Philip; Salter, Amy B

2017-09-06

Multiple imputation is a popular approach to handling missing data in medical research, yet little is known about its applicability for estimating the relative risk. Standard methods for imputing incomplete binary outcomes involve logistic regression or an assumption of multivariate normality, whereas relative risks are typically estimated using log binomial models. It is unclear whether misspecification of the imputation model in this setting could lead to biased parameter estimates. Using simulated data, we evaluated the performance of multiple imputation for handling missing data prior to estimating adjusted relative risks from a correctly specified multivariable log binomial model. We considered an arbitrary pattern of missing data in both outcome and exposure variables, with missing data induced under missing at random mechanisms. Focusing on standard model-based methods of multiple imputation, missing data were imputed using multivariate normal imputation or fully conditional specification with a logistic imputation model for the outcome. Multivariate normal imputation performed poorly in the simulation study, consistently producing estimates of the relative risk that were biased towards the null. Despite outperforming multivariate normal imputation, fully conditional specification also produced somewhat biased estimates, with greater bias observed for higher outcome prevalences and larger relative risks. Deleting imputed outcomes from analysis datasets did not improve the performance of fully conditional specification. Both multivariate normal imputation and fully conditional specification produced biased estimates of the relative risk, presumably since both use a misspecified imputation model. Based on simulation results, we recommend researchers use fully conditional specification rather than multivariate normal imputation and retain imputed outcomes in the analysis when estimating relative risks. However fully conditional specification is not without its shortcomings, and so further research is needed to identify optimal approaches for relative risk estimation within the multiple imputation framework.

Ipsative imputation for a 15-item Geriatric Depression Scale in community-dwelling elderly people.

PubMed

Imai, Hissei; Furukawa, Toshiaki A; Kasahara, Yoriko; Ishimoto, Yasuko; Kimura, Yumi; Fukutomi, Eriko; Chen, Wen-Ling; Tanaka, Mire; Sakamoto, Ryota; Wada, Taizo; Fujisawa, Michiko; Okumiya, Kiyohito; Matsubayashi, Kozo

2014-09-01

Missing data are inevitable in almost all medical studies. Imputation methods using the probabilistic model are common, but they cannot impute individual data and require special software. In contrast, the ipsative imputation method, which substitutes the missing items by the mean of the remaining items within the individual, is easy and does not need any special software, but it can provide individual scores. The aim of the present study was to evaluate the validity of the ipsative imputation method using data involving the 15-item Geriatric Depression Scale. Participants were community-dwelling elderly individuals (n = 1178). A structural equation model was constructed. The model fit indexes were calculated to assess the validity of the imputation method when it is used for individuals who were missing 20% of data or less and 40% of data or less, depending on whether we assumed that their correlation coefficients were the same as the dataset with no missing items. Finally, we compared path coefficients of the dataset imputed by ipsative imputation with those by multiple imputation. When compared with the assumption that the datasets differed, all of the model fit indexes were better under the assumption that the dataset without missing data is the same as that that was missing 20% of data or less. However, by the same assumption, the model fit indexes were worse in the dataset that was missing 40% of data or less. The path coefficients of the dataset imputed by ipsative imputation and by multiple imputation were compatible with each other if the proportion of missing items was 20% or less. Ipsative imputation appears to be a valid imputation method and can be used to impute data in studies using the 15-item Geriatric Depression Scale, if the percentage of its missing items is 20% or less. © 2014 The Authors. Psychogeriatrics © 2014 Japanese Psychogeriatric Society.

Binary variable multiple-model multiple imputation to address missing data mechanism uncertainty: Application to a smoking cessation trial

PubMed Central

Siddique, Juned; Harel, Ofer; Crespi, Catherine M.; Hedeker, Donald

2014-01-01

The true missing data mechanism is never known in practice. We present a method for generating multiple imputations for binary variables that formally incorporates missing data mechanism uncertainty. Imputations are generated from a distribution of imputation models rather than a single model, with the distribution reflecting subjective notions of missing data mechanism uncertainty. Parameter estimates and standard errors are obtained using rules for nested multiple imputation. Using simulation, we investigate the impact of missing data mechanism uncertainty on post-imputation inferences and show that incorporating this uncertainty can increase the coverage of parameter estimates. We apply our method to a longitudinal smoking cessation trial where nonignorably missing data were a concern. Our method provides a simple approach for formalizing subjective notions regarding nonresponse and can be implemented using existing imputation software. PMID:24634315

What impact do assumptions about missing data have on conclusions? A practical sensitivity analysis for a cancer survival registry.

PubMed

Smuk, M; Carpenter, J R; Morris, T P

2017-02-06

Within epidemiological and clinical research, missing data are a common issue and often over looked in publications. When the issue of missing observations is addressed it is usually assumed that the missing data are 'missing at random' (MAR). This assumption should be checked for plausibility, however it is untestable, thus inferences should be assessed for robustness to departures from missing at random. We highlight the method of pattern mixture sensitivity analysis after multiple imputation using colorectal cancer data as an example. We focus on the Dukes' stage variable which has the highest proportion of missing observations. First, we find the probability of being in each Dukes' stage given the MAR imputed dataset. We use these probabilities in a questionnaire to elicit prior beliefs from experts on what they believe the probability would be in the missing data. The questionnaire responses are then used in a Dirichlet draw to create a Bayesian 'missing not at random' (MNAR) prior to impute the missing observations. The model of interest is applied and inferences are compared to those from the MAR imputed data. The inferences were largely insensitive to departure from MAR. Inferences under MNAR suggested a smaller association between Dukes' stage and death, though the association remained positive and with similarly low p values. We conclude by discussing the positives and negatives of our method and highlight the importance of making people aware of the need to test the MAR assumption.

Improving cluster-based missing value estimation of DNA microarray data.

PubMed

Brás, Lígia P; Menezes, José C

2007-06-01

We present a modification of the weighted K-nearest neighbours imputation method (KNNimpute) for missing values (MVs) estimation in microarray data based on the reuse of estimated data. The method was called iterative KNN imputation (IKNNimpute) as the estimation is performed iteratively using the recently estimated values. The estimation efficiency of IKNNimpute was assessed under different conditions (data type, fraction and structure of missing data) by the normalized root mean squared error (NRMSE) and the correlation coefficients between estimated and true values, and compared with that of other cluster-based estimation methods (KNNimpute and sequential KNN). We further investigated the influence of imputation on the detection of differentially expressed genes using SAM by examining the differentially expressed genes that are lost after MV estimation. The performance measures give consistent results, indicating that the iterative procedure of IKNNimpute can enhance the prediction ability of cluster-based methods in the presence of high missing rates, in non-time series experiments and in data sets comprising both time series and non-time series data, because the information of the genes having MVs is used more efficiently and the iterative procedure allows refining the MV estimates. More importantly, IKNN has a smaller detrimental effect on the detection of differentially expressed genes.

Multiple Imputation for Incomplete Data in Epidemiologic Studies

PubMed Central

Harel, Ofer; Mitchell, Emily M; Perkins, Neil J; Cole, Stephen R; Tchetgen Tchetgen, Eric J; Sun, BaoLuo; Schisterman, Enrique F

2018-01-01

Abstract Epidemiologic studies are frequently susceptible to missing information. Omitting observations with missing variables remains a common strategy in epidemiologic studies, yet this simple approach can often severely bias parameter estimates of interest if the values are not missing completely at random. Even when missingness is completely random, complete-case analysis can reduce the efficiency of estimated parameters, because large amounts of available data are simply tossed out with the incomplete observations. Alternative methods for mitigating the influence of missing information, such as multiple imputation, are becoming an increasing popular strategy in order to retain all available information, reduce potential bias, and improve efficiency in parameter estimation. In this paper, we describe the theoretical underpinnings of multiple imputation, and we illustrate application of this method as part of a collaborative challenge to assess the performance of various techniques for dealing with missing data (Am J Epidemiol. 2018;187(3):568–575). We detail the steps necessary to perform multiple imputation on a subset of data from the Collaborative Perinatal Project (1959–1974), where the goal is to estimate the odds of spontaneous abortion associated with smoking during pregnancy. PMID:29165547

A stochastic multiple imputation algorithm for missing covariate data in tree-structured survival analysis.

PubMed

Wallace, Meredith L; Anderson, Stewart J; Mazumdar, Sati

2010-12-20

Missing covariate data present a challenge to tree-structured methodology due to the fact that a single tree model, as opposed to an estimated parameter value, may be desired for use in a clinical setting. To address this problem, we suggest a multiple imputation algorithm that adds draws of stochastic error to a tree-based single imputation method presented by Conversano and Siciliano (Technical Report, University of Naples, 2003). Unlike previously proposed techniques for accommodating missing covariate data in tree-structured analyses, our methodology allows the modeling of complex and nonlinear covariate structures while still resulting in a single tree model. We perform a simulation study to evaluate our stochastic multiple imputation algorithm when covariate data are missing at random and compare it to other currently used methods. Our algorithm is advantageous for identifying the true underlying covariate structure when complex data and larger percentages of missing covariate observations are present. It is competitive with other current methods with respect to prediction accuracy. To illustrate our algorithm, we create a tree-structured survival model for predicting time to treatment response in older, depressed adults. Copyright © 2010 John Wiley & Sons, Ltd.

Breast Cancer and Modifiable Lifestyle Factors in Argentinean Women: Addressing Missing Data in a Case-Control Study

PubMed Central

Coquet, Julia Becaria; Tumas, Natalia; Osella, Alberto Ruben; Tanzi, Matteo; Franco, Isabella; Diaz, Maria Del Pilar

2016-01-01

A number of studies have evidenced the effect of modifiable lifestyle factors such as diet, breastfeeding and nutritional status on breast cancer risk. However, none have addressed the missing data problem in nutritional epidemiologic research in South America. Missing data is a frequent problem in breast cancer studies and epidemiological settings in general. Estimates of effect obtained from these studies may be biased, if no appropriate method for handling missing data is applied. We performed Multiple Imputation for missing values on covariates in a breast cancer case-control study of Córdoba (Argentina) to optimize risk estimates. Data was obtained from a breast cancer case control study from 2008 to 2015 (318 cases, 526 controls). Complete case analysis and multiple imputation using chained equations were the methods applied to estimate the effects of a Traditional dietary pattern and other recognized factors associated with breast cancer. Physical activity and socioeconomic status were imputed. Logistic regression models were performed. When complete case analysis was performed only 31% of women were considered. Although a positive association of Traditional dietary pattern and breast cancer was observed from both approaches (complete case analysis OR=1.3, 95%CI=1.0-1.7; multiple imputation OR=1.4, 95%CI=1.2-1.7), effects of other covariates, like BMI and breastfeeding, were only identified when multiple imputation was considered. A Traditional dietary pattern, BMI and breastfeeding are associated with the occurrence of breast cancer in this Argentinean population when multiple imputation is appropriately performed. Multiple Imputation is suggested in Latin America’s epidemiologic studies to optimize effect estimates in the future. PMID:27892664

Data imputation analysis for Cosmic Rays time series

NASA Astrophysics Data System (ADS)

Fernandes, R. C.; Lucio, P. S.; Fernandez, J. H.

2017-05-01

The occurrence of missing data concerning Galactic Cosmic Rays time series (GCR) is inevitable since loss of data is due to mechanical and human failure or technical problems and different periods of operation of GCR stations. The aim of this study was to perform multiple dataset imputation in order to depict the observational dataset. The study has used the monthly time series of GCR Climax (CLMX) and Roma (ROME) from 1960 to 2004 to simulate scenarios of 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80% and 90% of missing data compared to observed ROME series, with 50 replicates. Then, the CLMX station as a proxy for allocation of these scenarios was used. Three different methods for monthly dataset imputation were selected: AMÉLIA II - runs the bootstrap Expectation Maximization algorithm, MICE - runs an algorithm via Multivariate Imputation by Chained Equations and MTSDI - an Expectation Maximization algorithm-based method for imputation of missing values in multivariate normal time series. The synthetic time series compared with the observed ROME series has also been evaluated using several skill measures as such as RMSE, NRMSE, Agreement Index, R, R2, F-test and t-test. The results showed that for CLMX and ROME, the R2 and R statistics were equal to 0.98 and 0.96, respectively. It was observed that increases in the number of gaps generate loss of quality of the time series. Data imputation was more efficient with MTSDI method, with negligible errors and best skill coefficients. The results suggest a limit of about 60% of missing data for imputation, for monthly averages, no more than this. It is noteworthy that CLMX, ROME and KIEL stations present no missing data in the target period. This methodology allowed reconstructing 43 time series.

Multi-task Gaussian process for imputing missing data in multi-trait and multi-environment trials.

PubMed

Hori, Tomoaki; Montcho, David; Agbangla, Clement; Ebana, Kaworu; Futakuchi, Koichi; Iwata, Hiroyoshi

2016-11-01

A method based on a multi-task Gaussian process using self-measuring similarity gave increased accuracy for imputing missing phenotypic data in multi-trait and multi-environment trials. Multi-environmental trial (MET) data often encounter the problem of missing data. Accurate imputation of missing data makes subsequent analysis more effective and the results easier to understand. Moreover, accurate imputation may help to reduce the cost of phenotyping for thinned-out lines tested in METs. METs are generally performed for multiple traits that are correlated to each other. Correlation among traits can be useful information for imputation, but single-trait-based methods cannot utilize information shared by traits that are correlated. In this paper, we propose imputation methods based on a multi-task Gaussian process (MTGP) using self-measuring similarity kernels reflecting relationships among traits, genotypes, and environments. This framework allows us to use genetic correlation among multi-trait multi-environment data and also to combine MET data and marker genotype data. We compared the accuracy of three MTGP methods and iterative regularized PCA using rice MET data. Two scenarios for the generation of missing data at various missing rates were considered. The MTGP performed a better imputation accuracy than regularized PCA, especially at high missing rates. Under the 'uniform' scenario, in which missing data arise randomly, inclusion of marker genotype data in the imputation increased the imputation accuracy at high missing rates. Under the 'fiber' scenario, in which missing data arise in all traits for some combinations between genotypes and environments, the inclusion of marker genotype data decreased the imputation accuracy for most traits while increasing the accuracy in a few traits remarkably. The proposed methods will be useful for solving the missing data problem in MET data.

Tackling Missing Data in Community Health Studies Using Additive LS-SVM Classifier.

PubMed

Wang, Guanjin; Deng, Zhaohong; Choi, Kup-Sze

2018-03-01

Missing data is a common issue in community health and epidemiological studies. Direct removal of samples with missing data can lead to reduced sample size and information bias, which deteriorates the significance of the results. While data imputation methods are available to deal with missing data, they are limited in performance and could introduce noises into the dataset. Instead of data imputation, a novel method based on additive least square support vector machine (LS-SVM) is proposed in this paper for predictive modeling when the input features of the model contain missing data. The method also determines simultaneously the influence of the features with missing values on the classification accuracy using the fast leave-one-out cross-validation strategy. The performance of the method is evaluated by applying it to predict the quality of life (QOL) of elderly people using health data collected in the community. The dataset involves demographics, socioeconomic status, health history, and the outcomes of health assessments of 444 community-dwelling elderly people, with 5% to 60% of data missing in some of the input features. The QOL is measured using a standard questionnaire of the World Health Organization. Results show that the proposed method outperforms four conventional methods for handling missing data-case deletion, feature deletion, mean imputation, and K-nearest neighbor imputation, with the average QOL prediction accuracy reaching 0.7418. It is potentially a promising technique for tackling missing data in community health research and other applications.

Spatial Copula Model for Imputing Traffic Flow Data from Remote Microwave Sensors.

PubMed

Ma, Xiaolei; Luan, Sen; Du, Bowen; Yu, Bin

2017-09-21

Issues of missing data have become increasingly serious with the rapid increase in usage of traffic sensors. Analyses of the Beijing ring expressway have showed that up to 50% of microwave sensors pose missing values. The imputation of missing traffic data must be urgently solved although a precise solution that cannot be easily achieved due to the significant number of missing portions. In this study, copula-based models are proposed for the spatial interpolation of traffic flow from remote traffic microwave sensors. Most existing interpolation methods only rely on covariance functions to depict spatial correlation and are unsuitable for coping with anomalies due to Gaussian consumption. Copula theory overcomes this issue and provides a connection between the correlation function and the marginal distribution function of traffic flow. To validate copula-based models, a comparison with three kriging methods is conducted. Results indicate that copula-based models outperform kriging methods, especially on roads with irregular traffic patterns. Copula-based models demonstrate significant potential to impute missing data in large-scale transportation networks.

Estimation of the Percentage of Newly Diagnosed HIV-Positive Persons Linked to HIV Medical Care in CDC-Funded HIV Testing Programs.

PubMed

Wang, Guoshen; Pan, Yi; Seth, Puja; Song, Ruiguang; Belcher, Lisa

2017-01-01

Missing data create challenges for determining progress made in linking HIV-positive persons to HIV medical care. Statistical methods are not used to address missing program data on linkage. In 2014, 61 health department jurisdictions were funded by Centers for Disease Control and Prevention (CDC) and submitted data on HIV testing, newly diagnosed HIV-positive persons, and linkage to HIV medical care. Missing or unusable data existed in our data set. A new approach using multiple imputation to address missing linkage data was proposed, and results were compared to the current approach that uses data with complete information. There were 12,472 newly diagnosed HIV-positive persons from CDC-funded HIV testing events in 2014. Using multiple imputation, 94.1% (95% confidence interval (CI): [93.7%, 94.6%]) of newly diagnosed persons were referred to HIV medical care, 88.6% (95% CI: [88.0%, 89.1%]) were linked to care within any time frame, and 83.6% (95% CI: [83.0%, 84.3%]) were linked to care within 90 days. Multiple imputation is recommended for addressing missing linkage data in future analyses when the missing percentage is high. The use of multiple imputation for missing values can result in a better understanding of how programs are performing on key HIV testing and HIV service delivery indicators.

3D-MICE: integration of cross-sectional and longitudinal imputation for multi-analyte longitudinal clinical data.

PubMed

Luo, Yuan; Szolovits, Peter; Dighe, Anand S; Baron, Jason M

2018-06-01

A key challenge in clinical data mining is that most clinical datasets contain missing data. Since many commonly used machine learning algorithms require complete datasets (no missing data), clinical analytic approaches often entail an imputation procedure to "fill in" missing data. However, although most clinical datasets contain a temporal component, most commonly used imputation methods do not adequately accommodate longitudinal time-based data. We sought to develop a new imputation algorithm, 3-dimensional multiple imputation with chained equations (3D-MICE), that can perform accurate imputation of missing clinical time series data. We extracted clinical laboratory test results for 13 commonly measured analytes (clinical laboratory tests). We imputed missing test results for the 13 analytes using 3 imputation methods: multiple imputation with chained equations (MICE), Gaussian process (GP), and 3D-MICE. 3D-MICE utilizes both MICE and GP imputation to integrate cross-sectional and longitudinal information. To evaluate imputation method performance, we randomly masked selected test results and imputed these masked results alongside results missing from our original data. We compared predicted results to measured results for masked data points. 3D-MICE performed significantly better than MICE and GP-based imputation in a composite of all 13 analytes, predicting missing results with a normalized root-mean-square error of 0.342, compared to 0.373 for MICE alone and 0.358 for GP alone. 3D-MICE offers a novel and practical approach to imputing clinical laboratory time series data. 3D-MICE may provide an additional tool for use as a foundation in clinical predictive analytics and intelligent clinical decision support.

Responsiveness-informed multiple imputation and inverse probability-weighting in cohort studies with missing data that are non-monotone or not missing at random.

PubMed

Doidge, James C

2018-02-01

Population-based cohort studies are invaluable to health research because of the breadth of data collection over time, and the representativeness of their samples. However, they are especially prone to missing data, which can compromise the validity of analyses when data are not missing at random. Having many waves of data collection presents opportunity for participants' responsiveness to be observed over time, which may be informative about missing data mechanisms and thus useful as an auxiliary variable. Modern approaches to handling missing data such as multiple imputation and maximum likelihood can be difficult to implement with the large numbers of auxiliary variables and large amounts of non-monotone missing data that occur in cohort studies. Inverse probability-weighting can be easier to implement but conventional wisdom has stated that it cannot be applied to non-monotone missing data. This paper describes two methods of applying inverse probability-weighting to non-monotone missing data, and explores the potential value of including measures of responsiveness in either inverse probability-weighting or multiple imputation. Simulation studies are used to compare methods and demonstrate that responsiveness in longitudinal studies can be used to mitigate bias induced by missing data, even when data are not missing at random.

Which missing value imputation method to use in expression profiles: a comparative study and two selection schemes.

PubMed

Brock, Guy N; Shaffer, John R; Blakesley, Richard E; Lotz, Meredith J; Tseng, George C

2008-01-10

Gene expression data frequently contain missing values, however, most down-stream analyses for microarray experiments require complete data. In the literature many methods have been proposed to estimate missing values via information of the correlation patterns within the gene expression matrix. Each method has its own advantages, but the specific conditions for which each method is preferred remains largely unclear. In this report we describe an extensive evaluation of eight current imputation methods on multiple types of microarray experiments, including time series, multiple exposures, and multiple exposures x time series data. We then introduce two complementary selection schemes for determining the most appropriate imputation method for any given data set. We found that the optimal imputation algorithms (LSA, LLS, and BPCA) are all highly competitive with each other, and that no method is uniformly superior in all the data sets we examined. The success of each method can also depend on the underlying "complexity" of the expression data, where we take complexity to indicate the difficulty in mapping the gene expression matrix to a lower-dimensional subspace. We developed an entropy measure to quantify the complexity of expression matrixes and found that, by incorporating this information, the entropy-based selection (EBS) scheme is useful for selecting an appropriate imputation algorithm. We further propose a simulation-based self-training selection (STS) scheme. This technique has been used previously for microarray data imputation, but for different purposes. The scheme selects the optimal or near-optimal method with high accuracy but at an increased computational cost. Our findings provide insight into the problem of which imputation method is optimal for a given data set. Three top-performing methods (LSA, LLS and BPCA) are competitive with each other. Global-based imputation methods (PLS, SVD, BPCA) performed better on mcroarray data with lower complexity, while neighbour-based methods (KNN, OLS, LSA, LLS) performed better in data with higher complexity. We also found that the EBS and STS schemes serve as complementary and effective tools for selecting the optimal imputation algorithm.

Application of Multiple Imputation for Missing Values in Three-Way Three-Mode Multi-Environment Trial Data

PubMed Central

Tian, Ting; McLachlan, Geoffrey J.; Dieters, Mark J.; Basford, Kaye E.

2015-01-01

It is a common occurrence in plant breeding programs to observe missing values in three-way three-mode multi-environment trial (MET) data. We proposed modifications of models for estimating missing observations for these data arrays, and developed a novel approach in terms of hierarchical clustering. Multiple imputation (MI) was used in four ways, multiple agglomerative hierarchical clustering, normal distribution model, normal regression model, and predictive mean match. The later three models used both Bayesian analysis and non-Bayesian analysis, while the first approach used a clustering procedure with randomly selected attributes and assigned real values from the nearest neighbour to the one with missing observations. Different proportions of data entries in six complete datasets were randomly selected to be missing and the MI methods were compared based on the efficiency and accuracy of estimating those values. The results indicated that the models using Bayesian analysis had slightly higher accuracy of estimation performance than those using non-Bayesian analysis but they were more time-consuming. However, the novel approach of multiple agglomerative hierarchical clustering demonstrated the overall best performances. PMID:26689369

Application of Multiple Imputation for Missing Values in Three-Way Three-Mode Multi-Environment Trial Data.

PubMed

Tian, Ting; McLachlan, Geoffrey J; Dieters, Mark J; Basford, Kaye E

2015-01-01

It is a common occurrence in plant breeding programs to observe missing values in three-way three-mode multi-environment trial (MET) data. We proposed modifications of models for estimating missing observations for these data arrays, and developed a novel approach in terms of hierarchical clustering. Multiple imputation (MI) was used in four ways, multiple agglomerative hierarchical clustering, normal distribution model, normal regression model, and predictive mean match. The later three models used both Bayesian analysis and non-Bayesian analysis, while the first approach used a clustering procedure with randomly selected attributes and assigned real values from the nearest neighbour to the one with missing observations. Different proportions of data entries in six complete datasets were randomly selected to be missing and the MI methods were compared based on the efficiency and accuracy of estimating those values. The results indicated that the models using Bayesian analysis had slightly higher accuracy of estimation performance than those using non-Bayesian analysis but they were more time-consuming. However, the novel approach of multiple agglomerative hierarchical clustering demonstrated the overall best performances.

Variable Selection in the Presence of Missing Data: Imputation-based Methods.

PubMed

Zhao, Yize; Long, Qi

2017-01-01

Variable selection plays an essential role in regression analysis as it identifies important variables that associated with outcomes and is known to improve predictive accuracy of resulting models. Variable selection methods have been widely investigated for fully observed data. However, in the presence of missing data, methods for variable selection need to be carefully designed to account for missing data mechanisms and statistical techniques used for handling missing data. Since imputation is arguably the most popular method for handling missing data due to its ease of use, statistical methods for variable selection that are combined with imputation are of particular interest. These methods, valid used under the assumptions of missing at random (MAR) and missing completely at random (MCAR), largely fall into three general strategies. The first strategy applies existing variable selection methods to each imputed dataset and then combine variable selection results across all imputed datasets. The second strategy applies existing variable selection methods to stacked imputed datasets. The third variable selection strategy combines resampling techniques such as bootstrap with imputation. Despite recent advances, this area remains under-developed and offers fertile ground for further research.

On Matrix Sampling and Imputation of Context Questionnaires with Implications for the Generation of Plausible Values in Large-Scale Assessments

ERIC Educational Resources Information Center

Kaplan, David; Su, Dan

2016-01-01

This article presents findings on the consequences of matrix sampling of context questionnaires for the generation of plausible values in large-scale assessments. Three studies are conducted. Study 1 uses data from PISA 2012 to examine several different forms of missing data imputation within the chained equations framework: predictive mean…

Multiple Imputation of Item Scores in Test and Questionnaire Data, and Influence on Psychometric Results

ERIC Educational Resources Information Center

van Ginkel, Joost R.; van der Ark, L. Andries; Sijtsma, Klaas

2007-01-01

The performance of five simple multiple imputation methods for dealing with missing data were compared. In addition, random imputation and multivariate normal imputation were used as lower and upper benchmark, respectively. Test data were simulated and item scores were deleted such that they were either missing completely at random, missing at…

Estimation of missing values in solar radiation data using piecewise interpolation methods: Case study at Penang city

NASA Astrophysics Data System (ADS)

Zainudin, Mohd Lutfi; Saaban, Azizan; Bakar, Mohd Nazari Abu

2015-12-01

The solar radiation values have been composed by automatic weather station using the device that namely pyranometer. The device is functions to records all the radiation values that have been dispersed, and these data are very useful for it experimental works and solar device's development. In addition, for modeling and designing on solar radiation system application is needed for complete data observation. Unfortunately, lack for obtained the complete solar radiation data frequently occur due to several technical problems, which mainly contributed by monitoring device. Into encountering this matter, estimation missing values in an effort to substitute absent values with imputed data. This paper aimed to evaluate several piecewise interpolation techniques likes linear, splines, cubic, and nearest neighbor into dealing missing values in hourly solar radiation data. Then, proposed an extendable work into investigating the potential used of cubic Bezier technique and cubic Said-ball method as estimator tools. As result, methods for cubic Bezier and Said-ball perform the best compare to another piecewise imputation technique.

Spatial Copula Model for Imputing Traffic Flow Data from Remote Microwave Sensors

PubMed Central

Ma, Xiaolei; Du, Bowen; Yu, Bin

2017-01-01

Issues of missing data have become increasingly serious with the rapid increase in usage of traffic sensors. Analyses of the Beijing ring expressway have showed that up to 50% of microwave sensors pose missing values. The imputation of missing traffic data must be urgently solved although a precise solution that cannot be easily achieved due to the significant number of missing portions. In this study, copula-based models are proposed for the spatial interpolation of traffic flow from remote traffic microwave sensors. Most existing interpolation methods only rely on covariance functions to depict spatial correlation and are unsuitable for coping with anomalies due to Gaussian consumption. Copula theory overcomes this issue and provides a connection between the correlation function and the marginal distribution function of traffic flow. To validate copula-based models, a comparison with three kriging methods is conducted. Results indicate that copula-based models outperform kriging methods, especially on roads with irregular traffic patterns. Copula-based models demonstrate significant potential to impute missing data in large-scale transportation networks. PMID:28934164

Identifying reprioritization response shift in a stroke caregiver population: a comparison of missing data methods.

PubMed

Sajobi, Tolulope T; Lix, Lisa M; Singh, Gurbakhshash; Lowerison, Mark; Engbers, Jordan; Mayo, Nancy E

2015-03-01

Response shift (RS) is an important phenomenon that influences the assessment of longitudinal changes in health-related quality of life (HRQOL) studies. Given that RS effects are often small, missing data due to attrition or item non-response can contribute to failure to detect RS effects. Since missing data are often encountered in longitudinal HRQOL data, effective strategies to deal with missing data are important to consider. This study aims to compare different imputation methods on the detection of reprioritization RS in the HRQOL of caregivers of stroke survivors. Data were from a Canadian multi-center longitudinal study of caregivers of stroke survivors over a one-year period. The Stroke Impact Scale physical function score at baseline, with a cutoff of 75, was used to measure patient stroke severity for the reprioritization RS analysis. Mean imputation, likelihood-based expectation-maximization imputation, and multiple imputation methods were compared in test procedures based on changes in relative importance weights to detect RS in SF-36 domains over a 6-month period. Monte Carlo simulation methods were used to compare the statistical powers of relative importance test procedures for detecting RS in incomplete longitudinal data under different missing data mechanisms and imputation methods. Of the 409 caregivers, 15.9 and 31.3 % of them had missing data at baseline and 6 months, respectively. There were no statistically significant changes in relative importance weights on any of the domains when complete-case analysis was adopted. But statistical significant changes were detected on physical functioning and/or vitality domains when mean imputation or EM imputation was adopted. There were also statistically significant changes in relative importance weights for physical functioning, mental health, and vitality domains when multiple imputation method was adopted. Our simulations revealed that relative importance test procedures were least powerful under complete-case analysis method and most powerful when a mean imputation or multiple imputation method was adopted for missing data, regardless of the missing data mechanism and proportion of missing data. Test procedures based on relative importance measures are sensitive to the type and amount of missing data and imputation method. Relative importance test procedures based on mean imputation and multiple imputation are recommended for detecting RS in incomplete data.

Hierarchical imputation of systematically and sporadically missing data: An approximate Bayesian approach using chained equations.

PubMed

Jolani, Shahab

2018-03-01

In health and medical sciences, multiple imputation (MI) is now becoming popular to obtain valid inferences in the presence of missing data. However, MI of clustered data such as multicenter studies and individual participant data meta-analysis requires advanced imputation routines that preserve the hierarchical structure of data. In clustered data, a specific challenge is the presence of systematically missing data, when a variable is completely missing in some clusters, and sporadically missing data, when it is partly missing in some clusters. Unfortunately, little is known about how to perform MI when both types of missing data occur simultaneously. We develop a new class of hierarchical imputation approach based on chained equations methodology that simultaneously imputes systematically and sporadically missing data while allowing for arbitrary patterns of missingness among them. Here, we use a random effect imputation model and adopt a simplification over fully Bayesian techniques such as Gibbs sampler to directly obtain draws of parameters within each step of the chained equations. We justify through theoretical arguments and extensive simulation studies that the proposed imputation methodology has good statistical properties in terms of bias and coverage rates of parameter estimates. An illustration is given in a case study with eight individual participant datasets. © 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

A Review of Missing Data Handling Methods in Education Research

ERIC Educational Resources Information Center

Cheema, Jehanzeb R.

2014-01-01

Missing data are a common occurrence in survey-based research studies in education, and the way missing values are handled can significantly affect the results of analyses based on such data. Despite known problems with performance of some missing data handling methods, such as mean imputation, many researchers in education continue to use those…

Evaluating the Use of Existing Data Sources, Probabilistic Linkage, and Multiple Imputation to Build Population-based Injury Databases Across Phases of Trauma Care

PubMed Central

Newgard, Craig; Malveau, Susan; Staudenmayer, Kristan; Wang, N. Ewen; Hsia, Renee Y.; Mann, N. Clay; Holmes, James F.; Kuppermann, Nathan; Haukoos, Jason S.; Bulger, Eileen M.; Dai, Mengtao; Cook, Lawrence J.

2012-01-01

Objectives The objective was to evaluate the process of using existing data sources, probabilistic linkage, and multiple imputation to create large population-based injury databases matched to outcomes. Methods This was a retrospective cohort study of injured children and adults transported by 94 emergency medical systems (EMS) agencies to 122 hospitals in seven regions of the western United States over a 36-month period (2006 to 2008). All injured patients evaluated by EMS personnel within specific geographic catchment areas were included, regardless of field disposition or outcome. The authors performed probabilistic linkage of EMS records to four hospital and postdischarge data sources (emergency department [ED] data, patient discharge data, trauma registries, and vital statistics files) and then handled missing values using multiple imputation. The authors compare and evaluate matched records, match rates (proportion of matches among eligible patients), and injury outcomes within and across sites. Results There were 381,719 injured patients evaluated by EMS personnel in the seven regions. Among transported patients, match rates ranged from 14.9% to 87.5% and were directly affected by the availability of hospital data sources and proportion of missing values for key linkage variables. For vital statistics records (1-year mortality), estimated match rates ranged from 88.0% to 98.7%. Use of multiple imputation (compared to complete case analysis) reduced bias for injury outcomes, although sample size, percentage missing, type of variable, and combined-site versus single-site imputation models all affected the resulting estimates and variance. Conclusions This project demonstrates the feasibility and describes the process of constructing population-based injury databases across multiple phases of care using existing data sources and commonly available analytic methods. Attention to key linkage variables and decisions for handling missing values can be used to increase match rates between data sources, minimize bias, and preserve sampling design. PMID:22506952

The Ability of Different Imputation Methods to Preserve the Significant Genes and Pathways in Cancer.

PubMed

Aghdam, Rosa; Baghfalaki, Taban; Khosravi, Pegah; Saberi Ansari, Elnaz

2017-12-01

Deciphering important genes and pathways from incomplete gene expression data could facilitate a better understanding of cancer. Different imputation methods can be applied to estimate the missing values. In our study, we evaluated various imputation methods for their performance in preserving significant genes and pathways. In the first step, 5% genes are considered in random for two types of ignorable and non-ignorable missingness mechanisms with various missing rates. Next, 10 well-known imputation methods were applied to the complete datasets. The significance analysis of microarrays (SAM) method was applied to detect the significant genes in rectal and lung cancers to showcase the utility of imputation approaches in preserving significant genes. To determine the impact of different imputation methods on the identification of important genes, the chi-squared test was used to compare the proportions of overlaps between significant genes detected from original data and those detected from the imputed datasets. Additionally, the significant genes are tested for their enrichment in important pathways, using the ConsensusPathDB. Our results showed that almost all the significant genes and pathways of the original dataset can be detected in all imputed datasets, indicating that there is no significant difference in the performance of various imputation methods tested. The source code and selected datasets are available on http://profiles.bs.ipm.ir/softwares/imputation_methods/. Copyright © 2017. Production and hosting by Elsevier B.V.

Multiple Imputation of Completely Missing Repeated Measures Data within Person from a Complex Sample: Application to Accelerometer Data in the National Health and Nutrition Examination Survey

PubMed Central

Liu, Benmei; Yu, Mandi; Graubard, Barry I; Troiano, Richard P; Schenker, Nathaniel

2016-01-01

The Physical Activity Monitor (PAM) component was introduced into the 2003-2004 National Health and Nutrition Examination Survey (NHANES) to collect objective information on physical activity including both movement intensity counts and ambulatory steps. Due to an error in the accelerometer device initialization process, the steps data were missing for all participants in several primary sampling units (PSUs), typically a single county or group of contiguous counties, who had intensity count data from their accelerometers. To avoid potential bias and loss in efficiency in estimation and inference involving the steps data, we considered methods to accurately impute the missing values for steps collected in the 2003-2004 NHANES. The objective was to come up with an efficient imputation method which minimized model-based assumptions. We adopted a multiple imputation approach based on Additive Regression, Bootstrapping and Predictive mean matching (ARBP) methods. This method fits alternative conditional expectation (ace) models, which use an automated procedure to estimate optimal transformations for both the predictor and response variables. This paper describes the approaches used in this imputation and evaluates the methods by comparing the distributions of the original and the imputed data. A simulation study using the observed data is also conducted as part of the model diagnostics. Finally some real data analyses are performed to compare the before and after imputation results. PMID:27488606

Comparing multiple imputation methods for systematically missing subject-level data.

PubMed

Kline, David; Andridge, Rebecca; Kaizar, Eloise

2017-06-01

When conducting research synthesis, the collection of studies that will be combined often do not measure the same set of variables, which creates missing data. When the studies to combine are longitudinal, missing data can occur on the observation-level (time-varying) or the subject-level (non-time-varying). Traditionally, the focus of missing data methods for longitudinal data has been on missing observation-level variables. In this paper, we focus on missing subject-level variables and compare two multiple imputation approaches: a joint modeling approach and a sequential conditional modeling approach. We find the joint modeling approach to be preferable to the sequential conditional approach, except when the covariance structure of the repeated outcome for each individual has homogenous variance and exchangeable correlation. Specifically, the regression coefficient estimates from an analysis incorporating imputed values based on the sequential conditional method are attenuated and less efficient than those from the joint method. Remarkably, the estimates from the sequential conditional method are often less efficient than a complete case analysis, which, in the context of research synthesis, implies that we lose efficiency by combining studies. Copyright © 2015 John Wiley & Sons, Ltd. Copyright © 2015 John Wiley & Sons, Ltd.

Review, evaluation, and discussion of the challenges of missing value imputation for mass spectrometry-based label-free global proteomics

DOE PAGES

Webb-Robertson, Bobbie-Jo M.; Wiberg, Holli K.; Matzke, Melissa M.; ...

2015-04-09

In this review, we apply selected imputation strategies to label-free liquid chromatography–mass spectrometry (LC–MS) proteomics datasets to evaluate the accuracy with respect to metrics of variance and classification. We evaluate several commonly used imputation approaches for individual merits and discuss the caveats of each approach with respect to the example LC–MS proteomics data. In general, local similarity-based approaches, such as the regularized expectation maximization and least-squares adaptive algorithms, yield the best overall performances with respect to metrics of accuracy and robustness. However, no single algorithm consistently outperforms the remaining approaches, and in some cases, performing classification without imputation sometimes yieldedmore » the most accurate classification. Thus, because of the complex mechanisms of missing data in proteomics, which also vary from peptide to protein, no individual method is a single solution for imputation. In summary, on the basis of the observations in this review, the goal for imputation in the field of computational proteomics should be to develop new approaches that work generically for this data type and new strategies to guide users in the selection of the best imputation for their dataset and analysis objectives.« less

Gaussian-based routines to impute categorical variables in health surveys.

PubMed

Yucel, Recai M; He, Yulei; Zaslavsky, Alan M

2011-12-20

The multivariate normal (MVN) distribution is arguably the most popular parametric model used in imputation and is available in most software packages (e.g., SAS PROC MI, R package norm). When it is applied to categorical variables as an approximation, practitioners often either apply simple rounding techniques for ordinal variables or create a distinct 'missing' category and/or disregard the nominal variable from the imputation phase. All of these practices can potentially lead to biased and/or uninterpretable inferences. In this work, we develop a new rounding methodology calibrated to preserve observed distributions to multiply impute missing categorical covariates. The major attractiveness of this method is its flexibility to use any 'working' imputation software, particularly those based on MVN, allowing practitioners to obtain usable imputations with small biases. A simulation study demonstrates the clear advantage of the proposed method in rounding ordinal variables and, in some scenarios, its plausibility in imputing nominal variables. We illustrate our methods on a widely used National Survey of Children with Special Health Care Needs where incomplete values on race posed a valid threat on inferences pertaining to disparities. Copyright © 2011 John Wiley & Sons, Ltd.

Missing Data and Multiple Imputation: An Unbiased Approach

NASA Technical Reports Server (NTRS)

Foy, M.; VanBaalen, M.; Wear, M.; Mendez, C.; Mason, S.; Meyers, V.; Alexander, D.; Law, J.

2014-01-01

The default method of dealing with missing data in statistical analyses is to only use the complete observations (complete case analysis), which can lead to unexpected bias when data do not meet the assumption of missing completely at random (MCAR). For the assumption of MCAR to be met, missingness cannot be related to either the observed or unobserved variables. A less stringent assumption, missing at random (MAR), requires that missingness not be associated with the value of the missing variable itself, but can be associated with the other observed variables. When data are truly MAR as opposed to MCAR, the default complete case analysis method can lead to biased results. There are statistical options available to adjust for data that are MAR, including multiple imputation (MI) which is consistent and efficient at estimating effects. Multiple imputation uses informing variables to determine statistical distributions for each piece of missing data. Then multiple datasets are created by randomly drawing on the distributions for each piece of missing data. Since MI is efficient, only a limited number, usually less than 20, of imputed datasets are required to get stable estimates. Each imputed dataset is analyzed using standard statistical techniques, and then results are combined to get overall estimates of effect. A simulation study will be demonstrated to show the results of using the default complete case analysis, and MI in a linear regression of MCAR and MAR simulated data. Further, MI was successfully applied to the association study of CO2 levels and headaches when initial analysis showed there may be an underlying association between missing CO2 levels and reported headaches. Through MI, we were able to show that there is a strong association between average CO2 levels and the risk of headaches. Each unit increase in CO2 (mmHg) resulted in a doubling in the odds of reported headaches.

Missing data imputation and haplotype phase inference for genome-wide association studies

PubMed Central

Browning, Sharon R.

2009-01-01

Imputation of missing data and the use of haplotype-based association tests can improve the power of genome-wide association studies (GWAS). In this article, I review methods for haplotype inference and missing data imputation, and discuss their application to GWAS. I discuss common features of the best algorithms for haplotype phase inference and missing data imputation in large-scale data sets, as well as some important differences between classes of methods, and highlight the methods that provide the highest accuracy and fastest computational performance. PMID:18850115

The rise of multiple imputation: a review of the reporting and implementation of the method in medical research.

PubMed

Hayati Rezvan, Panteha; Lee, Katherine J; Simpson, Julie A

2015-04-07

Missing data are common in medical research, which can lead to a loss in statistical power and potentially biased results if not handled appropriately. Multiple imputation (MI) is a statistical method, widely adopted in practice, for dealing with missing data. Many academic journals now emphasise the importance of reporting information regarding missing data and proposed guidelines for documenting the application of MI have been published. This review evaluated the reporting of missing data, the application of MI including the details provided regarding the imputation model, and the frequency of sensitivity analyses within the MI framework in medical research articles. A systematic review of articles published in the Lancet and New England Journal of Medicine between January 2008 and December 2013 in which MI was implemented was carried out. We identified 103 papers that used MI, with the number of papers increasing from 11 in 2008 to 26 in 2013. Nearly half of the papers specified the proportion of complete cases or the proportion with missing data by each variable. In the majority of the articles (86%) the imputed variables were specified. Of the 38 papers (37%) that stated the method of imputation, 20 used chained equations, 8 used multivariate normal imputation, and 10 used alternative methods. Very few articles (9%) detailed how they handled non-normally distributed variables during imputation. Thirty-nine papers (38%) stated the variables included in the imputation model. Less than half of the papers (46%) reported the number of imputations, and only two papers compared the distribution of imputed and observed data. Sixty-six papers presented the results from MI as a secondary analysis. Only three articles carried out a sensitivity analysis following MI to assess departures from the missing at random assumption, with details of the sensitivity analyses only provided by one article. This review outlined deficiencies in the documenting of missing data and the details provided about imputation. Furthermore, only a few articles performed sensitivity analyses following MI even though this is strongly recommended in guidelines. Authors are encouraged to follow the available guidelines and provide information on missing data and the imputation process.

Combining multiple imputation and meta-analysis with individual participant data

PubMed Central

Burgess, Stephen; White, Ian R; Resche-Rigon, Matthieu; Wood, Angela M

2013-01-01

Multiple imputation is a strategy for the analysis of incomplete data such that the impact of the missingness on the power and bias of estimates is mitigated. When data from multiple studies are collated, we can propose both within-study and multilevel imputation models to impute missing data on covariates. It is not clear how to choose between imputation models or how to combine imputation and inverse-variance weighted meta-analysis methods. This is especially important as often different studies measure data on different variables, meaning that we may need to impute data on a variable which is systematically missing in a particular study. In this paper, we consider a simulation analysis of sporadically missing data in a single covariate with a linear analysis model and discuss how the results would be applicable to the case of systematically missing data. We find in this context that ensuring the congeniality of the imputation and analysis models is important to give correct standard errors and confidence intervals. For example, if the analysis model allows between-study heterogeneity of a parameter, then we should incorporate this heterogeneity into the imputation model to maintain the congeniality of the two models. In an inverse-variance weighted meta-analysis, we should impute missing data and apply Rubin's rules at the study level prior to meta-analysis, rather than meta-analyzing each of the multiple imputations and then combining the meta-analysis estimates using Rubin's rules. We illustrate the results using data from the Emerging Risk Factors Collaboration. PMID:23703895

When and how should multiple imputation be used for handling missing data in randomised clinical trials - a practical guide with flowcharts.

PubMed

Jakobsen, Janus Christian; Gluud, Christian; Wetterslev, Jørn; Winkel, Per

2017-12-06

Missing data may seriously compromise inferences from randomised clinical trials, especially if missing data are not handled appropriately. The potential bias due to missing data depends on the mechanism causing the data to be missing, and the analytical methods applied to amend the missingness. Therefore, the analysis of trial data with missing values requires careful planning and attention. The authors had several meetings and discussions considering optimal ways of handling missing data to minimise the bias potential. We also searched PubMed (key words: missing data; randomi*; statistical analysis) and reference lists of known studies for papers (theoretical papers; empirical studies; simulation studies; etc.) on how to deal with missing data when analysing randomised clinical trials. Handling missing data is an important, yet difficult and complex task when analysing results of randomised clinical trials. We consider how to optimise the handling of missing data during the planning stage of a randomised clinical trial and recommend analytical approaches which may prevent bias caused by unavoidable missing data. We consider the strengths and limitations of using of best-worst and worst-best sensitivity analyses, multiple imputation, and full information maximum likelihood. We also present practical flowcharts on how to deal with missing data and an overview of the steps that always need to be considered during the analysis stage of a trial. We present a practical guide and flowcharts describing when and how multiple imputation should be used to handle missing data in randomised clinical.

Multiple Imputation based Clustering Validation (MIV) for Big Longitudinal Trial Data with Missing Values in eHealth.

PubMed

Zhang, Zhaoyang; Fang, Hua; Wang, Honggang

2016-06-01

Web-delivered trials are an important component in eHealth services. These trials, mostly behavior-based, generate big heterogeneous data that are longitudinal, high dimensional with missing values. Unsupervised learning methods have been widely applied in this area, however, validating the optimal number of clusters has been challenging. Built upon our multiple imputation (MI) based fuzzy clustering, MIfuzzy, we proposed a new multiple imputation based validation (MIV) framework and corresponding MIV algorithms for clustering big longitudinal eHealth data with missing values, more generally for fuzzy-logic based clustering methods. Specifically, we detect the optimal number of clusters by auto-searching and -synthesizing a suite of MI-based validation methods and indices, including conventional (bootstrap or cross-validation based) and emerging (modularity-based) validation indices for general clustering methods as well as the specific one (Xie and Beni) for fuzzy clustering. The MIV performance was demonstrated on a big longitudinal dataset from a real web-delivered trial and using simulation. The results indicate MI-based Xie and Beni index for fuzzy-clustering are more appropriate for detecting the optimal number of clusters for such complex data. The MIV concept and algorithms could be easily adapted to different types of clustering that could process big incomplete longitudinal trial data in eHealth services.

Multiple Imputation based Clustering Validation (MIV) for Big Longitudinal Trial Data with Missing Values in eHealth

PubMed Central

Zhang, Zhaoyang; Wang, Honggang

2016-01-01

Web-delivered trials are an important component in eHealth services. These trials, mostly behavior-based, generate big heterogeneous data that are longitudinal, high dimensional with missing values. Unsupervised learning methods have been widely applied in this area, however, validating the optimal number of clusters has been challenging. Built upon our multiple imputation (MI) based fuzzy clustering, MIfuzzy, we proposed a new multiple imputation based validation (MIV) framework and corresponding MIV algorithms for clustering big longitudinal eHealth data with missing values, more generally for fuzzy-logic based clustering methods. Specifically, we detect the optimal number of clusters by auto-searching and -synthesizing a suite of MI-based validation methods and indices, including conventional (bootstrap or cross-validation based) and emerging (modularity-based) validation indices for general clustering methods as well as the specific one (Xie and Beni) for fuzzy clustering. The MIV performance was demonstrated on a big longitudinal dataset from a real web-delivered trial and using simulation. The results indicate MI-based Xie and Beni index for fuzzy-clustering is more appropriate for detecting the optimal number of clusters for such complex data. The MIV concept and algorithms could be easily adapted to different types of clustering that could process big incomplete longitudinal trial data in eHealth services. PMID:27126063

Handling Missing Data in Structural Equation Models in R: A Replication Study for Applied Researchers

ERIC Educational Resources Information Center

Wolgast, Anett; Schwinger, Malte; Hahnel, Carolin; Stiensmeier-Pelster, Joachim

2017-01-01

Introduction: Multiple imputation (MI) is one of the most highly recommended methods for replacing missing values in research data. The scope of this paper is to demonstrate missing data handling in SEM by analyzing two modified data examples from educational psychology, and to give practical recommendations for applied researchers. Method: We…

Meta-analysis with missing study-level sample variance data.

PubMed

Chowdhry, Amit K; Dworkin, Robert H; McDermott, Michael P

2016-07-30

We consider a study-level meta-analysis with a normally distributed outcome variable and possibly unequal study-level variances, where the object of inference is the difference in means between a treatment and control group. A common complication in such an analysis is missing sample variances for some studies. A frequently used approach is to impute the weighted (by sample size) mean of the observed variances (mean imputation). Another approach is to include only those studies with variances reported (complete case analysis). Both mean imputation and complete case analysis are only valid under the missing-completely-at-random assumption, and even then the inverse variance weights produced are not necessarily optimal. We propose a multiple imputation method employing gamma meta-regression to impute the missing sample variances. Our method takes advantage of study-level covariates that may be used to provide information about the missing data. Through simulation studies, we show that multiple imputation, when the imputation model is correctly specified, is superior to competing methods in terms of confidence interval coverage probability and type I error probability when testing a specified group difference. Finally, we describe a similar approach to handling missing variances in cross-over studies. Copyright © 2016 John Wiley & Sons, Ltd. Copyright © 2016 John Wiley & Sons, Ltd.

Multiple imputation for IPD meta-analysis: allowing for heterogeneity and studies with missing covariates.

PubMed

Quartagno, M; Carpenter, J R

2016-07-30

Recently, multiple imputation has been proposed as a tool for individual patient data meta-analysis with sporadically missing observations, and it has been suggested that within-study imputation is usually preferable. However, such within study imputation cannot handle variables that are completely missing within studies. Further, if some of the contributing studies are relatively small, it may be appropriate to share information across studies when imputing. In this paper, we develop and evaluate a joint modelling approach to multiple imputation of individual patient data in meta-analysis, with an across-study probability distribution for the study specific covariance matrices. This retains the flexibility to allow for between-study heterogeneity when imputing while allowing (i) sharing information on the covariance matrix across studies when this is appropriate, and (ii) imputing variables that are wholly missing from studies. Simulation results show both equivalent performance to the within-study imputation approach where this is valid, and good results in more general, practically relevant, scenarios with studies of very different sizes, non-negligible between-study heterogeneity and wholly missing variables. We illustrate our approach using data from an individual patient data meta-analysis of hypertension trials. © 2015 The Authors. Statistics in Medicine Published by John Wiley & Sons Ltd. © 2015 The Authors. Statistics in Medicine Published by John Wiley & Sons Ltd.

Limitations in Using Multiple Imputation to Harmonize Individual Participant Data for Meta-Analysis.

PubMed

Siddique, Juned; de Chavez, Peter J; Howe, George; Cruden, Gracelyn; Brown, C Hendricks

2018-02-01

Individual participant data (IPD) meta-analysis is a meta-analysis in which the individual-level data for each study are obtained and used for synthesis. A common challenge in IPD meta-analysis is when variables of interest are measured differently in different studies. The term harmonization has been coined to describe the procedure of placing variables on the same scale in order to permit pooling of data from a large number of studies. Using data from an IPD meta-analysis of 19 adolescent depression trials, we describe a multiple imputation approach for harmonizing 10 depression measures across the 19 trials by treating those depression measures that were not used in a study as missing data. We then apply diagnostics to address the fit of our imputation model. Even after reducing the scale of our application, we were still unable to produce accurate imputations of the missing values. We describe those features of the data that made it difficult to harmonize the depression measures and provide some guidelines for using multiple imputation for harmonization in IPD meta-analysis.

Evaluating Imputation Algorithms for Low-Depth Genotyping-By-Sequencing (GBS) Data

PubMed Central

2016-01-01

Well-powered genomic studies require genome-wide marker coverage across many individuals. For non-model species with few genomic resources, high-throughput sequencing (HTS) methods, such as Genotyping-By-Sequencing (GBS), offer an inexpensive alternative to array-based genotyping. Although affordable, datasets derived from HTS methods suffer from sequencing error, alignment errors, and missing data, all of which introduce noise and uncertainty to variant discovery and genotype calling. Under such circumstances, meaningful analysis of the data is difficult. Our primary interest lies in the issue of how one can accurately infer or impute missing genotypes in HTS-derived datasets. Many of the existing genotype imputation algorithms and software packages were primarily developed by and optimized for the human genetics community, a field where a complete and accurate reference genome has been constructed and SNP arrays have, in large part, been the common genotyping platform. We set out to answer two questions: 1) can we use existing imputation methods developed by the human genetics community to impute missing genotypes in datasets derived from non-human species and 2) are these methods, which were developed and optimized to impute ascertained variants, amenable for imputation of missing genotypes at HTS-derived variants? We selected Beagle v.4, a widely used algorithm within the human genetics community with reportedly high accuracy, to serve as our imputation contender. We performed a series of cross-validation experiments, using GBS data collected from the species Manihot esculenta by the Next Generation (NEXTGEN) Cassava Breeding Project. NEXTGEN currently imputes missing genotypes in their datasets using a LASSO-penalized, linear regression method (denoted ‘glmnet’). We selected glmnet to serve as a benchmark imputation method for this reason. We obtained estimates of imputation accuracy by masking a subset of observed genotypes, imputing, and calculating the sample Pearson correlation between observed and imputed genotype dosages at the site and individual level; computation time served as a second metric for comparison. We then set out to examine factors affecting imputation accuracy, such as levels of missing data, read depth, minor allele frequency (MAF), and reference panel composition. PMID:27537694

Evaluating Imputation Algorithms for Low-Depth Genotyping-By-Sequencing (GBS) Data.

PubMed

Chan, Ariel W; Hamblin, Martha T; Jannink, Jean-Luc

2016-01-01

Well-powered genomic studies require genome-wide marker coverage across many individuals. For non-model species with few genomic resources, high-throughput sequencing (HTS) methods, such as Genotyping-By-Sequencing (GBS), offer an inexpensive alternative to array-based genotyping. Although affordable, datasets derived from HTS methods suffer from sequencing error, alignment errors, and missing data, all of which introduce noise and uncertainty to variant discovery and genotype calling. Under such circumstances, meaningful analysis of the data is difficult. Our primary interest lies in the issue of how one can accurately infer or impute missing genotypes in HTS-derived datasets. Many of the existing genotype imputation algorithms and software packages were primarily developed by and optimized for the human genetics community, a field where a complete and accurate reference genome has been constructed and SNP arrays have, in large part, been the common genotyping platform. We set out to answer two questions: 1) can we use existing imputation methods developed by the human genetics community to impute missing genotypes in datasets derived from non-human species and 2) are these methods, which were developed and optimized to impute ascertained variants, amenable for imputation of missing genotypes at HTS-derived variants? We selected Beagle v.4, a widely used algorithm within the human genetics community with reportedly high accuracy, to serve as our imputation contender. We performed a series of cross-validation experiments, using GBS data collected from the species Manihot esculenta by the Next Generation (NEXTGEN) Cassava Breeding Project. NEXTGEN currently imputes missing genotypes in their datasets using a LASSO-penalized, linear regression method (denoted 'glmnet'). We selected glmnet to serve as a benchmark imputation method for this reason. We obtained estimates of imputation accuracy by masking a subset of observed genotypes, imputing, and calculating the sample Pearson correlation between observed and imputed genotype dosages at the site and individual level; computation time served as a second metric for comparison. We then set out to examine factors affecting imputation accuracy, such as levels of missing data, read depth, minor allele frequency (MAF), and reference panel composition.

Statistical approaches to account for missing values in accelerometer data: Applications to modeling physical activity.

PubMed

Yue Xu, Selene; Nelson, Sandahl; Kerr, Jacqueline; Godbole, Suneeta; Patterson, Ruth; Merchant, Gina; Abramson, Ian; Staudenmayer, John; Natarajan, Loki

2018-04-01

Physical inactivity is a recognized risk factor for many chronic diseases. Accelerometers are increasingly used as an objective means to measure daily physical activity. One challenge in using these devices is missing data due to device nonwear. We used a well-characterized cohort of 333 overweight postmenopausal breast cancer survivors to examine missing data patterns of accelerometer outputs over the day. Based on these observed missingness patterns, we created psuedo-simulated datasets with realistic missing data patterns. We developed statistical methods to design imputation and variance weighting algorithms to account for missing data effects when fitting regression models. Bias and precision of each method were evaluated and compared. Our results indicated that not accounting for missing data in the analysis yielded unstable estimates in the regression analysis. Incorporating variance weights and/or subject-level imputation improved precision by >50%, compared to ignoring missing data. We recommend that these simple easy-to-implement statistical tools be used to improve analysis of accelerometer data.

Meta‐analysis of test accuracy studies using imputation for partial reporting of multiple thresholds

PubMed Central

Deeks, J.J.; Martin, E.C.; Riley, R.D.

2017-01-01

Introduction For tests reporting continuous results, primary studies usually provide test performance at multiple but often different thresholds. This creates missing data when performing a meta‐analysis at each threshold. A standard meta‐analysis (no imputation [NI]) ignores such missing data. A single imputation (SI) approach was recently proposed to recover missing threshold results. Here, we propose a new method that performs multiple imputation of the missing threshold results using discrete combinations (MIDC). Methods The new MIDC method imputes missing threshold results by randomly selecting from the set of all possible discrete combinations which lie between the results for 2 known bounding thresholds. Imputed and observed results are then synthesised at each threshold. This is repeated multiple times, and the multiple pooled results at each threshold are combined using Rubin's rules to give final estimates. We compared the NI, SI, and MIDC approaches via simulation. Results Both imputation methods outperform the NI method in simulations. There was generally little difference in the SI and MIDC methods, but the latter was noticeably better in terms of estimating the between‐study variances and generally gave better coverage, due to slightly larger standard errors of pooled estimates. Given selective reporting of thresholds, the imputation methods also reduced bias in the summary receiver operating characteristic curve. Simulations demonstrate the imputation methods rely on an equal threshold spacing assumption. A real example is presented. Conclusions The SI and, in particular, MIDC methods can be used to examine the impact of missing threshold results in meta‐analysis of test accuracy studies. PMID:29052347

A regressive methodology for estimating missing data in rainfall daily time series

NASA Astrophysics Data System (ADS)

Barca, E.; Passarella, G.

2009-04-01

The "presence" of gaps in environmental data time series represents a very common, but extremely critical problem, since it can produce biased results (Rubin, 1976). Missing data plagues almost all surveys. The problem is how to deal with missing data once it has been deemed impossible to recover the actual missing values. Apart from the amount of missing data, another issue which plays an important role in the choice of any recovery approach is the evaluation of "missingness" mechanisms. When data missing is conditioned by some other variable observed in the data set (Schafer, 1997) the mechanism is called MAR (Missing at Random). Otherwise, when the missingness mechanism depends on the actual value of the missing data, it is called NCAR (Not Missing at Random). This last is the most difficult condition to model. In the last decade interest arose in the estimation of missing data by using regression (single imputation). More recently multiple imputation has become also available, which returns a distribution of estimated values (Scheffer, 2002). In this paper an automatic methodology for estimating missing data is presented. In practice, given a gauging station affected by missing data (target station), the methodology checks the randomness of the missing data and classifies the "similarity" between the target station and the other gauging stations spread over the study area. Among different methods useful for defining the similarity degree, whose effectiveness strongly depends on the data distribution, the Spearman correlation coefficient was chosen. Once defined the similarity matrix, a suitable, nonparametric, univariate, and regressive method was applied in order to estimate missing data in the target station: the Theil method (Theil, 1950). Even though the methodology revealed to be rather reliable an improvement of the missing data estimation can be achieved by a generalization. A first possible improvement consists in extending the univariate technique to the multivariate approach. Another approach follows the paradigm of the "multiple imputation" (Rubin, 1987; Rubin, 1988), which consists in using a set of "similar stations" instead than the most similar. This way, a sort of estimation range can be determined allowing the introduction of uncertainty. Finally, time series can be grouped on the basis of monthly rainfall rates defining classes of wetness (i.e.: dry, moderately rainy and rainy), in order to achieve the estimation using homogeneous data subsets. We expect that integrating the methodology with these enhancements will certainly improve its reliability. The methodology was applied to the daily rainfall time series data registered in the Candelaro River Basin (Apulia - South Italy) from 1970 to 2001. REFERENCES D.B., Rubin, 1976. Inference and Missing Data. Biometrika 63 581-592 D.B. Rubin, 1987. Multiple Imputation for Nonresponce in Surveys, New York: John Wiley & Sons, Inc. D.B. Rubin, 1988. An overview of multiple imputation. In Survey Research Section, pp. 79-84, American Statistical Association, 1988. J.L., Schafer, 1997. Analysis of Incomplete Multivariate Data, Chapman & Hall. J., Scheffer, 2002. Dealing with Missing Data. Res. Lett. Inf. Math. Sci. 3, 153-160. Available online at http://www.massey.ac.nz/~wwiims/research/letters/ H. Theil, 1950. A rank-invariant method of linear and polynomial regression analysis. Indicationes Mathematicae, 12, pp.85-91.

Using full-cohort data in nested case-control and case-cohort studies by multiple imputation.

PubMed

Keogh, Ruth H; White, Ian R

2013-10-15

In many large prospective cohorts, expensive exposure measurements cannot be obtained for all individuals. Exposure-disease association studies are therefore often based on nested case-control or case-cohort studies in which complete information is obtained only for sampled individuals. However, in the full cohort, there may be a large amount of information on cheaply available covariates and possibly a surrogate of the main exposure(s), which typically goes unused. We view the nested case-control or case-cohort study plus the remainder of the cohort as a full-cohort study with missing data. Hence, we propose using multiple imputation (MI) to utilise information in the full cohort when data from the sub-studies are analysed. We use the fully observed data to fit the imputation models. We consider using approximate imputation models and also using rejection sampling to draw imputed values from the true distribution of the missing values given the observed data. Simulation studies show that using MI to utilise full-cohort information in the analysis of nested case-control and case-cohort studies can result in important gains in efficiency, particularly when a surrogate of the main exposure is available in the full cohort. In simulations, this method outperforms counter-matching in nested case-control studies and a weighted analysis for case-cohort studies, both of which use some full-cohort information. Approximate imputation models perform well except when there are interactions or non-linear terms in the outcome model, where imputation using rejection sampling works well. Copyright © 2013 John Wiley & Sons, Ltd.

Imputing data that are missing at high rates using a boosting algorithm

DOE Office of Scientific and Technical Information (OSTI.GOV)

Cauthen, Katherine Regina; Lambert, Gregory; Ray, Jaideep

Traditional multiple imputation approaches may perform poorly for datasets with high rates of missingness unless many m imputations are used. This paper implements an alternative machine learning-based approach to imputing data that are missing at high rates. Here, we use boosting to create a strong learner from a weak learner fitted to a dataset missing many observations. This approach may be applied to a variety of types of learners (models). The approach is demonstrated by application to a spatiotemporal dataset for predicting dengue outbreaks in India from meteorological covariates. A Bayesian spatiotemporal CAR model is boosted to produce imputations, andmore » the overall RMSE from a k-fold cross-validation is used to assess imputation accuracy.« less

Impact of Missing Data on the Detection of Differential Item Functioning: The Case of Mantel-Haenszel and Logistic Regression Analysis

ERIC Educational Resources Information Center

Robitzsch, Alexander; Rupp, Andre A.

2009-01-01

This article describes the results of a simulation study to investigate the impact of missing data on the detection of differential item functioning (DIF). Specifically, it investigates how four methods for dealing with missing data (listwise deletion, zero imputation, two-way imputation, response function imputation) interact with two methods of…

Relying on Your Own Best Judgment: Imputing Values to Missing Information in Decision Making.

ERIC Educational Resources Information Center

Johnson, Richard D.; And Others

Processes involved in making estimates of the value of missing information that could help in a decision making process were studied. Hypothetical purchases of ground beef were selected for the study as such purchases have the desirable property of quantifying both the price and quality. A total of 150 students at the University of Iowa rated the…

Genotyping by sequencing for genomic prediction in a soybean breeding population.

PubMed

Jarquín, Diego; Kocak, Kyle; Posadas, Luis; Hyma, Katie; Jedlicka, Joseph; Graef, George; Lorenz, Aaron

2014-08-29

Advances in genotyping technology, such as genotyping by sequencing (GBS), are making genomic prediction more attractive to reduce breeding cycle times and costs associated with phenotyping. Genomic prediction and selection has been studied in several crop species, but no reports exist in soybean. The objectives of this study were (i) evaluate prospects for genomic selection using GBS in a typical soybean breeding program and (ii) evaluate the effect of GBS marker selection and imputation on genomic prediction accuracy. To achieve these objectives, a set of soybean lines sampled from the University of Nebraska Soybean Breeding Program were genotyped using GBS and evaluated for yield and other agronomic traits at multiple Nebraska locations. Genotyping by sequencing scored 16,502 single nucleotide polymorphisms (SNPs) with minor-allele frequency (MAF) > 0.05 and percentage of missing values ≤ 5% on 301 elite soybean breeding lines. When SNPs with up to 80% missing values were included, 52,349 SNPs were scored. Prediction accuracy for grain yield, assessed using cross validation, was estimated to be 0.64, indicating good potential for using genomic selection for grain yield in soybean. Filtering SNPs based on missing data percentage had little to no effect on prediction accuracy, especially when random forest imputation was used to impute missing values. The highest accuracies were observed when random forest imputation was used on all SNPs, but differences were not significant. A standard additive G-BLUP model was robust; modeling additive-by-additive epistasis did not provide any improvement in prediction accuracy. The effect of training population size on accuracy began to plateau around 100, but accuracy steadily climbed until the largest possible size was used in this analysis. Including only SNPs with MAF > 0.30 provided higher accuracies when training populations were smaller. Using GBS for genomic prediction in soybean holds good potential to expedite genetic gain. Our results suggest that standard additive G-BLUP models can be used on unfiltered, imputed GBS data without loss in accuracy.

Missing in space: an evaluation of imputation methods for missing data in spatial analysis of risk factors for type II diabetes.

PubMed

Baker, Jannah; White, Nicole; Mengersen, Kerrie

2014-11-20

Spatial analysis is increasingly important for identifying modifiable geographic risk factors for disease. However, spatial health data from surveys are often incomplete, ranging from missing data for only a few variables, to missing data for many variables. For spatial analyses of health outcomes, selection of an appropriate imputation method is critical in order to produce the most accurate inferences. We present a cross-validation approach to select between three imputation methods for health survey data with correlated lifestyle covariates, using as a case study, type II diabetes mellitus (DM II) risk across 71 Queensland Local Government Areas (LGAs). We compare the accuracy of mean imputation to imputation using multivariate normal and conditional autoregressive prior distributions. Choice of imputation method depends upon the application and is not necessarily the most complex method. Mean imputation was selected as the most accurate method in this application. Selecting an appropriate imputation method for health survey data, after accounting for spatial correlation and correlation between covariates, allows more complete analysis of geographic risk factors for disease with more confidence in the results to inform public policy decision-making.

Comparison of Random Forest and Parametric Imputation Models for Imputing Missing Data Using MICE: A CALIBER Study

PubMed Central

Shah, Anoop D.; Bartlett, Jonathan W.; Carpenter, James; Nicholas, Owen; Hemingway, Harry

2014-01-01

Multivariate imputation by chained equations (MICE) is commonly used for imputing missing data in epidemiologic research. The “true” imputation model may contain nonlinearities which are not included in default imputation models. Random forest imputation is a machine learning technique which can accommodate nonlinearities and interactions and does not require a particular regression model to be specified. We compared parametric MICE with a random forest-based MICE algorithm in 2 simulation studies. The first study used 1,000 random samples of 2,000 persons drawn from the 10,128 stable angina patients in the CALIBER database (Cardiovascular Disease Research using Linked Bespoke Studies and Electronic Records; 2001–2010) with complete data on all covariates. Variables were artificially made “missing at random,” and the bias and efficiency of parameter estimates obtained using different imputation methods were compared. Both MICE methods produced unbiased estimates of (log) hazard ratios, but random forest was more efficient and produced narrower confidence intervals. The second study used simulated data in which the partially observed variable depended on the fully observed variables in a nonlinear way. Parameter estimates were less biased using random forest MICE, and confidence interval coverage was better. This suggests that random forest imputation may be useful for imputing complex epidemiologic data sets in which some patients have missing data. PMID:24589914

Comparison of random forest and parametric imputation models for imputing missing data using MICE: a CALIBER study.

PubMed

Shah, Anoop D; Bartlett, Jonathan W; Carpenter, James; Nicholas, Owen; Hemingway, Harry

2014-03-15

Multivariate imputation by chained equations (MICE) is commonly used for imputing missing data in epidemiologic research. The "true" imputation model may contain nonlinearities which are not included in default imputation models. Random forest imputation is a machine learning technique which can accommodate nonlinearities and interactions and does not require a particular regression model to be specified. We compared parametric MICE with a random forest-based MICE algorithm in 2 simulation studies. The first study used 1,000 random samples of 2,000 persons drawn from the 10,128 stable angina patients in the CALIBER database (Cardiovascular Disease Research using Linked Bespoke Studies and Electronic Records; 2001-2010) with complete data on all covariates. Variables were artificially made "missing at random," and the bias and efficiency of parameter estimates obtained using different imputation methods were compared. Both MICE methods produced unbiased estimates of (log) hazard ratios, but random forest was more efficient and produced narrower confidence intervals. The second study used simulated data in which the partially observed variable depended on the fully observed variables in a nonlinear way. Parameter estimates were less biased using random forest MICE, and confidence interval coverage was better. This suggests that random forest imputation may be useful for imputing complex epidemiologic data sets in which some patients have missing data.

Using Bayesian Imputation to Assess Racial and Ethnic Disparities in Pediatric Performance Measures.

PubMed

Brown, David P; Knapp, Caprice; Baker, Kimberly; Kaufmann, Meggen

2016-06-01

To analyze health care disparities in pediatric quality of care measures and determine the impact of data imputation. Five HEDIS measures are calculated based on 2012 administrative data for 145,652 children in two public insurance programs in Florida. The Bayesian Improved Surname and Geocoding (BISG) imputation method is used to impute missing race and ethnicity data for 42 percent of the sample (61,954 children). Models are estimated with and without the imputed race and ethnicity data. Dropping individuals with missing race and ethnicity data biases quality of care measures for minorities downward relative to nonminority children for several measures. These results provide further support for the importance of appropriately accounting for missing race and ethnicity data through imputation methods. © Health Research and Educational Trust.

Investigating the Effects of Imputation Methods for Modelling Gene Networks Using a Dynamic Bayesian Network from Gene Expression Data

PubMed Central

CHAI, Lian En; LAW, Chow Kuan; MOHAMAD, Mohd Saberi; CHONG, Chuii Khim; CHOON, Yee Wen; DERIS, Safaai; ILLIAS, Rosli Md

2014-01-01

Background: Gene expression data often contain missing expression values. Therefore, several imputation methods have been applied to solve the missing values, which include k-nearest neighbour (kNN), local least squares (LLS), and Bayesian principal component analysis (BPCA). However, the effects of these imputation methods on the modelling of gene regulatory networks from gene expression data have rarely been investigated and analysed using a dynamic Bayesian network (DBN). Methods: In the present study, we separately imputed datasets of the Escherichia coli S.O.S. DNA repair pathway and the Saccharomyces cerevisiae cell cycle pathway with kNN, LLS, and BPCA, and subsequently used these to generate gene regulatory networks (GRNs) using a discrete DBN. We made comparisons on the basis of previous studies in order to select the gene network with the least error. Results: We found that BPCA and LLS performed better on larger networks (based on the S. cerevisiae dataset), whereas kNN performed better on smaller networks (based on the E. coli dataset). Conclusion: The results suggest that the performance of each imputation method is dependent on the size of the dataset, and this subsequently affects the modelling of the resultant GRNs using a DBN. In addition, on the basis of these results, a DBN has the capacity to discover potential edges, as well as display interactions, between genes. PMID:24876803

Autoregressive-model-based missing value estimation for DNA microarray time series data.

PubMed

Choong, Miew Keen; Charbit, Maurice; Yan, Hong

2009-01-01

Missing value estimation is important in DNA microarray data analysis. A number of algorithms have been developed to solve this problem, but they have several limitations. Most existing algorithms are not able to deal with the situation where a particular time point (column) of the data is missing entirely. In this paper, we present an autoregressive-model-based missing value estimation method (ARLSimpute) that takes into account the dynamic property of microarray temporal data and the local similarity structures in the data. ARLSimpute is especially effective for the situation where a particular time point contains many missing values or where the entire time point is missing. Experiment results suggest that our proposed algorithm is an accurate missing value estimator in comparison with other imputation methods on simulated as well as real microarray time series datasets.

Making the most of missing values : object clustering with partial data in astronomy

NASA Technical Reports Server (NTRS)

Wagstaff, Kiri L.; Laidler, Victoria G.

2004-01-01

We demonstrate a clustering analysis algorithm, KSC, that a) uses all observed values and b) does not discard the partially observed objects. KSC uses soft constraints defined by the fully observed objects to assist in the grouping of objects with missing values. We present an analysis of objects taken from the Sloan Digital Sky Survey to demonstrate how imputing the values can be misleading and why the KSC approach can produce more appropriate results.

Multiple imputation of missing data in nested case-control and case-cohort studies.

PubMed

Keogh, Ruth H; Seaman, Shaun R; Bartlett, Jonathan W; Wood, Angela M

2018-06-05

The nested case-control and case-cohort designs are two main approaches for carrying out a substudy within a prospective cohort. This article adapts multiple imputation (MI) methods for handling missing covariates in full-cohort studies for nested case-control and case-cohort studies. We consider data missing by design and data missing by chance. MI analyses that make use of full-cohort data and MI analyses based on substudy data only are described, alongside an intermediate approach in which the imputation uses full-cohort data but the analysis uses only the substudy. We describe adaptations to two imputation methods: the approximate method (MI-approx) of White and Royston () and the "substantive model compatible" (MI-SMC) method of Bartlett et al. (). We also apply the "MI matched set" approach of Seaman and Keogh () to nested case-control studies, which does not require any full-cohort information. The methods are investigated using simulation studies and all perform well when their assumptions hold. Substantial gains in efficiency can be made by imputing data missing by design using the full-cohort approach or by imputing data missing by chance in analyses using the substudy only. The intermediate approach brings greater gains in efficiency relative to the substudy approach and is more robust to imputation model misspecification than the full-cohort approach. The methods are illustrated using the ARIC Study cohort. Supplementary Materials provide R and Stata code. © 2018, The International Biometric Society.

Assessment of predictive performance in incomplete data by combining internal validation and multiple imputation.

PubMed

Wahl, Simone; Boulesteix, Anne-Laure; Zierer, Astrid; Thorand, Barbara; van de Wiel, Mark A

2016-10-26

Missing values are a frequent issue in human studies. In many situations, multiple imputation (MI) is an appropriate missing data handling strategy, whereby missing values are imputed multiple times, the analysis is performed in every imputed data set, and the obtained estimates are pooled. If the aim is to estimate (added) predictive performance measures, such as (change in) the area under the receiver-operating characteristic curve (AUC), internal validation strategies become desirable in order to correct for optimism. It is not fully understood how internal validation should be combined with multiple imputation. In a comprehensive simulation study and in a real data set based on blood markers as predictors for mortality, we compare three combination strategies: Val-MI, internal validation followed by MI on the training and test parts separately, MI-Val, MI on the full data set followed by internal validation, and MI(-y)-Val, MI on the full data set omitting the outcome followed by internal validation. Different validation strategies, including bootstrap und cross-validation, different (added) performance measures, and various data characteristics are considered, and the strategies are evaluated with regard to bias and mean squared error of the obtained performance estimates. In addition, we elaborate on the number of resamples and imputations to be used, and adopt a strategy for confidence interval construction to incomplete data. Internal validation is essential in order to avoid optimism, with the bootstrap 0.632+ estimate representing a reliable method to correct for optimism. While estimates obtained by MI-Val are optimistically biased, those obtained by MI(-y)-Val tend to be pessimistic in the presence of a true underlying effect. Val-MI provides largely unbiased estimates, with a slight pessimistic bias with increasing true effect size, number of covariates and decreasing sample size. In Val-MI, accuracy of the estimate is more strongly improved by increasing the number of bootstrap draws rather than the number of imputations. With a simple integrated approach, valid confidence intervals for performance estimates can be obtained. When prognostic models are developed on incomplete data, Val-MI represents a valid strategy to obtain estimates of predictive performance measures.

Multiple imputation methods for bivariate outcomes in cluster randomised trials.

PubMed

DiazOrdaz, K; Kenward, M G; Gomes, M; Grieve, R

2016-09-10

Missing observations are common in cluster randomised trials. The problem is exacerbated when modelling bivariate outcomes jointly, as the proportion of complete cases is often considerably smaller than the proportion having either of the outcomes fully observed. Approaches taken to handling such missing data include the following: complete case analysis, single-level multiple imputation that ignores the clustering, multiple imputation with a fixed effect for each cluster and multilevel multiple imputation. We contrasted the alternative approaches to handling missing data in a cost-effectiveness analysis that uses data from a cluster randomised trial to evaluate an exercise intervention for care home residents. We then conducted a simulation study to assess the performance of these approaches on bivariate continuous outcomes, in terms of confidence interval coverage and empirical bias in the estimated treatment effects. Missing-at-random clustered data scenarios were simulated following a full-factorial design. Across all the missing data mechanisms considered, the multiple imputation methods provided estimators with negligible bias, while complete case analysis resulted in biased treatment effect estimates in scenarios where the randomised treatment arm was associated with missingness. Confidence interval coverage was generally in excess of nominal levels (up to 99.8%) following fixed-effects multiple imputation and too low following single-level multiple imputation. Multilevel multiple imputation led to coverage levels of approximately 95% throughout. © 2016 The Authors. Statistics in Medicine Published by John Wiley & Sons Ltd. © 2016 The Authors. Statistics in Medicine Published by John Wiley & Sons Ltd.

Evaluation of missing value methods for predicting ambient BTEX concentrations in two neighbouring cities in Southwestern Ontario Canada

NASA Astrophysics Data System (ADS)

Miller, Lindsay; Xu, Xiaohong; Wheeler, Amanda; Zhang, Tianchu; Hamadani, Mariam; Ejaz, Unam

2018-05-01

High density air monitoring campaigns provide spatial patterns of pollutant concentrations which are integral in exposure assessment. Such analysis can assist with the determination of links between air quality and health outcomes, however, problems due to missing data can threaten to compromise these studies. This research evaluates four methods; mean value imputation, inverse distance weighting (IDW), inter-species ratios, and regression, to address missing spatial concentration data ranging from one missing data point up to 50% missing data. BTEX (benzene, toluene, ethylbenzene, and xylenes) concentrations were measured in Windsor and Sarnia, Ontario in the fall of 2005. Concentrations and inter-species ratios were generally similar between the two cities. Benzene (B) was observed to be higher in Sarnia, whereas toluene (T) and the T/B ratios were higher in Windsor. Using these urban, industrialized cities as case studies, this research demonstrates that using inter-species ratios or regression of the data for which there is complete information, along with one measured concentration (i.e. benzene) to predict for missing concentrations (i.e. TEX) results in good agreement between predicted and measured values. In both cities, the general trend remains that best agreement is observed for the leave-one-out scenario, followed by 10% and 25% missing, and the least agreement for the 50% missing cases. In the absence of any known concentrations IDW can provide reasonable agreement between observed and estimated concentrations for the BTEX species, and was superior over mean value imputation which was not able to preserve the spatial trend. The proposed methods can be used to fill in missing data, while preserving the general characteristics and rank order of the data which are sufficient for epidemiologic studies.

Multiple Imputation of Cognitive Performance as a Repeatedly Measured Outcome

PubMed Central

Rawlings, Andreea M.; Sang, Yingying; Sharrett, A. Richey; Coresh, Josef; Griswold, Michael; Kucharska-Newton, Anna M.; Palta, Priya; Wruck, Lisa M.; Gross, Alden L.; Deal, Jennifer A.; Power, Melinda C.; Bandeen-Roche, Karen

2016-01-01

Background Longitudinal studies of cognitive performance are sensitive to dropout, as participants experiencing cognitive deficits are less likely to attend study visits, which may bias estimated associations between exposures of interest and cognitive decline. Multiple imputation is a powerful tool for handling missing data, however its use for missing cognitive outcome measures in longitudinal analyses remains limited. Methods We use multiple imputation by chained equations (MICE) to impute cognitive performance scores of participants who did not attend the 2011-2013 exam of the Atherosclerosis Risk in Communities Study. We examined the validity of imputed scores using observed and simulated data under varying assumptions. We examined differences in the estimated association between diabetes at baseline and 20-year cognitive decline with and without imputed values. Lastly, we discuss how different analytic methods (mixed models and models fit using generalized estimate equations) and choice of for whom to impute result in different estimands. Results Validation using observed data showed MICE produced unbiased imputations. Simulations showed a substantial reduction in the bias of the 20-year association between diabetes and cognitive decline comparing MICE (3-4% bias) to analyses of available data only (16-23% bias) in a construct where missingness was strongly informative but realistic. Associations between diabetes and 20-year cognitive decline were substantially stronger with MICE than in available-case analyses. Conclusions Our study suggests when informative data are available for non-examined participants, MICE can be an effective tool for imputing cognitive performance and improving assessment of cognitive decline, though careful thought should be given to target imputation population and analytic model chosen, as they may yield different estimands. PMID:27619926

Multiple Imputation For Combined-Survey Estimation With Incomplete Regressors In One But Not Both Surveys

PubMed Central

Rendall, Michael S.; Ghosh-Dastidar, Bonnie; Weden, Margaret M.; Baker, Elizabeth H.; Nazarov, Zafar

2013-01-01

Within-survey multiple imputation (MI) methods are adapted to pooled-survey regression estimation where one survey has more regressors, but typically fewer observations, than the other. This adaptation is achieved through: (1) larger numbers of imputations to compensate for the higher fraction of missing values; (2) model-fit statistics to check the assumption that the two surveys sample from a common universe; and (3) specificying the analysis model completely from variables present in the survey with the larger set of regressors, thereby excluding variables never jointly observed. In contrast to the typical within-survey MI context, cross-survey missingness is monotonic and easily satisfies the Missing At Random (MAR) assumption needed for unbiased MI. Large efficiency gains and substantial reduction in omitted variable bias are demonstrated in an application to sociodemographic differences in the risk of child obesity estimated from two nationally-representative cohort surveys. PMID:24223447

Imputation of Missing Genotypes From Sparse to High Density Using Long-Range Phasing

USDA-ARS?s Scientific Manuscript database

Related individuals in a population share long chromosome segments which trace to a common ancestor. We describe a long-range phasing algorithm that makes use of this property to phase whole chromosomes and simultaneously impute a large number of missing markers. We test our method by imputing marke...

A Comparison of Imputation Methods for Bayesian Factor Analysis Models

ERIC Educational Resources Information Center

Merkle, Edgar C.

2011-01-01

Imputation methods are popular for the handling of missing data in psychology. The methods generally consist of predicting missing data based on observed data, yielding a complete data set that is amiable to standard statistical analyses. In the context of Bayesian factor analysis, this article compares imputation under an unrestricted…

[Imputing missing data in public health: general concepts and application to dichotomous variables].

PubMed

Hernández, Gilma; Moriña, David; Navarro, Albert

The presence of missing data in collected variables is common in health surveys, but the subsequent imputation thereof at the time of analysis is not. Working with imputed data may have certain benefits regarding the precision of the estimators and the unbiased identification of associations between variables. The imputation process is probably still little understood by many non-statisticians, who view this process as highly complex and with an uncertain goal. To clarify these questions, this note aims to provide a straightforward, non-exhaustive overview of the imputation process to enable public health researchers ascertain its strengths. All this in the context of dichotomous variables which are commonplace in public health. To illustrate these concepts, an example in which missing data is handled by means of simple and multiple imputation is introduced. Copyright © 2017 SESPAS. Publicado por Elsevier España, S.L.U. All rights reserved.

Lazy collaborative filtering for data sets with missing values.

PubMed

Ren, Yongli; Li, Gang; Zhang, Jun; Zhou, Wanlei

2013-12-01

As one of the biggest challenges in research on recommender systems, the data sparsity issue is mainly caused by the fact that users tend to rate a small proportion of items from the huge number of available items. This issue becomes even more problematic for the neighborhood-based collaborative filtering (CF) methods, as there are even lower numbers of ratings available in the neighborhood of the query item. In this paper, we aim to address the data sparsity issue in the context of neighborhood-based CF. For a given query (user, item), a set of key ratings is first identified by taking the historical information of both the user and the item into account. Then, an auto-adaptive imputation (AutAI) method is proposed to impute the missing values in the set of key ratings. We present a theoretical analysis to show that the proposed imputation method effectively improves the performance of the conventional neighborhood-based CF methods. The experimental results show that our new method of CF with AutAI outperforms six existing recommendation methods in terms of accuracy.

Handling missing rows in multi-omics data integration: multiple imputation in multiple factor analysis framework.

PubMed

Voillet, Valentin; Besse, Philippe; Liaubet, Laurence; San Cristobal, Magali; González, Ignacio

2016-10-03

In omics data integration studies, it is common, for a variety of reasons, for some individuals to not be present in all data tables. Missing row values are challenging to deal with because most statistical methods cannot be directly applied to incomplete datasets. To overcome this issue, we propose a multiple imputation (MI) approach in a multivariate framework. In this study, we focus on multiple factor analysis (MFA) as a tool to compare and integrate multiple layers of information. MI involves filling the missing rows with plausible values, resulting in M completed datasets. MFA is then applied to each completed dataset to produce M different configurations (the matrices of coordinates of individuals). Finally, the M configurations are combined to yield a single consensus solution. We assessed the performance of our method, named MI-MFA, on two real omics datasets. Incomplete artificial datasets with different patterns of missingness were created from these data. The MI-MFA results were compared with two other approaches i.e., regularized iterative MFA (RI-MFA) and mean variable imputation (MVI-MFA). For each configuration resulting from these three strategies, the suitability of the solution was determined against the true MFA configuration obtained from the original data and a comprehensive graphical comparison showing how the MI-, RI- or MVI-MFA configurations diverge from the true configuration was produced. Two approaches i.e., confidence ellipses and convex hulls, to visualize and assess the uncertainty due to missing values were also described. We showed how the areas of ellipses and convex hulls increased with the number of missing individuals. A free and easy-to-use code was proposed to implement the MI-MFA method in the R statistical environment. We believe that MI-MFA provides a useful and attractive method for estimating the coordinates of individuals on the first MFA components despite missing rows. MI-MFA configurations were close to the true configuration even when many individuals were missing in several data tables. This method takes into account the uncertainty of MI-MFA configurations induced by the missing rows, thereby allowing the reliability of the results to be evaluated.

iVAR: a program for imputing missing data in multivariate time series using vector autoregressive models.

PubMed

Liu, Siwei; Molenaar, Peter C M

2014-12-01

This article introduces iVAR, an R program for imputing missing data in multivariate time series on the basis of vector autoregressive (VAR) models. We conducted a simulation study to compare iVAR with three methods for handling missing data: listwise deletion, imputation with sample means and variances, and multiple imputation ignoring time dependency. The results showed that iVAR produces better estimates for the cross-lagged coefficients than do the other three methods. We demonstrate the use of iVAR with an empirical example of time series electrodermal activity data and discuss the advantages and limitations of the program.

Investigation of Missing Responses in Implementation of Cognitive Diagnostic Models

ERIC Educational Resources Information Center

Dai, Shenghai

2017-01-01

This dissertation is aimed at investigating the impact of missing data and evaluating the performance of five selected methods for handling missing responses in the implementation of Cognitive Diagnostic Models (CDMs). The five methods are: a) treating missing data as incorrect (IN), b) person mean imputation (PM), c) two-way imputation (TW), d)…

Impact of pre-imputation SNP-filtering on genotype imputation results

PubMed Central

2014-01-01

Background Imputation of partially missing or unobserved genotypes is an indispensable tool for SNP data analyses. However, research and understanding of the impact of initial SNP-data quality control on imputation results is still limited. In this paper, we aim to evaluate the effect of different strategies of pre-imputation quality filtering on the performance of the widely used imputation algorithms MaCH and IMPUTE. Results We considered three scenarios: imputation of partially missing genotypes with usage of an external reference panel, without usage of an external reference panel, as well as imputation of completely un-typed SNPs using an external reference panel. We first created various datasets applying different SNP quality filters and masking certain percentages of randomly selected high-quality SNPs. We imputed these SNPs and compared the results between the different filtering scenarios by using established and newly proposed measures of imputation quality. While the established measures assess certainty of imputation results, our newly proposed measures focus on the agreement with true genotypes. These measures showed that pre-imputation SNP-filtering might be detrimental regarding imputation quality. Moreover, the strongest drivers of imputation quality were in general the burden of missingness and the number of SNPs used for imputation. We also found that using a reference panel always improves imputation quality of partially missing genotypes. MaCH performed slightly better than IMPUTE2 in most of our scenarios. Again, these results were more pronounced when using our newly defined measures of imputation quality. Conclusion Even a moderate filtering has a detrimental effect on the imputation quality. Therefore little or no SNP filtering prior to imputation appears to be the best strategy for imputing small to moderately sized datasets. Our results also showed that for these datasets, MaCH performs slightly better than IMPUTE2 in most scenarios at the cost of increased computing time. PMID:25112433

[Study on correction of data bias caused by different missing mechanisms in survey of medical expenditure among students enrolling in Urban Resident Basic Medical Insurance].

PubMed

Zhang, Haixia; Zhao, Junkang; Gu, Caijiao; Cui, Yan; Rong, Huiying; Meng, Fanlong; Wang, Tong

2015-05-01

The study of the medical expenditure and its influencing factors among the students enrolling in Urban Resident Basic Medical Insurance (URBMI) in Taiyuan indicated that non response bias and selection bias coexist in dependent variable of the survey data. Unlike previous studies only focused on one missing mechanism, a two-stage method to deal with two missing mechanisms simultaneously was suggested in this study, combining multiple imputation with sample selection model. A total of 1 190 questionnaires were returned by the students (or their parents) selected in child care settings, schools and universities in Taiyuan by stratified cluster random sampling in 2012. In the returned questionnaires, 2.52% existed not missing at random (NMAR) of dependent variable and 7.14% existed missing at random (MAR) of dependent variable. First, multiple imputation was conducted for MAR by using completed data, then sample selection model was used to correct NMAR in multiple imputation, and a multi influencing factor analysis model was established. Based on 1 000 times resampling, the best scheme of filling the random missing values is the predictive mean matching (PMM) method under the missing proportion. With this optimal scheme, a two stage survey was conducted. Finally, it was found that the influencing factors on annual medical expenditure among the students enrolling in URBMI in Taiyuan included population group, annual household gross income, affordability of medical insurance expenditure, chronic disease, seeking medical care in hospital, seeking medical care in community health center or private clinic, hospitalization, hospitalization canceled due to certain reason, self medication and acceptable proportion of self-paid medical expenditure. The two-stage method combining multiple imputation with sample selection model can deal with non response bias and selection bias effectively in dependent variable of the survey data.

A method to estimate the additional uncertainty in gap-filled NEE resulting from long gaps in the CO2 flux record

Treesearch

Andrew D. Richardson; David Y. Hollinger

2007-01-01

Missing values in any data set create problems for researchers. The process by which missing values are replaced, and the data set is made complete, is generally referred to as imputation. Within the eddy flux community, the term "gap filling" is more commonly applied. A major challenge is that random errors in measured data result in uncertainty in the gap-...

Use of Multiple Imputation to Estimate the Proportion of Respiratory Virus Detections Among Patients Hospitalized With Community-Acquired Pneumonia.

PubMed

Bozio, Catherine H; Flanders, W Dana; Finelli, Lyn; Bramley, Anna M; Reed, Carrie; Gandhi, Neel R; Vidal, Jorge E; Erdman, Dean; Levine, Min Z; Lindstrom, Stephen; Ampofo, Krow; Arnold, Sandra R; Self, Wesley H; Williams, Derek J; Grijalva, Carlos G; Anderson, Evan J; McCullers, Jonathan A; Edwards, Kathryn M; Pavia, Andrew T; Wunderink, Richard G; Jain, Seema

2018-04-01

Real-time polymerase chain reaction (PCR) on respiratory specimens and serology on paired blood specimens are used to determine the etiology of respiratory illnesses for research studies. However, convalescent serology is often not collected. We used multiple imputation to assign values for missing serology results to estimate virus-specific prevalence among pediatric and adult community-acquired pneumonia hospitalizations using data from an active population-based surveillance study. Presence of adenoviruses, human metapneumovirus, influenza viruses, parainfluenza virus types 1-3, and respiratory syncytial virus was defined by positive PCR on nasopharyngeal/oropharyngeal specimens or a 4-fold rise in paired serology. We performed multiple imputation by developing a multivariable regression model for each virus using data from patients with available serology results. We calculated absolute and relative differences in the proportion of each virus detected comparing the imputed to observed (nonimputed) results. Among 2222 children and 2259 adults, 98.8% and 99.5% had nasopharyngeal/oropharyngeal specimens and 43.2% and 37.5% had paired serum specimens, respectively. Imputed results increased viral etiology assignments by an absolute difference of 1.6%-4.4% and 0.8%-2.8% in children and adults, respectively; relative differences were 1.1-3.0 times higher. Multiple imputation can be used when serology results are missing, to refine virus-specific prevalence estimates, and these will likely increase estimates.

References for Haplotype Imputation in the Big Data Era

PubMed Central

Li, Wenzhi; Xu, Wei; Li, Qiling; Ma, Li; Song, Qing

2016-01-01

Imputation is a powerful in silico approach to fill in those missing values in the big datasets. This process requires a reference panel, which is a collection of big data from which the missing information can be extracted and imputed. Haplotype imputation requires ethnicity-matched references; a mismatched reference panel will significantly reduce the quality of imputation. However, currently existing big datasets cover only a small number of ethnicities, there is a lack of ethnicity-matched references for many ethnic populations in the world, which has hampered the data imputation of haplotypes and its downstream applications. To solve this issue, several approaches have been proposed and explored, including the mixed reference panel, the internal reference panel and genotype-converted reference panel. This review article provides the information and comparison between these approaches. Increasing evidence showed that not just one or two genetic elements dictate the gene activity and functions; instead, cis-interactions of multiple elements dictate gene activity. Cis-interactions require the interacting elements to be on the same chromosome molecule, therefore, haplotype analysis is essential for the investigation of cis-interactions among multiple genetic variants at different loci, and appears to be especially important for studying the common diseases. It will be valuable in a wide spectrum of applications from academic research, to clinical diagnosis, prevention, treatment, and pharmaceutical industry. PMID:27274952

1000 Genomes-based meta-analysis identifies 10 novel loci for kidney function

PubMed Central

Gorski, Mathias; van der Most, Peter J.; Teumer, Alexander; Chu, Audrey Y.; Li, Man; Mijatovic, Vladan; Nolte, Ilja M.; Cocca, Massimiliano; Taliun, Daniel; Gomez, Felicia; Li, Yong; Tayo, Bamidele; Tin, Adrienne; Feitosa, Mary F.; Aspelund, Thor; Attia, John; Biffar, Reiner; Bochud, Murielle; Boerwinkle, Eric; Borecki, Ingrid; Bottinger, Erwin P.; Chen, Ming-Huei; Chouraki, Vincent; Ciullo, Marina; Coresh, Josef; Cornelis, Marilyn C.; Curhan, Gary C.; d’Adamo, Adamo Pio; Dehghan, Abbas; Dengler, Laura; Ding, Jingzhong; Eiriksdottir, Gudny; Endlich, Karlhans; Enroth, Stefan; Esko, Tõnu; Franco, Oscar H.; Gasparini, Paolo; Gieger, Christian; Girotto, Giorgia; Gottesman, Omri; Gudnason, Vilmundur; Gyllensten, Ulf; Hancock, Stephen J.; Harris, Tamara B.; Helmer, Catherine; Höllerer, Simon; Hofer, Edith; Hofman, Albert; Holliday, Elizabeth G.; Homuth, Georg; Hu, Frank B.; Huth, Cornelia; Hutri-Kähönen, Nina; Hwang, Shih-Jen; Imboden, Medea; Johansson, Åsa; Kähönen, Mika; König, Wolfgang; Kramer, Holly; Krämer, Bernhard K.; Kumar, Ashish; Kutalik, Zoltan; Lambert, Jean-Charles; Launer, Lenore J.; Lehtimäki, Terho; de Borst, Martin; Navis, Gerjan; Swertz, Morris; Liu, Yongmei; Lohman, Kurt; Loos, Ruth J. F.; Lu, Yingchang; Lyytikäinen, Leo-Pekka; McEvoy, Mark A.; Meisinger, Christa; Meitinger, Thomas; Metspalu, Andres; Metzger, Marie; Mihailov, Evelin; Mitchell, Paul; Nauck, Matthias; Oldehinkel, Albertine J.; Olden, Matthias; WJH Penninx, Brenda; Pistis, Giorgio; Pramstaller, Peter P.; Probst-Hensch, Nicole; Raitakari, Olli T.; Rettig, Rainer; Ridker, Paul M.; Rivadeneira, Fernando; Robino, Antonietta; Rosas, Sylvia E.; Ruderfer, Douglas; Ruggiero, Daniela; Saba, Yasaman; Sala, Cinzia; Schmidt, Helena; Schmidt, Reinhold; Scott, Rodney J.; Sedaghat, Sanaz; Smith, Albert V.; Sorice, Rossella; Stengel, Benedicte; Stracke, Sylvia; Strauch, Konstantin; Toniolo, Daniela; Uitterlinden, Andre G.; Ulivi, Sheila; Viikari, Jorma S.; Völker, Uwe; Vollenweider, Peter; Völzke, Henry; Vuckovic, Dragana; Waldenberger, Melanie; Jin Wang, Jie; Yang, Qiong; Chasman, Daniel I.; Tromp, Gerard; Snieder, Harold; Heid, Iris M.; Fox, Caroline S.; Köttgen, Anna; Pattaro, Cristian; Böger, Carsten A.; Fuchsberger, Christian

2017-01-01

HapMap imputed genome-wide association studies (GWAS) have revealed >50 loci at which common variants with minor allele frequency >5% are associated with kidney function. GWAS using more complete reference sets for imputation, such as those from The 1000 Genomes project, promise to identify novel loci that have been missed by previous efforts. To investigate the value of such a more complete variant catalog, we conducted a GWAS meta-analysis of kidney function based on the estimated glomerular filtration rate (eGFR) in 110,517 European ancestry participants using 1000 Genomes imputed data. We identified 10 novel loci with p-value < 5 × 10−8 previously missed by HapMap-based GWAS. Six of these loci (HOXD8, ARL15, PIK3R1, EYA4, ASTN2, and EPB41L3) are tagged by common SNPs unique to the 1000 Genomes reference panel. Using pathway analysis, we identified 39 significant (FDR < 0.05) genes and 127 significantly (FDR < 0.05) enriched gene sets, which were missed by our previous analyses. Among those, the 10 identified novel genes are part of pathways of kidney development, carbohydrate metabolism, cardiac septum development and glucose metabolism. These results highlight the utility of re-imputing from denser reference panels, until whole-genome sequencing becomes feasible in large samples. PMID:28452372

1000 Genomes-based meta-analysis identifies 10 novel loci for kidney function.

PubMed

Gorski, Mathias; van der Most, Peter J; Teumer, Alexander; Chu, Audrey Y; Li, Man; Mijatovic, Vladan; Nolte, Ilja M; Cocca, Massimiliano; Taliun, Daniel; Gomez, Felicia; Li, Yong; Tayo, Bamidele; Tin, Adrienne; Feitosa, Mary F; Aspelund, Thor; Attia, John; Biffar, Reiner; Bochud, Murielle; Boerwinkle, Eric; Borecki, Ingrid; Bottinger, Erwin P; Chen, Ming-Huei; Chouraki, Vincent; Ciullo, Marina; Coresh, Josef; Cornelis, Marilyn C; Curhan, Gary C; d'Adamo, Adamo Pio; Dehghan, Abbas; Dengler, Laura; Ding, Jingzhong; Eiriksdottir, Gudny; Endlich, Karlhans; Enroth, Stefan; Esko, Tõnu; Franco, Oscar H; Gasparini, Paolo; Gieger, Christian; Girotto, Giorgia; Gottesman, Omri; Gudnason, Vilmundur; Gyllensten, Ulf; Hancock, Stephen J; Harris, Tamara B; Helmer, Catherine; Höllerer, Simon; Hofer, Edith; Hofman, Albert; Holliday, Elizabeth G; Homuth, Georg; Hu, Frank B; Huth, Cornelia; Hutri-Kähönen, Nina; Hwang, Shih-Jen; Imboden, Medea; Johansson, Åsa; Kähönen, Mika; König, Wolfgang; Kramer, Holly; Krämer, Bernhard K; Kumar, Ashish; Kutalik, Zoltan; Lambert, Jean-Charles; Launer, Lenore J; Lehtimäki, Terho; de Borst, Martin; Navis, Gerjan; Swertz, Morris; Liu, Yongmei; Lohman, Kurt; Loos, Ruth J F; Lu, Yingchang; Lyytikäinen, Leo-Pekka; McEvoy, Mark A; Meisinger, Christa; Meitinger, Thomas; Metspalu, Andres; Metzger, Marie; Mihailov, Evelin; Mitchell, Paul; Nauck, Matthias; Oldehinkel, Albertine J; Olden, Matthias; Wjh Penninx, Brenda; Pistis, Giorgio; Pramstaller, Peter P; Probst-Hensch, Nicole; Raitakari, Olli T; Rettig, Rainer; Ridker, Paul M; Rivadeneira, Fernando; Robino, Antonietta; Rosas, Sylvia E; Ruderfer, Douglas; Ruggiero, Daniela; Saba, Yasaman; Sala, Cinzia; Schmidt, Helena; Schmidt, Reinhold; Scott, Rodney J; Sedaghat, Sanaz; Smith, Albert V; Sorice, Rossella; Stengel, Benedicte; Stracke, Sylvia; Strauch, Konstantin; Toniolo, Daniela; Uitterlinden, Andre G; Ulivi, Sheila; Viikari, Jorma S; Völker, Uwe; Vollenweider, Peter; Völzke, Henry; Vuckovic, Dragana; Waldenberger, Melanie; Jin Wang, Jie; Yang, Qiong; Chasman, Daniel I; Tromp, Gerard; Snieder, Harold; Heid, Iris M; Fox, Caroline S; Köttgen, Anna; Pattaro, Cristian; Böger, Carsten A; Fuchsberger, Christian

2017-04-28

HapMap imputed genome-wide association studies (GWAS) have revealed >50 loci at which common variants with minor allele frequency >5% are associated with kidney function. GWAS using more complete reference sets for imputation, such as those from The 1000 Genomes project, promise to identify novel loci that have been missed by previous efforts. To investigate the value of such a more complete variant catalog, we conducted a GWAS meta-analysis of kidney function based on the estimated glomerular filtration rate (eGFR) in 110,517 European ancestry participants using 1000 Genomes imputed data. We identified 10 novel loci with p-value < 5 × 10 -8 previously missed by HapMap-based GWAS. Six of these loci (HOXD8, ARL15, PIK3R1, EYA4, ASTN2, and EPB41L3) are tagged by common SNPs unique to the 1000 Genomes reference panel. Using pathway analysis, we identified 39 significant (FDR < 0.05) genes and 127 significantly (FDR < 0.05) enriched gene sets, which were missed by our previous analyses. Among those, the 10 identified novel genes are part of pathways of kidney development, carbohydrate metabolism, cardiac septum development and glucose metabolism. These results highlight the utility of re-imputing from denser reference panels, until whole-genome sequencing becomes feasible in large samples.

Missing data in FFQs: making assumptions about item non-response.

PubMed

Lamb, Karen E; Olstad, Dana Lee; Nguyen, Cattram; Milte, Catherine; McNaughton, Sarah A

2017-04-01

FFQs are a popular method of capturing dietary information in epidemiological studies and may be used to derive dietary exposures such as nutrient intake or overall dietary patterns and diet quality. As FFQs can involve large numbers of questions, participants may fail to respond to all questions, leaving researchers to decide how to deal with missing data when deriving intake measures. The aim of the present commentary is to discuss the current practice for dealing with item non-response in FFQs and to propose a research agenda for reporting and handling missing data in FFQs. Single imputation techniques, such as zero imputation (assuming no consumption of the item) or mean imputation, are commonly used to deal with item non-response in FFQs. However, single imputation methods make strong assumptions about the missing data mechanism and do not reflect the uncertainty created by the missing data. This can lead to incorrect inference about associations between diet and health outcomes. Although the use of multiple imputation methods in epidemiology has increased, these have seldom been used in the field of nutritional epidemiology to address missing data in FFQs. We discuss methods for dealing with item non-response in FFQs, highlighting the assumptions made under each approach. Researchers analysing FFQs should ensure that missing data are handled appropriately and clearly report how missing data were treated in analyses. Simulation studies are required to enable systematic evaluation of the utility of various methods for handling item non-response in FFQs under different assumptions about the missing data mechanism.

Missing Oral Health-Related Data in the interRAI-HC - Associations with Selected Variables of General Health and the Effect of Multiple Imputation on the Relationship between Oral and General Health.

PubMed

Krausch-Hofmann, Stefanie; Bogaerts, Kris; Hofmann, Michael; de Almeida Mello, Johanna; Fávaro Moreira, Nádia Cristina; Lesaffre, Emmanuel; Declerck, Dominique; Declercq, Anja; Duyck, Joke

2015-01-01

Missing data within the comprehensive geriatric assessment of the interRAI suite of assessment instruments potentially imply the under-detection of conditions that require care as well as the risk of biased statistical results. Impaired oral health in older individuals has to be registered accurately as it causes pain and discomfort and is related to the general health status. This study was based on interRAI-Home Care (HC) baseline data from 7590 subjects (mean age 81.2 years, SD 6.9) in Belgium. It was investigated if missingness of the oral health-related items was associated with selected variables of general health. It was also determined if multiple imputation of missing data affected the associations between oral and general health. Multivariable logistic regression was used to determine if the prevalence of missingness in the oral health-related variables was associated with activities of daily life (ADLH), cognitive performance (CPS2) and depression (DRS). Associations between oral health and ADLH, CPS2 and DRS were determined, with missing data treated by 1. the complete-case technique and 2. by multiple imputation, and results were compared. The individual oral health-related variables had a similar proportion of missing values, ranging from 16.3% to 17.2%. The prevalence of missing data in all oral health-related variables was significantly associated with symptoms of depression (dental prosthesis use OR 1.66, CI 1.41-1.95; damaged teeth OR 1.74, CI 1.48-2.04; chewing problems OR 1.74, CI 1.47-2.05; dry mouth OR 1.65, CI 1.40-1.94). Missingness in damaged teeth (OR 1.27, CI 1.08-1.48), chewing problems (OR 1.22, CI 1.04-1.44) and dry mouth (OR 1.23, CI 1.05-1.44) occurred more frequently in cognitively impaired subjects. ADLH was not associated with the prevalence of missing data. When comparing the complete-case technique with the multiple imputation approach, nearly identical odds ratios characterized the associations between oral and general health. Cognitively impaired and depressive individuals had a higher risk of missing oral health-related information. Associations between oral health and ADLH, CPS2 and DRS were not influenced by multiple imputation of missing data. Further research should concentrate on the mechanisms that mediate the occurrence of missingness to develop preventative strategies.

Assessment of imputation methods using varying ecological information to fill the gaps in a tree functional trait database

NASA Astrophysics Data System (ADS)

Poyatos, Rafael; Sus, Oliver; Vilà-Cabrera, Albert; Vayreda, Jordi; Badiella, Llorenç; Mencuccini, Maurizio; Martínez-Vilalta, Jordi

2016-04-01

Plant functional traits are increasingly being used in ecosystem ecology thanks to the growing availability of large ecological databases. However, these databases usually contain a large fraction of missing data because measuring plant functional traits systematically is labour-intensive and because most databases are compilations of datasets with different sampling designs. As a result, within a given database, there is an inevitable variability in the number of traits available for each data entry and/or the species coverage in a given geographical area. The presence of missing data may severely bias trait-based analyses, such as the quantification of trait covariation or trait-environment relationships and may hamper efforts towards trait-based modelling of ecosystem biogeochemical cycles. Several data imputation (i.e. gap-filling) methods have been recently tested on compiled functional trait databases, but the performance of imputation methods applied to a functional trait database with a regular spatial sampling has not been thoroughly studied. Here, we assess the effects of data imputation on five tree functional traits (leaf biomass to sapwood area ratio, foliar nitrogen, maximum height, specific leaf area and wood density) in the Ecological and Forest Inventory of Catalonia, an extensive spatial database (covering 31900 km2). We tested the performance of species mean imputation, single imputation by the k-nearest neighbors algorithm (kNN) and a multiple imputation method, Multivariate Imputation with Chained Equations (MICE) at different levels of missing data (10%, 30%, 50%, and 80%). We also assessed the changes in imputation performance when additional predictors (species identity, climate, forest structure, spatial structure) were added in kNN and MICE imputations. We evaluated the imputed datasets using a battery of indexes describing departure from the complete dataset in trait distribution, in the mean prediction error, in the correlation matrix and in selected bivariate trait relationships. MICE yielded imputations which better preserved the variability and covariance structure of the data and provided an estimate of between-imputation uncertainty. We found that adding species identity as a predictor in MICE and kNN improved imputation for all traits, but adding climate did not lead to any appreciable improvement. However, forest structure and spatial structure did reduce imputation errors in maximum height and in leaf biomass to sapwood area ratios, respectively. Although species mean imputations showed the lowest error for 3 out the 5 studied traits, dataset-averaged errors were lowest for MICE imputations with all additional predictors, when missing data levels were 50% or lower. Species mean imputations always resulted in larger errors in the correlation matrix and appreciably altered the studied bivariate trait relationships. In conclusion, MICE imputations using species identity, climate, forest structure and spatial structure as predictors emerged as the most suitable method of the ones tested here, but it was also evident that imputation performance deteriorates at high levels of missing data (80%).

Multiple imputation methods for nonparametric inference on cumulative incidence with missing cause of failure

PubMed Central

Lee, Minjung; Dignam, James J.; Han, Junhee

2014-01-01

We propose a nonparametric approach for cumulative incidence estimation when causes of failure are unknown or missing for some subjects. Under the missing at random assumption, we estimate the cumulative incidence function using multiple imputation methods. We develop asymptotic theory for the cumulative incidence estimators obtained from multiple imputation methods. We also discuss how to construct confidence intervals for the cumulative incidence function and perform a test for comparing the cumulative incidence functions in two samples with missing cause of failure. Through simulation studies, we show that the proposed methods perform well. The methods are illustrated with data from a randomized clinical trial in early stage breast cancer. PMID:25043107

Attrition Bias Related to Missing Outcome Data: A Longitudinal Simulation Study.

PubMed

Lewin, Antoine; Brondeel, Ruben; Benmarhnia, Tarik; Thomas, Frédérique; Chaix, Basile

2018-01-01

Most longitudinal studies do not address potential selection biases due to selective attrition. Using empirical data and simulating additional attrition, we investigated the effectiveness of common approaches to handle missing outcome data from attrition in the association between individual education level and change in body mass index (BMI). Using data from the two waves of the French RECORD Cohort Study (N = 7,172), we first examined how inverse probability weighting (IPW) and multiple imputation handled missing outcome data from attrition in the observed data (stage 1). Second, simulating additional missing data in BMI at follow-up under various missing-at-random scenarios, we quantified the impact of attrition and assessed how multiple imputation performed compared to complete case analysis and to a perfectly specified IPW model as a gold standard (stage 2). With the observed data in stage 1, we found an inverse association between individual education and change in BMI, with complete case analysis, as well as with IPW and multiple imputation. When we simulated additional attrition under a missing-at-random pattern (stage 2), the bias increased with the magnitude of selective attrition, and multiple imputation was useless to address it. Our simulations revealed that selective attrition in the outcome heavily biased the association of interest. The present article contributes to raising awareness that for missing outcome data, multiple imputation does not do better than complete case analysis. More effort is thus needed during the design phase to understand attrition mechanisms by collecting information on the reasons for dropout.

The Effect of Missing Data Handling Methods on Goodness of Fit Indices in Confirmatory Factor Analysis

ERIC Educational Resources Information Center

Köse, Alper

2014-01-01

The primary objective of this study was to examine the effect of missing data on goodness of fit statistics in confirmatory factor analysis (CFA). For this aim, four missing data handling methods; listwise deletion, full information maximum likelihood, regression imputation and expectation maximization (EM) imputation were examined in terms of…

Handling Missing Data: Analysis of a Challenging Data Set Using Multiple Imputation

ERIC Educational Resources Information Center

Pampaka, Maria; Hutcheson, Graeme; Williams, Julian

2016-01-01

Missing data is endemic in much educational research. However, practices such as step-wise regression common in the educational research literature have been shown to be dangerous when significant data are missing, and multiple imputation (MI) is generally recommended by statisticians. In this paper, we provide a review of these advances and their…

Reporting the Use of Multiple Imputation for Missing Data in Higher Education Research

ERIC Educational Resources Information Center

Manly, Catherine A.; Wells, Ryan S.

2015-01-01

Higher education researchers using survey data often face decisions about handling missing data. Multiple imputation (MI) is considered by many statisticians to be the most appropriate technique for addressing missing data in many circumstances. In particular, it has been shown to be preferable to listwise deletion, which has historically been a…

A Study of Imputation Algorithms. Working Paper Series.

ERIC Educational Resources Information Center

Hu, Ming-xiu; Salvucci, Sameena

Many imputation techniques and imputation software packages have been developed over the years to deal with missing data. Different methods may work well under different circumstances, and it is advisable to conduct a sensitivity analysis when choosing an imputation method for a particular survey. This study reviewed about 30 imputation methods…

[Imputation methods for missing data in educational diagnostic evaluation].

PubMed

Fernández-Alonso, Rubén; Suárez-Álvarez, Javier; Muñiz, José

2012-02-01

In the diagnostic evaluation of educational systems, self-reports are commonly used to collect data, both cognitive and orectic. For various reasons, in these self-reports, some of the students' data are frequently missing. The main goal of this research is to compare the performance of different imputation methods for missing data in the context of the evaluation of educational systems. On an empirical database of 5,000 subjects, 72 conditions were simulated: three levels of missing data, three types of loss mechanisms, and eight methods of imputation. The levels of missing data were 5%, 10%, and 20%. The loss mechanisms were set at: Missing completely at random, moderately conditioned, and strongly conditioned. The eight imputation methods used were: listwise deletion, replacement by the mean of the scale, by the item mean, the subject mean, the corrected subject mean, multiple regression, and Expectation-Maximization (EM) algorithm, with and without auxiliary variables. The results indicate that the recovery of the data is more accurate when using an appropriate combination of different methods of recovering lost data. When a case is incomplete, the mean of the subject works very well, whereas for completely lost data, multiple imputation with the EM algorithm is recommended. The use of this combination is especially recommended when data loss is greater and its loss mechanism is more conditioned. Lastly, the results are discussed, and some future lines of research are analyzed.

A Youth Compendium of Physical Activities: Activity Codes and Metabolic Intensities

PubMed Central

BUTTE, NANCY F.; WATSON, KATHLEEN B.; RIDLEY, KATE; ZAKERI, ISSA F.; MCMURRAY, ROBERT G.; PFEIFFER, KARIN A.; CROUTER, SCOTT E.; HERRMANN, STEPHEN D.; BASSETT, DAVID R.; LONG, ALEXANDER; BERHANE, ZEKARIAS; TROST, STEWART G.; AINSWORTH, BARBARA E.; BERRIGAN, DAVID; FULTON, JANET E.

2018-01-01

ABSTRACT Purpose A Youth Compendium of Physical Activities (Youth Compendium) was developed to estimate the energy costs of physical activities using data on youth only. Methods On the basis of a literature search and pooled data of energy expenditure measurements in youth, the energy costs of 196 activities were compiled in 16 activity categories to form a Youth Compendium of Physical Activities. To estimate the intensity of each activity, measured oxygen consumption (V˙O2) was divided by basal metabolic rate (Schofield age-, sex-, and mass-specific equations) to produce a youth MET (METy). A mixed linear model was developed for each activity category to impute missing values for age ranges with no observations for a specific activity. Results This Youth Compendium consists of METy values for 196 specific activities classified into 16 major categories for four age-groups, 6–9, 10–12, 13–15, and 16–18 yr. METy values in this Youth Compendium were measured (51%) or imputed (49%) from youth data. Conclusion This Youth Compendium of Physical Activities uses pediatric data exclusively, addresses the age dependency of METy, and imputes missing METy values and thus represents advancement in physical activity research and practice. This Youth Compendium will be a valuable resource for stakeholders interested in evaluating interventions, programs, and policies designed to assess and encourage physical activity in youth. PMID:28938248

An original imputation technique of missing data for assessing exposure of newborns to perchlorate in drinking water.

PubMed

Caron, Alexandre; Clement, Guillaume; Heyman, Christophe; Aernout, Eva; Chazard, Emmanuel; Le Tertre, Alain

2015-01-01

Incompleteness of epidemiological databases is a major drawback when it comes to analyzing data. We conceived an epidemiological study to assess the association between newborn thyroid function and the exposure to perchlorates found in the tap water of the mother's home. Only 9% of newborn's exposure to perchlorate was known. The aim of our study was to design, test and evaluate an original method for imputing perchlorate exposure of newborns based on their maternity of birth. In a first database, an exhaustive collection of newborn's thyroid function measured during a systematic neonatal screening was collected. In this database the municipality of residence of the newborn's mother was only available for 2012. Between 2004 and 2011, the closest data available was the municipality of the maternity of birth. Exposure was assessed using a second database which contained the perchlorate levels for each municipality. We computed the catchment area of every maternity ward based on the French nationwide exhaustive database of inpatient stay. Municipality, and consequently perchlorate exposure, was imputed by a weighted draw in the catchment area. Missing values for remaining covariates were imputed by chained equation. A linear mixture model was computed on each imputed dataset. We compared odds ratios (ORs) and 95% confidence intervals (95% CI) estimated on real versus imputed 2012 data. The same model was then carried out for the whole imputed database. The ORs estimated on 36,695 observations by our multiple imputation method are comparable to the real 2012 data. On the 394,979 observations of the whole database, the ORs remain stable but the 95% CI tighten considerably. The model estimates computed on imputed data are similar to those calculated on real data. The main advantage of multiple imputation is to provide unbiased estimate of the ORs while maintaining their variances. Thus, our method will be used to increase the statistical power of future studies by including all 394,979 newborns.

Use of Multiple Imputation to Estimate the Proportion of Respiratory Virus Detections Among Patients Hospitalized With Community-Acquired Pneumonia

PubMed Central

Bozio, Catherine H; Flanders, W Dana; Finelli, Lyn; Bramley, Anna M; Reed, Carrie; Gandhi, Neel R; Vidal, Jorge E; Erdman, Dean; Levine, Min Z; Lindstrom, Stephen; Ampofo, Krow; Arnold, Sandra R; Self, Wesley H; Williams, Derek J; Grijalva, Carlos G; Anderson, Evan J; McCullers, Jonathan A; Edwards, Kathryn M; Pavia, Andrew T; Wunderink, Richard G; Jain, Seema

2018-01-01

Abstract Background Real-time polymerase chain reaction (PCR) on respiratory specimens and serology on paired blood specimens are used to determine the etiology of respiratory illnesses for research studies. However, convalescent serology is often not collected. We used multiple imputation to assign values for missing serology results to estimate virus-specific prevalence among pediatric and adult community-acquired pneumonia hospitalizations using data from an active population-based surveillance study. Methods Presence of adenoviruses, human metapneumovirus, influenza viruses, parainfluenza virus types 1–3, and respiratory syncytial virus was defined by positive PCR on nasopharyngeal/oropharyngeal specimens or a 4-fold rise in paired serology. We performed multiple imputation by developing a multivariable regression model for each virus using data from patients with available serology results. We calculated absolute and relative differences in the proportion of each virus detected comparing the imputed to observed (nonimputed) results. Results Among 2222 children and 2259 adults, 98.8% and 99.5% had nasopharyngeal/oropharyngeal specimens and 43.2% and 37.5% had paired serum specimens, respectively. Imputed results increased viral etiology assignments by an absolute difference of 1.6%–4.4% and 0.8%–2.8% in children and adults, respectively; relative differences were 1.1–3.0 times higher. Conclusions Multiple imputation can be used when serology results are missing, to refine virus-specific prevalence estimates, and these will likely increase estimates.

Multiple imputation of covariates by fully conditional specification: Accommodating the substantive model

PubMed Central

Seaman, Shaun R; White, Ian R; Carpenter, James R

2015-01-01

Missing covariate data commonly occur in epidemiological and clinical research, and are often dealt with using multiple imputation. Imputation of partially observed covariates is complicated if the substantive model is non-linear (e.g. Cox proportional hazards model), or contains non-linear (e.g. squared) or interaction terms, and standard software implementations of multiple imputation may impute covariates from models that are incompatible with such substantive models. We show how imputation by fully conditional specification, a popular approach for performing multiple imputation, can be modified so that covariates are imputed from models which are compatible with the substantive model. We investigate through simulation the performance of this proposal, and compare it with existing approaches. Simulation results suggest our proposal gives consistent estimates for a range of common substantive models, including models which contain non-linear covariate effects or interactions, provided data are missing at random and the assumed imputation models are correctly specified and mutually compatible. Stata software implementing the approach is freely available. PMID:24525487

Hydrologic Response to Climate Change: Missing Precipitation Data Matters for Computed Timing Trends

NASA Astrophysics Data System (ADS)

Daniels, B.

2016-12-01

This work demonstrates the derivation of climate timing statistics and applying them to determine resulting hydroclimate impacts. Long-term daily precipitation observations from 50 California stations were used to compute climate trends of precipitation event Intensity, event Duration and Pause between events. Each precipitation event trend was then applied as input to a PRMS hydrology model which showed hydrology changes to recharge, baseflow, streamflow, etc. An important concern was precipitation uncertainty induced by missing observation values and causing errors in quantification of precipitation trends. Many standard statistical techniques such as ARIMA and simple endogenous or even exogenous imputation were applied but failed to help resolve these uncertainties. What helped resolve these uncertainties was use of multiple imputation techniques. This involved fitting of Weibull probability distributions to multiple imputed values for the three precipitation trends.Permutation resampling techniques using Monte Carlo processing were then applied to the multiple imputation values to derive significance p-values for each trend. Significance at the 95% level for Intensity was found for 11 of the 50 stations, Duration from 16 of the 50, and Pause from 19, of which 12 were 99% significant. The significance weighted trends for California are Intensity -4.61% per decade, Duration +3.49% per decade, and Pause +3.58% per decade. Two California basins with PRMS hydrologic models were studied: Feather River in the northern Sierra Nevada mountains and the central coast Soquel-Aptos. Each local trend was changed without changing the other trends or the total precipitation. Feather River Basin's critical supply to Lake Oroville and the State Water Project benefited from a total streamflow increase of 1.5%. The Soquel-Aptos Basin water supply was impacted by a total groundwater recharge decrease of -7.5% and streamflow decrease of -3.2%.

Quantifying the impact of fixed effects modeling of clusters in multiple imputation for cluster randomized trials

PubMed Central

Andridge, Rebecca. R.

2011-01-01

In cluster randomized trials (CRTs), identifiable clusters rather than individuals are randomized to study groups. Resulting data often consist of a small number of clusters with correlated observations within a treatment group. Missing data often present a problem in the analysis of such trials, and multiple imputation (MI) has been used to create complete data sets, enabling subsequent analysis with well-established analysis methods for CRTs. We discuss strategies for accounting for clustering when multiply imputing a missing continuous outcome, focusing on estimation of the variance of group means as used in an adjusted t-test or ANOVA. These analysis procedures are congenial to (can be derived from) a mixed effects imputation model; however, this imputation procedure is not yet available in commercial statistical software. An alternative approach that is readily available and has been used in recent studies is to include fixed effects for cluster, but the impact of using this convenient method has not been studied. We show that under this imputation model the MI variance estimator is positively biased and that smaller ICCs lead to larger overestimation of the MI variance. Analytical expressions for the bias of the variance estimator are derived in the case of data missing completely at random (MCAR), and cases in which data are missing at random (MAR) are illustrated through simulation. Finally, various imputation methods are applied to data from the Detroit Middle School Asthma Project, a recent school-based CRT, and differences in inference are compared. PMID:21259309

A structured framework for assessing sensitivity to missing data assumptions in longitudinal clinical trials.

PubMed

Mallinckrodt, C H; Lin, Q; Molenberghs, M

2013-01-01

The objective of this research was to demonstrate a framework for drawing inference from sensitivity analyses of incomplete longitudinal clinical trial data via a re-analysis of data from a confirmatory clinical trial in depression. A likelihood-based approach that assumed missing at random (MAR) was the primary analysis. Robustness to departure from MAR was assessed by comparing the primary result to those from a series of analyses that employed varying missing not at random (MNAR) assumptions (selection models, pattern mixture models and shared parameter models) and to MAR methods that used inclusive models. The key sensitivity analysis used multiple imputation assuming that after dropout the trajectory of drug-treated patients was that of placebo treated patients with a similar outcome history (placebo multiple imputation). This result was used as the worst reasonable case to define the lower limit of plausible values for the treatment contrast. The endpoint contrast from the primary analysis was - 2.79 (p = .013). In placebo multiple imputation, the result was - 2.17. Results from the other sensitivity analyses ranged from - 2.21 to - 3.87 and were symmetrically distributed around the primary result. Hence, no clear evidence of bias from missing not at random data was found. In the worst reasonable case scenario, the treatment effect was 80% of the magnitude of the primary result. Therefore, it was concluded that a treatment effect existed. The structured sensitivity framework of using a worst reasonable case result based on a controlled imputation approach with transparent and debatable assumptions supplemented a series of plausible alternative models under varying assumptions was useful in this specific situation and holds promise as a generally useful framework. Copyright © 2012 John Wiley & Sons, Ltd.

Impact of Missing Data for Body Mass Index in an Epidemiologic Study.

PubMed

Razzaghi, Hilda; Tinker, Sarah C; Herring, Amy H; Howards, Penelope P; Waller, D Kim; Johnson, Candice Y

2016-07-01

Objective To assess the potential impact of missing data on body mass index (BMI) on the association between prepregnancy obesity and specific birth defects. Methods Data from the National Birth Defects Prevention Study (NBDPS) were analyzed. We assessed the factors associated with missing BMI data among mothers of infants without birth defects. Four analytic methods were then used to assess the impact of missing BMI data on the association between maternal prepregnancy obesity and three birth defects; spina bifida, gastroschisis, and cleft lip with/without cleft palate. The analytic methods were: (1) complete case analysis; (2) assignment of missing values to either obese or normal BMI; (3) multiple imputation; and (4) probabilistic sensitivity analysis. Logistic regression was used to estimate crude and adjusted odds ratios (aOR) and 95 % confidence intervals (CI). Results Of NBDPS control mothers 4.6 % were missing BMI data, and most of the missing values were attributable to missing height (~90 %). Missing BMI data was associated with birth outside of the US (aOR 8.6; 95 % CI 5.5, 13.4), interview in Spanish (aOR 2.4; 95 % CI 1.8, 3.2), Hispanic ethnicity (aOR 2.0; 95 % CI 1.2, 3.4), and <12 years education (aOR 2.3; 95 % CI 1.7, 3.1). Overall the results of the multiple imputation and probabilistic sensitivity analysis were similar to the complete case analysis. Conclusions Although in some scenarios missing BMI data can bias the magnitude of association, it does not appear likely to have impacted conclusions from a traditional complete case analysis of these data.

Multiple Imputation for Multivariate Missing-Data Problems: A Data Analyst's Perspective.

ERIC Educational Resources Information Center

Schafer, Joseph L.; Olsen, Maren K.

1998-01-01

The key ideas of multiple imputation for multivariate missing data problems are reviewed. Software programs available for this analysis are described, and their use is illustrated with data from the Adolescent Alcohol Prevention Trial (W. Hansen and J. Graham, 1991). (SLD)

Estimating Interaction Effects With Incomplete Predictor Variables

PubMed Central

Enders, Craig K.; Baraldi, Amanda N.; Cham, Heining

2014-01-01

The existing missing data literature does not provide a clear prescription for estimating interaction effects with missing data, particularly when the interaction involves a pair of continuous variables. In this article, we describe maximum likelihood and multiple imputation procedures for this common analysis problem. We outline 3 latent variable model specifications for interaction analyses with missing data. These models apply procedures from the latent variable interaction literature to analyses with a single indicator per construct (e.g., a regression analysis with scale scores). We also discuss multiple imputation for interaction effects, emphasizing an approach that applies standard imputation procedures to the product of 2 raw score predictors. We thoroughly describe the process of probing interaction effects with maximum likelihood and multiple imputation. For both missing data handling techniques, we outline centering and transformation strategies that researchers can implement in popular software packages, and we use a series of real data analyses to illustrate these methods. Finally, we use computer simulations to evaluate the performance of the proposed techniques. PMID:24707955

Adaptation of clinical prediction models for application in local settings.

PubMed

Kappen, Teus H; Vergouwe, Yvonne; van Klei, Wilton A; van Wolfswinkel, Leo; Kalkman, Cor J; Moons, Karel G M

2012-01-01

When planning to use a validated prediction model in new patients, adequate performance is not guaranteed. For example, changes in clinical practice over time or a different case mix than the original validation population may result in inaccurate risk predictions. To demonstrate how clinical information can direct updating a prediction model and development of a strategy for handling missing predictor values in clinical practice. A previously derived and validated prediction model for postoperative nausea and vomiting was updated using a data set of 1847 patients. The update consisted of 1) changing the definition of an existing predictor, 2) reestimating the regression coefficient of a predictor, and 3) adding a new predictor to the model. The updated model was then validated in a new series of 3822 patients. Furthermore, several imputation models were considered to handle real-time missing values, so that possible missing predictor values could be anticipated during actual model use. Differences in clinical practice between our local population and the original derivation population guided the update strategy of the prediction model. The predictive accuracy of the updated model was better (c statistic, 0.68; calibration slope, 1.0) than the original model (c statistic, 0.62; calibration slope, 0.57). Inclusion of logistical variables in the imputation models, besides observed patient characteristics, contributed to a strategy to deal with missing predictor values at the time of risk calculation. Extensive knowledge of local, clinical processes provides crucial information to guide the process of adapting a prediction model to new clinical practices.

Gastrointestinal symptomatology as a predictor of severe outcomes of invasive group A streptococcal infections.

PubMed

Khateeb, O M; Osborne, D; Mulla, Z D

2010-04-01

Invasive group A streptococcal (GAS) disease is a condition of clinical and public health significance. We conducted epidemiological analyses to determine if the presence of gastrointestinal (GI) complaints (diarrhea and/or vomiting) early in the course of invasive GAS disease is associated with either of two severe outcomes: GAS necrotizing fasciitis, or hospital mortality. Subjects were hospitalized for invasive GAS disease throughout the state of Florida, USA, during a 4-year period. Multiple imputation using the Markov chain Monte Carlo method was used to replace missing values with plausible values. Excluding cases with missing data resulted in a sample size of 138 invasive GAS patients (the complete subject analysis) while the imputed datasets contained 257 records. GI symptomatology within 48 h of hospital admission was not associated with hospital mortality in either the complete subject analysis [adjusted odds ratio (aOR) 0.86, 95% confidence interval (CI) 0.31-2.39] or in the imputed datasets. GI symptoms were significantly associated with GAS necrotizing fasciitis in the complete subject analysis (aOR 4.64, 95% CI 1.18-18.23) and in the imputed datasets but only in patients aged <55 years. The common cause of GI symptoms and necrotizing fasciitis may be streptococcal exotoxins. Clinicians who are treating young individuals presumed to be in the early stages of invasive GAS disease should take note of GI symptoms and remain vigilant for the development of a GAS necrotizing soft-tissue infection.

The roles of nearest neighbor methods in imputing missing data in forest inventory and monitoring databases

Treesearch

Bianca N. I. Eskelson; Hailemariam Temesgen; Valerie Lemay; Tara M. Barrett; Nicholas L. Crookston; Andrew T. Hudak

2009-01-01

Almost universally, forest inventory and monitoring databases are incomplete, ranging from missing data for only a few records and a few variables, common for small land areas, to missing data for many observations and many variables, common for large land areas. For a wide variety of applications, nearest neighbor (NN) imputation methods have been developed to fill in...

LinkImpute: Fast and Accurate Genotype Imputation for Nonmodel Organisms

PubMed Central

Money, Daniel; Gardner, Kyle; Migicovsky, Zoë; Schwaninger, Heidi; Zhong, Gan-Yuan; Myles, Sean

2015-01-01

Obtaining genome-wide genotype data from a set of individuals is the first step in many genomic studies, including genome-wide association and genomic selection. All genotyping methods suffer from some level of missing data, and genotype imputation can be used to fill in the missing data and improve the power of downstream analyses. Model organisms like human and cattle benefit from high-quality reference genomes and panels of reference genotypes that aid in imputation accuracy. In nonmodel organisms, however, genetic and physical maps often are either of poor quality or are completely absent, and there are no panels of reference genotypes available. There is therefore a need for imputation methods designed specifically for nonmodel organisms in which genomic resources are poorly developed and marker order is unreliable or unknown. Here we introduce LinkImpute, a software package based on a k-nearest neighbor genotype imputation method, LD-kNNi, which is designed for unordered markers. No physical or genetic maps are required, and it is designed to work on unphased genotype data from heterozygous species. It exploits the fact that markers useful for imputation often are not physically close to the missing genotype but rather distributed throughout the genome. Using genotyping-by-sequencing data from diverse and heterozygous accessions of apples, grapes, and maize, we compare LD-kNNi with several genotype imputation methods and show that LD-kNNi is fast, comparable in accuracy to the best-existing methods, and exhibits the least bias in allele frequency estimates. PMID:26377960

Gap-filling a spatially explicit plant trait database: comparing imputation methods and different levels of environmental information

NASA Astrophysics Data System (ADS)

Poyatos, Rafael; Sus, Oliver; Badiella, Llorenç; Mencuccini, Maurizio; Martínez-Vilalta, Jordi

2018-05-01

The ubiquity of missing data in plant trait databases may hinder trait-based analyses of ecological patterns and processes. Spatially explicit datasets with information on intraspecific trait variability are rare but offer great promise in improving our understanding of functional biogeography. At the same time, they offer specific challenges in terms of data imputation. Here we compare statistical imputation approaches, using varying levels of environmental information, for five plant traits (leaf biomass to sapwood area ratio, leaf nitrogen content, maximum tree height, leaf mass per area and wood density) in a spatially explicit plant trait dataset of temperate and Mediterranean tree species (Ecological and Forest Inventory of Catalonia, IEFC, dataset for Catalonia, north-east Iberian Peninsula, 31 900 km2). We simulated gaps at different missingness levels (10-80 %) in a complete trait matrix, and we used overall trait means, species means, k nearest neighbours (kNN), ordinary and regression kriging, and multivariate imputation using chained equations (MICE) to impute missing trait values. We assessed these methods in terms of their accuracy and of their ability to preserve trait distributions, multi-trait correlation structure and bivariate trait relationships. The relatively good performance of mean and species mean imputations in terms of accuracy masked a poor representation of trait distributions and multivariate trait structure. Species identity improved MICE imputations for all traits, whereas forest structure and topography improved imputations for some traits. No method performed best consistently for the five studied traits, but, considering all traits and performance metrics, MICE informed by relevant ecological variables gave the best results. However, at higher missingness (> 30 %), species mean imputations and regression kriging tended to outperform MICE for some traits. MICE informed by relevant ecological variables allowed us to fill the gaps in the IEFC incomplete dataset (5495 plots) and quantify imputation uncertainty. Resulting spatial patterns of the studied traits in Catalan forests were broadly similar when using species means, regression kriging or the best-performing MICE application, but some important discrepancies were observed at the local level. Our results highlight the need to assess imputation quality beyond just imputation accuracy and show that including environmental information in statistical imputation approaches yields more plausible imputations in spatially explicit plant trait datasets.

SparRec: An effective matrix completion framework of missing data imputation for GWAS

NASA Astrophysics Data System (ADS)

Jiang, Bo; Ma, Shiqian; Causey, Jason; Qiao, Linbo; Hardin, Matthew Price; Bitts, Ian; Johnson, Daniel; Zhang, Shuzhong; Huang, Xiuzhen

2016-10-01

Genome-wide association studies present computational challenges for missing data imputation, while the advances of genotype technologies are generating datasets of large sample sizes with sample sets genotyped on multiple SNP chips. We present a new framework SparRec (Sparse Recovery) for imputation, with the following properties: (1) The optimization models of SparRec, based on low-rank and low number of co-clusters of matrices, are different from current statistics methods. While our low-rank matrix completion (LRMC) model is similar to Mendel-Impute, our matrix co-clustering factorization (MCCF) model is completely new. (2) SparRec, as other matrix completion methods, is flexible to be applied to missing data imputation for large meta-analysis with different cohorts genotyped on different sets of SNPs, even when there is no reference panel. This kind of meta-analysis is very challenging for current statistics based methods. (3) SparRec has consistent performance and achieves high recovery accuracy even when the missing data rate is as high as 90%. Compared with Mendel-Impute, our low-rank based method achieves similar accuracy and efficiency, while the co-clustering based method has advantages in running time. The testing results show that SparRec has significant advantages and competitive performance over other state-of-the-art existing statistics methods including Beagle and fastPhase.

An Introduction to Modern Missing Data Analyses

ERIC Educational Resources Information Center

Baraldi, Amanda N.; Enders, Craig K.

2010-01-01

A great deal of recent methodological research has focused on two modern missing data analysis methods: maximum likelihood and multiple imputation. These approaches are advantageous to traditional techniques (e.g. deletion and mean imputation techniques) because they require less stringent assumptions and mitigate the pitfalls of traditional…

Nearest neighbor imputation using spatial–temporal correlations in wireless sensor networks

PubMed Central

Li, YuanYuan; Parker, Lynne E.

2016-01-01

Missing data is common in Wireless Sensor Networks (WSNs), especially with multi-hop communications. There are many reasons for this phenomenon, such as unstable wireless communications, synchronization issues, and unreliable sensors. Unfortunately, missing data creates a number of problems for WSNs. First, since most sensor nodes in the network are battery-powered, it is too expensive to have the nodes retransmit missing data across the network. Data re-transmission may also cause time delays when detecting abnormal changes in an environment. Furthermore, localized reasoning techniques on sensor nodes (such as machine learning algorithms to classify states of the environment) are generally not robust enough to handle missing data. Since sensor data collected by a WSN is generally correlated in time and space, we illustrate how replacing missing sensor values with spatially and temporally correlated sensor values can significantly improve the network’s performance. However, our studies show that it is important to determine which nodes are spatially and temporally correlated with each other. Simple techniques based on Euclidean distance are not sufficient for complex environmental deployments. Thus, we have developed a novel Nearest Neighbor (NN) imputation method that estimates missing data in WSNs by learning spatial and temporal correlations between sensor nodes. To improve the search time, we utilize a kd-tree data structure, which is a non-parametric, data-driven binary search tree. Instead of using traditional mean and variance of each dimension for kd-tree construction, and Euclidean distance for kd-tree search, we use weighted variances and weighted Euclidean distances based on measured percentages of missing data. We have evaluated this approach through experiments on sensor data from a volcano dataset collected by a network of Crossbow motes, as well as experiments using sensor data from a highway traffic monitoring application. Our experimental results show that our proposed 𝒦-NN imputation method has a competitive accuracy with state-of-the-art Expectation–Maximization (EM) techniques, while using much simpler computational techniques, thus making it suitable for use in resource-constrained WSNs. PMID:28435414

Evaluation of two-fold fully conditional specification multiple imputation for longitudinal electronic health record data

PubMed Central

Welch, Catherine A; Petersen, Irene; Bartlett, Jonathan W; White, Ian R; Marston, Louise; Morris, Richard W; Nazareth, Irwin; Walters, Kate; Carpenter, James

2014-01-01

Most implementations of multiple imputation (MI) of missing data are designed for simple rectangular data structures ignoring temporal ordering of data. Therefore, when applying MI to longitudinal data with intermittent patterns of missing data, some alternative strategies must be considered. One approach is to divide data into time blocks and implement MI independently at each block. An alternative approach is to include all time blocks in the same MI model. With increasing numbers of time blocks, this approach is likely to break down because of co-linearity and over-fitting. The new two-fold fully conditional specification (FCS) MI algorithm addresses these issues, by only conditioning on measurements, which are local in time. We describe and report the results of a novel simulation study to critically evaluate the two-fold FCS algorithm and its suitability for imputation of longitudinal electronic health records. After generating a full data set, approximately 70% of selected continuous and categorical variables were made missing completely at random in each of ten time blocks. Subsequently, we applied a simple time-to-event model. We compared efficiency of estimated coefficients from a complete records analysis, MI of data in the baseline time block and the two-fold FCS algorithm. The results show that the two-fold FCS algorithm maximises the use of data available, with the gain relative to baseline MI depending on the strength of correlations within and between variables. Using this approach also increases plausibility of the missing at random assumption by using repeated measures over time of variables whose baseline values may be missing. PMID:24782349

Imputation of missing genotypes from sparse to high density using long-range phasing

USDA-ARS?s Scientific Manuscript database

Related individuals share potentially long chromosome segments that trace to a common ancestor. A phasing algorithm (ChromoPhase) that utilizes this characteristic of finite populations was developed to phase large sections of a chromosome. In addition to phasing, ChromoPhase imputes missing genotyp...

Bias Characterization in Probabilistic Genotype Data and Improved Signal Detection with Multiple Imputation

PubMed Central

Palmer, Cameron; Pe’er, Itsik

2016-01-01

Missing data are an unavoidable component of modern statistical genetics. Different array or sequencing technologies cover different single nucleotide polymorphisms (SNPs), leading to a complicated mosaic pattern of missingness where both individual genotypes and entire SNPs are sporadically absent. Such missing data patterns cannot be ignored without introducing bias, yet cannot be inferred exclusively from nonmissing data. In genome-wide association studies, the accepted solution to missingness is to impute missing data using external reference haplotypes. The resulting probabilistic genotypes may be analyzed in the place of genotype calls. A general-purpose paradigm, called Multiple Imputation (MI), is known to model uncertainty in many contexts, yet it is not widely used in association studies. Here, we undertake a systematic evaluation of existing imputed data analysis methods and MI. We characterize biases related to uncertainty in association studies, and find that bias is introduced both at the imputation level, when imputation algorithms generate inconsistent genotype probabilities, and at the association level, when analysis methods inadequately model genotype uncertainty. We find that MI performs at least as well as existing methods or in some cases much better, and provides a straightforward paradigm for adapting existing genotype association methods to uncertain data. PMID:27310603

METHODS FOR CLUSTERING TIME SERIES DATA ACQUIRED FROM MOBILE HEALTH APPS.

PubMed

Tignor, Nicole; Wang, Pei; Genes, Nicholas; Rogers, Linda; Hershman, Steven G; Scott, Erick R; Zweig, Micol; Yvonne Chan, Yu-Feng; Schadt, Eric E

2017-01-01

In our recent Asthma Mobile Health Study (AMHS), thousands of asthma patients across the country contributed medical data through the iPhone Asthma Health App on a daily basis for an extended period of time. The collected data included daily self-reported asthma symptoms, symptom triggers, and real time geographic location information. The AMHS is just one of many studies occurring in the context of now many thousands of mobile health apps aimed at improving wellness and better managing chronic disease conditions, leveraging the passive and active collection of data from mobile, handheld smart devices. The ability to identify patient groups or patterns of symptoms that might predict adverse outcomes such as asthma exacerbations or hospitalizations from these types of large, prospectively collected data sets, would be of significant general interest. However, conventional clustering methods cannot be applied to these types of longitudinally collected data, especially survey data actively collected from app users, given heterogeneous patterns of missing values due to: 1) varying survey response rates among different users, 2) varying survey response rates over time of each user, and 3) non-overlapping periods of enrollment among different users. To handle such complicated missing data structure, we proposed a probability imputation model to infer missing data. We also employed a consensus clustering strategy in tandem with the multiple imputation procedure. Through simulation studies under a range of scenarios reflecting real data conditions, we identified favorable performance of the proposed method over other strategies that impute the missing value through low-rank matrix completion. When applying the proposed new method to study asthma triggers and symptoms collected as part of the AMHS, we identified several patient groups with distinct phenotype patterns. Further validation of the methods described in this paper might be used to identify clinically important patterns in large data sets with complicated missing data structure, improving the ability to use such data sets to identify at-risk populations for potential intervention.

Correcting bias due to missing stage data in the non-parametric estimation of stage-specific net survival for colorectal cancer using multiple imputation.

PubMed

Falcaro, Milena; Carpenter, James R

2017-06-01

Population-based net survival by tumour stage at diagnosis is a key measure in cancer surveillance. Unfortunately, data on tumour stage are often missing for a non-negligible proportion of patients and the mechanism giving rise to the missingness is usually anything but completely at random. In this setting, restricting analysis to the subset of complete records gives typically biased results. Multiple imputation is a promising practical approach to the issues raised by the missing data, but its use in conjunction with the Pohar-Perme method for estimating net survival has not been formally evaluated. We performed a resampling study using colorectal cancer population-based registry data to evaluate the ability of multiple imputation, used along with the Pohar-Perme method, to deliver unbiased estimates of stage-specific net survival and recover missing stage information. We created 1000 independent data sets, each containing 5000 patients. Stage data were then made missing at random under two scenarios (30% and 50% missingness). Complete records analysis showed substantial bias and poor confidence interval coverage. Across both scenarios our multiple imputation strategy virtually eliminated the bias and greatly improved confidence interval coverage. In the presence of missing stage data complete records analysis often gives severely biased results. We showed that combining multiple imputation with the Pohar-Perme estimator provides a valid practical approach for the estimation of stage-specific colorectal cancer net survival. As usual, when the percentage of missing data is high the results should be interpreted cautiously and sensitivity analyses are recommended. Copyright © 2017 Elsevier Ltd. All rights reserved.

Latent Factor Models and Analyses for Operator Response Times

DTIC Science & Technology

1990-09-01

since imputation is based on presumption of model correctness, and the missing and non-standard values are imputed using the presumed model. The same...oN 0’a w’ w0. NI. 0. . INOC - .I O CO.1 0-CN NO.ION I.3-.03.-.. 0 0 0 0 oOO , 0 0 0 - w0 0 0 0 0 0 In 0 0 0 0 0 0 0 w 0 0 0 0 0 0 0 0 0 0 0 0aO . o w

Explicating the Conditions Under Which Multilevel Multiple Imputation Mitigates Bias Resulting from Random Coefficient-Dependent Missing Longitudinal Data.

PubMed

Gottfredson, Nisha C; Sterba, Sonya K; Jackson, Kristina M

2017-01-01

Random coefficient-dependent (RCD) missingness is a non-ignorable mechanism through which missing data can arise in longitudinal designs. RCD, for which we cannot test, is a problematic form of missingness that occurs if subject-specific random effects correlate with propensity for missingness or dropout. Particularly when covariate missingness is a problem, investigators typically handle missing longitudinal data by using single-level multiple imputation procedures implemented with long-format data, which ignores within-person dependency entirely, or implemented with wide-format (i.e., multivariate) data, which ignores some aspects of within-person dependency. When either of these standard approaches to handling missing longitudinal data is used, RCD missingness leads to parameter bias and incorrect inference. We explain why multilevel multiple imputation (MMI) should alleviate bias induced by a RCD missing data mechanism under conditions that contribute to stronger determinacy of random coefficients. We evaluate our hypothesis with a simulation study. Three design factors are considered: intraclass correlation (ICC; ranging from .25 to .75), number of waves (ranging from 4 to 8), and percent of missing data (ranging from 20 to 50%). We find that MMI greatly outperforms the single-level wide-format (multivariate) method for imputation under a RCD mechanism. For the MMI analyses, bias was most alleviated when the ICC is high, there were more waves of data, and when there was less missing data. Practical recommendations for handling longitudinal missing data are suggested.

Relative efficiency of joint-model and full-conditional-specification multiple imputation when conditional models are compatible: The general location model.

PubMed

Seaman, Shaun R; Hughes, Rachael A

2018-06-01

Estimating the parameters of a regression model of interest is complicated by missing data on the variables in that model. Multiple imputation is commonly used to handle these missing data. Joint model multiple imputation and full-conditional specification multiple imputation are known to yield imputed data with the same asymptotic distribution when the conditional models of full-conditional specification are compatible with that joint model. We show that this asymptotic equivalence of imputation distributions does not imply that joint model multiple imputation and full-conditional specification multiple imputation will also yield asymptotically equally efficient inference about the parameters of the model of interest, nor that they will be equally robust to misspecification of the joint model. When the conditional models used by full-conditional specification multiple imputation are linear, logistic and multinomial regressions, these are compatible with a restricted general location joint model. We show that multiple imputation using the restricted general location joint model can be substantially more asymptotically efficient than full-conditional specification multiple imputation, but this typically requires very strong associations between variables. When associations are weaker, the efficiency gain is small. Moreover, full-conditional specification multiple imputation is shown to be potentially much more robust than joint model multiple imputation using the restricted general location model to mispecification of that model when there is substantial missingness in the outcome variable.

Strategies for Dealing with Missing Accelerometer Data.

PubMed

Stephens, Samantha; Beyene, Joseph; Tremblay, Mark S; Faulkner, Guy; Pullnayegum, Eleanor; Feldman, Brian M

2018-05-01

Missing data is a universal research problem that can affect studies examining the relationship between physical activity measured with accelerometers and health outcomes. Statistical techniques are available to deal with missing data; however, available techniques have not been synthesized. A scoping review was conducted to summarize the advantages and disadvantages of identified methods of dealing with missing data from accelerometers. Missing data poses a threat to the validity and interpretation of trials using physical activity data from accelerometry. Imputation using multiple imputation techniques is recommended to deal with missing data and improve the validity and interpretation of studies using accelerometry. Copyright © 2018 Elsevier Inc. All rights reserved.

LinkImpute: Fast and Accurate Genotype Imputation for Nonmodel Organisms.

PubMed

Money, Daniel; Gardner, Kyle; Migicovsky, Zoë; Schwaninger, Heidi; Zhong, Gan-Yuan; Myles, Sean

2015-09-15

Obtaining genome-wide genotype data from a set of individuals is the first step in many genomic studies, including genome-wide association and genomic selection. All genotyping methods suffer from some level of missing data, and genotype imputation can be used to fill in the missing data and improve the power of downstream analyses. Model organisms like human and cattle benefit from high-quality reference genomes and panels of reference genotypes that aid in imputation accuracy. In nonmodel organisms, however, genetic and physical maps often are either of poor quality or are completely absent, and there are no panels of reference genotypes available. There is therefore a need for imputation methods designed specifically for nonmodel organisms in which genomic resources are poorly developed and marker order is unreliable or unknown. Here we introduce LinkImpute, a software package based on a k-nearest neighbor genotype imputation method, LD-kNNi, which is designed for unordered markers. No physical or genetic maps are required, and it is designed to work on unphased genotype data from heterozygous species. It exploits the fact that markers useful for imputation often are not physically close to the missing genotype but rather distributed throughout the genome. Using genotyping-by-sequencing data from diverse and heterozygous accessions of apples, grapes, and maize, we compare LD-kNNi with several genotype imputation methods and show that LD-kNNi is fast, comparable in accuracy to the best-existing methods, and exhibits the least bias in allele frequency estimates. Copyright © 2015 Money et al.

Multiple Imputation of Multilevel Missing Data-Rigor versus Simplicity

ERIC Educational Resources Information Center

Drechsler, Jörg

2015-01-01

Multiple imputation is widely accepted as the method of choice to address item-nonresponse in surveys. However, research on imputation strategies for the hierarchical structures that are typically found in the data in educational contexts is still limited. While a multilevel imputation model should be preferred from a theoretical point of view if…

Covariate Selection for Multilevel Models with Missing Data

PubMed Central

Marino, Miguel; Buxton, Orfeu M.; Li, Yi

2017-01-01

Missing covariate data hampers variable selection in multilevel regression settings. Current variable selection techniques for multiply-imputed data commonly address missingness in the predictors through list-wise deletion and stepwise-selection methods which are problematic. Moreover, most variable selection methods are developed for independent linear regression models and do not accommodate multilevel mixed effects regression models with incomplete covariate data. We develop a novel methodology that is able to perform covariate selection across multiply-imputed data for multilevel random effects models when missing data is present. Specifically, we propose to stack the multiply-imputed data sets from a multiple imputation procedure and to apply a group variable selection procedure through group lasso regularization to assess the overall impact of each predictor on the outcome across the imputed data sets. Simulations confirm the advantageous performance of the proposed method compared with the competing methods. We applied the method to reanalyze the Healthy Directions-Small Business cancer prevention study, which evaluated a behavioral intervention program targeting multiple risk-related behaviors in a working-class, multi-ethnic population. PMID:28239457

How to deal with missing longitudinal data in cost of illness analysis in Alzheimer's disease-suggestions from the GERAS observational study.

PubMed

Belger, Mark; Haro, Josep Maria; Reed, Catherine; Happich, Michael; Kahle-Wrobleski, Kristin; Argimon, Josep Maria; Bruno, Giuseppe; Dodel, Richard; Jones, Roy W; Vellas, Bruno; Wimo, Anders

2016-07-18

Missing data are a common problem in prospective studies with a long follow-up, and the volume, pattern and reasons for missing data may be relevant when estimating the cost of illness. We aimed to evaluate the effects of different methods for dealing with missing longitudinal cost data and for costing caregiver time on total societal costs in Alzheimer's disease (AD). GERAS is an 18-month observational study of costs associated with AD. Total societal costs included patient health and social care costs, and caregiver health and informal care costs. Missing data were classified as missing completely at random (MCAR), missing at random (MAR) or missing not at random (MNAR). Simulation datasets were generated from baseline data with 10-40 % missing total cost data for each missing data mechanism. Datasets were also simulated to reflect the missing cost data pattern at 18 months using MAR and MNAR assumptions. Naïve and multiple imputation (MI) methods were applied to each dataset and results compared with complete GERAS 18-month cost data. Opportunity and replacement cost approaches were used for caregiver time, which was costed with and without supervision included and with time for working caregivers only being costed. Total costs were available for 99.4 % of 1497 patients at baseline. For MCAR datasets, naïve methods performed as well as MI methods. For MAR, MI methods performed better than naïve methods. All imputation approaches were poor for MNAR data. For all approaches, percentage bias increased with missing data volume. For datasets reflecting 18-month patterns, a combination of imputation methods provided more accurate cost estimates (e.g. bias: -1 % vs -6 % for single MI method), although different approaches to costing caregiver time had a greater impact on estimated costs (29-43 % increase over base case estimate). Methods used to impute missing cost data in AD will impact on accuracy of cost estimates although varying approaches to costing informal caregiver time has the greatest impact on total costs. Tailoring imputation methods to the reason for missing data will further our understanding of the best analytical approach for studies involving cost outcomes.

LinkImputeR: user-guided genotype calling and imputation for non-model organisms.

PubMed

Money, Daniel; Migicovsky, Zoë; Gardner, Kyle; Myles, Sean

2017-07-10

Genomic studies such as genome-wide association and genomic selection require genome-wide genotype data. All existing technologies used to create these data result in missing genotypes, which are often then inferred using genotype imputation software. However, existing imputation methods most often make use only of genotypes that are successfully inferred after having passed a certain read depth threshold. Because of this, any read information for genotypes that did not pass the threshold, and were thus set to missing, is ignored. Most genomic studies also choose read depth thresholds and quality filters without investigating their effects on the size and quality of the resulting genotype data. Moreover, almost all genotype imputation methods require ordered markers and are therefore of limited utility in non-model organisms. Here we introduce LinkImputeR, a software program that exploits the read count information that is normally ignored, and makes use of all available DNA sequence information for the purposes of genotype calling and imputation. It is specifically designed for non-model organisms since it requires neither ordered markers nor a reference panel of genotypes. Using next-generation DNA sequence (NGS) data from apple, cannabis and grape, we quantify the effect of varying read count and missingness thresholds on the quantity and quality of genotypes generated from LinkImputeR. We demonstrate that LinkImputeR can increase the number of genotype calls by more than an order of magnitude, can improve genotyping accuracy by several percent and can thus improve the power of downstream analyses. Moreover, we show that the effects of quality and read depth filters can differ substantially between data sets and should therefore be investigated on a per-study basis. By exploiting DNA sequence data that is normally ignored during genotype calling and imputation, LinkImputeR can significantly improve both the quantity and quality of genotype data generated from NGS technologies. It enables the user to quickly and easily examine the effects of varying thresholds and filters on the number and quality of the resulting genotype calls. In this manner, users can decide on thresholds that are most suitable for their purposes. We show that LinkImputeR can significantly augment the value and utility of NGS data sets, especially in non-model organisms with poor genomic resources.

Nuclear Forensics Analysis with Missing and Uncertain Data

DOE PAGES

Langan, Roisin T.; Archibald, Richard K.; Lamberti, Vincent

2015-10-05

We have applied a new imputation-based method for analyzing incomplete data, called Monte Carlo Bayesian Database Generation (MCBDG), to the Spent Fuel Isotopic Composition (SFCOMPO) database. About 60% of the entries are absent for SFCOMPO. The method estimates missing values of a property from a probability distribution created from the existing data for the property, and then generates multiple instances of the completed database for training a machine learning algorithm. Uncertainty in the data is represented by an empirical or an assumed error distribution. The method makes few assumptions about the underlying data, and compares favorably against results obtained bymore » replacing missing information with constant values.« less

Causes and Consequences of Missing Health-Related Quality of Life Assessments in Patients Who Undergo Mechanical Circulatory Support Implantation: Insights From INTERMACS (Interagency Registry for Mechanically Assisted Circulatory Support).

PubMed

Grady, Kathleen L; Jones, Philip G; Cristian-Andrei, Adin; Naftel, David C; Myers, Susan; Dew, Mary Amanda; Idrissi, Katharine; Weidner, Gerdi; Wissman, Sherri A; Kirklin, James K; Spertus, John A

2017-12-01

Missing health-related quality of life (HRQOL) data in longitudinal studies can reduce precision and power and bias results. Using INTERMACS (Interagency Registry for Mechanically Assisted Circulatory Support), we sought to identify factors associated with missing HRQOL data, examine the impact of these factors on estimated HRQOL assuming missing at random missingness, and perform sensitivity analyses to examine missing not at random (MNAR) missingness because of illness severity. INTERMACS patients (n=3248) with a preimplantation profile of 1 (critical cardiogenic shock) or 2 (progressive decline) were assessed with the EQ-5D-3L visual analog scale and Kansas City Cardiomyopathy Questionnaire-12 summary scores pre-implantation and 3 months postoperatively. Mean and median observed and missing at random-imputed HRQOL scores were calculated, followed by sensitivity analyses. Independent factors associated with HRQOL scores and missing HRQOL assessments were determined using multivariable regression. Independent factors associated with preimplantation and 3-month HRQOL scores, and with the likelihood of missing HRQOL assessments, revealed few correlates of HRQOL and missing assessments ( R 2 range, 4.7%-11.9%). For patients with INTERMACS profiles 1 and 2 and INTERMACS profile 1 alone, missing at random-imputed mean and median HRQOL scores were similar to observed scores, before and 3 months after implantation, whereas MNAR-imputed mean scores were lower (≥5 points) at baseline but not at 3 months. We recommend use of sensitivity analyses using an MNAR imputation strategy for longitudinal studies when missingness is attributable to illness severity. Conduct of MNAR sensitivity analyses may be less critical after mechanical circulatory support implant, when there are likely fewer MNAR data. © 2017 American Heart Association, Inc.

Reading Profiles in Multi-Site Data With Missingness.

PubMed

Eckert, Mark A; Vaden, Kenneth I; Gebregziabher, Mulugeta

2018-01-01

Children with reading disability exhibit varied deficits in reading and cognitive abilities that contribute to their reading comprehension problems. Some children exhibit primary deficits in phonological processing, while others can exhibit deficits in oral language and executive functions that affect comprehension. This behavioral heterogeneity is problematic when missing data prevent the characterization of different reading profiles, which often occurs in retrospective data sharing initiatives without coordinated data collection. Here we show that reading profiles can be reliably identified based on Random Forest classification of incomplete behavioral datasets, after the missForest method is used to multiply impute missing values. Results from simulation analyses showed that reading profiles could be accurately classified across degrees of missingness (e.g., ∼5% classification error for 30% missingness across the sample). The application of missForest to a real multi-site dataset with missingness ( n = 924) showed that reading disability profiles significantly and consistently differed in reading and cognitive abilities for cases with and without missing data. The results of validation analyses indicated that the reading profiles (cases with and without missing data) exhibited significant differences for an independent set of behavioral variables that were not used to classify reading profiles. Together, the results show how multiple imputation can be applied to the classification of cases with missing data and can increase the integrity of results from multi-site open access datasets.

Assessing and comparison of different machine learning methods in parent-offspring trios for genotype imputation.

PubMed

Mikhchi, Abbas; Honarvar, Mahmood; Kashan, Nasser Emam Jomeh; Aminafshar, Mehdi

2016-06-21

Genotype imputation is an important tool for prediction of unknown genotypes for both unrelated individuals and parent-offspring trios. Several imputation methods are available and can either employ universal machine learning methods, or deploy algorithms dedicated to infer missing genotypes. In this research the performance of eight machine learning methods: Support Vector Machine, K-Nearest Neighbors, Extreme Learning Machine, Radial Basis Function, Random Forest, AdaBoost, LogitBoost, and TotalBoost compared in terms of the imputation accuracy, computation time and the factors affecting imputation accuracy. The methods employed using real and simulated datasets to impute the un-typed SNPs in parent-offspring trios. The tested methods show that imputation of parent-offspring trios can be accurate. The Random Forest and Support Vector Machine were more accurate than the other machine learning methods. The TotalBoost performed slightly worse than the other methods.The running times were different between methods. The ELM was always most fast algorithm. In case of increasing the sample size, the RBF requires long imputation time.The tested methods in this research can be an alternative for imputation of un-typed SNPs in low missing rate of data. However, it is recommended that other machine learning methods to be used for imputation. Copyright © 2016 Elsevier Ltd. All rights reserved.

Accounting for Dependence Induced by Weighted KNN Imputation in Paired Samples, Motivated by a Colorectal Cancer Study

PubMed Central

Suyundikov, Anvar; Stevens, John R.; Corcoran, Christopher; Herrick, Jennifer; Wolff, Roger K.; Slattery, Martha L.

2015-01-01

Missing data can arise in bioinformatics applications for a variety of reasons, and imputation methods are frequently applied to such data. We are motivated by a colorectal cancer study where miRNA expression was measured in paired tumor-normal samples of hundreds of patients, but data for many normal samples were missing due to lack of tissue availability. We compare the precision and power performance of several imputation methods, and draw attention to the statistical dependence induced by K-Nearest Neighbors (KNN) imputation. This imputation-induced dependence has not previously been addressed in the literature. We demonstrate how to account for this dependence, and show through simulation how the choice to ignore or account for this dependence affects both power and type I error rate control. PMID:25849489

Recurrent Neural Networks for Multivariate Time Series with Missing Values.

PubMed

Che, Zhengping; Purushotham, Sanjay; Cho, Kyunghyun; Sontag, David; Liu, Yan

2018-04-17

Multivariate time series data in practical applications, such as health care, geoscience, and biology, are characterized by a variety of missing values. In time series prediction and other related tasks, it has been noted that missing values and their missing patterns are often correlated with the target labels, a.k.a., informative missingness. There is very limited work on exploiting the missing patterns for effective imputation and improving prediction performance. In this paper, we develop novel deep learning models, namely GRU-D, as one of the early attempts. GRU-D is based on Gated Recurrent Unit (GRU), a state-of-the-art recurrent neural network. It takes two representations of missing patterns, i.e., masking and time interval, and effectively incorporates them into a deep model architecture so that it not only captures the long-term temporal dependencies in time series, but also utilizes the missing patterns to achieve better prediction results. Experiments of time series classification tasks on real-world clinical datasets (MIMIC-III, PhysioNet) and synthetic datasets demonstrate that our models achieve state-of-the-art performance and provide useful insights for better understanding and utilization of missing values in time series analysis.

What are the appropriate methods for analyzing patient-reported outcomes in randomized trials when data are missing?

PubMed

Hamel, J F; Sebille, V; Le Neel, T; Kubis, G; Boyer, F C; Hardouin, J B

2017-12-01

Subjective health measurements using Patient Reported Outcomes (PRO) are increasingly used in randomized trials, particularly for patient groups comparisons. Two main types of analytical strategies can be used for such data: Classical Test Theory (CTT) and Item Response Theory models (IRT). These two strategies display very similar characteristics when data are complete, but in the common case when data are missing, whether IRT or CTT would be the most appropriate remains unknown and was investigated using simulations. We simulated PRO data such as quality of life data. Missing responses to items were simulated as being completely random, depending on an observable covariate or on an unobserved latent trait. The considered CTT-based methods allowed comparing scores using complete-case analysis, personal mean imputations or multiple-imputations based on a two-way procedure. The IRT-based method was the Wald test on a Rasch model including a group covariate. The IRT-based method and the multiple-imputations-based method for CTT displayed the highest observed power and were the only unbiased method whatever the kind of missing data. Online software and Stata® modules compatibles with the innate mi impute suite are provided for performing such analyses. Traditional procedures (listwise deletion and personal mean imputations) should be avoided, due to inevitable problems of biases and lack of power.

Missing data methods for dealing with missing items in quality of life questionnaires. A comparison by simulation of personal mean score, full information maximum likelihood, multiple imputation, and hot deck techniques applied to the SF-36 in the French 2003 decennial health survey.

PubMed

Peyre, Hugo; Leplège, Alain; Coste, Joël

2011-03-01

Missing items are common in quality of life (QoL) questionnaires and present a challenge for research in this field. It remains unclear which of the various methods proposed to deal with missing data performs best in this context. We compared personal mean score, full information maximum likelihood, multiple imputation, and hot deck techniques using various realistic simulation scenarios of item missingness in QoL questionnaires constructed within the framework of classical test theory. Samples of 300 and 1,000 subjects were randomly drawn from the 2003 INSEE Decennial Health Survey (of 23,018 subjects representative of the French population and having completed the SF-36) and various patterns of missing data were generated according to three different item non-response rates (3, 6, and 9%) and three types of missing data (Little and Rubin's "missing completely at random," "missing at random," and "missing not at random"). The missing data methods were evaluated in terms of accuracy and precision for the analysis of one descriptive and one association parameter for three different scales of the SF-36. For all item non-response rates and types of missing data, multiple imputation and full information maximum likelihood appeared superior to the personal mean score and especially to hot deck in terms of accuracy and precision; however, the use of personal mean score was associated with insignificant bias (relative bias <2%) in all studied situations. Whereas multiple imputation and full information maximum likelihood are confirmed as reference methods, the personal mean score appears nonetheless appropriate for dealing with items missing from completed SF-36 questionnaires in most situations of routine use. These results can reasonably be extended to other questionnaires constructed according to classical test theory.

VIGAN: Missing View Imputation with Generative Adversarial Networks.

PubMed

Shang, Chao; Palmer, Aaron; Sun, Jiangwen; Chen, Ko-Shin; Lu, Jin; Bi, Jinbo

2017-01-01

In an era when big data are becoming the norm, there is less concern with the quantity but more with the quality and completeness of the data. In many disciplines, data are collected from heterogeneous sources, resulting in multi-view or multi-modal datasets. The missing data problem has been challenging to address in multi-view data analysis. Especially, when certain samples miss an entire view of data, it creates the missing view problem. Classic multiple imputations or matrix completion methods are hardly effective here when no information can be based on in the specific view to impute data for such samples. The commonly-used simple method of removing samples with a missing view can dramatically reduce sample size, thus diminishing the statistical power of a subsequent analysis. In this paper, we propose a novel approach for view imputation via generative adversarial networks (GANs), which we name by VIGAN. This approach first treats each view as a separate domain and identifies domain-to-domain mappings via a GAN using randomly-sampled data from each view, and then employs a multi-modal denoising autoencoder (DAE) to reconstruct the missing view from the GAN outputs based on paired data across the views. Then, by optimizing the GAN and DAE jointly, our model enables the knowledge integration for domain mappings and view correspondences to effectively recover the missing view. Empirical results on benchmark datasets validate the VIGAN approach by comparing against the state of the art. The evaluation of VIGAN in a genetic study of substance use disorders further proves the effectiveness and usability of this approach in life science.

A review of the handling of missing longitudinal outcome data in clinical trials

PubMed Central

2014-01-01

The aim of this review was to establish the frequency with which trials take into account missingness, and to discover what methods trialists use for adjustment in randomised controlled trials with longitudinal measurements. Failing to address the problems that can arise from missing outcome data can result in misleading conclusions. Missing data should be addressed as a means of a sensitivity analysis of the complete case analysis results. One hundred publications of randomised controlled trials with longitudinal measurements were selected randomly from trial publications from the years 2005 to 2012. Information was extracted from these trials, including whether reasons for dropout were reported, what methods were used for handing the missing data, whether there was any explanation of the methods for missing data handling, and whether a statistician was involved in the analysis. The main focus of the review was on missing data post dropout rather than missing interim data. Of all the papers in the study, 9 (9%) had no missing data. More than half of the papers included in the study failed to make any attempt to explain the reasons for their choice of missing data handling method. Of the papers with clear missing data handling methods, 44 papers (50%) used adequate methods of missing data handling, whereas 30 (34%) of the papers used missing data methods which may not have been appropriate. In the remaining 17 papers (19%), it was difficult to assess the validity of the methods used. An imputation method was used in 18 papers (20%). Multiple imputation methods were introduced in 1987 and are an efficient way of accounting for missing data in general, and yet only 4 papers used these methods. Out of the 18 papers which used imputation, only 7 displayed the results as a sensitivity analysis of the complete case analysis results. 61% of the papers that used an imputation explained the reasons for their chosen method. Just under a third of the papers made no reference to reasons for missing outcome data. There was little consistency in reporting of missing data within longitudinal trials. PMID:24947664

Missing continuous outcomes under covariate dependent missingness in cluster randomised trials

PubMed Central

Diaz-Ordaz, Karla; Bartlett, Jonathan W

2016-01-01

Attrition is a common occurrence in cluster randomised trials which leads to missing outcome data. Two approaches for analysing such trials are cluster-level analysis and individual-level analysis. This paper compares the performance of unadjusted cluster-level analysis, baseline covariate adjusted cluster-level analysis and linear mixed model analysis, under baseline covariate dependent missingness in continuous outcomes, in terms of bias, average estimated standard error and coverage probability. The methods of complete records analysis and multiple imputation are used to handle the missing outcome data. We considered four scenarios, with the missingness mechanism and baseline covariate effect on outcome either the same or different between intervention groups. We show that both unadjusted cluster-level analysis and baseline covariate adjusted cluster-level analysis give unbiased estimates of the intervention effect only if both intervention groups have the same missingness mechanisms and there is no interaction between baseline covariate and intervention group. Linear mixed model and multiple imputation give unbiased estimates under all four considered scenarios, provided that an interaction of intervention and baseline covariate is included in the model when appropriate. Cluster mean imputation has been proposed as a valid approach for handling missing outcomes in cluster randomised trials. We show that cluster mean imputation only gives unbiased estimates when missingness mechanism is the same between the intervention groups and there is no interaction between baseline covariate and intervention group. Multiple imputation shows overcoverage for small number of clusters in each intervention group. PMID:27177885

Missing continuous outcomes under covariate dependent missingness in cluster randomised trials.

PubMed

Hossain, Anower; Diaz-Ordaz, Karla; Bartlett, Jonathan W

2017-06-01

Attrition is a common occurrence in cluster randomised trials which leads to missing outcome data. Two approaches for analysing such trials are cluster-level analysis and individual-level analysis. This paper compares the performance of unadjusted cluster-level analysis, baseline covariate adjusted cluster-level analysis and linear mixed model analysis, under baseline covariate dependent missingness in continuous outcomes, in terms of bias, average estimated standard error and coverage probability. The methods of complete records analysis and multiple imputation are used to handle the missing outcome data. We considered four scenarios, with the missingness mechanism and baseline covariate effect on outcome either the same or different between intervention groups. We show that both unadjusted cluster-level analysis and baseline covariate adjusted cluster-level analysis give unbiased estimates of the intervention effect only if both intervention groups have the same missingness mechanisms and there is no interaction between baseline covariate and intervention group. Linear mixed model and multiple imputation give unbiased estimates under all four considered scenarios, provided that an interaction of intervention and baseline covariate is included in the model when appropriate. Cluster mean imputation has been proposed as a valid approach for handling missing outcomes in cluster randomised trials. We show that cluster mean imputation only gives unbiased estimates when missingness mechanism is the same between the intervention groups and there is no interaction between baseline covariate and intervention group. Multiple imputation shows overcoverage for small number of clusters in each intervention group.

A comparison of multiple imputation methods for incomplete longitudinal binary data.

PubMed

Yamaguchi, Yusuke; Misumi, Toshihiro; Maruo, Kazushi

2018-01-01

Longitudinal binary data are commonly encountered in clinical trials. Multiple imputation is an approach for getting a valid estimation of treatment effects under an assumption of missing at random mechanism. Although there are a variety of multiple imputation methods for the longitudinal binary data, a limited number of researches have reported on relative performances of the methods. Moreover, when focusing on the treatment effect throughout a period that has often been used in clinical evaluations of specific disease areas, no definite investigations comparing the methods have been available. We conducted an extensive simulation study to examine comparative performances of six multiple imputation methods available in the SAS MI procedure for longitudinal binary data, where two endpoints of responder rates at a specified time point and throughout a period were assessed. The simulation study suggested that results from naive approaches of a single imputation with non-responders and a complete case analysis could be very sensitive against missing data. The multiple imputation methods using a monotone method and a full conditional specification with a logistic regression imputation model were recommended for obtaining unbiased and robust estimations of the treatment effect. The methods were illustrated with data from a mental health research.

A suggested approach for imputation of missing dietary data for young children in daycare.

PubMed

Stevens, June; Ou, Fang-Shu; Truesdale, Kimberly P; Zeng, Donglin; Vaughn, Amber E; Pratt, Charlotte; Ward, Dianne S

2015-01-01

Parent-reported 24-h diet recalls are an accepted method of estimating intake in young children. However, many children eat while at childcare making accurate proxy reports by parents difficult. The goal of this study was to demonstrate a method to impute missing weekday lunch and daytime snack nutrient data for daycare children and to explore the concurrent predictive and criterion validity of the method. Data were from children aged 2-5 years in the My Parenting SOS project (n=308; 870 24-h diet recalls). Mixed models were used to simultaneously predict breakfast, dinner, and evening snacks (B+D+ES); lunch; and daytime snacks for all children after adjusting for age, sex, and body mass index (BMI). From these models, we imputed the missing weekday daycare lunches by interpolation using the mean lunch to B+D+ES [L/(B+D+ES)] ratio among non-daycare children on weekdays and the L/(B+D+ES) ratio for all children on weekends. Daytime snack data were used to impute snacks. The reported mean (± standard deviation) weekday intake was lower for daycare children [725 (±324) kcal] compared to non-daycare children [1,048 (±463) kcal]. Weekend intake for all children was 1,173 (±427) kcal. After imputation, weekday caloric intake for daycare children was 1,230 (±409) kcal. Daily intakes that included imputed data were associated with age and sex but not with BMI. This work indicates that imputation is a promising method for improving the precision of daily nutrient data from young children.

Ascertainment bias from imputation methods evaluation in wheat.

PubMed

Brandariz, Sofía P; González Reymúndez, Agustín; Lado, Bettina; Malosetti, Marcos; Garcia, Antonio Augusto Franco; Quincke, Martín; von Zitzewitz, Jarislav; Castro, Marina; Matus, Iván; Del Pozo, Alejandro; Castro, Ariel J; Gutiérrez, Lucía

2016-10-04

Whole-genome genotyping techniques like Genotyping-by-sequencing (GBS) are being used for genetic studies such as Genome-Wide Association (GWAS) and Genomewide Selection (GS), where different strategies for imputation have been developed. Nevertheless, imputation error may lead to poor performance (i.e. smaller power or higher false positive rate) when complete data is not required as it is for GWAS, and each marker is taken at a time. The aim of this study was to compare the performance of GWAS analysis for Quantitative Trait Loci (QTL) of major and minor effect using different imputation methods when no reference panel is available in a wheat GBS panel. In this study, we compared the power and false positive rate of dissecting quantitative traits for imputed and not-imputed marker score matrices in: (1) a complete molecular marker barley panel array, and (2) a GBS wheat panel with missing data. We found that there is an ascertainment bias in imputation method comparisons. Simulating over a complete matrix and creating missing data at random proved that imputation methods have a poorer performance. Furthermore, we found that when QTL were simulated with imputed data, the imputation methods performed better than the not-imputed ones. On the other hand, when QTL were simulated with not-imputed data, the not-imputed method and one of the imputation methods performed better for dissecting quantitative traits. Moreover, larger differences between imputation methods were detected for QTL of major effect than QTL of minor effect. We also compared the different marker score matrices for GWAS analysis in a real wheat phenotype dataset, and we found minimal differences indicating that imputation did not improve the GWAS performance when a reference panel was not available. Poorer performance was found in GWAS analysis when an imputed marker score matrix was used, no reference panel is available, in a wheat GBS panel.

Statistical primer: how to deal with missing data in scientific research?

PubMed

Papageorgiou, Grigorios; Grant, Stuart W; Takkenberg, Johanna J M; Mokhles, Mostafa M

2018-05-10

Missing data are a common challenge encountered in research which can compromise the results of statistical inference when not handled appropriately. This paper aims to introduce basic concepts of missing data to a non-statistical audience, list and compare some of the most popular approaches for handling missing data in practice and provide guidelines and recommendations for dealing with and reporting missing data in scientific research. Complete case analysis and single imputation are simple approaches for handling missing data and are popular in practice, however, in most cases they are not guaranteed to provide valid inferences. Multiple imputation is a robust and general alternative which is appropriate for data missing at random, surpassing the disadvantages of the simpler approaches, but should always be conducted with care. The aforementioned approaches are illustrated and compared in an example application using Cox regression.

Multiple imputation strategies for zero-inflated cost data in economic evaluations: which method works best?

PubMed

MacNeil Vroomen, Janet; Eekhout, Iris; Dijkgraaf, Marcel G; van Hout, Hein; de Rooij, Sophia E; Heymans, Martijn W; Bosmans, Judith E

2016-11-01

Cost and effect data often have missing data because economic evaluations are frequently added onto clinical studies where cost data are rarely the primary outcome. The objective of this article was to investigate which multiple imputation strategy is most appropriate to use for missing cost-effectiveness data in a randomized controlled trial. Three incomplete data sets were generated from a complete reference data set with 17, 35 and 50 % missing data in effects and costs. The strategies evaluated included complete case analysis (CCA), multiple imputation with predictive mean matching (MI-PMM), MI-PMM on log-transformed costs (log MI-PMM), and a two-step MI. Mean cost and effect estimates, standard errors and incremental net benefits were compared with the results of the analyses on the complete reference data set. The CCA, MI-PMM, and the two-step MI strategy diverged from the results for the reference data set when the amount of missing data increased. In contrast, the estimates of the Log MI-PMM strategy remained stable irrespective of the amount of missing data. MI provided better estimates than CCA in all scenarios. With low amounts of missing data the MI strategies appeared equivalent but we recommend using the log MI-PMM with missing data greater than 35 %.

Considerations of multiple imputation approaches for handling missing data in clinical trials.

PubMed

Quan, Hui; Qi, Li; Luo, Xiaodong; Darchy, Loic

2018-07-01

Missing data exist in all clinical trials and missing data issue is a very serious issue in terms of the interpretability of the trial results. There is no universally applicable solution for all missing data problems. Methods used for handling missing data issue depend on the circumstances particularly the assumptions on missing data mechanisms. In recent years, if the missing at random mechanism cannot be assumed, conservative approaches such as the control-based and returning to baseline multiple imputation approaches are applied for dealing with the missing data issues. In this paper, we focus on the variability in data analysis of these approaches. As demonstrated by examples, the choice of the variability can impact the conclusion of the analysis. Besides the methods for continuous endpoints, we also discuss methods for binary and time to event endpoints as well as consideration for non-inferiority assessment. Copyright © 2018. Published by Elsevier Inc.

DOE Office of Scientific and Technical Information (OSTI.GOV)

Langan, Roisin T.; Archibald, Richard K.; Lamberti, Vincent

We have applied a new imputation-based method for analyzing incomplete data, called Monte Carlo Bayesian Database Generation (MCBDG), to the Spent Fuel Isotopic Composition (SFCOMPO) database. About 60% of the entries are absent for SFCOMPO. The method estimates missing values of a property from a probability distribution created from the existing data for the property, and then generates multiple instances of the completed database for training a machine learning algorithm. Uncertainty in the data is represented by an empirical or an assumed error distribution. The method makes few assumptions about the underlying data, and compares favorably against results obtained bymore » replacing missing information with constant values.« less

TRANSPOSABLE REGULARIZED COVARIANCE MODELS WITH AN APPLICATION TO MISSING DATA IMPUTATION

PubMed Central

Allen, Genevera I.; Tibshirani, Robert

2015-01-01

Missing data estimation is an important challenge with high-dimensional data arranged in the form of a matrix. Typically this data matrix is transposable, meaning that either the rows, columns or both can be treated as features. To model transposable data, we present a modification of the matrix-variate normal, the mean-restricted matrix-variate normal, in which the rows and columns each have a separate mean vector and covariance matrix. By placing additive penalties on the inverse covariance matrices of the rows and columns, these so called transposable regularized covariance models allow for maximum likelihood estimation of the mean and non-singular covariance matrices. Using these models, we formulate EM-type algorithms for missing data imputation in both the multivariate and transposable frameworks. We present theoretical results exploiting the structure of our transposable models that allow these models and imputation methods to be applied to high-dimensional data. Simulations and results on microarray data and the Netflix data show that these imputation techniques often outperform existing methods and offer a greater degree of flexibility. PMID:26877823

TRANSPOSABLE REGULARIZED COVARIANCE MODELS WITH AN APPLICATION TO MISSING DATA IMPUTATION.

PubMed

Allen, Genevera I; Tibshirani, Robert

2010-06-01

Missing data estimation is an important challenge with high-dimensional data arranged in the form of a matrix. Typically this data matrix is transposable , meaning that either the rows, columns or both can be treated as features. To model transposable data, we present a modification of the matrix-variate normal, the mean-restricted matrix-variate normal , in which the rows and columns each have a separate mean vector and covariance matrix. By placing additive penalties on the inverse covariance matrices of the rows and columns, these so called transposable regularized covariance models allow for maximum likelihood estimation of the mean and non-singular covariance matrices. Using these models, we formulate EM-type algorithms for missing data imputation in both the multivariate and transposable frameworks. We present theoretical results exploiting the structure of our transposable models that allow these models and imputation methods to be applied to high-dimensional data. Simulations and results on microarray data and the Netflix data show that these imputation techniques often outperform existing methods and offer a greater degree of flexibility.

Selection-Fusion Approach for Classification of Datasets with Missing Values

PubMed Central

Ghannad-Rezaie, Mostafa; Soltanian-Zadeh, Hamid; Ying, Hao; Dong, Ming

2010-01-01

This paper proposes a new approach based on missing value pattern discovery for classifying incomplete data. This approach is particularly designed for classification of datasets with a small number of samples and a high percentage of missing values where available missing value treatment approaches do not usually work well. Based on the pattern of the missing values, the proposed approach finds subsets of samples for which most of the features are available and trains a classifier for each subset. Then, it combines the outputs of the classifiers. Subset selection is translated into a clustering problem, allowing derivation of a mathematical framework for it. A trade off is established between the computational complexity (number of subsets) and the accuracy of the overall classifier. To deal with this trade off, a numerical criterion is proposed for the prediction of the overall performance. The proposed method is applied to seven datasets from the popular University of California, Irvine data mining archive and an epilepsy dataset from Henry Ford Hospital, Detroit, Michigan (total of eight datasets). Experimental results show that classification accuracy of the proposed method is superior to those of the widely used multiple imputations method and four other methods. They also show that the level of superiority depends on the pattern and percentage of missing values. PMID:20212921

Treating pre-instrumental data as "missing" data: using a tree-ring-based paleoclimate record and imputations to reconstruct streamflow in the Missouri River Basin

NASA Astrophysics Data System (ADS)

Ho, M. W.; Lall, U.; Cook, E. R.

2015-12-01

Advances in paleoclimatology in the past few decades have provided opportunities to expand the temporal perspective of the hydrological and climatological variability across the world. The North American region is particularly fortunate in this respect where a relatively dense network of high resolution paleoclimate proxy records have been assembled. One such network is the annually-resolved Living Blended Drought Atlas (LBDA): a paleoclimate reconstruction of the Palmer Drought Severity Index (PDSI) that covers North America on a 0.5° × 0.5° grid based on tree-ring chronologies. However, the use of the LBDA to assess North American streamflow variability requires a model by which streamflow may be reconstructed. Paleoclimate reconstructions have typically used models that first seek to quantify the relationship between the paleoclimate variable and the environmental variable of interest before extrapolating the relationship back in time. In contrast, the pre-instrumental streamflow is here considered as "missing" data. A method of imputing the "missing" streamflow data, prior to the instrumental record, is applied through multiple imputation using chained equations for streamflow in the Missouri River Basin. In this method, the distribution of the instrumental streamflow and LBDA is used to estimate sets of plausible values for the "missing" streamflow data resulting in a ~600 year-long streamflow reconstruction. Past research into external climate forcings, oceanic-atmospheric variability and its teleconnections, and assessments of rare multi-centennial instrumental records demonstrate that large temporal oscillations in hydrological conditions are unlikely to be captured in most instrumental records. The reconstruction of multi-centennial records of streamflow will enable comprehensive assessments of current and future water resource infrastructure and operations under the existing scope of natural climate variability.

Genetic variance estimation with imputed variants finds negligible missing heritability for human height and body mass index

PubMed Central

Yang, Jian; Bakshi, Andrew; Zhu, Zhihong; Hemani, Gibran; Vinkhuyzen, Anna A.E.; Lee, Sang Hong; Robinson, Matthew R.; Perry, John R.B.; Nolte, Ilja M.; van Vliet-Ostaptchouk, Jana V.; Snieder, Harold; Esko, Tonu; Milani, Lili; Mägi, Reedik; Metspalu, Andres; Hamsten, Anders; Magnusson, Patrik K.E.; Pedersen, Nancy L.; Ingelsson, Erik; Soranzo, Nicole; Keller, Matthew C.; Wray, Naomi R.; Goddard, Michael E.; Visscher, Peter M.

2015-01-01

We propose a method (GREML-LDMS) to estimate heritability for human complex traits in unrelated individuals using whole-genome sequencing (WGS) data. We demonstrate using simulations based on WGS data that ~97% and ~68% of variation at common and rare variants, respectively, can be captured by imputation. Using the GREML-LDMS method, we estimate from 44,126 unrelated individuals that all ~17M imputed variants explain 56% (s.e. = 2.3%) of variance for height and 27% (s.e. = 2.5%) for body mass index (BMI), and find evidence that height- and BMI-associated variants have been under natural selection. Considering imperfect tagging of imputation and potential overestimation of heritability from previous family-based studies, heritability is likely to be 60–70% for height and 30–40% for BMI. Therefore, missing heritability is small for both traits. For further gene discovery of complex traits, a design with SNP arrays followed by imputation is more cost-effective than WGS at current prices. PMID:26323059

Data Imputation in Epistatic MAPs by Network-Guided Matrix Completion

PubMed Central

Žitnik, Marinka; Zupan, Blaž

2015-01-01

Abstract Epistatic miniarray profile (E-MAP) is a popular large-scale genetic interaction discovery platform. E-MAPs benefit from quantitative output, which makes it possible to detect subtle interactions with greater precision. However, due to the limits of biotechnology, E-MAP studies fail to measure genetic interactions for up to 40% of gene pairs in an assay. Missing measurements can be recovered by computational techniques for data imputation, in this way completing the interaction profiles and enabling downstream analysis algorithms that could otherwise be sensitive to missing data values. We introduce a new interaction data imputation method called network-guided matrix completion (NG-MC). The core part of NG-MC is low-rank probabilistic matrix completion that incorporates prior knowledge presented as a collection of gene networks. NG-MC assumes that interactions are transitive, such that latent gene interaction profiles inferred by NG-MC depend on the profiles of their direct neighbors in gene networks. As the NG-MC inference algorithm progresses, it propagates latent interaction profiles through each of the networks and updates gene network weights toward improved prediction. In a study with four different E-MAP data assays and considered protein–protein interaction and gene ontology similarity networks, NG-MC significantly surpassed existing alternative techniques. Inclusion of information from gene networks also allowed NG-MC to predict interactions for genes that were not included in original E-MAP assays, a task that could not be considered by current imputation approaches. PMID:25658751

Is structural stigma's effect on the mortality of sexual minorities robust? A failure to replicate the results of a published study.

PubMed

Regnerus, Mark

2017-09-01

The study of stigma's influence on health has surged in recent years. Hatzenbuehler et al.'s (2014) study of structural stigma's effect on mortality revealed an average of 12 years' shorter life expectancy for sexual minorities who resided in communities thought to exhibit high levels of anti-gay prejudice, using data from the 1988-2002 administrations of the US General Social Survey linked to mortality outcome data in the 2008 National Death Index. In the original study, the key predictor variable (structural stigma) led to results suggesting the profound negative influence of structural stigma on the mortality of sexual minorities. Attempts to replicate the study, in order to explore alternative hypotheses, repeatedly failed to generate the original study's key finding on structural stigma. Efforts to discern the source of the disparity in results revealed complications in the multiple imputation process for missing values of the components of structural stigma. This prompted efforts at replication using 10 different imputation approaches. Efforts to replicate Hatzenbuehler et al.'s (2014) key finding on structural stigma's notable influence on the premature mortality of sexual minorities, including a more refined imputation strategy than described in the original study, failed. No data imputation approach yielded parameters that supported the original study's conclusions. Alternative hypotheses, which originally motivated the present study, revealed little new information. Ten different approaches to multiple imputation of missing data yielded none in which the effect of structural stigma on the mortality of sexual minorities was statistically significant. Minimally, the original study's structural stigma variable (and hence its key result) is so sensitive to subjective measurement decisions as to be rendered unreliable. Copyright © 2016 The Author. Published by Elsevier Ltd.. All rights reserved.

Probability genotype imputation method and integrated weighted lasso for QTL identification.

PubMed

Demetrashvili, Nino; Van den Heuvel, Edwin R; Wit, Ernst C

2013-12-30

Many QTL studies have two common features: (1) often there is missing marker information, (2) among many markers involved in the biological process only a few are causal. In statistics, the second issue falls under the headings "sparsity" and "causal inference". The goal of this work is to develop a two-step statistical methodology for QTL mapping for markers with binary genotypes. The first step introduces a novel imputation method for missing genotypes. Outcomes of the proposed imputation method are probabilities which serve as weights to the second step, namely in weighted lasso. The sparse phenotype inference is employed to select a set of predictive markers for the trait of interest. Simulation studies validate the proposed methodology under a wide range of realistic settings. Furthermore, the methodology outperforms alternative imputation and variable selection methods in such studies. The methodology was applied to an Arabidopsis experiment, containing 69 markers for 165 recombinant inbred lines of a F8 generation. The results confirm previously identified regions, however several new markers are also found. On the basis of the inferred ROC behavior these markers show good potential for being real, especially for the germination trait Gmax. Our imputation method shows higher accuracy in terms of sensitivity and specificity compared to alternative imputation method. Also, the proposed weighted lasso outperforms commonly practiced multiple regression as well as the traditional lasso and adaptive lasso with three weighting schemes. This means that under realistic missing data settings this methodology can be used for QTL identification.

Predictors of clinical outcome in pediatric oligodendroglioma: meta-analysis of individual patient data and multiple imputation.

PubMed

Wang, Kevin Yuqi; Vankov, Emilian R; Lin, Doris Da May

2018-02-01

OBJECTIVE Oligodendroglioma is a rare primary CNS neoplasm in the pediatric population, and only a limited number of studies in the literature have characterized this entity. Existing studies are limited by small sample sizes and discrepant interstudy findings in identified prognostic factors. In the present study, the authors aimed to increase the statistical power in evaluating for potential prognostic factors of pediatric oligodendrogliomas and sought to reconcile the discrepant findings present among existing studies by performing an individual-patient-data (IPD) meta-analysis and using multiple imputation to address data not directly available from existing studies. METHODS A systematic search was performed, and all studies found to be related to pediatric oligodendrogliomas and associated outcomes were screened for inclusion. Each study was searched for specific demographic and clinical characteristics of each patient and the duration of event-free survival (EFS) and overall survival (OS). Given that certain demographic and clinical information of each patient was not available within all studies, a multivariable imputation via chained equations model was used to impute missing data after the mechanism of missing data was determined. The primary end points of interest were hazard ratios for EFS and OS, as calculated by the Cox proportional-hazards model. Both univariate and multivariate analyses were performed. The multivariate model was adjusted for age, sex, tumor grade, mixed pathologies, extent of resection, chemotherapy, radiation therapy, tumor location, and initial presentation. A p value of less than 0.05 was considered statistically significant. RESULTS A systematic search identified 24 studies with both time-to-event and IPD characteristics available, and a total of 237 individual cases were available for analysis. A median of 19.4% of the values among clinical, demographic, and outcome variables in the compiled 237 cases were missing. Multivariate Cox regression analysis revealed subtotal resection (p = 0.007 [EFS] and 0.043 [OS]), initial presentation of headache (p = 0.006 [EFS] and 0.004 [OS]), mixed pathologies (p = 0.005 [EFS] and 0.049 [OS]), and location of the tumor in the parietal lobe (p = 0.044 [EFS] and 0.030 [OS]) to be significant predictors of tumor progression or recurrence and death. CONCLUSIONS The use of IPD meta-analysis provides a valuable means for increasing statistical power in investigations of disease entities with a very low incidence. Missing data are common in research, and multiple imputation is a flexible and valid approach for addressing this issue, when it is used conscientiously. Undergoing subtotal resection, having a parietal tumor, having tumors with mixed pathologies, and suffering headaches at the time of diagnosis portended a poorer prognosis in pediatric patients with oligodendroglioma.

Data estimation and prediction for natural resources public data

Treesearch

Hans T. Schreuder; Robin M. Reich

1998-01-01

A key product of both Forest Inventory and Analysis (FIA) of the USDA Forest Service and the Natural Resources Inventory (NRI) of the Natural Resources Conservation Service is a scientific data base that should be defensible in court. Multiple imputation procedures (MIPs) have been proposed both for missing value estimation and prediction of non-remeasured cells in...

Handling missing values in the MDS-UPDRS.

PubMed

Goetz, Christopher G; Luo, Sheng; Wang, Lu; Tilley, Barbara C; LaPelle, Nancy R; Stebbins, Glenn T

2015-10-01

This study was undertaken to define the number of missing values permissible to render valid total scores for each Movement Disorder Society Unified Parkinson's Disease Rating Scale (MDS-UPDRS) part. To handle missing values, imputation strategies serve as guidelines to reject an incomplete rating or create a surrogate score. We tested a rigorous, scale-specific, data-based approach to handling missing values for the MDS-UPDRS. From two large MDS-UPDRS datasets, we sequentially deleted item scores, either consistently (same items) or randomly (different items) across all subjects. Lin's Concordance Correlation Coefficient (CCC) compared scores calculated without missing values with prorated scores based on sequentially increasing missing values. The maximal number of missing values retaining a CCC greater than 0.95 determined the threshold for rendering a valid prorated score. A second confirmatory sample was selected from the MDS-UPDRS international translation program. To provide valid part scores applicable across all Hoehn and Yahr (H&Y) stages when the same items are consistently missing, one missing item from Part I, one from Part II, three from Part III, but none from Part IV can be allowed. To provide valid part scores applicable across all H&Y stages when random item entries are missing, one missing item from Part I, two from Part II, seven from Part III, but none from Part IV can be allowed. All cutoff values were confirmed in the validation sample. These analyses are useful for constructing valid surrogate part scores for MDS-UPDRS when missing items fall within the identified threshold and give scientific justification for rejecting partially completed ratings that fall below the threshold. © 2015 International Parkinson and Movement Disorder Society.

Multiple imputation as one tool to provide longitudinal databases for modelling human height and weight development.

PubMed

Aßmann, C

2016-06-01

Besides large efforts regarding field work, provision of valid databases requires statistical and informational infrastructure to enable long-term access to longitudinal data sets on height, weight and related issues. To foster use of longitudinal data sets within the scientific community, provision of valid databases has to address data-protection regulations. It is, therefore, of major importance to hinder identifiability of individuals from publicly available databases. To reach this goal, one possible strategy is to provide a synthetic database to the public allowing for pretesting strategies for data analysis. The synthetic databases can be established using multiple imputation tools. Given the approval of the strategy, verification is based on the original data. Multiple imputation by chained equations is illustrated to facilitate provision of synthetic databases as it allows for capturing a wide range of statistical interdependencies. Also missing values, typically occurring within longitudinal databases for reasons of item non-response, can be addressed via multiple imputation when providing databases. The provision of synthetic databases using multiple imputation techniques is one possible strategy to ensure data protection, increase visibility of longitudinal databases and enhance the analytical potential.

Best practices for missing data management in counseling psychology.

PubMed

Schlomer, Gabriel L; Bauman, Sheri; Card, Noel A

2010-01-01

This article urges counseling psychology researchers to recognize and report how missing data are handled, because consumers of research cannot accurately interpret findings without knowing the amount and pattern of missing data or the strategies that were used to handle those data. Patterns of missing data are reviewed, and some of the common strategies for dealing with them are described. The authors provide an illustration in which data were simulated and evaluate 3 methods of handling missing data: mean substitution, multiple imputation, and full information maximum likelihood. Results suggest that mean substitution is a poor method for handling missing data, whereas both multiple imputation and full information maximum likelihood are recommended alternatives to this approach. The authors suggest that researchers fully consider and report the amount and pattern of missing data and the strategy for handling those data in counseling psychology research and that editors advise researchers of this expectation.

Missing CD4+ cell response in randomized clinical trials of maraviroc and dolutegravir.

PubMed

Cuffe, Robert; Barnett, Carly; Granier, Catherine; Machida, Mitsuaki; Wang, Cunshan; Roger, James

2015-10-01

Missing data can compromise inferences from clinical trials, yet the topic has received little attention in the clinical trial community. Shortcomings in commonly used methods used to analyze studies with missing data (complete case, last- or baseline-observation carried forward) have been highlighted in a recent Food and Drug Administration-sponsored report. This report recommends how to mitigate the issues associated with missing data. We present an example of the proposed concepts using data from recent clinical trials. CD4+ cell count data from the previously reported SINGLE and MOTIVATE studies of dolutegravir and maraviroc were analyzed using a variety of statistical methods to explore the impact of missing data. Four methodologies were used: complete case analysis, simple imputation, mixed models for repeated measures, and multiple imputation. We compared the sensitivity of conclusions to the volume of missing data and to the assumptions underpinning each method. Rates of missing data were greater in the MOTIVATE studies (35%-68% premature withdrawal) than in SINGLE (12%-20%). The sensitivity of results to assumptions about missing data was related to volume of missing data. Estimates of treatment differences by various analysis methods ranged across a 61 cells/mm3 window in MOTIVATE and a 22 cells/mm3 window in SINGLE. Where missing data are anticipated, analyses require robust statistical and clinical debate of the necessary but unverifiable underlying statistical assumptions. Multiple imputation makes these assumptions transparent, can accommodate a broad range of scenarios, and is a natural analysis for clinical trials in HIV with missing data.

On the value of Mendelian laws of segregation in families: data quality control, imputation and beyond

PubMed Central

Blue, Elizabeth Marchani; Sun, Lei; Tintle, Nathan L.; Wijsman, Ellen M.

2014-01-01

When analyzing family data, we dream of perfectly informative data, even whole genome sequences (WGS) for all family members. Reality intervenes, and we find next-generation sequence (NGS) data have error, and are often too expensive or impossible to collect on everyone. Genetic Analysis Workshop 18 groups “Quality Control” and “Dropping WGS through families using GWAS framework” focused on finding, correcting, and using errors within the available sequence and family data, developing methods to infer and analyze missing sequence data among relatives, and testing for linkage and association with simulated blood pressure. We found that single nucleotide polymorphisms, NGS, and imputed data are generally concordant, but that errors are particularly likely at rare variants, homozygous genotypes, within regions with repeated sequences or structural variants, and within sequence data imputed from unrelateds. Admixture complicated identification of cryptic relatedness, but information from Mendelian transmission improved error detection and provided an estimate of the de novo mutation rate. Both genotype and pedigree errors had an adverse effect on subsequent analyses. Computationally fast rules-based imputation was accurate, but could not cover as many loci or subjects as more computationally demanding probability-based methods. Incorporating population-level data into pedigree-based imputation methods improved results. Observed data outperformed imputed data in association testing, but imputed data were also useful. We discuss the strengths and weaknesses of existing methods, and suggest possible future directions. Topics include improving communication between those performing data collection and analysis, establishing thresholds for and improving imputation quality, and incorporating error into imputation and analytical models. PMID:25112184

Sensitivity to imputation models and assumptions in receiver operating characteristic analysis with incomplete data

PubMed Central

Karakaya, Jale; Karabulut, Erdem; Yucel, Recai M.

2015-01-01

Modern statistical methods using incomplete data have been increasingly applied in a wide variety of substantive problems. Similarly, receiver operating characteristic (ROC) analysis, a method used in evaluating diagnostic tests or biomarkers in medical research, has also been increasingly popular problem in both its development and application. While missing-data methods have been applied in ROC analysis, the impact of model mis-specification and/or assumptions (e.g. missing at random) underlying the missing data has not been thoroughly studied. In this work, we study the performance of multiple imputation (MI) inference in ROC analysis. Particularly, we investigate parametric and non-parametric techniques for MI inference under common missingness mechanisms. Depending on the coherency of the imputation model with the underlying data generation mechanism, our results show that MI generally leads to well-calibrated inferences under ignorable missingness mechanisms. PMID:26379316

Filling gaps in large ecological databases: consequences for the study of global-scale plant functional trait patterns

NASA Astrophysics Data System (ADS)

Schrodt, Franziska; Shan, Hanhuai; Fazayeli, Farideh; Karpatne, Anuj; Kattge, Jens; Banerjee, Arindam; Reichstein, Markus; Reich, Peter

2013-04-01

With the advent of remotely sensed data and coordinated efforts to create global databases, the ecological community has progressively become more data-intensive. However, in contrast to other disciplines, statistical ways of handling these large data sets, especially the gaps which are inherent to them, are lacking. Widely used theoretical approaches, for example model averaging based on Akaike's information criterion (AIC), are sensitive to missing values. Yet, the most common way of handling sparse matrices - the deletion of cases with missing data (complete case analysis) - is known to severely reduce statistical power as well as inducing biased parameter estimates. In order to address these issues, we present novel approaches to gap filling in large ecological data sets using matrix factorization techniques. Factorization based matrix completion was developed in a recommender system context and has since been widely used to impute missing data in fields outside the ecological community. Here, we evaluate the effectiveness of probabilistic matrix factorization techniques for imputing missing data in ecological matrices using two imputation techniques. Hierarchical Probabilistic Matrix Factorization (HPMF) effectively incorporates hierarchical phylogenetic information (phylogenetic group, family, genus, species and individual plant) into the trait imputation. Advanced Hierarchical Probabilistic Matrix Factorization (aHPMF) on the other hand includes climate and soil information into the matrix factorization by regressing the environmental variables against residuals of the HPMF. One unique opportunity opened up by aHPMF is out-of-sample prediction, where traits can be predicted for specific species at locations different to those sampled in the past. This has potentially far-reaching consequences for the study of global-scale plant functional trait patterns. We test the accuracy and effectiveness of HPMF and aHPMF in filling sparse matrices, using the TRY database of plant functional traits (http://www.try-db.org). TRY is one of the largest global compilations of plant trait databases (750 traits of 1 million plants), encompassing data on morphological, anatomical, biochemical, phenological and physiological features of plants. However, despite of unprecedented coverage, the TRY database is still very sparse, severely limiting joint trait analyses. Plant traits are the key to understanding how plants as primary producers adjust to changes in environmental conditions and in turn influence them. Forming the basis for Dynamic Global Vegetation Models (DGVMs), plant traits are also fundamental in global change studies for predicting future ecosystem changes. It is thus imperative that missing data is imputed in as accurate and precise a way as possible. In this study, we show the advantages and disadvantages of applying probabilistic matrix factorization techniques in incorporating hierarchical and environmental information for the prediction of missing plant traits as compared to conventional imputation techniques such as the complete case and mean approaches. We will discuss the implications of using gap-filled data for global-scale studies of plant functional trait - environment relationship as opposed to the above-mentioned conventional techniques, using examples of out-of-sample predictions of foliar Nitrogen across several species' ranges and biomes.

Can We Spin Straw Into Gold? An Evaluation of Immigrant Legal Status Imputation Approaches

PubMed Central

Van Hook, Jennifer; Bachmeier, James D.; Coffman, Donna; Harel, Ofer

2014-01-01

Researchers have developed logical, demographic, and statistical strategies for imputing immigrants’ legal status, but these methods have never been empirically assessed. We used Monte Carlo simulations to test whether, and under what conditions, legal status imputation approaches yield unbiased estimates of the association of unauthorized status with health insurance coverage. We tested five methods under a range of missing data scenarios. Logical and demographic imputation methods yielded biased estimates across all missing data scenarios. Statistical imputation approaches yielded unbiased estimates only when unauthorized status was jointly observed with insurance coverage; when this condition was not met, these methods overestimated insurance coverage for unauthorized relative to legal immigrants. We next showed how bias can be reduced by incorporating prior information about unauthorized immigrants. Finally, we demonstrated the utility of the best-performing statistical method for increasing power. We used it to produce state/regional estimates of insurance coverage among unauthorized immigrants in the Current Population Survey, a data source that contains no direct measures of immigrants’ legal status. We conclude that commonly employed legal status imputation approaches are likely to produce biased estimates, but data and statistical methods exist that could substantially reduce these biases. PMID:25511332

Instrumental Variable Methods for Continuous Outcomes That Accommodate Nonignorable Missing Baseline Values.

PubMed

Ertefaie, Ashkan; Flory, James H; Hennessy, Sean; Small, Dylan S

2017-06-15

Instrumental variable (IV) methods provide unbiased treatment effect estimation in the presence of unmeasured confounders under certain assumptions. To provide valid estimates of treatment effect, treatment effect confounders that are associated with the IV (IV-confounders) must be included in the analysis, and not including observations with missing values may lead to bias. Missing covariate data are particularly problematic when the probability that a value is missing is related to the value itself, which is known as nonignorable missingness. In such cases, imputation-based methods are biased. Using health-care provider preference as an IV method, we propose a 2-step procedure with which to estimate a valid treatment effect in the presence of baseline variables with nonignorable missing values. First, the provider preference IV value is estimated by performing a complete-case analysis using a random-effects model that includes IV-confounders. Second, the treatment effect is estimated using a 2-stage least squares IV approach that excludes IV-confounders with missing values. Simulation results are presented, and the method is applied to an analysis comparing the effects of sulfonylureas versus metformin on body mass index, where the variables baseline body mass index and glycosylated hemoglobin have missing values. Our result supports the association of sulfonylureas with weight gain. © The Author 2017. Published by Oxford University Press on behalf of the Johns Hopkins Bloomberg School of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Reporting and dealing with missing quality of life data in RCTs: has the picture changed in the last decade?

PubMed

Fielding, S; Ogbuagu, A; Sivasubramaniam, S; MacLennan, G; Ramsay, C R

2016-12-01

Missing data are a major problem in the analysis of data from randomised trials affecting power and potentially producing biased treatment effects. Specifically focussing on quality of life outcomes, we aimed to report the amount of missing data, whether imputation was used and what methods and was the missing mechanism discussed from four leading medical journals and compare the picture to our previous review nearly a decade ago. A random selection (50 %) of all RCTS published during 2013-2014 in BMJ, JAMA, Lancet and NEJM was obtained. RCTs reported in research letters, cluster RCTs, non-randomised designs, review articles and meta-analysis were excluded. We included 87 RCTs in the review of which 35 % the amount of missing primary QoL data was unclear, 31 (36 %) used imputation. Only 23 % discussed the missing data mechanism. Nearly half used complete case analysis. Reporting was more unclear for secondary QoL outcomes. Compared to the previous review, multiple imputation was used more prominently but mainly in sensitivity analysis. Inadequate reporting and handling of missing QoL data in RCTs are still an issue. There is a large gap between statistical methods research relating to missing data and the use of the methods in applications. A sensitivity analysis should be undertaken to explore the sensitivity of the main results to different missing data assumptions. Medical journals can help to improve the situation by requiring higher standards of reporting and analytical methods to deal with missing data, and by issuing guidance to authors on expected standard.

The use of multiple imputation for the accurate measurements of individual feed intake by electronic feeders.

PubMed

Jiao, S; Tiezzi, F; Huang, Y; Gray, K A; Maltecca, C

2016-02-01

Obtaining accurate individual feed intake records is the key first step in achieving genetic progress toward more efficient nutrient utilization in pigs. Feed intake records collected by electronic feeding systems contain errors (erroneous and abnormal values exceeding certain cutoff criteria), which are due to feeder malfunction or animal-feeder interaction. In this study, we examined the use of a novel data-editing strategy involving multiple imputation to minimize the impact of errors and missing values on the quality of feed intake data collected by an electronic feeding system. Accuracy of feed intake data adjustment obtained from the conventional linear mixed model (LMM) approach was compared with 2 alternative implementations of multiple imputation by chained equation, denoted as MI (multiple imputation) and MICE (multiple imputation by chained equation). The 3 methods were compared under 3 scenarios, where 5, 10, and 20% feed intake error rates were simulated. Each of the scenarios was replicated 5 times. Accuracy of the alternative error adjustment was measured as the correlation between the true daily feed intake (DFI; daily feed intake in the testing period) or true ADFI (the mean DFI across testing period) and the adjusted DFI or adjusted ADFI. In the editing process, error cutoff criteria are used to define if a feed intake visit contains errors. To investigate the possibility that the error cutoff criteria may affect any of the 3 methods, the simulation was repeated with 2 alternative error cutoff values. Multiple imputation methods outperformed the LMM approach in all scenarios with mean accuracies of 96.7, 93.5, and 90.2% obtained with MI and 96.8, 94.4, and 90.1% obtained with MICE compared with 91.0, 82.6, and 68.7% using LMM for DFI. Similar results were obtained for ADFI. Furthermore, multiple imputation methods consistently performed better than LMM regardless of the cutoff criteria applied to define errors. In conclusion, multiple imputation is proposed as a more accurate and flexible method for error adjustments in feed intake data collected by electronic feeders.

Method variation in the impact of missing data on response shift detection.

PubMed

Schwartz, Carolyn E; Sajobi, Tolulope T; Verdam, Mathilde G E; Sebille, Veronique; Lix, Lisa M; Guilleux, Alice; Sprangers, Mirjam A G

2015-03-01

Missing data due to attrition or item non-response can result in biased estimates and loss of power in longitudinal quality-of-life (QOL) research. The impact of missing data on response shift (RS) detection is relatively unknown. This overview article synthesizes the findings of three methods tested in this special section regarding the impact of missing data patterns on RS detection in incomplete longitudinal data. The RS detection methods investigated include: (1) Relative importance analysis to detect reprioritization RS in stroke caregivers; (2) Oort's structural equation modeling (SEM) to detect recalibration, reprioritization, and reconceptualization RS in cancer patients; and (3) Rasch-based item-response theory-based (IRT) models as compared to SEM models to detect recalibration and reprioritization RS in hospitalized chronic disease patients. Each method dealt with missing data differently, either with imputation (1), attrition-based multi-group analysis (2), or probabilistic analysis that is robust to missingness due to the specific objectivity property (3). Relative importance analyses were sensitive to the type and amount of missing data and imputation method, with multiple imputation showing the largest RS effects. The attrition-based multi-group SEM revealed differential effects of both the changes in health-related QOL and the occurrence of response shift by attrition stratum, and enabled a more complete interpretation of findings. The IRT RS algorithm found evidence of small recalibration and reprioritization effects in General Health, whereas SEM mostly evidenced small recalibration effects. These differences may be due to differences between the two methods in handling of missing data. Missing data imputation techniques result in different conclusions about the presence of reprioritization RS using the relative importance method, while the attrition-based SEM approach highlighted different recalibration and reprioritization RS effects by attrition group. The IRT analyses detected more recalibration and reprioritization RS effects than SEM, presumably due to IRT's robustness to missing data. Future research should apply simulation techniques in order to make conclusive statements about the impacts of missing data according to the type and amount of RS.

A Time-Series Water Level Forecasting Model Based on Imputation and Variable Selection Method.

PubMed

Yang, Jun-He; Cheng, Ching-Hsue; Chan, Chia-Pan

2017-01-01

Reservoirs are important for households and impact the national economy. This paper proposed a time-series forecasting model based on estimating a missing value followed by variable selection to forecast the reservoir's water level. This study collected data from the Taiwan Shimen Reservoir as well as daily atmospheric data from 2008 to 2015. The two datasets are concatenated into an integrated dataset based on ordering of the data as a research dataset. The proposed time-series forecasting model summarily has three foci. First, this study uses five imputation methods to directly delete the missing value. Second, we identified the key variable via factor analysis and then deleted the unimportant variables sequentially via the variable selection method. Finally, the proposed model uses a Random Forest to build the forecasting model of the reservoir's water level. This was done to compare with the listing method under the forecasting error. These experimental results indicate that the Random Forest forecasting model when applied to variable selection with full variables has better forecasting performance than the listing model. In addition, this experiment shows that the proposed variable selection can help determine five forecast methods used here to improve the forecasting capability.

Recovering incomplete data using Statistical Multiple Imputations (SMI): a case study in environmental chemistry.

PubMed

Mercer, Theresa G; Frostick, Lynne E; Walmsley, Anthony D

2011-10-15

This paper presents a statistical technique that can be applied to environmental chemistry data where missing values and limit of detection levels prevent the application of statistics. A working example is taken from an environmental leaching study that was set up to determine if there were significant differences in levels of leached arsenic (As), chromium (Cr) and copper (Cu) between lysimeters containing preservative treated wood waste and those containing untreated wood. Fourteen lysimeters were setup and left in natural conditions for 21 weeks. The resultant leachate was analysed by ICP-OES to determine the As, Cr and Cu concentrations. However, due to the variation inherent in each lysimeter combined with the limits of detection offered by ICP-OES, the collected quantitative data was somewhat incomplete. Initial data analysis was hampered by the number of 'missing values' in the data. To recover the dataset, the statistical tool of Statistical Multiple Imputation (SMI) was applied, and the data was re-analysed successfully. It was demonstrated that using SMI did not affect the variance in the data, but facilitated analysis of the complete dataset. Copyright © 2011 Elsevier B.V. All rights reserved.

Genetic variance estimation with imputed variants finds negligible missing heritability for human height and body mass index.

PubMed

Yang, Jian; Bakshi, Andrew; Zhu, Zhihong; Hemani, Gibran; Vinkhuyzen, Anna A E; Lee, Sang Hong; Robinson, Matthew R; Perry, John R B; Nolte, Ilja M; van Vliet-Ostaptchouk, Jana V; Snieder, Harold; Esko, Tonu; Milani, Lili; Mägi, Reedik; Metspalu, Andres; Hamsten, Anders; Magnusson, Patrik K E; Pedersen, Nancy L; Ingelsson, Erik; Soranzo, Nicole; Keller, Matthew C; Wray, Naomi R; Goddard, Michael E; Visscher, Peter M

2015-10-01

We propose a method (GREML-LDMS) to estimate heritability for human complex traits in unrelated individuals using whole-genome sequencing data. We demonstrate using simulations based on whole-genome sequencing data that ∼97% and ∼68% of variation at common and rare variants, respectively, can be captured by imputation. Using the GREML-LDMS method, we estimate from 44,126 unrelated individuals that all ∼17 million imputed variants explain 56% (standard error (s.e.) = 2.3%) of variance for height and 27% (s.e. = 2.5%) of variance for body mass index (BMI), and we find evidence that height- and BMI-associated variants have been under natural selection. Considering the imperfect tagging of imputation and potential overestimation of heritability from previous family-based studies, heritability is likely to be 60-70% for height and 30-40% for BMI. Therefore, the missing heritability is small for both traits. For further discovery of genes associated with complex traits, a study design with SNP arrays followed by imputation is more cost-effective than whole-genome sequencing at current prices.

Methods for estimating missing human skeletal element osteometric dimensions employed in the revised fully technique for estimating stature.

PubMed

Auerbach, Benjamin M

2011-05-01

One of the greatest limitations to the application of the revised Fully anatomical stature estimation method is the inability to measure some of the skeletal elements required in its calculation. These element dimensions cannot be obtained due to taphonomic factors, incomplete excavation, or disease processes, and result in missing data. This study examines methods of imputing these missing dimensions using observable Fully measurements from the skeleton and the accuracy of incorporating these missing element estimations into anatomical stature reconstruction. These are further assessed against stature estimations obtained from mathematical regression formulae for the lower limb bones (femur and tibia). Two thousand seven hundred and seventeen North and South American indigenous skeletons were measured, and subsets of these with observable Fully dimensions were used to simulate missing elements and create estimation methods and equations. Comparisons were made directly between anatomically reconstructed statures and mathematically derived statures, as well as with anatomically derived statures with imputed missing dimensions. These analyses demonstrate that, while mathematical stature estimations are more accurate, anatomical statures incorporating missing dimensions are not appreciably less accurate and are more precise. The anatomical stature estimation method using imputed missing dimensions is supported. Missing element estimation, however, is limited to the vertebral column (only when lumbar vertebrae are present) and to talocalcaneal height (only when femora and tibiae are present). Crania, entire vertebral columns, and femoral or tibial lengths cannot be reliably estimated. Further discussion of the applicability of these methods is discussed. Copyright © 2011 Wiley-Liss, Inc.

Impact of Missing Data on Person-Model Fit and Person Trait Estimation

ERIC Educational Resources Information Center

Zhang, Bo; Walker, Cindy M.

2008-01-01

The purpose of this research was to examine the effects of missing data on person-model fit and person trait estimation in tests with dichotomous items. Under the missing-completely-at-random framework, four missing data treatment techniques were investigated including pairwise deletion, coding missing responses as incorrect, hotdeck imputation,…

Methods for Mediation Analysis with Missing Data

ERIC Educational Resources Information Center

Zhang, Zhiyong; Wang, Lijuan

2013-01-01

Despite wide applications of both mediation models and missing data techniques, formal discussion of mediation analysis with missing data is still rare. We introduce and compare four approaches to dealing with missing data in mediation analysis including list wise deletion, pairwise deletion, multiple imputation (MI), and a two-stage maximum…

Heart rate reactivity and current post-traumatic stress disorder when data are missing.

PubMed

Jeon-Slaughter, Haekyung; Tucker, Phebe; Pfefferbaum, Betty; North, Carol S; de Andrade, Bernardo Borba; Neas, Barbara

2011-08-01

This study demonstrates that auxiliary and exclusion criteria variables increase the effectiveness of missing imputation in correcting underestimation of physiologic reactivity in relation to post-traumatic stress disorder (PTSD) caused by deleting cases with missing physiologic data. This study used data from survivors of the 1995 Oklahoma City bombing and imputed missing heart rate data using auxiliary and exclusion criteria variables. Logistic regression was used to examine heart rate reactivity in relation to current PTSD. Of 113 survivors who participated in the bombing study's 7-year follow-up interview, 42 (37%) had missing data on heart rate reactivity due to exclusion criteria (medical illness or use of cardiovascular or psychotropic medications) or non-participation. Logistic regression results based on imputed heart rate data using exclusion criteria and auxiliary (the presence of any current PTSD arousal symptoms) variables showed that survivors with current bombing-related PTSD had significantly higher heart rates at baseline and recovered more slowly back to baseline heart rate during resting periods than survivors without current PTSD, while results based on complete cases failed to show significant correlations between current PTSD and heart rates at any assessment points. Suggested methods yielded an otherwise undetectable link between physiology and current PTSD. © 2011 The Authors. Psychiatry and Clinical Neurosciences © 2011 Japanese Society of Psychiatry and Neurology.

Missing data handling in non-inferiority and equivalence trials: A systematic review.

PubMed

Rabe, Brooke A; Day, Simon; Fiero, Mallorie H; Bell, Melanie L

2018-05-25

Non-inferiority (NI) and equivalence clinical trials test whether a new treatment is therapeutically no worse than, or equivalent to, an existing standard of care. Missing data in clinical trials have been shown to reduce statistical power and potentially bias estimates of effect size; however, in NI and equivalence trials, they present additional issues. For instance, they may decrease sensitivity to differences between treatment groups and bias toward the alternative hypothesis of NI (or equivalence). Our primary aim was to review the extent of and methods for handling missing data (model-based methods, single imputation, multiple imputation, complete case), the analysis sets used (Intention-To-Treat, Per-Protocol, or both), and whether sensitivity analyses were used to explore departures from assumptions about the missing data. We conducted a systematic review of NI and equivalence trials published between May 2015 and April 2016 by searching the PubMed database. Articles were reviewed primarily by 2 reviewers, with 6 articles reviewed by both reviewers to establish consensus. Of 109 selected articles, 93% reported some missing data in the primary outcome. Among those, 50% reported complete case analysis, and 28% reported single imputation approaches for handling missing data. Only 32% reported conducting analyses of both intention-to-treat and per-protocol populations. Only 11% conducted any sensitivity analyses to test assumptions with respect to missing data. Missing data are common in NI and equivalence trials, and they are often handled by methods which may bias estimates and lead to incorrect conclusions. Copyright © 2018 John Wiley & Sons, Ltd.

DrImpute: imputing dropout events in single cell RNA sequencing data.

PubMed

Gong, Wuming; Kwak, Il-Youp; Pota, Pruthvi; Koyano-Nakagawa, Naoko; Garry, Daniel J

2018-06-08

The single cell RNA sequencing (scRNA-seq) technique begin a new era by allowing the observation of gene expression at the single cell level. However, there is also a large amount of technical and biological noise. Because of the low number of RNA transcriptomes and the stochastic nature of the gene expression pattern, there is a high chance of missing nonzero entries as zero, which are called dropout events. We develop DrImpute to impute dropout events in scRNA-seq data. We show that DrImpute has significantly better performance on the separation of the dropout zeros from true zeros than existing imputation algorithms. We also demonstrate that DrImpute can significantly improve the performance of existing tools for clustering, visualization and lineage reconstruction of nine published scRNA-seq datasets. DrImpute can serve as a very useful addition to the currently existing statistical tools for single cell RNA-seq analysis. DrImpute is implemented in R and is available at https://github.com/gongx030/DrImpute .

Propensity score analysis with partially observed covariates: How should multiple imputation be used?

PubMed

Leyrat, Clémence; Seaman, Shaun R; White, Ian R; Douglas, Ian; Smeeth, Liam; Kim, Joseph; Resche-Rigon, Matthieu; Carpenter, James R; Williamson, Elizabeth J

2017-01-01

Inverse probability of treatment weighting is a popular propensity score-based approach to estimate marginal treatment effects in observational studies at risk of confounding bias. A major issue when estimating the propensity score is the presence of partially observed covariates. Multiple imputation is a natural approach to handle missing data on covariates: covariates are imputed and a propensity score analysis is performed in each imputed dataset to estimate the treatment effect. The treatment effect estimates from each imputed dataset are then combined to obtain an overall estimate. We call this method MIte. However, an alternative approach has been proposed, in which the propensity scores are combined across the imputed datasets (MIps). Therefore, there are remaining uncertainties about how to implement multiple imputation for propensity score analysis: (a) should we apply Rubin's rules to the inverse probability of treatment weighting treatment effect estimates or to the propensity score estimates themselves? (b) does the outcome have to be included in the imputation model? (c) how should we estimate the variance of the inverse probability of treatment weighting estimator after multiple imputation? We studied the consistency and balancing properties of the MIte and MIps estimators and performed a simulation study to empirically assess their performance for the analysis of a binary outcome. We also compared the performance of these methods to complete case analysis and the missingness pattern approach, which uses a different propensity score model for each pattern of missingness, and a third multiple imputation approach in which the propensity score parameters are combined rather than the propensity scores themselves (MIpar). Under a missing at random mechanism, complete case and missingness pattern analyses were biased in most cases for estimating the marginal treatment effect, whereas multiple imputation approaches were approximately unbiased as long as the outcome was included in the imputation model. Only MIte was unbiased in all the studied scenarios and Rubin's rules provided good variance estimates for MIte. The propensity score estimated in the MIte approach showed good balancing properties. In conclusion, when using multiple imputation in the inverse probability of treatment weighting context, MIte with the outcome included in the imputation model is the preferred approach.

The Estimation of Gestational Age at Birth in Database Studies.

PubMed

Eberg, Maria; Platt, Robert W; Filion, Kristian B

2017-11-01

Studies on the safety of prenatal medication use require valid estimation of the pregnancy duration. However, gestational age is often incompletely recorded in administrative and clinical databases. Our objective was to compare different approaches to estimating the pregnancy duration. Using data from the Clinical Practice Research Datalink and Hospital Episode Statistics, we examined the following four approaches to estimating missing gestational age: (1) generalized estimating equations for longitudinal data; (2) multiple imputation; (3) estimation based on fetal birth weight and sex; and (4) conventional approaches that assigned a fixed value (39 weeks for all or 39 weeks for full term and 35 weeks for preterm). The gestational age recorded in Hospital Episode Statistics was considered the gold standard. We conducted a simulation study comparing the described approaches in terms of estimated bias and mean square error. A total of 25,929 infants from 22,774 mothers were included in our "gold standard" cohort. The smallest average absolute bias was observed for the generalized estimating equation that included birth weight, while the largest absolute bias occurred when assigning 39-week gestation to all those with missing values. The smallest mean square errors were detected with generalized estimating equations while multiple imputation had the highest mean square errors. The use of generalized estimating equations resulted in the most accurate estimation of missing gestational age when birth weight information was available. In the absence of birth weight, assignment of fixed gestational age based on term/preterm status may be the optimal approach.

A Probability Based Framework for Testing the Missing Data Mechanism

ERIC Educational Resources Information Center

Lin, Johnny Cheng-Han

2013-01-01

Many methods exist for imputing missing data but fewer methods have been proposed to test the missing data mechanism. Little (1988) introduced a multivariate chi-square test for the missing completely at random data mechanism (MCAR) that compares observed means for each pattern with expectation-maximization (EM) estimated means. As an alternative,…

Is missing geographic positioning system data in accelerometry studies a problem, and is imputation the solution?

PubMed Central

Meseck, Kristin; Jankowska, Marta M.; Schipperijn, Jasper; Natarajan, Loki; Godbole, Suneeta; Carlson, Jordan; Takemoto, Michelle; Crist, Katie; Kerr, Jacqueline

2016-01-01

The main purpose of the present study was to assess the impact of global positioning system (GPS) signal lapse on physical activity analyses, discover any existing associations between missing GPS data and environmental and demographics attributes, and to determine whether imputation is an accurate and viable method for correcting GPS data loss. Accelerometer and GPS data of 782 participants from 8 studies were pooled to represent a range of lifestyles and interactions with the built environment. Periods of GPS signal lapse were identified and extracted. Generalised linear mixed models were run with the number of lapses and the length of lapses as outcomes. The signal lapses were imputed using a simple ruleset, and imputation was validated against person-worn camera imagery. A final generalised linear mixed model was used to identify the difference between the amount of GPS minutes pre- and post-imputation for the activity categories of sedentary, light, and moderate-to-vigorous physical activity. Over 17% of the dataset was comprised of GPS data lapses. No strong associations were found between increasing lapse length and number of lapses and the demographic and built environment variables. A significant difference was found between the pre- and post-imputation minutes for each activity category. No demographic or environmental bias was found for length or number of lapses, but imputation of GPS data may make a significant difference for inclusion of physical activity data that occurred during a lapse. Imputing GPS data lapses is a viable technique for returning spatial context to accelerometer data and improving the completeness of the dataset. PMID:27245796

Handling Missing Data With Multilevel Structural Equation Modeling and Full Information Maximum Likelihood Techniques.

PubMed

Schminkey, Donna L; von Oertzen, Timo; Bullock, Linda

2016-08-01

With increasing access to population-based data and electronic health records for secondary analysis, missing data are common. In the social and behavioral sciences, missing data frequently are handled with multiple imputation methods or full information maximum likelihood (FIML) techniques, but healthcare researchers have not embraced these methodologies to the same extent and more often use either traditional imputation techniques or complete case analysis, which can compromise power and introduce unintended bias. This article is a review of options for handling missing data, concluding with a case study demonstrating the utility of multilevel structural equation modeling using full information maximum likelihood (MSEM with FIML) to handle large amounts of missing data. MSEM with FIML is a parsimonious and hypothesis-driven strategy to cope with large amounts of missing data without compromising power or introducing bias. This technique is relevant for nurse researchers faced with ever-increasing amounts of electronic data and decreasing research budgets. © 2016 Wiley Periodicals, Inc. © 2016 Wiley Periodicals, Inc.

Substance dependence among those without symptoms of substance abuse in the World Mental Health Survey.

PubMed

Lago, Luise; Glantz, Meyer D; Kessler, Ronald C; Sampson, Nancy A; Al-Hamzawi, Ali; Florescu, Silvia; Moskalewicz, Jacek; Murphy, Sam; Navarro-Mateu, Fernando; Torres de Galvis, Yolanda; Viana, Maria Carmen; Xavier, Miguel; Degenhardt, Louisa

2017-09-01

The World Health Organization (WHO) World Mental Health (WMH) Survey Initiative uses the Composite International Diagnostic Interview (CIDI). The first 13 surveys only assessed substance dependence among respondents with a history of substance abuse; later surveys also assessed substance dependence without symptoms of abuse. We compared results across the two sets of surveys to assess implications of the revised logic and develop an imputation model for missing values of lifetime dependence in the earlier surveys. Lifetime dependence without symptoms of abuse was low in the second set of surveys (0.3% alcohol, 0.2% drugs). Regression-based imputation models were built in random half-samples of the new surveys and validated in the other half. There were minimal differences for imputed and actual reported cases in the validation dataset for age, gender and quantity; more mental disorders and days out of role were found in the imputed cases. Concordance between imputed and observed dependence cases in the full sample was high for alcohol [sensitivity 88.0%, specificity 99.8%, total classification accuracy (TCA) 99.5%, area under the curve (AUC) 0.94] and drug dependence (sensitivity 100.0%, specificity 99.8%, TCA 99.8%, AUC 1.00). This provides cross-national evidence of the small degree to which lifetime dependence occurs without symptoms of abuse. Imputation of substance dependence in the earlier WMH surveys improved estimates of dependence. Copyright © 2017 John Wiley & Sons, Ltd.

Warpgroup: increased precision of metabolomic data processing by consensus integration bound analysis

PubMed Central

Mahieu, Nathaniel G.; Spalding, Jonathan L.; Patti, Gary J.

2016-01-01

Motivation: Current informatic techniques for processing raw chromatography/mass spectrometry data break down under several common, non-ideal conditions. Importantly, hydrophilic liquid interaction chromatography (a key separation technology for metabolomics) produces data which are especially challenging to process. We identify three critical points of failure in current informatic workflows: compound specific drift, integration region variance, and naive missing value imputation. We implement the Warpgroup algorithm to address these challenges. Results: Warpgroup adds peak subregion detection, consensus integration bound detection, and intelligent missing value imputation steps to the conventional informatic workflow. When compared with the conventional workflow, Warpgroup made major improvements to the processed data. The coefficient of variation for peaks detected in replicate injections of a complex Escherichia Coli extract were halved (a reduction of 19%). Integration regions across samples were much more robust. Additionally, many signals lost by the conventional workflow were ‘rescued’ by the Warpgroup refinement, thereby resulting in greater analyte coverage in the processed data. Availability and implementation: Warpgroup is an open source R package available on GitHub at github.com/nathaniel-mahieu/warpgroup. The package includes example data and XCMS compatibility wrappers for ease of use. Supplementary information: Supplementary data are available at Bioinformatics online. Contact: nathaniel.mahieu@wustl.edu or gjpattij@wustl.edu PMID:26424859

Novel methods to optimize genotypic imputation for low-coverage, next-generation sequence data in crop plants

USDA-ARS?s Scientific Manuscript database

Next-generation sequencing technology such as genotyping-by-sequencing (GBS) made low-cost, but often low-coverage, whole-genome sequencing widely available. Extensive inbreeding in crop plants provides an untapped, high quality source of phased haplotypes for imputing missing genotypes. We introduc...

Predictive model for survival in patients with gastric cancer.

PubMed

Goshayeshi, Ladan; Hoseini, Benyamin; Yousefli, Zahra; Khooie, Alireza; Etminani, Kobra; Esmaeilzadeh, Abbas; Golabpour, Amin

2017-12-01

Gastric cancer is one of the most prevalent cancers in the world. Characterized by poor prognosis, it is a frequent cause of cancer in Iran. The aim of the study was to design a predictive model of survival time for patients suffering from gastric cancer. This was a historical cohort conducted between 2011 and 2016. Study population were 277 patients suffering from gastric cancer. Data were gathered from the Iranian Cancer Registry and the laboratory of Emam Reza Hospital in Mashhad, Iran. Patients or their relatives underwent interviews where it was needed. Missing values were imputed by data mining techniques. Fifteen factors were analyzed. Survival was addressed as a dependent variable. Then, the predictive model was designed by combining both genetic algorithm and logistic regression. Matlab 2014 software was used to combine them. Of the 277 patients, only survival of 80 patients was available whose data were used for designing the predictive model. Mean ?SD of missing values for each patient was 4.43?.41 combined predictive model achieved 72.57% accuracy. Sex, birth year, age at diagnosis time, age at diagnosis time of patients' family, family history of gastric cancer, and family history of other gastrointestinal cancers were six parameters associated with patient survival. The study revealed that imputing missing values by data mining techniques have a good accuracy. And it also revealed six parameters extracted by genetic algorithm effect on the survival of patients with gastric cancer. Our combined predictive model, with a good accuracy, is appropriate to forecast the survival of patients suffering from Gastric cancer. So, we suggest policy makers and specialists to apply it for prediction of patients' survival.

Model specification and bootstrapping for multiply imputed data: An application to count models for the frequency of alcohol use

PubMed Central

Comulada, W. Scott

2015-01-01

Stata’s mi commands provide powerful tools to conduct multiple imputation in the presence of ignorable missing data. In this article, I present Stata code to extend the capabilities of the mi commands to address two areas of statistical inference where results are not easily aggregated across imputed datasets. First, mi commands are restricted to covariate selection. I show how to address model fit to correctly specify a model. Second, the mi commands readily aggregate model-based standard errors. I show how standard errors can be bootstrapped for situations where model assumptions may not be met. I illustrate model specification and bootstrapping on frequency counts for the number of times that alcohol was consumed in data with missing observations from a behavioral intervention. PMID:26973439

A Review of Methods for Missing Data.

ERIC Educational Resources Information Center

Pigott, Therese D.

2001-01-01

Reviews methods for handling missing data in a research study. Model-based methods, such as maximum likelihood using the EM algorithm and multiple imputation, hold more promise than ad hoc methods. Although model-based methods require more specialized computer programs and assumptions about the nature of missing data, these methods are appropriate…

Epidemiologic Evaluation of Measurement Data in the Presence of Detection Limits

PubMed Central

Lubin, Jay H.; Colt, Joanne S.; Camann, David; Davis, Scott; Cerhan, James R.; Severson, Richard K.; Bernstein, Leslie; Hartge, Patricia

2004-01-01

Quantitative measurements of environmental factors greatly improve the quality of epidemiologic studies but can pose challenges because of the presence of upper or lower detection limits or interfering compounds, which do not allow for precise measured values. We consider the regression of an environmental measurement (dependent variable) on several covariates (independent variables). Various strategies are commonly employed to impute values for interval-measured data, including assignment of one-half the detection limit to nondetected values or of “fill-in” values randomly selected from an appropriate distribution. On the basis of a limited simulation study, we found that the former approach can be biased unless the percentage of measurements below detection limits is small (5–10%). The fill-in approach generally produces unbiased parameter estimates but may produce biased variance estimates and thereby distort inference when 30% or more of the data are below detection limits. Truncated data methods (e.g., Tobit regression) and multiple imputation offer two unbiased approaches for analyzing measurement data with detection limits. If interest resides solely on regression parameters, then Tobit regression can be used. If individualized values for measurements below detection limits are needed for additional analysis, such as relative risk regression or graphical display, then multiple imputation produces unbiased estimates and nominal confidence intervals unless the proportion of missing data is extreme. We illustrate various approaches using measurements of pesticide residues in carpet dust in control subjects from a case–control study of non-Hodgkin lymphoma. PMID:15579415

Multiple imputation of missing covariates for the Cox proportional hazards cure model

PubMed Central

Beesley, Lauren J; Bartlett, Jonathan W; Wolf, Gregory T; Taylor, Jeremy M G

2016-01-01

We explore several approaches for imputing partially observed covariates when the outcome of interest is a censored event time and when there is an underlying subset of the population that will never experience the event of interest. We call these subjects “cured,” and we consider the case where the data are modeled using a Cox proportional hazards (CPH) mixture cure model. We study covariate imputation approaches using fully conditional specification (FCS). We derive the exact conditional distribution and suggest a sampling scheme for imputing partially observed covariates in the CPH cure model setting. We also propose several approximations to the exact distribution that are simpler and more convenient to use for imputation. A simulation study demonstrates that the proposed imputation approaches outperform existing imputation approaches for survival data without a cure fraction in terms of bias in estimating CPH cure model parameters. We apply our multiple imputation techniques to a study of patients with head and neck cancer. PMID:27439726

Addressing continuous data measured with different instruments for participants excluded from trial analysis: a guide for systematic reviewers.

PubMed

Ebrahim, Shanil; Johnston, Bradley C; Akl, Elie A; Mustafa, Reem A; Sun, Xin; Walter, Stephen D; Heels-Ansdell, Diane; Alonso-Coello, Pablo; Guyatt, Gordon H

2014-05-01

We previously developed an approach to address the impact of missing participant data in meta-analyses of continuous variables in trials that used the same measurement instrument. We extend this approach to meta-analyses including trials that use different instruments to measure the same construct. We reviewed the available literature, conducted an iterative consultative process, and developed an approach involving a complete-case analysis complemented by sensitivity analyses that apply a series of increasingly stringent assumptions about results in patients with missing continuous outcome data. Our approach involves choosing the reference measurement instrument; converting scores from different instruments to the units of the reference instrument; developing four successively more stringent imputation strategies for addressing missing participant data; calculating a pooled mean difference for the complete-case analysis and imputation strategies; calculating the proportion of patients who experienced an important treatment effect; and judging the impact of the imputation strategies on the confidence in the estimate of effect. We applied our approach to an example systematic review of respiratory rehabilitation for chronic obstructive pulmonary disease. Our extended approach provides quantitative guidance for addressing missing participant data in systematic reviews of trials using different instruments to measure the same construct. Copyright © 2014 Elsevier Inc. All rights reserved.

Generating Multiple Imputations for Matrix Sampling Data Analyzed with Item Response Models.

ERIC Educational Resources Information Center

Thomas, Neal; Gan, Nianci

1997-01-01

Describes and assesses missing data methods currently used to analyze data from matrix sampling designs implemented by the National Assessment of Educational Progress. Several improved methods are developed, and these models are evaluated using an EM algorithm to obtain maximum likelihood estimates followed by multiple imputation of complete data…

Analyzing partially missing confounder information in comparative effectiveness and safety research of therapeutics.

PubMed

Toh, Sengwee; García Rodríguez, Luis A; Hernán, Miguel A

2012-05-01

Electronic healthcare databases are commonly used in comparative effectiveness and safety research of therapeutics. Many databases now include additional confounder information in a subset of the study population through data linkage or data collection. We described and compared existing methods for analyzing such datasets. Using data from The Health Improvement Network and the relation between non-steroidal anti-inflammatory drugs and upper gastrointestinal bleeding as an example, we employed several methods to handle partially missing confounder information. The crude odds ratio (OR) of upper gastrointestinal bleeding was 1.50 (95% confidence interval: 0.98, 2.28) among selective cyclo-oxygenase-2 inhibitor initiators (n = 43 569) compared with traditional non-steroidal anti-inflammatory drug initiators (n = 411 616). The OR dropped to 0.81 (0.52, 1.27) upon adjustment for confounders recorded for all patients. When further considering three additional variables missing in 22% of the study population (smoking, alcohol consumption, body mass index), the OR was between 0.80 and 0.83 for the missing-category approach, the missing-indicator approach, single imputation by the most common category, multiple imputation by chained equations, and propensity score calibration. The OR was 0.65 (0.39, 1.09) and 0.67 (0.38, 1.16) for the unweighted and the inverse probability weighted complete-case analysis, respectively. Existing methods for handling partially missing confounder data require different assumptions and may produce different results. The unweighted complete-case analysis, the missing-category/indicator approach, and single imputation require often unrealistic assumptions and should be avoided. In this study, differences across methods were not substantial, likely due to relatively low proportion of missingness and weak confounding effect by the three additional variables upon adjustment for other variables. Copyright © 2012 John Wiley & Sons, Ltd.

Is using multiple imputation better than complete case analysis for estimating a prevalence (risk) difference in randomized controlled trials when binary outcome observations are missing?

PubMed

Mukaka, Mavuto; White, Sarah A; Terlouw, Dianne J; Mwapasa, Victor; Kalilani-Phiri, Linda; Faragher, E Brian

2016-07-22

Missing outcomes can seriously impair the ability to make correct inferences from randomized controlled trials (RCTs). Complete case (CC) analysis is commonly used, but it reduces sample size and is perceived to lead to reduced statistical efficiency of estimates while increasing the potential for bias. As multiple imputation (MI) methods preserve sample size, they are generally viewed as the preferred analytical approach. We examined this assumption, comparing the performance of CC and MI methods to determine risk difference (RD) estimates in the presence of missing binary outcomes. We conducted simulation studies of 5000 simulated data sets with 50 imputations of RCTs with one primary follow-up endpoint at different underlying levels of RD (3-25 %) and missing outcomes (5-30 %). For missing at random (MAR) or missing completely at random (MCAR) outcomes, CC method estimates generally remained unbiased and achieved precision similar to or better than MI methods, and high statistical coverage. Missing not at random (MNAR) scenarios yielded invalid inferences with both methods. Effect size estimate bias was reduced in MI methods by always including group membership even if this was unrelated to missingness. Surprisingly, under MAR and MCAR conditions in the assessed scenarios, MI offered no statistical advantage over CC methods. While MI must inherently accompany CC methods for intention-to-treat analyses, these findings endorse CC methods for per protocol risk difference analyses in these conditions. These findings provide an argument for the use of the CC approach to always complement MI analyses, with the usual caveat that the validity of the mechanism for missingness be thoroughly discussed. More importantly, researchers should strive to collect as much data as possible.

Evaluating the validity of multiple imputation for missing physiological data in the national trauma data bank.

PubMed

Moore, Lynne; Hanley, James A; Lavoie, André; Turgeon, Alexis

2009-05-01

The National Trauma Data Bank (NTDB) is plagued by the problem of missing physiological data. The Glasgow Coma Scale score, Respiratory Rate and Systolic Blood Pressure are an essential part of risk adjustment strategies for trauma system evaluation and clinical research. Missing data on these variables may compromise the feasibility and the validity of trauma group comparisons. To evaluate the validity of Multiple Imputation (MI) for completing missing physiological data in the National Trauma Data Bank (NTDB), by assessing the impact of MI on 1) frequency distributions, 2) associations with mortality, and 3) risk adjustment. Analyses were based on 170,956 NTDB observations with complete physiological data (observed data set). Missing physiological data were artificially imposed on this data set and then imputed using MI (MI data set). To assess the impact of MI on risk adjustment, 100 pairs of hospitals were randomly selected with replacement and compared using adjusted Odds Ratios (OR) of mortality. OR generated by the observed data set were then compared to those generated by the MI data set. Frequency distributions and associations with mortality were preserved following MI. The median absolute difference between adjusted OR of mortality generated by the observed data set and by the MI data set was 3.6% (inter-quartile range: 2.4%-6.1%). This study suggests that, provided it is implemented with care, MI of missing physiological data in the NTDB leads to valid frequency distributions, preserves associations with mortality, and does not compromise risk adjustment in inter-hospital comparisons of mortality.

Missing Data Imputation versus Full Information Maximum Likelihood with Second-Level Dependencies

ERIC Educational Resources Information Center

Larsen, Ross

2011-01-01

Missing data in the presence of upper level dependencies in multilevel models have never been thoroughly examined. Whereas first-level subjects are independent over time, the second-level subjects might exhibit nonzero covariances over time. This study compares 2 missing data techniques in the presence of a second-level dependency: multiple…

A systematic review of randomised controlled trials in rheumatoid arthritis: the reporting and handling of missing data in composite outcomes.

PubMed

Ibrahim, Fowzia; Tom, Brian D M; Scott, David L; Prevost, Andrew Toby

2016-06-02

Most reported outcome measures in rheumatoid arthritis (RA) trials are composite, whose components comprise single measures that are combined into one outcome. The aims of this review were to assess the range of missing data rates in primary composite outcomes and to document the current practice for handling and reporting missing data in published RA trials compared to the Consolidated Standards of Reporting Trials (CONSORT) recommendations. A systematic search for randomised controlled trials was conducted for RA trials published between 2008 and 2013 in four rheumatology and four high impact general medical journals. A total of 51 trials with a composite primary outcome were identified, of which 38 (75 %) used the binary American College of Rheumatology responder index and 13 (25 %) used the Disease Activity Score for 28 joints (DAS28). Forty-four trials (86 %) reported on an intention-to-treat analysis population, while 7 trials (14 %) analysed according to a modified intention-to-treat population. Missing data rates for the primary composite outcome ranged from 2-53 % and were above 30 % in 9 trials, 20-30 % in 11 trials, 10-20 % in 18 trials and below 10 % in 13 trials. Thirty-eight trials (75 %) used non-responder imputation and 10 (20 %) used last observation carried forward to impute missing composite outcome data at the primary time point. The rate of dropout was on average 61 % times higher in the placebo group compared to the treatment group in the 34 placebo controlled trials (relative rate 1.61, 95 % CI: 1.29, 2.02). Thirty-seven trials (73 %) did not report the use of sensitivity analyses to assess the handling of missing data in the primary analysis as recommended by CONSORT guidelines. This review highlights an improvement in rheumatology trial practice since the revision of CONSORT guidelines, in terms of power calculation and participant's flow diagram. However, there is a need to improve the handling and reporting of missing composite outcome data and their components in RA trials. In particular, sensitivity analyses need to be more widely used in RA trials because imputation is widespread and generally uses single imputation methods, and in this area the missing data rates are commonly differentially higher in the placebo group.

Multivariate missing data in hydrology - Review and applications

NASA Astrophysics Data System (ADS)

Ben Aissia, Mohamed-Aymen; Chebana, Fateh; Ouarda, Taha B. M. J.

2017-12-01

Water resources planning and management require complete data sets of a number of hydrological variables, such as flood peaks and volumes. However, hydrologists are often faced with the problem of missing data (MD) in hydrological databases. Several methods are used to deal with the imputation of MD. During the last decade, multivariate approaches have gained popularity in the field of hydrology, especially in hydrological frequency analysis (HFA). However, treating the MD remains neglected in the multivariate HFA literature whereas the focus has been mainly on the modeling component. For a complete analysis and in order to optimize the use of data, MD should also be treated in the multivariate setting prior to modeling and inference. Imputation of MD in the multivariate hydrological framework can have direct implications on the quality of the estimation. Indeed, the dependence between the series represents important additional information that can be included in the imputation process. The objective of the present paper is to highlight the importance of treating MD in multivariate hydrological frequency analysis by reviewing and applying multivariate imputation methods and by comparing univariate and multivariate imputation methods. An application is carried out for multiple flood attributes on three sites in order to evaluate the performance of the different methods based on the leave-one-out procedure. The results indicate that, the performance of imputation methods can be improved by adopting the multivariate setting, compared to mean substitution and interpolation methods, especially when using the copula-based approach.

A Kriging based spatiotemporal approach for traffic volume data imputation

PubMed Central

Han, Lee D.; Liu, Xiaohan; Pu, Li; Chin, Shih-miao; Hwang, Ho-ling

2018-01-01

Along with the rapid development of Intelligent Transportation Systems, traffic data collection technologies have progressed fast. The emergence of innovative data collection technologies such as remote traffic microwave sensor, Bluetooth sensor, GPS-based floating car method, and automated license plate recognition, has significantly increased the variety and volume of traffic data. Despite the development of these technologies, the missing data issue is still a problem that poses great challenge for data based applications such as traffic forecasting, real-time incident detection, dynamic route guidance, and massive evacuation optimization. A thorough literature review suggests most current imputation models either focus on the temporal nature of the traffic data and fail to consider the spatial information of neighboring locations or assume the data follow a certain distribution. These two issues reduce the imputation accuracy and limit the use of the corresponding imputation methods respectively. As a result, this paper presents a Kriging based data imputation approach that is able to fully utilize the spatiotemporal correlation in the traffic data and that does not assume the data follow any distribution. A set of scenarios with different missing rates are used to evaluate the performance of the proposed method. The performance of the proposed method was compared with that of two other widely used methods, historical average and K-nearest neighborhood. Comparison results indicate that the proposed method has the highest imputation accuracy and is more flexible compared to other methods. PMID:29664928

DOE Office of Scientific and Technical Information (OSTI.GOV)

Zainudin, Mohd Lutfi, E-mail: mdlutfi07@gmail.com; Institut Matematik Kejuruteraan; Saaban, Azizan, E-mail: azizan.s@uum.edu.my

The solar radiation values have been composed by automatic weather station using the device that namely pyranometer. The device is functions to records all the radiation values that have been dispersed, and these data are very useful for it experimental works and solar device’s development. In addition, for modeling and designing on solar radiation system application is needed for complete data observation. Unfortunately, lack for obtained the complete solar radiation data frequently occur due to several technical problems, which mainly contributed by monitoring device. Into encountering this matter, estimation missing values in an effort to substitute absent values with imputedmore » data. This paper aimed to evaluate several piecewise interpolation techniques likes linear, splines, cubic, and nearest neighbor into dealing missing values in hourly solar radiation data. Then, proposed an extendable work into investigating the potential used of cubic Bezier technique and cubic Said-ball method as estimator tools. As result, methods for cubic Bezier and Said-ball perform the best compare to another piecewise imputation technique.« less

Estimating the dose response relationship for occupational radiation exposure measured with minimum detection level.

PubMed

Xue, Xiaonan; Shore, Roy E; Ye, Xiangyang; Kim, Mimi Y

2004-10-01

Occupational exposures are often recorded as zero when the exposure is below the minimum detection level (BMDL). This can lead to an underestimation of the doses received by individuals and can lead to biased estimates of risk in occupational epidemiologic studies. The extent of the exposure underestimation is increased with the magnitude of the minimum detection level (MDL) and the frequency of monitoring. This paper uses multiple imputation methods to impute values for the missing doses due to BMDL. A Gibbs sampling algorithm is developed to implement the method, which is applied to two distinct scenarios: when dose information is available for each measurement (but BMDL is recorded as zero or some other arbitrary value), or when the dose information available represents the summation of a series of measurements (e.g., only yearly cumulative exposure is available but based on, say, weekly measurements). Then the average of the multiple imputed exposure realizations for each individual is used to obtain an unbiased estimate of the relative risk associated with exposure. Simulation studies are used to evaluate the performance of the estimators. As an illustration, the method is applied to a sample of historical occupational radiation exposure data from the Oak Ridge National Laboratory.

Using beta coefficients to impute missing correlations in meta-analysis research: Reasons for caution.

PubMed

Roth, Philip L; Le, Huy; Oh, In-Sue; Van Iddekinge, Chad H; Bobko, Philip

2018-06-01

Meta-analysis has become a well-accepted method for synthesizing empirical research about a given phenomenon. Many meta-analyses focus on synthesizing correlations across primary studies, but some primary studies do not report correlations. Peterson and Brown (2005) suggested that researchers could use standardized regression weights (i.e., beta coefficients) to impute missing correlations. Indeed, their beta estimation procedures (BEPs) have been used in meta-analyses in a wide variety of fields. In this study, the authors evaluated the accuracy of BEPs in meta-analysis. We first examined how use of BEPs might affect results from a published meta-analysis. We then developed a series of Monte Carlo simulations that systematically compared the use of existing correlations (that were not missing) to data sets that incorporated BEPs (that impute missing correlations from corresponding beta coefficients). These simulations estimated ρ̄ (mean population correlation) and SDρ (true standard deviation) across a variety of meta-analytic conditions. Results from both the existing meta-analysis and the Monte Carlo simulations revealed that BEPs were associated with potentially large biases when estimating ρ̄ and even larger biases when estimating SDρ. Using only existing correlations often substantially outperformed use of BEPs and virtually never performed worse than BEPs. Overall, the authors urge a return to the standard practice of using only existing correlations in meta-analysis. (PsycINFO Database Record (c) 2018 APA, all rights reserved).

The impact of missing trauma data on predicting massive transfusion

PubMed Central

Trickey, Amber W.; Fox, Erin E.; del Junco, Deborah J.; Ning, Jing; Holcomb, John B.; Brasel, Karen J.; Cohen, Mitchell J.; Schreiber, Martin A.; Bulger, Eileen M.; Phelan, Herb A.; Alarcon, Louis H.; Myers, John G.; Muskat, Peter; Cotton, Bryan A.; Wade, Charles E.; Rahbar, Mohammad H.

2013-01-01

INTRODUCTION Missing data are inherent in clinical research and may be especially problematic for trauma studies. This study describes a sensitivity analysis to evaluate the impact of missing data on clinical risk prediction algorithms. Three blood transfusion prediction models were evaluated utilizing an observational trauma dataset with valid missing data. METHODS The PRospective Observational Multi-center Major Trauma Transfusion (PROMMTT) study included patients requiring ≥ 1 unit of red blood cells (RBC) at 10 participating U.S. Level I trauma centers from July 2009 – October 2010. Physiologic, laboratory, and treatment data were collected prospectively up to 24h after hospital admission. Subjects who received ≥ 10 RBC units within 24h of admission were classified as massive transfusion (MT) patients. Correct classification percentages for three MT prediction models were evaluated using complete case analysis and multiple imputation. A sensitivity analysis for missing data was conducted to determine the upper and lower bounds for correct classification percentages. RESULTS PROMMTT enrolled 1,245 subjects. MT was received by 297 patients (24%). Missing percentage ranged from 2.2% (heart rate) to 45% (respiratory rate). Proportions of complete cases utilized in the MT prediction models ranged from 41% to 88%. All models demonstrated similar correct classification percentages using complete case analysis and multiple imputation. In the sensitivity analysis, correct classification upper-lower bound ranges per model were 4%, 10%, and 12%. Predictive accuracy for all models using PROMMTT data was lower than reported in the original datasets. CONCLUSIONS Evaluating the accuracy clinical prediction models with missing data can be misleading, especially with many predictor variables and moderate levels of missingness per variable. The proposed sensitivity analysis describes the influence of missing data on risk prediction algorithms. Reporting upper/lower bounds for percent correct classification may be more informative than multiple imputation, which provided similar results to complete case analysis in this study. PMID:23778514

Traffic speed data imputation method based on tensor completion.

PubMed

Ran, Bin; Tan, Huachun; Feng, Jianshuai; Liu, Ying; Wang, Wuhong

2015-01-01

Traffic speed data plays a key role in Intelligent Transportation Systems (ITS); however, missing traffic data would affect the performance of ITS as well as Advanced Traveler Information Systems (ATIS). In this paper, we handle this issue by a novel tensor-based imputation approach. Specifically, tensor pattern is adopted for modeling traffic speed data and then High accurate Low Rank Tensor Completion (HaLRTC), an efficient tensor completion method, is employed to estimate the missing traffic speed data. This proposed method is able to recover missing entries from given entries, which may be noisy, considering severe fluctuation of traffic speed data compared with traffic volume. The proposed method is evaluated on Performance Measurement System (PeMS) database, and the experimental results show the superiority of the proposed approach over state-of-the-art baseline approaches.

Traffic Speed Data Imputation Method Based on Tensor Completion

PubMed Central

Ran, Bin; Feng, Jianshuai; Liu, Ying; Wang, Wuhong

2015-01-01

Traffic speed data plays a key role in Intelligent Transportation Systems (ITS); however, missing traffic data would affect the performance of ITS as well as Advanced Traveler Information Systems (ATIS). In this paper, we handle this issue by a novel tensor-based imputation approach. Specifically, tensor pattern is adopted for modeling traffic speed data and then High accurate Low Rank Tensor Completion (HaLRTC), an efficient tensor completion method, is employed to estimate the missing traffic speed data. This proposed method is able to recover missing entries from given entries, which may be noisy, considering severe fluctuation of traffic speed data compared with traffic volume. The proposed method is evaluated on Performance Measurement System (PeMS) database, and the experimental results show the superiority of the proposed approach over state-of-the-art baseline approaches. PMID:25866501

Imputation of Test Scores in the National Education Longitudinal Study of 1988 (NELS:88). Working Paper Series.

ERIC Educational Resources Information Center

Bokossa, Maxime C.; Huang, Gary G.

This report describes the imputation procedures used to deal with missing data in the National Education Longitudinal Study of 1988 (NELS:88), the only current National Center for Education Statistics (NCES) dataset that contains scores from cognitive tests given the same set of students at multiple time points. As is inevitable, cognitive test…

Guidebook for Imputation of Missing Data. Technical Report No. 17.

ERIC Educational Resources Information Center

Wise, Lauress L.; McLaughlin, Donald H.

This guidebook is designed for data analysts who are working with computer data files that contain records with incomplete data. It indicates choices the analyst must make and the criteria for making those choices in regard to the following questions: (1) What resources are available for performing the imputation? (2) How big is the data file? (3)…

Cost effectiveness of the addition of a comprehensive CT scan to the abdomen and pelvis for the detection of cancer after unprovoked venous thromboembolism.

PubMed

Coyle, Kathryn; Carrier, Marc; Lazo-Langner, Alejandro; Shivakumar, Sudeep; Zarychanski, Ryan; Tagalakis, Vicky; Solymoss, Susan; Routhier, Nathalie; Douketis, James; Coyle, Douglas

2017-03-01

Unprovoked venous thromboembolism (VTE) can be the first manifestation of cancer. It is unclear if extensive screening for occult cancer including a comprehensive computed tomography (CT) scan of the abdomen/pelvis is cost-effective in this patient population. To assess the health care related costs, number of missed cancer cases and health related utility values of a limited screening strategy with and without the addition of a comprehensive CT scan of the abdomen/pelvis and to identify to what extent testing should be done in these circumstances to allow early detection of occult cancers. Cost effectiveness analysis using data that was collected alongside the SOME randomized controlled trial which compared an extensive occult cancer screening including a CT of the abdomen/pelvis to a more limited screening strategy in patients with a first unprovoked VTE, was used for the current analyses. Analyses were conducted with a one-year time horizon from a Canadian health care perspective. Primary analysis was based on complete cases, with sensitivity analysis using appropriate multiple imputation methods to account for missing data. Data from a total of 854 patients with a first unprovoked VTE were included in these analyses. The addition of a comprehensive CT scan was associated with higher costs ($551 CDN) with no improvement in utility values or number of missed cancers. Results were consistent when adopting multiple imputation methods. The addition of a comprehensive CT scan of the abdomen/pelvis for the screening of occult cancer in patients with unprovoked VTE is not cost effective, as it is both more costly and not more effective in detecting occult cancer. Copyright © 2017 Elsevier Ltd. All rights reserved.

Identification of significant features by the Global Mean Rank test.

PubMed

Klammer, Martin; Dybowski, J Nikolaj; Hoffmann, Daniel; Schaab, Christoph

2014-01-01

With the introduction of omics-technologies such as transcriptomics and proteomics, numerous methods for the reliable identification of significantly regulated features (genes, proteins, etc.) have been developed. Experimental practice requires these tests to successfully deal with conditions such as small numbers of replicates, missing values, non-normally distributed expression levels, and non-identical distributions of features. With the MeanRank test we aimed at developing a test that performs robustly under these conditions, while favorably scaling with the number of replicates. The test proposed here is a global one-sample location test, which is based on the mean ranks across replicates, and internally estimates and controls the false discovery rate. Furthermore, missing data is accounted for without the need of imputation. In extensive simulations comparing MeanRank to other frequently used methods, we found that it performs well with small and large numbers of replicates, feature dependent variance between replicates, and variable regulation across features on simulation data and a recent two-color microarray spike-in dataset. The tests were then used to identify significant changes in the phosphoproteomes of cancer cells induced by the kinase inhibitors erlotinib and 3-MB-PP1 in two independently published mass spectrometry-based studies. MeanRank outperformed the other global rank-based methods applied in this study. Compared to the popular Significance Analysis of Microarrays and Linear Models for Microarray methods, MeanRank performed similar or better. Furthermore, MeanRank exhibits more consistent behavior regarding the degree of regulation and is robust against the choice of preprocessing methods. MeanRank does not require any imputation of missing values, is easy to understand, and yields results that are easy to interpret. The software implementing the algorithm is freely available for academic and commercial use.

Economic Analysis of a Multi-Site Prevention Program: Assessment of Program Costs and Characterizing Site-level Variability

PubMed Central

Corso, Phaedra S.; Ingels, Justin B.; Kogan, Steven M.; Foster, E. Michael; Chen, Yi-Fu; Brody, Gene H.

2013-01-01

Programmatic cost analyses of preventive interventions commonly have a number of methodological difficulties. To determine the mean total costs and properly characterize variability, one often has to deal with small sample sizes, skewed distributions, and especially missing data. Standard approaches for dealing with missing data such as multiple imputation may suffer from a small sample size, a lack of appropriate covariates, or too few details around the method used to handle the missing data. In this study, we estimate total programmatic costs for a prevention trial evaluating the Strong African American Families-Teen program. This intervention focuses on the prevention of substance abuse and risky sexual behavior. To account for missing data in the assessment of programmatic costs we compare multiple imputation to probabilistic sensitivity analysis. The latter approach uses collected cost data to create a distribution around each input parameter. We found that with the multiple imputation approach, the mean (95% confidence interval) incremental difference was $2149 ($397, $3901). With the probabilistic sensitivity analysis approach, the incremental difference was $2583 ($778, $4346). Although the true cost of the program is unknown, probabilistic sensitivity analysis may be a more viable alternative for capturing variability in estimates of programmatic costs when dealing with missing data, particularly with small sample sizes and the lack of strong predictor variables. Further, the larger standard errors produced by the probabilistic sensitivity analysis method may signal its ability to capture more of the variability in the data, thus better informing policymakers on the potentially true cost of the intervention. PMID:23299559

Economic analysis of a multi-site prevention program: assessment of program costs and characterizing site-level variability.

PubMed

Corso, Phaedra S; Ingels, Justin B; Kogan, Steven M; Foster, E Michael; Chen, Yi-Fu; Brody, Gene H

2013-10-01

Programmatic cost analyses of preventive interventions commonly have a number of methodological difficulties. To determine the mean total costs and properly characterize variability, one often has to deal with small sample sizes, skewed distributions, and especially missing data. Standard approaches for dealing with missing data such as multiple imputation may suffer from a small sample size, a lack of appropriate covariates, or too few details around the method used to handle the missing data. In this study, we estimate total programmatic costs for a prevention trial evaluating the Strong African American Families-Teen program. This intervention focuses on the prevention of substance abuse and risky sexual behavior. To account for missing data in the assessment of programmatic costs we compare multiple imputation to probabilistic sensitivity analysis. The latter approach uses collected cost data to create a distribution around each input parameter. We found that with the multiple imputation approach, the mean (95 % confidence interval) incremental difference was $2,149 ($397, $3,901). With the probabilistic sensitivity analysis approach, the incremental difference was $2,583 ($778, $4,346). Although the true cost of the program is unknown, probabilistic sensitivity analysis may be a more viable alternative for capturing variability in estimates of programmatic costs when dealing with missing data, particularly with small sample sizes and the lack of strong predictor variables. Further, the larger standard errors produced by the probabilistic sensitivity analysis method may signal its ability to capture more of the variability in the data, thus better informing policymakers on the potentially true cost of the intervention.

The Use of Multiple Imputation for Missing Data in Uniform DIF Analysis: Power and Type I Error Rates

ERIC Educational Resources Information Center

Finch, Holmes

2011-01-01

Methods of uniform differential item functioning (DIF) detection have been extensively studied in the complete data case. However, less work has been done examining the performance of these methods when missing item responses are present. Research that has been done in this regard appears to indicate that treating missing item responses as…

Postmodeling Sensitivity Analysis to Detect the Effect of Missing Data Mechanisms

ERIC Educational Resources Information Center

Jamshidian, Mortaza; Mata, Matthew

2008-01-01

Incomplete or missing data is a common problem in almost all areas of empirical research. It is well known that simple and ad hoc methods such as complete case analysis or mean imputation can lead to biased and/or inefficient estimates. The method of maximum likelihood works well; however, when the missing data mechanism is not one of missing…

Handling missing Mini-Mental State Examination (MMSE) values: Results from a cross-sectional long-term-care study.

PubMed

Godin, Judith; Keefe, Janice; Andrew, Melissa K

2017-04-01

Missing values are commonly encountered on the Mini Mental State Examination (MMSE), particularly when administered to frail older people. This presents challenges for MMSE scoring in research settings. We sought to describe missingness in MMSEs administered in long-term-care facilities (LTCF) and to compare and contrast approaches to dealing with missing items. As part of the Care and Construction project in Nova Scotia, Canada, LTCF residents completed an MMSE. Different methods of dealing with missing values (e.g., use of raw scores, raw scores/number of items attempted, scale-level multiple imputation [MI], and blended approaches) are compared to item-level MI. The MMSE was administered to 320 residents living in 23 LTCF. The sample was predominately female (73%), and 38% of participants were aged >85 years. At least one item was missing from 122 (38.2%) of the MMSEs. Data were not Missing Completely at Random (MCAR), χ 2 (1110) = 1,351, p < 0.001. Using raw scores for those missing <6 items in combination with scale-level MI resulted in the regression coefficients and standard errors closest to item-level MI. Patterns of missing items often suggest systematic problems, such as trouble with manual dexterity, literacy, or visual impairment. While these observations may be relatively easy to take into account in clinical settings, non-random missingness presents challenges for research and must be considered in statistical analyses. We present suggestions for dealing with missing MMSE data based on the extent of missingness and the goal of analyses. Copyright © 2016 The Authors. Production and hosting by Elsevier B.V. All rights reserved.

Comparison of data analysis strategies for intent-to-treat analysis in pre-test-post-test designs with substantial dropout rates.

PubMed

Salim, Agus; Mackinnon, Andrew; Christensen, Helen; Griffiths, Kathleen

2008-09-30

The pre-test-post-test design (PPD) is predominant in trials of psychotherapeutic treatments. Missing data due to withdrawals present an even bigger challenge in assessing treatment effectiveness under the PPD than under designs with more observations since dropout implies an absence of information about response to treatment. When confronted with missing data, often it is reasonable to assume that the mechanism underlying missingness is related to observed but not to unobserved outcomes (missing at random, MAR). Previous simulation and theoretical studies have shown that, under MAR, modern techniques such as maximum-likelihood (ML) based methods and multiple imputation (MI) can be used to produce unbiased estimates of treatment effects. In practice, however, ad hoc methods such as last observation carried forward (LOCF) imputation and complete-case (CC) analysis continue to be used. In order to better understand the behaviour of these methods in the PPD, we compare the performance of traditional approaches (LOCF, CC) and theoretically sound techniques (MI, ML), under various MAR mechanisms. We show that the LOCF method is seriously biased and conclude that its use should be abandoned. Complete-case analysis produces unbiased estimates only when the dropout mechanism does not depend on pre-test values even when dropout is related to fixed covariates including treatment group (covariate-dependent: CD). However, CC analysis is generally biased under MAR. The magnitude of the bias is largest when the correlation of post- and pre-test is relatively low.

Missing portion sizes in FFQ--alternatives to use of standard portions.

PubMed

Køster-Rasmussen, Rasmus; Siersma, Volkert; Halldorsson, Thorhallur I; de Fine Olivarius, Niels; Henriksen, Jan E; Heitmann, Berit L

2015-08-01

Standard portions or substitution of missing portion sizes with medians may generate bias when quantifying the dietary intake from FFQ. The present study compared four different methods to include portion sizes in FFQ. We evaluated three stochastic methods for imputation of portion sizes based on information about anthropometry, sex, physical activity and age. Energy intakes computed with standard portion sizes, defined as sex-specific medians (median), or with portion sizes estimated with multinomial logistic regression (MLR), 'comparable categories' (Coca) or k-nearest neighbours (KNN) were compared with a reference based on self-reported portion sizes (quantified by a photographic food atlas embedded in the FFQ). The Danish Health Examination Survey 2007-2008. The study included 3728 adults with complete portion size data. Compared with the reference, the root-mean-square errors of the mean daily total energy intake (in kJ) computed with portion sizes estimated by the four methods were (men; women): median (1118; 1061), MLR (1060; 1051), Coca (1230; 1146), KNN (1281; 1181). The equivalent biases (mean error) were (in kJ): median (579; 469), MLR (248; 178), Coca (234; 188), KNN (-340; 218). The methods MLR and Coca provided the best agreement with the reference. The stochastic methods allowed for estimation of meaningful portion sizes by conditioning on information about physiology and they were suitable for multiple imputation. We propose to use MLR or Coca to substitute missing portion size values or when portion sizes needs to be included in FFQ without portion size data.

Applied Missing Data Analysis. Methodology in the Social Sciences Series

ERIC Educational Resources Information Center

Enders, Craig K.

2010-01-01

Walking readers step by step through complex concepts, this book translates missing data techniques into something that applied researchers and graduate students can understand and utilize in their own research. Enders explains the rationale and procedural details for maximum likelihood estimation, Bayesian estimation, multiple imputation, and…

Gap-filling methods to impute eddy covariance flux data by preserving variance.

NASA Astrophysics Data System (ADS)

Kunwor, S.; Staudhammer, C. L.; Starr, G.; Loescher, H. W.

2015-12-01

To represent carbon dynamics, in terms of exchange of CO2 between the terrestrial ecosystem and the atmosphere, eddy covariance (EC) data has been collected using eddy flux towers from various sites across globe for more than two decades. However, measurements from EC data are missing for various reasons: precipitation, routine maintenance, or lack of vertical turbulence. In order to have estimates of net ecosystem exchange of carbon dioxide (NEE) with high precision and accuracy, robust gap-filling methods to impute missing data are required. While the methods used so far have provided robust estimates of the mean value of NEE, little attention has been paid to preserving the variance structures embodied by the flux data. Preserving the variance of these data will provide unbiased and precise estimates of NEE over time, which mimic natural fluctuations. We used a non-linear regression approach with moving windows of different lengths (15, 30, and 60-days) to estimate non-linear regression parameters for one year of flux data from a long-leaf pine site at the Joseph Jones Ecological Research Center. We used as our base the Michaelis-Menten and Van't Hoff functions. We assessed the potential physiological drivers of these parameters with linear models using micrometeorological predictors. We then used a parameter prediction approach to refine the non-linear gap-filling equations based on micrometeorological conditions. This provides us an opportunity to incorporate additional variables, such as vapor pressure deficit (VPD) and volumetric water content (VWC) into the equations. Our preliminary results indicate that improvements in gap-filling can be gained with a 30-day moving window with additional micrometeorological predictors (as indicated by lower root mean square error (RMSE) of the predicted values of NEE). Our next steps are to use these parameter predictions from moving windows to gap-fill the data with and without incorporation of potential driver variables of the parameters traditionally used. Then, comparisons of the predicted values from these methods and 'traditional' gap-filling methods (using 12 fixed monthly windows) will be assessed to show the scale of preserving variance. Further, this method will be applied to impute artificially created gaps for analyzing if variance is preserved.

Normal Theory Two-Stage ML Estimator When Data Are Missing at the Item Level

PubMed Central

Savalei, Victoria; Rhemtulla, Mijke

2017-01-01

In many modeling contexts, the variables in the model are linear composites of the raw items measured for each participant; for instance, regression and path analysis models rely on scale scores, and structural equation models often use parcels as indicators of latent constructs. Currently, no analytic estimation method exists to appropriately handle missing data at the item level. Item-level multiple imputation (MI), however, can handle such missing data straightforwardly. In this article, we develop an analytic approach for dealing with item-level missing data—that is, one that obtains a unique set of parameter estimates directly from the incomplete data set and does not require imputations. The proposed approach is a variant of the two-stage maximum likelihood (TSML) methodology, and it is the analytic equivalent of item-level MI. We compare the new TSML approach to three existing alternatives for handling item-level missing data: scale-level full information maximum likelihood, available-case maximum likelihood, and item-level MI. We find that the TSML approach is the best analytic approach, and its performance is similar to item-level MI. We recommend its implementation in popular software and its further study. PMID:29276371

Normal Theory Two-Stage ML Estimator When Data Are Missing at the Item Level.

PubMed

Savalei, Victoria; Rhemtulla, Mijke

2017-08-01

In many modeling contexts, the variables in the model are linear composites of the raw items measured for each participant; for instance, regression and path analysis models rely on scale scores, and structural equation models often use parcels as indicators of latent constructs. Currently, no analytic estimation method exists to appropriately handle missing data at the item level. Item-level multiple imputation (MI), however, can handle such missing data straightforwardly. In this article, we develop an analytic approach for dealing with item-level missing data-that is, one that obtains a unique set of parameter estimates directly from the incomplete data set and does not require imputations. The proposed approach is a variant of the two-stage maximum likelihood (TSML) methodology, and it is the analytic equivalent of item-level MI. We compare the new TSML approach to three existing alternatives for handling item-level missing data: scale-level full information maximum likelihood, available-case maximum likelihood, and item-level MI. We find that the TSML approach is the best analytic approach, and its performance is similar to item-level MI. We recommend its implementation in popular software and its further study.

Principled Approaches to Missing Data in Epidemiologic Studies

PubMed Central

Perkins, Neil J; Cole, Stephen R; Harel, Ofer; Tchetgen Tchetgen, Eric J; Sun, BaoLuo; Mitchell, Emily M; Schisterman, Enrique F

2018-01-01

Abstract Principled methods with which to appropriately analyze missing data have long existed; however, broad implementation of these methods remains challenging. In this and 2 companion papers (Am J Epidemiol. 2018;187(3):576–584 and Am J Epidemiol. 2018;187(3):585–591), we discuss issues pertaining to missing data in the epidemiologic literature. We provide details regarding missing-data mechanisms and nomenclature and encourage the conduct of principled analyses through a detailed comparison of multiple imputation and inverse probability weighting. Data from the Collaborative Perinatal Project, a multisite US study conducted from 1959 to 1974, are used to create a masked data-analytical challenge with missing data induced by known mechanisms. We illustrate the deleterious effects of missing data with naive methods and show how principled methods can sometimes mitigate such effects. For example, when data were missing at random, naive methods showed a spurious protective effect of smoking on the risk of spontaneous abortion (odds ratio (OR) = 0.43, 95% confidence interval (CI): 0.19, 0.93), while implementation of principled methods multiple imputation (OR = 1.30, 95% CI: 0.95, 1.77) or augmented inverse probability weighting (OR = 1.40, 95% CI: 1.00, 1.97) provided estimates closer to the “true” full-data effect (OR = 1.31, 95% CI: 1.05, 1.64). We call for greater acknowledgement of and attention to missing data and for the broad use of principled missing-data methods in epidemiologic research. PMID:29165572

Principled Approaches to Missing Data in Epidemiologic Studies.

PubMed

Perkins, Neil J; Cole, Stephen R; Harel, Ofer; Tchetgen Tchetgen, Eric J; Sun, BaoLuo; Mitchell, Emily M; Schisterman, Enrique F

2018-03-01

Principled methods with which to appropriately analyze missing data have long existed; however, broad implementation of these methods remains challenging. In this and 2 companion papers (Am J Epidemiol. 2018;187(3):576-584 and Am J Epidemiol. 2018;187(3):585-591), we discuss issues pertaining to missing data in the epidemiologic literature. We provide details regarding missing-data mechanisms and nomenclature and encourage the conduct of principled analyses through a detailed comparison of multiple imputation and inverse probability weighting. Data from the Collaborative Perinatal Project, a multisite US study conducted from 1959 to 1974, are used to create a masked data-analytical challenge with missing data induced by known mechanisms. We illustrate the deleterious effects of missing data with naive methods and show how principled methods can sometimes mitigate such effects. For example, when data were missing at random, naive methods showed a spurious protective effect of smoking on the risk of spontaneous abortion (odds ratio (OR) = 0.43, 95% confidence interval (CI): 0.19, 0.93), while implementation of principled methods multiple imputation (OR = 1.30, 95% CI: 0.95, 1.77) or augmented inverse probability weighting (OR = 1.40, 95% CI: 1.00, 1.97) provided estimates closer to the "true" full-data effect (OR = 1.31, 95% CI: 1.05, 1.64). We call for greater acknowledgement of and attention to missing data and for the broad use of principled missing-data methods in epidemiologic research.

The relationship between visibility aid use and motor vehicle related injuries among bicyclists presenting to emergency departments.

PubMed

Hagel, B E; Romanow, N T R; Morgunov, N; Embree, T; Couperthwaite, A B; Voaklander, D; Rowe, B H

2014-04-01

Little is known about the effectiveness of visibility aids (VAs; e.g., reflectors, lights, fluorescent clothing) in reducing the risk of a bicyclist-motor-vehicle (MV) collision. To determine if VAs reduce the risk of a bicyclist-MV collision. Cases were bicyclists struck by a MV and assessed at Calgary and Edmonton, Alberta, Canada, emergency departments (EDs) from May 2008 to October 2010. Controls were bicyclists with non-MV injuries. Participants were interviewed about their personal and injury characteristics, including use of VAs. Injury information was collected from charts. Odds ratios (ORs) and 95% confidence intervals (CIs) were estimated for VAs during daylight and dark conditions, and adjusted for confounders using logistic regression. Missing values were imputed using chained equations and adjusted OR estimates from the imputed data were calculated. There were 2403 injured bicyclists including 278 cases. After adjusting for age, sex, type of bicycling (commuting vs. recreational) and bicyclist speed, white compared with black (OR 0.52; 95% CI 0.28, 0.95), and bicyclist self-reported light compared with dark coloured (OR 0.67; 95% CI 0.49, 0.92) upper body clothing reduced the odds of a MV collision during daylight. After imputing missing values, white compared with black (OR 0.57; 95% CI: 0.32, 0.99) and bicyclist self-reported light compared with dark coloured (OR 0.71; 95% CI 0.52, 0.97) upper body clothing remained protective against MV collision in daylight conditions. During dark conditions, crude estimates indicated that reflective clothing or other items, red/orange/yellow front upper body clothing compared with black, fluorescent clothing, headlights and tail lights were estimated to increase the odds of a MV collision. An imputed adjusted analysis revealed that red/orange/yellow front upper body clothing colour (OR 4.11; 95% CI 1.06, 15.99) and tail lights (OR 2.54; 95% CI: 1.06, 6.07) remained the only significant risk factors for MV collisions. One or more visibility aids reduced the odds of a bicyclist MV collision resulting in hospitalization. Bicyclist clothing choice may be important in reducing the risk of MV collision. The protective effect of visibility aids varies based on light conditions, and non-bicyclist risk factors also need to be considered. Copyright © 2014 Elsevier Ltd. All rights reserved.

Principle Component Analysis with Incomplete Data: A simulation of R pcaMethods package in Constructing an Environmental Quality Index with Missing Data

EPA Science Inventory

Missing data is a common problem in the application of statistical techniques. In principal component analysis (PCA), a technique for dimensionality reduction, incomplete data points are either discarded or imputed using interpolation methods. Such approaches are less valid when ...

Residential proximity to electromagnetic field sources and birth weight: Minimizing residual confounding using multiple imputation and propensity score matching.

PubMed

de Vocht, Frank; Lee, Brian

2014-08-01

Studies have suggested that residential exposure to extremely low frequency (50 Hz) electromagnetic fields (ELF-EMF) from high voltage cables, overhead power lines, electricity substations or towers are associated with reduced birth weight and may be associated with adverse birth outcomes or even miscarriages. We previously conducted a study of 140,356 singleton live births between 2004 and 2008 in Northwest England, which suggested that close residential proximity (≤ 50 m) to ELF-EMF sources was associated with reduced average birth weight of 212 g (95%CI: -395 to -29 g) but not with statistically significant increased risks for other adverse perinatal outcomes. However, the cohort was limited by missing data for most potentially confounding variables including maternal smoking during pregnancy, which was only available for a small subgroup, while also residual confounding could not be excluded. This study, using the same cohort, was conducted to minimize the effects of these problems using multiple imputation to address missing data and propensity score matching to minimize residual confounding. Missing data were imputed using multiple imputation using chained equations to generate five datasets. For each dataset 115 exposed women (residing ≤ 50 m from a residential ELF-EMF source) were propensity score matched to 1150 unexposed women. After doubly robust confounder adjustment, close proximity to a residential ELF-EMF source remained associated with a reduction in birth weight of -116 g (95% confidence interval: -224:-7 g). No effect was found for proximity ≤ 100 m compared to women living further away. These results indicate that although the effect size was about half of the effect previously reported, close maternal residential proximity to sources of ELF-EMF remained associated with suboptimal fetal growth. Copyright © 2014 Elsevier Ltd. All rights reserved.

DOE Office of Scientific and Technical Information (OSTI.GOV)

Webb-Robertson, Bobbie-Jo M.; Wiberg, Holli K.; Matzke, Melissa M.

In this review, we apply selected imputation strategies to label-free liquid chromatography–mass spectrometry (LC–MS) proteomics datasets to evaluate the accuracy with respect to metrics of variance and classification. We evaluate several commonly used imputation approaches for individual merits and discuss the caveats of each approach with respect to the example LC–MS proteomics data. In general, local similarity-based approaches, such as the regularized expectation maximization and least-squares adaptive algorithms, yield the best overall performances with respect to metrics of accuracy and robustness. However, no single algorithm consistently outperforms the remaining approaches, and in some cases, performing classification without imputation sometimes yieldedmore » the most accurate classification. Thus, because of the complex mechanisms of missing data in proteomics, which also vary from peptide to protein, no individual method is a single solution for imputation. In summary, on the basis of the observations in this review, the goal for imputation in the field of computational proteomics should be to develop new approaches that work generically for this data type and new strategies to guide users in the selection of the best imputation for their dataset and analysis objectives.« less

Fast and accurate genotype imputation in genome-wide association studies through pre-phasing

PubMed Central

Howie, Bryan; Fuchsberger, Christian; Stephens, Matthew; Marchini, Jonathan; Abecasis, Gonçalo R.

2013-01-01

Sequencing efforts, including the 1000 Genomes Project and disease-specific efforts, are producing large collections of haplotypes that can be used for genotype imputation in genome-wide association studies (GWAS). Imputing from these reference panels can help identify new risk alleles, but the use of large panels with existing methods imposes a high computational burden. To keep imputation broadly accessible, we introduce a strategy called “pre-phasing” that maintains the accuracy of leading methods while cutting computational costs by orders of magnitude. In brief, we first statistically estimate the haplotypes for each GWAS individual (“pre-phasing”) and then impute missing genotypes into these estimated haplotypes. This reduces the computational cost because: (i) the GWAS samples must be phased only once, whereas standard methods would implicitly re-phase with each reference panel update; (ii) it is much faster to match a phased GWAS haplotype to one reference haplotype than to match unphased GWAS genotypes to a pair of reference haplotypes. This strategy will be particularly valuable for repeated imputation as reference panels evolve. PMID:22820512

A comparison of selected parametric and imputation methods for estimating snag density and snag quality attributes

USGS Publications Warehouse

Eskelson, Bianca N.I.; Hagar, Joan; Temesgen, Hailemariam

2012-01-01

Snags (standing dead trees) are an essential structural component of forests. Because wildlife use of snags depends on size and decay stage, snag density estimation without any information about snag quality attributes is of little value for wildlife management decision makers. Little work has been done to develop models that allow multivariate estimation of snag density by snag quality class. Using climate, topography, Landsat TM data, stand age and forest type collected for 2356 forested Forest Inventory and Analysis plots in western Washington and western Oregon, we evaluated two multivariate techniques for their abilities to estimate density of snags by three decay classes. The density of live trees and snags in three decay classes (D1: recently dead, little decay; D2: decay, without top, some branches and bark missing; D3: extensive decay, missing bark and most branches) with diameter at breast height (DBH) ≥ 12.7 cm was estimated using a nonparametric random forest nearest neighbor imputation technique (RF) and a parametric two-stage model (QPORD), for which the number of trees per hectare was estimated with a Quasipoisson model in the first stage and the probability of belonging to a tree status class (live, D1, D2, D3) was estimated with an ordinal regression model in the second stage. The presence of large snags with DBH ≥ 50 cm was predicted using a logistic regression and RF imputation. Because of the more homogenous conditions on private forest lands, snag density by decay class was predicted with higher accuracies on private forest lands than on public lands, while presence of large snags was more accurately predicted on public lands, owing to the higher prevalence of large snags on public lands. RF outperformed the QPORD model in terms of percent accurate predictions, while QPORD provided smaller root mean square errors in predicting snag density by decay class. The logistic regression model achieved more accurate presence/absence classification of large snags than the RF imputation approach. Adjusting the decision threshold to account for unequal size for presence and absence classes is more straightforward for the logistic regression than for the RF imputation approach. Overall, model accuracies were poor in this study, which can be attributed to the poor predictive quality of the explanatory variables and the large range of forest types and geographic conditions observed in the data.

DAnTE: a statistical tool for quantitative analysis of –omics data

DOE Office of Scientific and Technical Information (OSTI.GOV)

Polpitiya, Ashoka D.; Qian, Weijun; Jaitly, Navdeep

2008-05-03

DAnTE (Data Analysis Tool Extension) is a statistical tool designed to address challenges unique to quantitative bottom-up, shotgun proteomics data. This tool has also been demonstrated for microarray data and can easily be extended to other high-throughput data types. DAnTE features selected normalization methods, missing value imputation algorithms, peptide to protein rollup methods, an extensive array of plotting functions, and a comprehensive ANOVA scheme that can handle unbalanced data and random effects. The Graphical User Interface (GUI) is designed to be very intuitive and user friendly.

Penalized regression procedures for variable selection in the potential outcomes framework

PubMed Central

Ghosh, Debashis; Zhu, Yeying; Coffman, Donna L.

2015-01-01

A recent topic of much interest in causal inference is model selection. In this article, we describe a framework in which to consider penalized regression approaches to variable selection for causal effects. The framework leads to a simple ‘impute, then select’ class of procedures that is agnostic to the type of imputation algorithm as well as penalized regression used. It also clarifies how model selection involves a multivariate regression model for causal inference problems, and that these methods can be applied for identifying subgroups in which treatment effects are homogeneous. Analogies and links with the literature on machine learning methods, missing data and imputation are drawn. A difference LASSO algorithm is defined, along with its multiple imputation analogues. The procedures are illustrated using a well-known right heart catheterization dataset. PMID:25628185

An imputed genotype resource for the laboratory mouse

PubMed Central

Szatkiewicz, Jin P.; Beane, Glen L.; Ding, Yueming; Hutchins, Lucie; de Villena, Fernando Pardo-Manuel; Churchill, Gary A.

2009-01-01

We have created a high-density SNP resource encompassing 7.87 million polymorphic loci across 49 inbred mouse strains of the laboratory mouse by combining data available from public databases and training a hidden Markov model to impute missing genotypes in the combined data. The strong linkage disequilibrium found in dense sets of SNP markers in the laboratory mouse provides the basis for accurate imputation. Using genotypes from eight independent SNP resources, we empirically validated the quality of the imputed genotypes and demonstrate that they are highly reliable for most inbred strains. The imputed SNP resource will be useful for studies of natural variation and complex traits. It will facilitate association study designs by providing high density SNP genotypes for large numbers of mouse strains. We anticipate that this resource will continue to evolve as new genotype data become available for laboratory mouse strains. The data are available for bulk download or query at http://cgd.jax.org/. PMID:18301946

Combining item response theory with multiple imputation to equate health assessment questionnaires.

PubMed

Gu, Chenyang; Gutman, Roee

2017-09-01

The assessment of patients' functional status across the continuum of care requires a common patient assessment tool. However, assessment tools that are used in various health care settings differ and cannot be easily contrasted. For example, the Functional Independence Measure (FIM) is used to evaluate the functional status of patients who stay in inpatient rehabilitation facilities, the Minimum Data Set (MDS) is collected for all patients who stay in skilled nursing facilities, and the Outcome and Assessment Information Set (OASIS) is collected if they choose home health care provided by home health agencies. All three instruments or questionnaires include functional status items, but the specific items, rating scales, and instructions for scoring different activities vary between the different settings. We consider equating different health assessment questionnaires as a missing data problem, and propose a variant of predictive mean matching method that relies on Item Response Theory (IRT) models to impute unmeasured item responses. Using real data sets, we simulated missing measurements and compared our proposed approach to existing methods for missing data imputation. We show that, for all of the estimands considered, and in most of the experimental conditions that were examined, the proposed approach provides valid inferences, and generally has better coverages, relatively smaller biases, and shorter interval estimates. The proposed method is further illustrated using a real data set. © 2016, The International Biometric Society.

Missing Data in the Field of Otorhinolaryngology and Head & Neck Surgery: Need for Improvement.

PubMed

Netten, Anouk P; Dekker, Friedo W; Rieffe, Carolien; Soede, Wim; Briaire, Jeroen J; Frijns, Johan H M

Clinical studies are often facing missing data. Data can be missing for various reasons, for example, patients moved, certain measurements are only administered in high-risk groups, and patients are unable to attend clinic because of their health status. There are various ways to handle these missing data (e.g., complete cases analyses, mean substitution). Each of these techniques potentially influences both the analyses and the results of a study. The first aim of this structured review was to analyze how often researchers in the field of otorhinolaryngology/head & neck surgery report missing data. The second aim was to systematically describe how researchers handle missing data in their analyses. The third aim was to provide a solution on how to deal with missing data by means of the multiple imputation technique. With this review, we aim to contribute to a higher quality of reporting in otorhinolaryngology research. Clinical studies among the 398 most recently published research articles in three major journals in the field of otorhinolaryngology/head & neck surgery were analyzed based on how researchers reported and handled missing data. Of the 316 clinical studies, 85 studies reported some form of missing data. Of those 85, only a small number (12 studies, 3.8%) actively handled the missingness in their data. The majority of researchers exclude incomplete cases, which results in biased outcomes and a drop in statistical power. Within otorhinolaryngology research, missing data are largely ignored and underreported, and consequently, handled inadequately. This has major impact on the results and conclusions drawn from this research. Based on the outcomes of this review, we provide solutions on how to deal with missing data. To illustrate, we clarify the use of multiple imputation techniques, which recently became widely available in standard statistical programs.

Substance Dependence among those without symptoms of Substance Abuse in the World Mental Health Survey

PubMed Central

Lago, Luise; Glantz, Meyer; Kessler, Ronald C.; Sampson, Nancy; Al-Hamzawi, Ali; Florescu, Silvia; Moskalewicz, Jacek; Murphy, Sam; Navarro-Mateu, Fernando; de Galvis, Yolanda Torres; Viana, Maria Carmen; Xavier, Miguel; Degenhardt, Louisa

2017-01-01

The WHO World Mental Health (WMH) Survey Initiative uses the Composite International Diagnostic Interview (CIDI) for data collection and operationalizes diagnoses using DSM-IV criteria. The first 13 WMH surveys used CIDI version 3.0, which only assessed substance dependence among respondents with a history of substance abuse, while subsequent surveys also assessed substance dependence without symptoms of abuse. The aim of the current report is to compare results across the two sets of surveys to assess the implications of the revised skip logic and develop an imputation model for missing values of lifetime dependence without symptoms of abuse in the earlier surveys. Prevalence of lifetime dependence without symptoms of abuse was low in the second set of WMH surveys (0.3% for alcohol and 0.2% for drugs). Regression-based imputation models were built in random half-samples of the new surveys and validated in the other half-samples. There were minimal difference in the distributions of imputed and reported cases in the validation dataset for important correlates such as age, gender and quantity though higher numbers of additional mental disorders and number of days out of role were found in the imputed than reported cases. Concordance between imputed and observed estimates of total lifetime dependence in the full sample was high both for alcohol dependence (sensitivity 88.0%, specificity 99.8%, TCA 99.5% and AUC 0.94) and drug dependence (sensitivity 100.0%, specificity 99.8%, TCA 99.8% and AUC 1.00). This study provides cross-national evidence of the degree to which each of lifetime alcohol dependence and lifetime drug dependence occur without symptoms of abuse. Additionally imputation of substance dependence without symptoms of abuse in the earlier WMH surveys will result in improved estimates of lifetime prevalence for comparison with other epidemiological studies both cross-nationally and over time. PMID:28211594

Classification models for identification of at-risk groups for incident memory complaints.

PubMed

van den Kommer, Tessa N; Comijs, Hannie C; Rijs, Kelly J; Heymans, Martijn W; van Boxtel, Martin P J; Deeg, Dorly J H

2014-02-01

Memory complaints in older adults may be a precursor of measurable cognitive decline. Causes for these complaints may vary across age groups. The goal of this study was to develop classification models for the early identification of persons at risk for memory complaints using a broad range of characteristics. Two age groups were studied, 55-65 years old (N = 1,416.8) and 65-75 years old (N = 471) using data from the Longitudinal Aging Study Amsterdam. Participants reporting memory complaints at baseline were excluded. Data on predictors of memory complaints were collected at baseline and analyzed using logistic regression analyses. Multiple imputation was applied to handle the missing data; missing data due to mortality were not imputed. In persons aged 55-65 years, 14.4% reported memory complaints after three years of follow-up. Persons using medication, who were former smokers and had insufficient/poor hearing, were at the highest risk of developing memory complaints, i.e., a predictive value of 33.3%. In persons 65-75 years old, the incidence of memory complaints was 22.5%. Persons with a low sense of mastery, who reported having pain, were at the highest risk of memory complaints resulting in a final predictive value of 56.9%. In the subsample of persons without a low sense of mastery who (almost) never visited organizations and had a low level of memory performance, 46.8% reported memory complaints at follow-up. The classification models led to the identification of specific target groups at risk for memory complaints. Suggestions for person-tailored interventions may be based on these risk profiles.

Statistical analysis and handling of missing data in cluster randomized trials: a systematic review.

PubMed

Fiero, Mallorie H; Huang, Shuang; Oren, Eyal; Bell, Melanie L

2016-02-09

Cluster randomized trials (CRTs) randomize participants in groups, rather than as individuals and are key tools used to assess interventions in health research where treatment contamination is likely or if individual randomization is not feasible. Two potential major pitfalls exist regarding CRTs, namely handling missing data and not accounting for clustering in the primary analysis. The aim of this review was to evaluate approaches for handling missing data and statistical analysis with respect to the primary outcome in CRTs. We systematically searched for CRTs published between August 2013 and July 2014 using PubMed, Web of Science, and PsycINFO. For each trial, two independent reviewers assessed the extent of the missing data and method(s) used for handling missing data in the primary and sensitivity analyses. We evaluated the primary analysis and determined whether it was at the cluster or individual level. Of the 86 included CRTs, 80 (93%) trials reported some missing outcome data. Of those reporting missing data, the median percent of individuals with a missing outcome was 19% (range 0.5 to 90%). The most common way to handle missing data in the primary analysis was complete case analysis (44, 55%), whereas 18 (22%) used mixed models, six (8%) used single imputation, four (5%) used unweighted generalized estimating equations, and two (2%) used multiple imputation. Fourteen (16%) trials reported a sensitivity analysis for missing data, but most assumed the same missing data mechanism as in the primary analysis. Overall, 67 (78%) trials accounted for clustering in the primary analysis. High rates of missing outcome data are present in the majority of CRTs, yet handling missing data in practice remains suboptimal. Researchers and applied statisticians should carry out appropriate missing data methods, which are valid under plausible assumptions in order to increase statistical power in trials and reduce the possibility of bias. Sensitivity analysis should be performed, with weakened assumptions regarding the missing data mechanism to explore the robustness of results reported in the primary analysis.

Across-Platform Imputation of DNA Methylation Levels Incorporating Nonlocal Information Using Penalized Functional Regression.

PubMed

Zhang, Guosheng; Huang, Kuan-Chieh; Xu, Zheng; Tzeng, Jung-Ying; Conneely, Karen N; Guan, Weihua; Kang, Jian; Li, Yun

2016-05-01

DNA methylation is a key epigenetic mark involved in both normal development and disease progression. Recent advances in high-throughput technologies have enabled genome-wide profiling of DNA methylation. However, DNA methylation profiling often employs different designs and platforms with varying resolution, which hinders joint analysis of methylation data from multiple platforms. In this study, we propose a penalized functional regression model to impute missing methylation data. By incorporating functional predictors, our model utilizes information from nonlocal probes to improve imputation quality. Here, we compared the performance of our functional model to linear regression and the best single probe surrogate in real data and via simulations. Specifically, we applied different imputation approaches to an acute myeloid leukemia dataset consisting of 194 samples and our method showed higher imputation accuracy, manifested, for example, by a 94% relative increase in information content and up to 86% more CpG sites passing post-imputation filtering. Our simulated association study further demonstrated that our method substantially improves the statistical power to identify trait-associated methylation loci. These findings indicate that the penalized functional regression model is a convenient and valuable imputation tool for methylation data, and it can boost statistical power in downstream epigenome-wide association study (EWAS). © 2016 WILEY PERIODICALS, INC.

Investigating Factorial Invariance of Latent Variables Across Populations When Manifest Variables Are Missing Completely

PubMed Central

Widaman, Keith F.; Grimm, Kevin J.; Early, Dawnté R.; Robins, Richard W.; Conger, Rand D.

2013-01-01

Difficulties arise in multiple-group evaluations of factorial invariance if particular manifest variables are missing completely in certain groups. Ad hoc analytic alternatives can be used in such situations (e.g., deleting manifest variables), but some common approaches, such as multiple imputation, are not viable. At least 3 solutions to this problem are viable: analyzing differing sets of variables across groups, using pattern mixture approaches, and a new method using random number generation. The latter solution, proposed in this article, is to generate pseudo-random normal deviates for all observations for manifest variables that are missing completely in a given sample and then to specify multiple-group models in a way that respects the random nature of these values. An empirical example is presented in detail comparing the 3 approaches. The proposed solution can enable quantitative comparisons at the latent variable level between groups using programs that require the same number of manifest variables in each group. PMID:24019738

Analyzing the effect of selected control policy measures and sociodemographic factors on alcoholic beverage consumption in Europe within the AMPHORA project: statistical methods.

PubMed

Baccini, Michela; Carreras, Giulia

2014-10-01

This paper describes the methods used to investigate variations in total alcoholic beverage consumption as related to selected control intervention policies and other socioeconomic factors (unplanned factors) within 12 European countries involved in the AMPHORA project. The analysis presented several critical points: presence of missing values, strong correlation among the unplanned factors, long-term waves or trends in both the time series of alcohol consumption and the time series of the main explanatory variables. These difficulties were addressed by implementing a multiple imputation procedure for filling in missing values, then specifying for each country a multiple regression model which accounted for time trend, policy measures and a limited set of unplanned factors, selected in advance on the basis of sociological and statistical considerations are addressed. This approach allowed estimating the "net" effect of the selected control policies on alcohol consumption, but not the association between each unplanned factor and the outcome.

Estimating soil moisture exceedance probability from antecedent rainfall

NASA Astrophysics Data System (ADS)

Cronkite-Ratcliff, C.; Kalansky, J.; Stock, J. D.; Collins, B. D.

2016-12-01

The first storms of the rainy season in coastal California, USA, add moisture to soils but rarely trigger landslides. Previous workers proposed that antecedent rainfall, the cumulative seasonal rain from October 1 onwards, had to exceed specific amounts in order to trigger landsliding. Recent monitoring of soil moisture upslope of historic landslides in the San Francisco Bay Area shows that storms can cause positive pressure heads once soil moisture values exceed a threshold of volumetric water content (VWC). We propose that antecedent rainfall could be used to estimate the probability that VWC exceeds this threshold. A major challenge to estimating the probability of exceedance is that rain gauge records are frequently incomplete. We developed a stochastic model to impute (infill) missing hourly precipitation data. This model uses nearest neighbor-based conditional resampling of the gauge record using data from nearby rain gauges. Using co-located VWC measurements, imputed data can be used to estimate the probability that VWC exceeds a specific threshold for a given antecedent rainfall. The stochastic imputation model can also provide an estimate of uncertainty in the exceedance probability curve. Here we demonstrate the method using soil moisture and precipitation data from several sites located throughout Northern California. Results show a significant variability between sites in the sensitivity of VWC exceedance probability to antecedent rainfall.

Performance of bias-correction methods for exposure measurement error using repeated measurements with and without missing data.

PubMed

Batistatou, Evridiki; McNamee, Roseanne

2012-12-10

It is known that measurement error leads to bias in assessing exposure effects, which can however, be corrected if independent replicates are available. For expensive replicates, two-stage (2S) studies that produce data 'missing by design', may be preferred over a single-stage (1S) study, because in the second stage, measurement of replicates is restricted to a sample of first-stage subjects. Motivated by an occupational study on the acute effect of carbon black exposure on respiratory morbidity, we compare the performance of several bias-correction methods for both designs in a simulation study: an instrumental variable method (EVROS IV) based on grouping strategies, which had been recommended especially when measurement error is large, the regression calibration and the simulation extrapolation methods. For the 2S design, either the problem of 'missing' data was ignored or the 'missing' data were imputed using multiple imputations. Both in 1S and 2S designs, in the case of small or moderate measurement error, regression calibration was shown to be the preferred approach in terms of root mean square error. For 2S designs, regression calibration as implemented by Stata software is not recommended in contrast to our implementation of this method; the 'problematic' implementation of regression calibration although substantially improved with use of multiple imputations. The EVROS IV method, under a good/fairly good grouping, outperforms the regression calibration approach in both design scenarios when exposure mismeasurement is severe. Both in 1S and 2S designs with moderate or large measurement error, simulation extrapolation severely failed to correct for bias. Copyright © 2012 John Wiley & Sons, Ltd.

Measuring the Association Between Body Mass Index and All-Cause Mortality in the Presence of Missing Data: Analyses From the Scottish National Diabetes Register.

PubMed

Read, Stephanie H; Lewis, Steff C; Halbesma, Nynke; Wild, Sarah H

2017-04-15

Incorrectly handling missing data can lead to imprecise and biased estimates. We describe the effect of applying different approaches to handling missing data in an analysis of the association between body mass index and all-cause mortality among people with type 2 diabetes. We used data from the Scottish diabetes register that were linked to hospital admissions data and death registrations. The analysis was based on people diagnosed with type 2 diabetes between 2004 and 2011, with follow-up until May 31, 2014. The association between body mass index and mortality was investigated using Cox proportional hazards models. Findings were compared using 4 different missing-data methods: complete-case analysis, 2 multiple-imputation models, and nearest-neighbor imputation. There were 124,451 cases of type 2 diabetes, among which there were 17,085 deaths during 787,275 person-years of follow-up. Patients with missing data (24.8%) had higher mortality than those without missing data (adjusted hazard ratio = 1.36, 95% confidence interval: 1.31, 1.41). A U-shaped relationship between body mass index and mortality was observed, with the lowest hazard ratios occurring among moderately obese people, regardless of the chosen approach for handling missing data. Missing data may affect absolute and relative risk estimates differently and should be considered in analyses of routinely collected data. © The Author 2017. Published by Oxford University Press on behalf of the Johns Hopkins Bloomberg School of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Time perspective and well-being: Swedish survey questionnaires and data.

PubMed

Garcia, Danilo; Nima, Ali Al; Lindskär, Erik

2016-12-01

The data pertains 448 Swedes' responses to questionnaires on time perspective (Zimbardo Time Perspective Inventory), temporal life satisfaction (Temporal Satisfaction with Life Scale), affect (Positive Affect and Negative Affect Schedule), and psychological well-being (Ryff׳s Scales of Psychological Well-Being-short version). The data was collected among university students and individuals at a training facility (see U. Sailer, P. Rosenberg, A.A. Nima, A. Gamble, T. Gärling, T. Archer, D. Garcia, 2014; [1]). Since there were no differences in any of the other background variables, but exercise frequency, all subsequent analyses were conducted on the 448 participants as one single sample. In this article we include the Swedish versions of the questionnaires used to operationalize the time perspective and well-being variables. The data is available, SPSS file, as Supplementary material in this article. We used the Expectation-Maximization Algorithm to input missing values. Little׳s Chi-Square test for Missing Completely at Random showed a χ (2)=67.25 (df=53, p=.09) for men and χ (2)=77.65 (df=72, p=.31) for women. These values suggested that the Expectation-Maximization Algorithm was suitable to use on this data for missing data imputation.

Practical methods for dealing with 'not applicable' item responses in the AMC Linear Disability Score project

PubMed Central

Holman, Rebecca; Glas, Cees AW; Lindeboom, Robert; Zwinderman, Aeilko H; de Haan, Rob J

2004-01-01

Background Whenever questionnaires are used to collect data on constructs, such as functional status or health related quality of life, it is unlikely that all respondents will respond to all items. This paper examines ways of dealing with responses in a 'not applicable' category to items included in the AMC Linear Disability Score (ALDS) project item bank. Methods The data examined in this paper come from the responses of 392 respondents to 32 items and form part of the calibration sample for the ALDS item bank. The data are analysed using the one-parameter logistic item response theory model. The four practical strategies for dealing with this type of response are: cold deck imputation; hot deck imputation; treating the missing responses as if these items had never been offered to those individual patients; and using a model which takes account of the 'tendency to respond to items'. Results The item and respondent population parameter estimates were very similar for the strategies involving hot deck imputation; treating the missing responses as if these items had never been offered to those individual patients; and using a model which takes account of the 'tendency to respond to items'. The estimates obtained using the cold deck imputation method were substantially different. Conclusions The cold deck imputation method was not considered suitable for use in the ALDS item bank. The other three methods described can be usefully implemented in the ALDS item bank, depending on the purpose of the data analysis to be carried out. These three methods may be useful for other data sets examining similar constructs, when item response theory based methods are used. PMID:15200681

A methodology for treating missing data applied to daily rainfall data in the Candelaro River Basin (Italy).

PubMed

Lo Presti, Rossella; Barca, Emanuele; Passarella, Giuseppe

2010-01-01

Environmental time series are often affected by the "presence" of missing data, but when dealing statistically with data, the need to fill in the gaps estimating the missing values must be considered. At present, a large number of statistical techniques are available to achieve this objective; they range from very simple methods, such as using the sample mean, to very sophisticated ones, such as multiple imputation. A brand new methodology for missing data estimation is proposed, which tries to merge the obvious advantages of the simplest techniques (e.g. their vocation to be easily implemented) with the strength of the newest techniques. The proposed method consists in the application of two consecutive stages: once it has been ascertained that a specific monitoring station is affected by missing data, the "most similar" monitoring stations are identified among neighbouring stations on the basis of a suitable similarity coefficient; in the second stage, a regressive method is applied in order to estimate the missing data. In this paper, four different regressive methods are applied and compared, in order to determine which is the most reliable for filling in the gaps, using rainfall data series measured in the Candelaro River Basin located in South Italy.

Forecasting air quality time series using deep learning.

PubMed

Freeman, Brian S; Taylor, Graham; Gharabaghi, Bahram; Thé, Jesse

2018-04-13

This paper presents one of the first applications of deep learning (DL) techniques to predict air pollution time series. Air quality management relies extensively on time series data captured at air monitoring stations as the basis of identifying population exposure to airborne pollutants and determining compliance with local ambient air standards. In this paper, 8 hr averaged surface ozone (O 3 ) concentrations were predicted using deep learning consisting of a recurrent neural network (RNN) with long short-term memory (LSTM). Hourly air quality and meteorological data were used to train and forecast values up to 72 hours with low error rates. The LSTM was able to forecast the duration of continuous O 3 exceedances as well. Prior to training the network, the dataset was reviewed for missing data and outliers. Missing data were imputed using a novel technique that averaged gaps less than eight time steps with incremental steps based on first-order differences of neighboring time periods. Data were then used to train decision trees to evaluate input feature importance over different time prediction horizons. The number of features used to train the LSTM model was reduced from 25 features to 5 features, resulting in improved accuracy as measured by Mean Absolute Error (MAE). Parameter sensitivity analysis identified look-back nodes associated with the RNN proved to be a significant source of error if not aligned with the prediction horizon. Overall, MAE's less than 2 were calculated for predictions out to 72 hours. Novel deep learning techniques were used to train an 8-hour averaged ozone forecast model. Missing data and outliers within the captured data set were replaced using a new imputation method that generated calculated values closer to the expected value based on the time and season. Decision trees were used to identify input variables with the greatest importance. The methods presented in this paper allow air managers to forecast long range air pollution concentration while only monitoring key parameters and without transforming the data set in its entirety, thus allowing real time inputs and continuous prediction.

Neurological and developmental outcome in extremely preterm children born in England in 1995 and 2006: the EPICure studies.

PubMed

Moore, Tamanna; Hennessy, Enid M; Myles, Jonathan; Johnson, Samantha J; Draper, Elizabeth S; Costeloe, Kate L; Marlow, Neil

2012-12-04

To determine outcomes at age 3 years in babies born before 27 completed weeks' gestation in 2006, and to evaluate changes in outcome since 1995 for babies born between 22 and 25 weeks' gestation. Prospective national cohort studies, EPICure and EPICure 2. Hospital and home based evaluations, England. 1031 surviving babies born in 2006 before 27 completed weeks' gestation. Outcomes for 584 babies born at 22-25 weeks' gestation were compared with those of 260 surviving babies of the same gestational age born in 1995. Survival to age 3 years, impairment (2008 consensus definitions), and developmental scores. Multiple imputation was used to account for the high proportion of missing data in the 2006 cohort. Of the 576 babies evaluated after birth in 2006, 13.4% (n=77) were categorised as having severe impairment and 11.8% (n=68) moderate impairment. The prevalence of neurodevelopmental impairment was significantly associated with length of gestation, with greater impairment as gestational age decreased: 45% at 22-23 weeks, 30% at 24 weeks, 25% at 25 weeks, and 20% at 26 weeks (P<0.001). Cerebral palsy was present in 83 (14%) survivors. Mean developmental quotients were lower than those of the general population (normal values 100 (SD 15)) and showed a direct relation with gestational age: 80 (SD 21) at 22-23 weeks, 87 (19) at 24 weeks, 88 (19) at 25 weeks, and 91 (18) at 26 weeks. These results did not differ significantly after imputation. Comparing imputed outcomes between the 2006 and 1995 cohorts, the proportion of survivors born between 22 and 25 weeks' gestation with severe disability, using 1995 definitions, was 18% (95% confidence interval 14% to 24%) in 1995 and 19% (14% to 23%) in 2006. Fewer survivors had shunted hydrocephalus or seizures. Survival of babies admitted for neonatal care increased from 39% (35% to 43%) in 1995 to 52% (49% to 55%) in 2006, an increase of 13% (8% to 18%), and survival without disability increased from 23% (20% to 26%) in 1995 to 34% (31% to 37%) in 2006, an increase of 11% (6% to 16%). Survival and impairment in early childhood are both closely related to gestational age for babies born at less than 27 weeks' gestation. Using multiple imputation to account for the high proportion of missing values, a higher proportion of babies admitted for neonatal care now survive without disability, particularly those born at gestational ages 24 and 25 weeks.

Missing data in forest ecology and management: advances in quantitative methods [Preface

Treesearch

Tara Barrett; Matti Maltomo

2012-01-01

In recent years, substantial progress has been made for handling missing data issues for applications in forest ecology and management, particularly in the area of imputation techniques. A session on this topic was held at the XXlll IUFRO World Congress in Seoul, South Korea, on August 23-28, 2010, resulting in this special issue of six papers that address recent...

A SPATIOTEMPORAL APPROACH FOR HIGH RESOLUTION TRAFFIC FLOW IMPUTATION

DOE Office of Scientific and Technical Information (OSTI.GOV)

Han, Lee; Chin, Shih-Miao; Hwang, Ho-Ling

Along with the rapid development of Intelligent Transportation Systems (ITS), traffic data collection technologies have been evolving dramatically. The emergence of innovative data collection technologies such as Remote Traffic Microwave Sensor (RTMS), Bluetooth sensor, GPS-based Floating Car method, automated license plate recognition (ALPR) (1), etc., creates an explosion of traffic data, which brings transportation engineering into the new era of Big Data. However, despite the advance of technologies, the missing data issue is still inevitable and has posed great challenges for research such as traffic forecasting, real-time incident detection and management, dynamic route guidance, and massive evacuation optimization, because themore » degree of success of these endeavors depends on the timely availability of relatively complete and reasonably accurate traffic data. A thorough literature review suggests most current imputation models, if not all, focus largely on the temporal nature of the traffic data and fail to consider the fact that traffic stream characteristics at a certain location are closely related to those at neighboring locations and utilize these correlations for data imputation. To this end, this paper presents a Kriging based spatiotemporal data imputation approach that is able to fully utilize the spatiotemporal information underlying in traffic data. Imputation performance of the proposed approach was tested using simulated scenarios and achieved stable imputation accuracy. Moreover, the proposed Kriging imputation model is more flexible compared to current models.« less

The current practice of handling and reporting missing outcome data in eight widely used PROMs in RCT publications: a review of the current literature.

PubMed

Rombach, Ines; Rivero-Arias, Oliver; Gray, Alastair M; Jenkinson, Crispin; Burke, Órlaith

2016-07-01

Patient-reported outcome measures (PROMs) are designed to assess patients' perceived health states or health-related quality of life. However, PROMs are susceptible to missing data, which can affect the validity of conclusions from randomised controlled trials (RCTs). This review aims to assess current practice in the handling, analysis and reporting of missing PROMs outcome data in RCTs compared to contemporary methodology and guidance. This structured review of the literature includes RCTs with a minimum of 50 participants per arm. Studies using the EQ-5D-3L, EORTC QLQ-C30, SF-12 and SF-36 were included if published in 2013; those using the less commonly implemented HUI, OHS, OKS and PDQ were included if published between 2009 and 2013. The review included 237 records (4-76 per relevant PROM). Complete case analysis and single imputation were commonly used in 33 and 15 % of publications, respectively. Multiple imputation was reported for 9 % of the PROMs reviewed. The majority of publications (93 %) failed to describe the assumed missing data mechanism, while low numbers of papers reported methods to minimise missing data (23 %), performed sensitivity analyses (22 %) or discussed the potential influence of missing data on results (16 %). Considerable discrepancy exists between approved methodology and current practice in handling, analysis and reporting of missing PROMs outcome data in RCTs. Greater awareness is needed for the potential biases introduced by inappropriate handling of missing data, as well as the importance of sensitivity analysis and clear reporting to enable appropriate assessments of treatment effects and conclusions from RCTs.

Controlled pattern imputation for sensitivity analysis of longitudinal binary and ordinal outcomes with nonignorable dropout.

PubMed

Tang, Yongqiang

2018-04-30

The controlled imputation method refers to a class of pattern mixture models that have been commonly used as sensitivity analyses of longitudinal clinical trials with nonignorable dropout in recent years. These pattern mixture models assume that participants in the experimental arm after dropout have similar response profiles to the control participants or have worse outcomes than otherwise similar participants who remain on the experimental treatment. In spite of its popularity, the controlled imputation has not been formally developed for longitudinal binary and ordinal outcomes partially due to the lack of a natural multivariate distribution for such endpoints. In this paper, we propose 2 approaches for implementing the controlled imputation for binary and ordinal data based respectively on the sequential logistic regression and the multivariate probit model. Efficient Markov chain Monte Carlo algorithms are developed for missing data imputation by using the monotone data augmentation technique for the sequential logistic regression and a parameter-expanded monotone data augmentation scheme for the multivariate probit model. We assess the performance of the proposed procedures by simulation and the analysis of a schizophrenia clinical trial and compare them with the fully conditional specification, last observation carried forward, and baseline observation carried forward imputation methods. Copyright © 2018 John Wiley & Sons, Ltd.

Imputing missing data via sparse reconstruction techniques.

DOT National Transportation Integrated Search

2017-06-01

The State of Texas does not currently have an automated approach for estimating volumes for links without counts. This research project proposes the development of an automated system to efficiently estimate the traffic volumes on uncounted links, in...

The Contribution of Missed Clinic Visits to Disparities in HIV Viral Load Outcomes

PubMed Central

Westfall, Andrew O.; Gardner, Lytt I.; Giordano, Thomas P.; Wilson, Tracey E.; Drainoni, Mari-Lynn; Keruly, Jeanne C.; Rodriguez, Allan E.; Malitz, Faye; Batey, D. Scott; Mugavero, Michael J.

2015-01-01

Objectives. We explored the contribution of missed primary HIV care visits (“no-show”) to observed disparities in virological failure (VF) among Black persons and persons with injection drug use (IDU) history. Methods. We used patient-level data from 6 academic clinics, before the Centers for Disease Control and Prevention and Health Resources and Services Administration Retention in Care intervention. We employed staged multivariable logistic regression and multivariable models stratified by no-show visit frequency to evaluate the association of sociodemographic factors with VF. We used multiple imputations to assign missing viral load values. Results. Among 10 053 patients (mean age = 46 years; 35% female; 64% Black; 15% with IDU history), 31% experienced VF. Although Black patients and patients with IDU history were significantly more likely to experience VF in initial analyses, race and IDU parameter estimates were attenuated after sequential addition of no-show frequency. In stratified models, race and IDU were not statistically significantly associated with VF at any no-show level. Conclusions. Because missed clinic visits contributed to observed differences in viral load outcomes among Black and IDU patients, achieving an improved understanding of differential visit attendance is imperative to reducing disparities in HIV. PMID:26270301

A guide to missing data for the pediatric nephrologist.

PubMed

Larkins, Nicholas G; Craig, Jonathan C; Teixeira-Pinto, Armando

2018-03-13

Missing data is an important and common source of bias in clinical research. Readers should be alert to and consider the impact of missing data when reading studies. Beyond preventing missing data in the first place, through good study design and conduct, there are different strategies available to handle data containing missing observations. Complete case analysis is often biased unless data are missing completely at random. Better methods of handling missing data include multiple imputation and models using likelihood-based estimation. With advancing computing power and modern statistical software, these methods are within the reach of clinician-researchers under guidance of a biostatistician. As clinicians reading papers, we need to continue to update our understanding of statistical methods, so that we understand the limitations of these techniques and can critically interpret literature.

On normality, ethnicity, and missing values in quantitative trait locus mapping

PubMed Central

Labbe, Aurélie; Wormald, Hanna

2005-01-01

Background This paper deals with the detection of significant linkage for quantitative traits using a variance components approach. Microsatellite markers were obtained for the Genetic Analysis Workshop 14 Collaborative Study on the Genetics of Alcoholism data. Ethnic heterogeneity, highly skewed quantitative measures, and a high rate of missing values are all present in this dataset and well known to impact upon linkage analysis. This makes it a good candidate for investigation. Results As expected, we observed a number of changes in LOD scores, especially for chromosomes 1, 7, and 18, along with the three factors studied. A dramatic example of such changes can be found in chromosome 7. Highly significant linkage to one of the quantitative traits became insignificant when a proper normalizing transformation of the trait was used and when analysis was carried out on an ethnically homogeneous subset of the original pedigrees. Conclusion In agreement with existing literature, transforming a trait to ensure normality using a Box-Cox transformation is highly recommended in order to avoid false-positive linkages. Furthermore, pedigrees should be sorted by ethnic groups and analyses should be carried out separately. Finally, one should be aware that the inclusion of covariates with a high rate of missing values reduces considerably the number of subjects included in the model. In such a case, the loss in power may be large. Imputation methods are then recommended. PMID:16451664

Human papillomavirus in oropharyngeal cancer in Canada: analysis of 5 comprehensive cancer centres using multiple imputation

PubMed Central

Habbous, Steven; Chu, Karen P.; Lau, Harold; Schorr, Melissa; Belayneh, Mathieos; Ha, Michael N.; Murray, Scott; O’Sullivan, Brian; Huang, Shao Hui; Snow, Stephanie; Parliament, Matthew; Hao, Desiree; Cheung, Winson Y.; Xu, Wei; Liu, Geoffrey

2017-01-01

BACKGROUND: The incidence of oropharyngeal cancer has risen over the past 2 decades. This rise has been attributed to human papillomavirus (HPV), but information on temporal trends in incidence of HPV-associated cancers across Canada is limited. METHODS: We collected social, clinical and demographic characteristics and p16 protein status (p16-positive or p16-negative, using this immunohistochemistry variable as a surrogate marker of HPV status) for 3643 patients with oropharyngeal cancer diagnosed between 2000 and 2012 at comprehensive cancer centres in British Columbia (6 centres), Edmonton, Calgary, Toronto and Halifax. We used receiver operating characteristic curves and multiple imputation to estimate the p16 status for missing values. We chose a best-imputation probability cut point on the basis of accuracy in samples with known p16 status and through an independent relation between p16 status and overall survival. We used logistic and Cox proportional hazard regression. RESULTS: We found no temporal changes in p16-positive status initially, but there was significant selection bias, with p16 testing significantly more likely to be performed in males, lifetime never-smokers, patients with tonsillar or base-of-tongue tumours and those with nodal involvement (p < 0.05 for each variable). We used the following variables associated with p16-positive status for multiple imputation: male sex, tonsillar or base-of-tongue tumours, smaller tumours, nodal involvement, less smoking and lower alcohol consumption (p < 0.05 for each variable). Using sensitivity analyses, we showed that different imputation probability cut points for p16-positive status each identified a rise from 2000 to 2012, with the best-probability cut point identifying an increase from 47.3% in 2000 to 73.7% in 2012 (p < 0.001). INTERPRETATION: Across multiple centres in Canada, there was a steady rise in the proportion of oropharyngeal cancers attributable to HPV from 2000 to 2012. PMID:28808115

Estimates of alcohol involvement in fatal crashes : new alcohol methodology

DOT National Transportation Integrated Search

2002-01-01

The National Highway Traffic Safety Administration (NHTSA) has adopted a new method to : estimate missing blood alcohol concentration (BAC) test result data. This new method, multiple : imputation, will be used by NHTSAs National Center for Statis...

Multiple imputation for cure rate quantile regression with censored data.

PubMed

Wu, Yuanshan; Yin, Guosheng

2017-03-01

The main challenge in the context of cure rate analysis is that one never knows whether censored subjects are cured or uncured, or whether they are susceptible or insusceptible to the event of interest. Considering the susceptible indicator as missing data, we propose a multiple imputation approach to cure rate quantile regression for censored data with a survival fraction. We develop an iterative algorithm to estimate the conditionally uncured probability for each subject. By utilizing this estimated probability and Bernoulli sample imputation, we can classify each subject as cured or uncured, and then employ the locally weighted method to estimate the quantile regression coefficients with only the uncured subjects. Repeating the imputation procedure multiple times and taking an average over the resultant estimators, we obtain consistent estimators for the quantile regression coefficients. Our approach relaxes the usual global linearity assumption, so that we can apply quantile regression to any particular quantile of interest. We establish asymptotic properties for the proposed estimators, including both consistency and asymptotic normality. We conduct simulation studies to assess the finite-sample performance of the proposed multiple imputation method and apply it to a lung cancer study as an illustration. © 2016, The International Biometric Society.

[Prevention and handling of missing data in clinical trials].

PubMed

Jiang, Zhi-wei; Li, Chan-juan; Wang, Ling; Xia, Jie-lai

2015-11-01

Missing data is a common but unavoidable issue in clinical trials. It not only lowers the trial power, but brings the bias to the trial results. Therefore, on one hand, the missing data handling methods are employed in data analysis. On the other hand, it is vital to prevent the missing data in the trials. Prevention of missing data should take the first place. From the perspective of data, firstly, some measures should be taken at the stages of protocol design, data collection and data check to enhance the patients' compliance and reduce the unnecessary missing data. Secondly, the causes of confirmed missing data in the trials should be notified and recorded in detail, which are very important to determine the mechanism of missing data and choose the suitable missing data handling methods, e.g., last observation carried forward (LOCF); multiple imputation (MI); mixed-effect model repeated measure (MMRM), etc.

Statistical robustness of machine-learning estimates for characterizing a groundwater-surface water system, Southland, New Zealand

NASA Astrophysics Data System (ADS)

Friedel, M. J.; Daughney, C.

2016-12-01

The development of a successful surface-groundwater management strategy depends on the quality of data provided for analysis. This study evaluates the statistical robustness when using a modified self-organizing map (MSOM) technique to estimate missing values for three hypersurface models: synoptic groundwater-surface water hydrochemistry, time-series of groundwater-surface water hydrochemistry, and mixed-survey (combination of groundwater-surface water hydrochemistry and lithologies) hydrostratigraphic unit data. These models of increasing complexity are developed and validated based on observations from the Southland region of New Zealand. In each case, the estimation method is sufficiently robust to cope with groundwater-surface water hydrochemistry vagaries due to sample size and extreme data insufficiency, even when >80% of the data are missing. The estimation of surface water hydrochemistry time series values enabled the evaluation of seasonal variation, and the imputation of lithologies facilitated the evaluation of hydrostratigraphic controls on groundwater-surface water interaction. The robust statistical results for groundwater-surface water models of increasing data complexity provide justification to apply the MSOM technique in other regions of New Zealand and abroad.

A comparison of genomic selection models across time in interior spruce (Picea engelmannii × glauca) using unordered SNP imputation methods

PubMed Central

Ratcliffe, B; El-Dien, O G; Klápště, J; Porth, I; Chen, C; Jaquish, B; El-Kassaby, Y A

2015-01-01

Genomic selection (GS) potentially offers an unparalleled advantage over traditional pedigree-based selection (TS) methods by reducing the time commitment required to carry out a single cycle of tree improvement. This quality is particularly appealing to tree breeders, where lengthy improvement cycles are the norm. We explored the prospect of implementing GS for interior spruce (Picea engelmannii × glauca) utilizing a genotyped population of 769 trees belonging to 25 open-pollinated families. A series of repeated tree height measurements through ages 3–40 years permitted the testing of GS methods temporally. The genotyping-by-sequencing (GBS) platform was used for single nucleotide polymorphism (SNP) discovery in conjunction with three unordered imputation methods applied to a data set with 60% missing information. Further, three diverse GS models were evaluated based on predictive accuracy (PA), and their marker effects. Moderate levels of PA (0.31–0.55) were observed and were of sufficient capacity to deliver improved selection response over TS. Additionally, PA varied substantially through time accordingly with spatial competition among trees. As expected, temporal PA was well correlated with age-age genetic correlation (r=0.99), and decreased substantially with increasing difference in age between the training and validation populations (0.04–0.47). Moreover, our imputation comparisons indicate that k-nearest neighbor and singular value decomposition yielded a greater number of SNPs and gave higher predictive accuracies than imputing with the mean. Furthermore, the ridge regression (rrBLUP) and BayesCπ (BCπ) models both yielded equal, and better PA than the generalized ridge regression heteroscedastic effect model for the traits evaluated. PMID:26126540

A comparison of genomic selection models across time in interior spruce (Picea engelmannii × glauca) using unordered SNP imputation methods.

PubMed

Ratcliffe, B; El-Dien, O G; Klápště, J; Porth, I; Chen, C; Jaquish, B; El-Kassaby, Y A

2015-12-01

Genomic selection (GS) potentially offers an unparalleled advantage over traditional pedigree-based selection (TS) methods by reducing the time commitment required to carry out a single cycle of tree improvement. This quality is particularly appealing to tree breeders, where lengthy improvement cycles are the norm. We explored the prospect of implementing GS for interior spruce (Picea engelmannii × glauca) utilizing a genotyped population of 769 trees belonging to 25 open-pollinated families. A series of repeated tree height measurements through ages 3-40 years permitted the testing of GS methods temporally. The genotyping-by-sequencing (GBS) platform was used for single nucleotide polymorphism (SNP) discovery in conjunction with three unordered imputation methods applied to a data set with 60% missing information. Further, three diverse GS models were evaluated based on predictive accuracy (PA), and their marker effects. Moderate levels of PA (0.31-0.55) were observed and were of sufficient capacity to deliver improved selection response over TS. Additionally, PA varied substantially through time accordingly with spatial competition among trees. As expected, temporal PA was well correlated with age-age genetic correlation (r=0.99), and decreased substantially with increasing difference in age between the training and validation populations (0.04-0.47). Moreover, our imputation comparisons indicate that k-nearest neighbor and singular value decomposition yielded a greater number of SNPs and gave higher predictive accuracies than imputing with the mean. Furthermore, the ridge regression (rrBLUP) and BayesCπ (BCπ) models both yielded equal, and better PA than the generalized ridge regression heteroscedastic effect model for the traits evaluated.

Parameter estimation in Cox models with missing failure indicators and the OPPERA study.

PubMed

Brownstein, Naomi C; Cai, Jianwen; Slade, Gary D; Bair, Eric

2015-12-30

In a prospective cohort study, examining all participants for incidence of the condition of interest may be prohibitively expensive. For example, the "gold standard" for diagnosing temporomandibular disorder (TMD) is a physical examination by a trained clinician. In large studies, examining all participants in this manner is infeasible. Instead, it is common to use questionnaires to screen for incidence of TMD and perform the "gold standard" examination only on participants who screen positively. Unfortunately, some participants may leave the study before receiving the "gold standard" examination. Within the framework of survival analysis, this results in missing failure indicators. Motivated by the Orofacial Pain: Prospective Evaluation and Risk Assessment (OPPERA) study, a large cohort study of TMD, we propose a method for parameter estimation in survival models with missing failure indicators. We estimate the probability of being an incident case for those lacking a "gold standard" examination using logistic regression. These estimated probabilities are used to generate multiple imputations of case status for each missing examination that are combined with observed data in appropriate regression models. The variance introduced by the procedure is estimated using multiple imputation. The method can be used to estimate both regression coefficients in Cox proportional hazard models as well as incidence rates using Poisson regression. We simulate data with missing failure indicators and show that our method performs as well as or better than competing methods. Finally, we apply the proposed method to data from the OPPERA study. Copyright © 2015 John Wiley & Sons, Ltd.

Gap-filling meteorological variables with Empirical Orthogonal Functions

NASA Astrophysics Data System (ADS)

Graf, Alexander

2017-04-01

Gap-filling or modelling surface-atmosphere fluxes critically depends on an, ideally continuous, availability of their meteorological driver variables, such as e.g. air temperature, humidity, radiation, wind speed and precipitation. Unlike for eddy-covariance-based fluxes, data gaps are not unavoidable for these measurements. Nevertheless, missing or erroneous data can occur in practice due to instrument or power failures, disturbance, and temporary sensor or station dismounting for e.g. agricultural management or maintenance. If stations with similar measurements are available nearby, using their data for imputation (i.e. estimating missing data) either directly, after an elevation correction or via linear regression, is usually preferred over linear interpolation or monthly mean diurnal cycles. The popular implementation of regional networks of (partly low-cost) stations increases both, the need and the potential, for such neighbour-based imputation methods. For repeated satellite imagery, Beckers and Rixen (2003) suggested an imputation method based on empirical orthogonal functions (EOFs). While exploiting the same linear relations between time series at different observation points as regression, it is able to use information from all observation points to simultaneously estimate missing data at all observation points, provided that never all observations are missing at the same time. Briefly, the method uses the ability of the first few EOFs of a data matrix to reconstruct a noise-reduced version of this matrix; iterating missing data points from an initial guess (the column-wise averages) to an optimal version determined by cross-validation. The poster presents and discusses lessons learned from adapting and applying this methodology to station data. Several years of 10-minute averages of air temperature, pressure and humidity, incoming shortwave, longwave and photosynthetically active radiation, wind speed and precipitation, measured by a regional (70 km by 20 km by 650 m elevation difference) network of 18 sites, were treated by various modifications of the method. The performance per variable and as a function of methodology, such as e.g. number of used EOFs and method to determine its optimum, period length and data transformation, is assessed by cross-validation. Beckers, J.-M., Rixen, M. (2003): EOF calculations and data filling from incomplete oceanographic datasets. J. Atmos. Ocean. Tech. 20, 1839-1856.

Accuracy of estimation of genomic breeding values in pigs using low-density genotypes and imputation.

PubMed

Badke, Yvonne M; Bates, Ronald O; Ernst, Catherine W; Fix, Justin; Steibel, Juan P

2014-04-16

Genomic selection has the potential to increase genetic progress. Genotype imputation of high-density single-nucleotide polymorphism (SNP) genotypes can improve the cost efficiency of genomic breeding value (GEBV) prediction for pig breeding. Consequently, the objectives of this work were to: (1) estimate accuracy of genomic evaluation and GEBV for three traits in a Yorkshire population and (2) quantify the loss of accuracy of genomic evaluation and GEBV when genotypes were imputed under two scenarios: a high-cost, high-accuracy scenario in which only selection candidates were imputed from a low-density platform and a low-cost, low-accuracy scenario in which all animals were imputed using a small reference panel of haplotypes. Phenotypes and genotypes obtained with the PorcineSNP60 BeadChip were available for 983 Yorkshire boars. Genotypes of selection candidates were masked and imputed using tagSNP in the GeneSeek Genomic Profiler (10K). Imputation was performed with BEAGLE using 128 or 1800 haplotypes as reference panels. GEBV were obtained through an animal-centric ridge regression model using de-regressed breeding values as response variables. Accuracy of genomic evaluation was estimated as the correlation between estimated breeding values and GEBV in a 10-fold cross validation design. Accuracy of genomic evaluation using observed genotypes was high for all traits (0.65-0.68). Using genotypes imputed from a large reference panel (accuracy: R(2) = 0.95) for genomic evaluation did not significantly decrease accuracy, whereas a scenario with genotypes imputed from a small reference panel (R(2) = 0.88) did show a significant decrease in accuracy. Genomic evaluation based on imputed genotypes in selection candidates can be implemented at a fraction of the cost of a genomic evaluation using observed genotypes and still yield virtually the same accuracy. On the other side, using a very small reference panel of haplotypes to impute training animals and candidates for selection results in lower accuracy of genomic evaluation.

Multinomial Logistic Regression & Bootstrapping for Bayesian Estimation of Vertical Facies Prediction in Heterogeneous Sandstone Reservoirs

NASA Astrophysics Data System (ADS)

Al-Mudhafar, W. J.

2013-12-01

Precisely prediction of rock facies leads to adequate reservoir characterization by improving the porosity-permeability relationships to estimate the properties in non-cored intervals. It also helps to accurately identify the spatial facies distribution to perform an accurate reservoir model for optimal future reservoir performance. In this paper, the facies estimation has been done through Multinomial logistic regression (MLR) with respect to the well logs and core data in a well in upper sandstone formation of South Rumaila oil field. The entire independent variables are gamma rays, formation density, water saturation, shale volume, log porosity, core porosity, and core permeability. Firstly, Robust Sequential Imputation Algorithm has been considered to impute the missing data. This algorithm starts from a complete subset of the dataset and estimates sequentially the missing values in an incomplete observation by minimizing the determinant of the covariance of the augmented data matrix. Then, the observation is added to the complete data matrix and the algorithm continues with the next observation with missing values. The MLR has been chosen to estimate the maximum likelihood and minimize the standard error for the nonlinear relationships between facies & core and log data. The MLR is used to predict the probabilities of the different possible facies given each independent variable by constructing a linear predictor function having a set of weights that are linearly combined with the independent variables by using a dot product. Beta distribution of facies has been considered as prior knowledge and the resulted predicted probability (posterior) has been estimated from MLR based on Baye's theorem that represents the relationship between predicted probability (posterior) with the conditional probability and the prior knowledge. To assess the statistical accuracy of the model, the bootstrap should be carried out to estimate extra-sample prediction error by randomly drawing datasets with replacement from the training data. Each sample has the same size of the original training set and it can be conducted N times to produce N bootstrap datasets to re-fit the model accordingly to decrease the squared difference between the estimated and observed categorical variables (facies) leading to decrease the degree of uncertainty.

Astronaut Preflight Cardiovascular Variables Associated with Vascular Compliance are Highly Correlated with Post-Flight Eye Outcome Measures in the Visual Impairment Intracranial Pressure (VIIP) Syndrome Following Long Duration Spaceflight

NASA Technical Reports Server (NTRS)

Otto, Christian; Ploutz-Snyder, R.

2015-01-01

The detection of the first VIIP case occurred in 2005, and adequate eye outcome measures were available for 31 (67.4%) of the 46 long duration US crewmembers who had flown on the ISS since its first crewed mission in 2000. Therefore, this analysis is limited to a subgroup (22 males and 9 females). A "cardiovascular profile" for each astronaut was compiled by examining twelve individual parameters; eleven of these were preflight variables: systolic blood pressure, pulse pressure, body mass index, percentage body fat, LDL, HDL, triglycerides, use of anti-lipid medication, fasting serum glucose, and maximal oxygen uptake in ml/kg. Each of these variables was averaged across three preflight annual physical exams. Astronaut age prior to the long duration mission, and inflight salt intake was also included in the analysis. The group of cardiovascular variables for each crew member was compared with seven VIIP eye outcome variables collected during the immediate post-flight period: anterior-posterior axial length of the globe measured by ultrasound and optical biometry; optic nerve sheath diameter, optic nerve diameter, and optic nerve to sheath ratio- each measured by ultrasound and magnetic resonance imaging (MRI), intraocular pressure (IOP), change in manifest refraction, mean retinal nerve fiber layer (RNFL) on optical coherence tomography (OCT), and RNFL of the inferior and superior retinal quadrants. Since most of the VIIP eye outcome measures were added sequentially beginning in 2005, as knowledge of the syndrome improved, data were unavailable for 22.0% of the outcome measurements. To address the missing data, we employed multivariate multiple imputation techniques with predictive mean matching methods to accumulate 200 separate imputed datasets for analysis. We were able to impute data for the 22.0% of missing VIIP eye outcomes. We then applied Rubin's rules for collapsing the statistical results across our 200 multiply imputed data sets to assess the canonical correlation between the eye outcomes and the twelve astronaut cardiovascular variables available for all 31 subjects. Results: A highly significant canonical correlation was observed among the canonical solutions (p<.00001), with an average best canonical correlation of.97. The results suggest a strong association between astronauts' measures of cardiovascular health and the seven eye outcomes of the VIIP syndrome used in this analysis. Furthermore, the "joint test" revealed a significant difference in cardiovascular profile between male and female astronauts (Prob > F = 0.00001). Overall, female astronauts demonstrated a significantly healthier cardiovascular status. Individually, the female astronauts had significantly healthier profiles on seven of twelve cardiovascular variables than the men (p values ranging from <0.0001 to <0.05). Male astronauts did not demonstrate significantly healthier values on any of the twelve cardiovascular variables measured

Analysis of partially observed clustered data using generalized estimating equations and multiple imputation

PubMed Central

Aloisio, Kathryn M.; Swanson, Sonja A.; Micali, Nadia; Field, Alison; Horton, Nicholas J.

2015-01-01

Clustered data arise in many settings, particularly within the social and biomedical sciences. As an example, multiple–source reports are commonly collected in child and adolescent psychiatric epidemiologic studies where researchers use various informants (e.g. parent and adolescent) to provide a holistic view of a subject’s symptomatology. Fitzmaurice et al. (1995) have described estimation of multiple source models using a standard generalized estimating equation (GEE) framework. However, these studies often have missing data due to additional stages of consent and assent required. The usual GEE is unbiased when missingness is Missing Completely at Random (MCAR) in the sense of Little and Rubin (2002). This is a strong assumption that may not be tenable. Other options such as weighted generalized estimating equations (WEEs) are computationally challenging when missingness is non–monotone. Multiple imputation is an attractive method to fit incomplete data models while only requiring the less restrictive Missing at Random (MAR) assumption. Previously estimation of partially observed clustered data was computationally challenging however recent developments in Stata have facilitated their use in practice. We demonstrate how to utilize multiple imputation in conjunction with a GEE to investigate the prevalence of disordered eating symptoms in adolescents reported by parents and adolescents as well as factors associated with concordance and prevalence. The methods are motivated by the Avon Longitudinal Study of Parents and their Children (ALSPAC), a cohort study that enrolled more than 14,000 pregnant mothers in 1991–92 and has followed the health and development of their children at regular intervals. While point estimates were fairly similar to the GEE under MCAR, the MAR model had smaller standard errors, while requiring less stringent assumptions regarding missingness. PMID:25642154

Evaluation of techniques for handling missing cost-to-charge ratios in the USA Nationwide Inpatient Sample: a simulation study.

PubMed

Yu, Tzy-Chyi; Zhou, Huanxue

2015-09-01

Evaluate performance of techniques used to handle missing cost-to-charge ratio (CCR) data in the USA Healthcare Cost and Utilization Project's Nationwide Inpatient Sample. Four techniques to replace missing CCR data were evaluated: deleting discharges with missing CCRs (complete case analysis), reweighting as recommended by Healthcare Cost and Utilization Project, reweighting by adjustment cells and hot deck imputation by adjustment cells. Bias and root mean squared error of these techniques on hospital cost were evaluated in five disease cohorts. Similar mean cost estimates would be obtained with any of the four techniques when the percentage of missing data is low (<10%). When total cost is the outcome of interest, a reweighting technique to avoid underestimation from dropping observations with missing data should be adopted.

A multiple imputation strategy for sequential multiple assignment randomized trials

PubMed Central

Shortreed, Susan M.; Laber, Eric; Stroup, T. Scott; Pineau, Joelle; Murphy, Susan A.

2014-01-01

Sequential multiple assignment randomized trials (SMARTs) are increasingly being used to inform clinical and intervention science. In a SMART, each patient is repeatedly randomized over time. Each randomization occurs at a critical decision point in the treatment course. These critical decision points often correspond to milestones in the disease process or other changes in a patient’s health status. Thus, the timing and number of randomizations may vary across patients and depend on evolving patient-specific information. This presents unique challenges when analyzing data from a SMART in the presence of missing data. This paper presents the first comprehensive discussion of missing data issues typical of SMART studies: we describe five specific challenges, and propose a flexible imputation strategy to facilitate valid statistical estimation and inference using incomplete data from a SMART. To illustrate these contributions, we consider data from the Clinical Antipsychotic Trial of Intervention and Effectiveness (CATIE), one of the most well-known SMARTs to date. PMID:24919867

The treatment of missing data in a large cardiovascular clinical outcomes study.

PubMed

Little, Roderick J; Wang, Julia; Sun, Xiang; Tian, Hong; Suh, Eun-Young; Lee, Michael; Sarich, Troy; Oppenheimer, Leonard; Plotnikov, Alexei; Wittes, Janet; Cook-Bruns, Nancy; Burton, Paul; Gibson, C Michael; Mohanty, Surya

2016-06-01

The potential impact of missing data on the results of clinical trials has received heightened attention recently. A National Research Council study provides recommendations for limiting missing data in clinical trial design and conduct, and principles for analysis, including the need for sensitivity analyses to assess robustness of findings to alternative assumptions about the missing data. A Food and Drug Administration advisory committee raised missing data as a serious concern in their review of results from the ATLAS ACS 2 TIMI 51 study, a large clinical trial that assessed rivaroxaban for its ability to reduce the risk of cardiovascular death, myocardial infarction or stroke in patients with acute coronary syndrome. This case study describes a variety of measures that were taken to address concerns about the missing data. A range of analyses are described to assess the potential impact of missing data on conclusions. In particular, measures of the amount of missing data are discussed, and the fraction of missing information from multiple imputation is proposed as an alternative measure. The sensitivity analysis in the National Research Council study is modified in the context of survival analysis where some individuals are lost to follow-up. The impact of deviations from ignorable censoring is assessed by differentially increasing the hazard of the primary outcome in the treatment groups and multiply imputing events between dropout and the end of the study. Tipping-point analyses are described, where the deviation from ignorable censoring that results in a reversal of significance of the treatment effect is determined. A study to determine the vital status of participants lost to follow-up was also conducted, and the results of including this additional information are assessed. Sensitivity analyses suggest that findings of the ATLAS ACS 2 TIMI 51 study are robust to missing data; this robustness is reinforced by the follow-up study, since inclusion of data from this study had little impact on the study conclusions. Missing data are a serious problem in clinical trials. The methods presented here, namely, the sensitivity analyses, the follow-up study to determine survival of missing cases, and the proposed measurement of missing data via the fraction of missing information, have potential application in other studies involving survival analysis where missing data are a concern. © The Author(s) 2016.

A Hot-Deck Multiple Imputation Procedure for Gaps in Longitudinal Recurrent Event Histories

PubMed Central

Wang, Chia-Ning; Little, Roderick; Nan, Bin; Harlow, Siobán D.

2012-01-01

Summary We propose a regression-based hot deck multiple imputation method for gaps of missing data in longitudinal studies, where subjects experience a recurrent event process and a terminal event. Examples are repeated asthma episodes and death, or menstrual periods and the menopause, as in our motivating application. Research interest concerns the onset time of a marker event, defined by the recurrent-event process, or the duration from this marker event to the final event. Gaps in the recorded event history make it difficult to determine the onset time of the marker event, and hence, the duration from onset to the final event. Simple approaches such as jumping gap times or dropping cases with gaps have obvious limitations. We propose a procedure for imputing information in the gaps by substituting information in the gap from a matched individual with a completely recorded history in the corresponding interval. Predictive Mean Matching is used to incorporate information on longitudinal characteristics of the repeated process and the final event time. Multiple imputation is used to propagate imputation uncertainty. The procedure is applied to an important data set for assessing the timing and duration of the menopausal transition. The performance of the proposed method is assessed by a simulation study. PMID:21361886

Hierarchical Bayesian modelling of gene expression time series across irregularly sampled replicates and clusters.

PubMed

Hensman, James; Lawrence, Neil D; Rattray, Magnus

2013-08-20

Time course data from microarrays and high-throughput sequencing experiments require simple, computationally efficient and powerful statistical models to extract meaningful biological signal, and for tasks such as data fusion and clustering. Existing methodologies fail to capture either the temporal or replicated nature of the experiments, and often impose constraints on the data collection process, such as regularly spaced samples, or similar sampling schema across replications. We propose hierarchical Gaussian processes as a general model of gene expression time-series, with application to a variety of problems. In particular, we illustrate the method's capacity for missing data imputation, data fusion and clustering.The method can impute data which is missing both systematically and at random: in a hold-out test on real data, performance is significantly better than commonly used imputation methods. The method's ability to model inter- and intra-cluster variance leads to more biologically meaningful clusters. The approach removes the necessity for evenly spaced samples, an advantage illustrated on a developmental Drosophila dataset with irregular replications. The hierarchical Gaussian process model provides an excellent statistical basis for several gene-expression time-series tasks. It has only a few additional parameters over a regular GP, has negligible additional complexity, is easily implemented and can be integrated into several existing algorithms. Our experiments were implemented in python, and are available from the authors' website: http://staffwww.dcs.shef.ac.uk/people/J.Hensman/.

Using ecological propensity score to adjust for missing confounders in small area studies.

PubMed

Wang, Yingbo; Pirani, Monica; Hansell, Anna L; Richardson, Sylvia; Blangiardo, Marta

2017-11-09

Small area ecological studies are commonly used in epidemiology to assess the impact of area level risk factors on health outcomes when data are only available in an aggregated form. However, the resulting estimates are often biased due to unmeasured confounders, which typically are not available from the standard administrative registries used for these studies. Extra information on confounders can be provided through external data sets such as surveys or cohorts, where the data are available at the individual level rather than at the area level; however, such data typically lack the geographical coverage of administrative registries. We develop a framework of analysis which combines ecological and individual level data from different sources to provide an adjusted estimate of area level risk factors which is less biased. Our method (i) summarizes all available individual level confounders into an area level scalar variable, which we call ecological propensity score (EPS), (ii) implements a hierarchical structured approach to impute the values of EPS whenever they are missing, and (iii) includes the estimated and imputed EPS into the ecological regression linking the risk factors to the health outcome. Through a simulation study, we show that integrating individual level data into small area analyses via EPS is a promising method to reduce the bias intrinsic in ecological studies due to unmeasured confounders; we also apply the method to a real case study to evaluate the effect of air pollution on coronary heart disease hospital admissions in Greater London. © The Author 2017. Published by Oxford University Press.

Principled Missing Data Treatments.

PubMed

Lang, Kyle M; Little, Todd D

2018-04-01

We review a number of issues regarding missing data treatments for intervention and prevention researchers. Many of the common missing data practices in prevention research are still, unfortunately, ill-advised (e.g., use of listwise and pairwise deletion, insufficient use of auxiliary variables). Our goal is to promote better practice in the handling of missing data. We review the current state of missing data methodology and recent missing data reporting in prevention research. We describe antiquated, ad hoc missing data treatments and discuss their limitations. We discuss two modern, principled missing data treatments: multiple imputation and full information maximum likelihood, and we offer practical tips on how to best employ these methods in prevention research. The principled missing data treatments that we discuss are couched in terms of how they improve causal and statistical inference in the prevention sciences. Our recommendations are firmly grounded in missing data theory and well-validated statistical principles for handling the missing data issues that are ubiquitous in biosocial and prevention research. We augment our broad survey of missing data analysis with references to more exhaustive resources.

A Cautious Note on Auxiliary Variables That Can Increase Bias in Missing Data Problems.

PubMed

Thoemmes, Felix; Rose, Norman

2014-01-01

The treatment of missing data in the social sciences has changed tremendously during the last decade. Modern missing data techniques such as multiple imputation and full-information maximum likelihood are used much more frequently. These methods assume that data are missing at random. One very common approach to increase the likelihood that missing at random is achieved consists of including many covariates as so-called auxiliary variables. These variables are either included based on data considerations or in an inclusive fashion; that is, taking all available auxiliary variables. In this article, we point out that there are some instances in which auxiliary variables exhibit the surprising property of increasing bias in missing data problems. In a series of focused simulation studies, we highlight some situations in which this type of biasing behavior can occur. We briefly discuss possible ways how one can avoid selecting bias-inducing covariates as auxiliary variables.

Quality Assurance Decisions with Air Models: A Case Study of ImputatIon of Missing Input Data Using EPA's Multi-Layer Model

EPA Science Inventory

Abstract Environmental models are frequently used within regulatory and policy frameworks to estimate environmental metrics that are difficult or impossible to physically measure. As important decision tools, the uncertainty associated with the model outputs should impact their ...

Effect of Ferric Carboxymaltose on Exercise Capacity in Patients With Chronic Heart Failure and Iron Deficiency.

PubMed

van Veldhuisen, Dirk J; Ponikowski, Piotr; van der Meer, Peter; Metra, Marco; Böhm, Michael; Doletsky, Artem; Voors, Adriaan A; Macdougall, Iain C; Anker, Stefan D; Roubert, Bernard; Zakin, Lorraine; Cohen-Solal, Alain

2017-10-10

Iron deficiency is common in patients with heart failure (HF) and is associated with reduced exercise capacity and poor outcomes. Whether correction of iron deficiency with (intravenous) ferric carboxymaltose (FCM) affects peak oxygen consumption [peak VO 2 ], an objective measure of exercise intolerance in HF, has not been examined. We studied patients with systolic HF (left ventricular ejection fraction ≤45%) and mild to moderate symptoms despite optimal HF medication. Patients were randomized 1:1 to treatment with FCM for 24 weeks or standard of care. The primary end point was the change in peak VO 2 from baseline to 24 weeks. Secondary end points included the effect on hematinic and cardiac biomarkers, quality of life, and safety. For the primary analysis, patients who died had a value of 0 imputed for 24-week peak VO 2 . Additional sensitivity analyses were performed to determine the impact of imputation of missing peak VO 2 data. A total of 172 patients with HF were studied and received FCM (n=86) or standard of care (control group, n=86). At baseline, the groups were well matched; mean age was 64 years, 75% were male, mean left ventricular ejection fraction was 32%, and peak VO 2 was 13.5 mL/min/kg. FCM significantly increased serum ferritin and transferrin saturation. At 24 weeks, peak VO 2 had decreased in the control group (least square means -1.19±0.389 mL/min/kg) but was maintained on FCM (-0.16±0.387 mL/min/kg; P =0.020 between groups). In a sensitivity analysis, in which missing data were not imputed, peak VO 2 at 24 weeks decreased by -0.63±0.375 mL/min/kg in the control group and by -0.16±0.373 mL/min/kg in the FCM group; P =0.23 between groups). Patients' global assessment and functional class as assessed by the New York Heart Association improved on FCM versus standard of care. Treatment with intravenous FCM in patients with HF and iron deficiency improves iron stores. Although a favorable effect on peak VO 2 was observed on FCM, compared with standard of care in the primary analysis, this effect was highly sensitive to the imputation strategy for peak VO 2 among patients who died. Whether FCM is associated with an improved outcome in these high-risk patients needs further study. URL: http://www.clinicaltrials.gov. Unique identifier: NCT01394562. © 2017 The Authors.

Effect of Ferric Carboxymaltose on Exercise Capacity in Patients With Chronic Heart Failure and Iron Deficiency

PubMed Central

Ponikowski, Piotr; van der Meer, Peter; Metra, Marco; Böhm, Michael; Doletsky, Artem; Voors, Adriaan A.; Macdougall, Iain C.; Anker, Stefan D.; Roubert, Bernard; Zakin, Lorraine; Cohen-Solal, Alain

2017-01-01

Background: Iron deficiency is common in patients with heart failure (HF) and is associated with reduced exercise capacity and poor outcomes. Whether correction of iron deficiency with (intravenous) ferric carboxymaltose (FCM) affects peak oxygen consumption [peak VO2], an objective measure of exercise intolerance in HF, has not been examined. Methods: We studied patients with systolic HF (left ventricular ejection fraction ≤45%) and mild to moderate symptoms despite optimal HF medication. Patients were randomized 1:1 to treatment with FCM for 24 weeks or standard of care. The primary end point was the change in peak VO2 from baseline to 24 weeks. Secondary end points included the effect on hematinic and cardiac biomarkers, quality of life, and safety. For the primary analysis, patients who died had a value of 0 imputed for 24-week peak VO2. Additional sensitivity analyses were performed to determine the impact of imputation of missing peak VO2 data. Results: A total of 172 patients with HF were studied and received FCM (n=86) or standard of care (control group, n=86). At baseline, the groups were well matched; mean age was 64 years, 75% were male, mean left ventricular ejection fraction was 32%, and peak VO2 was 13.5 mL/min/kg. FCM significantly increased serum ferritin and transferrin saturation. At 24 weeks, peak VO2 had decreased in the control group (least square means −1.19±0.389 mL/min/kg) but was maintained on FCM (−0.16±0.387 mL/min/kg; P=0.020 between groups). In a sensitivity analysis, in which missing data were not imputed, peak VO2 at 24 weeks decreased by −0.63±0.375 mL/min/kg in the control group and by −0.16±0.373 mL/min/kg in the FCM group; P=0.23 between groups). Patients’ global assessment and functional class as assessed by the New York Heart Association improved on FCM versus standard of care. Conclusions: Treatment with intravenous FCM in patients with HF and iron deficiency improves iron stores. Although a favorable effect on peak VO2 was observed on FCM, compared with standard of care in the primary analysis, this effect was highly sensitive to the imputation strategy for peak VO2 among patients who died. Whether FCM is associated with an improved outcome in these high-risk patients needs further study. Clinical Trial Registration: URL: http://www.clinicaltrials.gov. Unique identifier: NCT01394562. PMID:28701470

Estimating the effects of maternal education on child dental caries using marginal structural models: The Longitudinal Study of Indigenous Australian Children.

PubMed

Ju, Xiangqun; Jamieson, Lisa M; Mejia, Gloria C

2016-12-01

To estimate the effect of mothers' education on Indigenous Australian children's dental caries experience while controlling for the mediating effect of children's sweet food intake. The Longitudinal Study of Indigenous Children is a study of two representative cohorts of Indigenous Australian children, aged from 6 months to 2 years (baby cohort) and from 3.5 to 5 years (child cohort) at baseline. The children's primary caregiver undertook a face-to-face interview in 2008 and repeated annually for the next 4 years. Data included household demographics, child health (nutrition information and dental health), maternal conditions and highest qualification levels. Mother's educational level was classified into four categories: 0-9 years, 10 years, 11-12 years and >12 years. Children's mean sweet food intake was categorized as <20%, 20-30%, and >30%. After multiple imputation of missing values, a marginal structural model with stabilized inverse probability weights was used to estimate the direct effect of mothers' education level on children's dental decay experience. From 2008 to 2012, complete data on 1720 mother-child dyads were available. Dental caries experience for children was 42.3% over the 5-year period. The controlled direct effect estimates of mother's education on child dental caries were 1.21 (95% CI: 1.01-1.45), 1.03 (95% CI: 0.91-1.18) and 1.07 (95% CI: 0.93-1.22); after multiple imputation of missing values, the effects were 1.21 (95% CI: 1.05-1.39), 1.06 (95% CI: 0.94-1.19) and 1.06 (95% CI: 0.95-1.19), comparing '0-9', '10' and '11-12' years to > 12 years of education. Mothers' education level had a direct effect on children's dental decay experience that was not mediated by sweet food intake and other risk factors when estimated using a marginal structural model. © 2016 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd.

Results of Database Studies in Spine Surgery Can Be Influenced by Missing Data.

PubMed

Basques, Bryce A; McLynn, Ryan P; Fice, Michael P; Samuel, Andre M; Lukasiewicz, Adam M; Bohl, Daniel D; Ahn, Junyoung; Singh, Kern; Grauer, Jonathan N

2017-12-01

National databases are increasingly being used for research in spine surgery; however, one limitation of such databases that has received sparse mention is the frequency of missing data. Studies using these databases often do not emphasize the percentage of missing data for each variable used and do not specify how patients with missing data are incorporated into analyses. This study uses the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) database to examine whether different treatments of missing data can influence the results of spine studies. (1) What is the frequency of missing data fields for demographics, medical comorbidities, preoperative laboratory values, operating room times, and length of stay recorded in ACS-NSQIP? (2) Using three common approaches to handling missing data, how frequently do those approaches agree in terms of finding particular variables to be associated with adverse events? (3) Do different approaches to handling missing data influence the outcomes and effect sizes of an analysis testing for an association with these variables with occurrence of adverse events? Patients who underwent spine surgery between 2005 and 2013 were identified from the ACS-NSQIP database. A total of 88,471 patients undergoing spine surgery were identified. The most common procedures were anterior cervical discectomy and fusion, lumbar decompression, and lumbar fusion. Demographics, comorbidities, and perioperative laboratory values were tabulated for each patient, and the percent of missing data was noted for each variable. These variables were tested for an association with "any adverse event" using three separate multivariate regressions that used the most common treatments for missing data. In the first regression, patients with any missing data were excluded. In the second regression, missing data were treated as a negative or "reference" value; for continuous variables, the mean of each variable's reference range was computed and imputed. In the third regression, any variables with > 10% rate of missing data were removed from the regression; among variables with ≤ 10% missing data, individual cases with missing values were excluded. The results of these regressions were compared to determine how the different treatments of missing data could affect the results of spine studies using the ACS-NSQIP database. Of the 88,471 patients, as many as 4441 (5%) had missing elements among demographic data, 69,184 (72%) among comorbidities, 70,892 (80%) among preoperative laboratory values, and 56,551 (64%) among operating room times. Considering the three different treatments of missing data, we found different risk factors for adverse events. Of 44 risk factors found to be associated with adverse events in any analysis, only 15 (34%) of these risk factors were common among the three regressions. The second treatment of missing data (assuming "normal" value) found the most risk factors (40) to be associated with any adverse event, whereas the first treatment (deleting patients with missing data) found the fewest associations at 20. Among the risk factors associated with any adverse event, the 10 with the greatest effect size (odds ratio) by each regression were ranked. Of the 15 variables in the top 10 for any regression, six of these were common among all three lists. Differing treatments of missing data can influence the results of spine studies using the ACS-NSQIP. The current study highlights the importance of considering how such missing data are handled. Until there are better guidelines on the best approaches to handle missing data, investigators should report how missing data were handled to increase the quality and transparency of orthopaedic database research. Readers of large database studies should note whether handling of missing data was addressed and consider potential bias with high rates or unspecified or weak methods for handling missing data.

Multiple Imputation in Two-Stage Cluster Samples Using The Weighted Finite Population Bayesian Bootstrap.

PubMed

Zhou, Hanzhi; Elliott, Michael R; Raghunathan, Trivellore E

2016-06-01

Multistage sampling is often employed in survey samples for cost and convenience. However, accounting for clustering features when generating datasets for multiple imputation is a nontrivial task, particularly when, as is often the case, cluster sampling is accompanied by unequal probabilities of selection, necessitating case weights. Thus, multiple imputation often ignores complex sample designs and assumes simple random sampling when generating imputations, even though failing to account for complex sample design features is known to yield biased estimates and confidence intervals that have incorrect nominal coverage. In this article, we extend a recently developed, weighted, finite-population Bayesian bootstrap procedure to generate synthetic populations conditional on complex sample design data that can be treated as simple random samples at the imputation stage, obviating the need to directly model design features for imputation. We develop two forms of this method: one where the probabilities of selection are known at the first and second stages of the design, and the other, more common in public use files, where only the final weight based on the product of the two probabilities is known. We show that this method has advantages in terms of bias, mean square error, and coverage properties over methods where sample designs are ignored, with little loss in efficiency, even when compared with correct fully parametric models. An application is made using the National Automotive Sampling System Crashworthiness Data System, a multistage, unequal probability sample of U.S. passenger vehicle crashes, which suffers from a substantial amount of missing data in "Delta-V," a key crash severity measure.

Multiple Imputation in Two-Stage Cluster Samples Using The Weighted Finite Population Bayesian Bootstrap

PubMed Central

Zhou, Hanzhi; Elliott, Michael R.; Raghunathan, Trivellore E.

2017-01-01

Multistage sampling is often employed in survey samples for cost and convenience. However, accounting for clustering features when generating datasets for multiple imputation is a nontrivial task, particularly when, as is often the case, cluster sampling is accompanied by unequal probabilities of selection, necessitating case weights. Thus, multiple imputation often ignores complex sample designs and assumes simple random sampling when generating imputations, even though failing to account for complex sample design features is known to yield biased estimates and confidence intervals that have incorrect nominal coverage. In this article, we extend a recently developed, weighted, finite-population Bayesian bootstrap procedure to generate synthetic populations conditional on complex sample design data that can be treated as simple random samples at the imputation stage, obviating the need to directly model design features for imputation. We develop two forms of this method: one where the probabilities of selection are known at the first and second stages of the design, and the other, more common in public use files, where only the final weight based on the product of the two probabilities is known. We show that this method has advantages in terms of bias, mean square error, and coverage properties over methods where sample designs are ignored, with little loss in efficiency, even when compared with correct fully parametric models. An application is made using the National Automotive Sampling System Crashworthiness Data System, a multistage, unequal probability sample of U.S. passenger vehicle crashes, which suffers from a substantial amount of missing data in “Delta-V,” a key crash severity measure. PMID:29226161

Adjusting HIV prevalence estimates for non-participation: an application to demographic surveillance

PubMed Central

McGovern, Mark E.; Marra, Giampiero; Radice, Rosalba; Canning, David; Newell, Marie-Louise; Bärnighausen, Till

2015-01-01

Introduction HIV testing is a cornerstone of efforts to combat the HIV epidemic, and testing conducted as part of surveillance provides invaluable data on the spread of infection and the effectiveness of campaigns to reduce the transmission of HIV. However, participation in HIV testing can be low, and if respondents systematically select not to be tested because they know or suspect they are HIV positive (and fear disclosure), standard approaches to deal with missing data will fail to remove selection bias. We implemented Heckman-type selection models, which can be used to adjust for missing data that are not missing at random, and established the extent of selection bias in a population-based HIV survey in an HIV hyperendemic community in rural South Africa. Methods We used data from a population-based HIV survey carried out in 2009 in rural KwaZulu-Natal, South Africa. In this survey, 5565 women (35%) and 2567 men (27%) provided blood for an HIV test. We accounted for missing data using interviewer identity as a selection variable which predicted consent to HIV testing but was unlikely to be independently associated with HIV status. Our approach involved using this selection variable to examine the HIV status of residents who would ordinarily refuse to test, except that they were allocated a persuasive interviewer. Our copula model allows for flexibility when modelling the dependence structure between HIV survey participation and HIV status. Results For women, our selection model generated an HIV prevalence estimate of 33% (95% CI 27–40) for all people eligible to consent to HIV testing in the survey. This estimate is higher than the estimate of 24% generated when only information from respondents who participated in testing is used in the analysis, and the estimate of 27% when imputation analysis is used to predict missing data on HIV status. For men, we found an HIV prevalence of 25% (95% CI 15–35) using the selection model, compared to 16% among those who participated in testing, and 18% estimated with imputation. We provide new confidence intervals that correct for the fact that the relationship between testing and HIV status is unknown and requires estimation. Conclusions We confirm the feasibility and value of adopting selection models to account for missing data in population-based HIV surveys and surveillance systems. Elements of survey design, such as interviewer identity, present the opportunity to adopt this approach in routine applications. Where non-participation is high, true confidence intervals are much wider than those generated by standard approaches to dealing with missing data suggest. PMID:26613900

Sensitivity Analysis of Multiple Informant Models When Data are Not Missing at Random

PubMed Central

Blozis, Shelley A.; Ge, Xiaojia; Xu, Shu; Natsuaki, Misaki N.; Shaw, Daniel S.; Neiderhiser, Jenae; Scaramella, Laura; Leve, Leslie; Reiss, David

2014-01-01

Missing data are common in studies that rely on multiple informant data to evaluate relationships among variables for distinguishable individuals clustered within groups. Estimation of structural equation models using raw data allows for incomplete data, and so all groups may be retained even if only one member of a group contributes data. Statistical inference is based on the assumption that data are missing completely at random or missing at random. Importantly, whether or not data are missing is assumed to be independent of the missing data. A saturated correlates model that incorporates correlates of the missingness or the missing data into an analysis and multiple imputation that may also use such correlates offer advantages over the standard implementation of SEM when data are not missing at random because these approaches may result in a data analysis problem for which the missingness is ignorable. This paper considers these approaches in an analysis of family data to assess the sensitivity of parameter estimates to assumptions about missing data, a strategy that may be easily implemented using SEM software. PMID:25221420

Accuracy of Range Restriction Correction with Multiple Imputation in Small and Moderate Samples: A Simulation Study

ERIC Educational Resources Information Center

Pfaffel, Andreas; Spiel, Christiane

2016-01-01

Approaches to correcting correlation coefficients for range restriction have been developed under the framework of large sample theory. The accuracy of missing data techniques for correcting correlation coefficients for range restriction has thus far only been investigated with relatively large samples. However, researchers and evaluators are…

Properties of different density genotypes used in dairy cattle evaluation

USDA-ARS?s Scientific Manuscript database

Dairy cattle breeders have used a 50K chip since April 2008 and a less expensive, lower density (3K) chip since September 2010 in genomic selection. Evaluations from 3K are less reliable because genotype calls are less accurate and missing markers are imputed. After excluding genotypes with < 90% ca...

A Statistical Model for Misreported Binary Outcomes in Clustered RCTs of Education Interventions

ERIC Educational Resources Information Center

Schochet, Peter Z.

2013-01-01

In randomized control trials (RCTs) of educational interventions, there is a growing literature on impact estimation methods to adjust for missing student outcome data using such methods as multiple imputation, the construction of nonresponse weights, casewise deletion, and maximum likelihood methods (see, for example, Allison, 2002; Graham, 2009;…

Comparison of Imputation Procedures for Replacing Denied-access Plots

Treesearch

Susan L. King

2005-01-01

In forest inventories, missing plots are caused by hazardous terrain, inaccessible locations, or denied access. Maryland had a large number of denied-access plots in the latest periodic inventory conducted by the Northeastern Forest Inventory and Analysis unit. The denial pattern, which can introduce error into the estimates, was investigated by dropping the 1999...

Bayesian Approaches to Imputation, Hypothesis Testing, and Parameter Estimation

ERIC Educational Resources Information Center

Ross, Steven J.; Mackey, Beth

2015-01-01

This chapter introduces three applications of Bayesian inference to common and novel issues in second language research. After a review of the critiques of conventional hypothesis testing, our focus centers on ways Bayesian inference can be used for dealing with missing data, for testing theory-driven substantive hypotheses without a default null…

The impact of democracy and media freedom on under-5 mortality, 1961-2011.

PubMed

Wigley, Simon; Akkoyunlu-Wigley, Arzu

2017-10-01

Do democracies produce better health outcomes for children than autocracies? We argue that (1) democratic governments have an incentive to reduce child mortality among low-income families and (2) that media freedom enhances their ability to deliver mortality-reducing resources to the poorest. A panel of 167 countries for the years 1961-2011 is used to test those two theoretical claims. We find that level of democracy is negatively associated with under-5 mortality, and that that negative association is greater in the presence of media freedom. These results are robust to the inclusion of country and year fixed effects, time-varying control variables, and the multiple imputation of missing values. Copyright © 2017 Elsevier Ltd. All rights reserved.

Little Evidence That Time in Child Care Causes Externalizing Problems During Early Childhood in Norway

PubMed Central

Zachrisson, Henrik Daae; Dearing, Eric; Lekhal, Ratib; Toppelberg, Claudio O.

2012-01-01

Associations between maternal reports of hours in child care and children’s externalizing problems at 18 and 36 months of age were examined in a population-based Norwegian sample (n = 75,271). Within a sociopolitical context of homogenously high-quality child care, there was little evidence that high quantity of care causes externalizing problems. Using conventional approaches to handling selection bias and listwise deletion for substantial attrition in this sample, more hours in care predicted higher problem levels, yet with small effect sizes. The finding, however, was not robust to using multiple imputation for missing values. Moreover, when sibling and individual fixed-effects models for handling selection bias were used, no relation between hours and problems was evident. PMID:23311645

Imputation of variants from the 1000 Genomes Project modestly improves known associations and can identify low-frequency variant-phenotype associations undetected by HapMap based imputation.

PubMed

Wood, Andrew R; Perry, John R B; Tanaka, Toshiko; Hernandez, Dena G; Zheng, Hou-Feng; Melzer, David; Gibbs, J Raphael; Nalls, Michael A; Weedon, Michael N; Spector, Tim D; Richards, J Brent; Bandinelli, Stefania; Ferrucci, Luigi; Singleton, Andrew B; Frayling, Timothy M

2013-01-01

Genome-wide association (GWA) studies have been limited by the reliance on common variants present on microarrays or imputable from the HapMap Project data. More recently, the completion of the 1000 Genomes Project has provided variant and haplotype information for several million variants derived from sequencing over 1,000 individuals. To help understand the extent to which more variants (including low frequency (1% ≤ MAF <5%) and rare variants (<1%)) can enhance previously identified associations and identify novel loci, we selected 93 quantitative circulating factors where data was available from the InCHIANTI population study. These phenotypes included cytokines, binding proteins, hormones, vitamins and ions. We selected these phenotypes because many have known strong genetic associations and are potentially important to help understand disease processes. We performed a genome-wide scan for these 93 phenotypes in InCHIANTI. We identified 21 signals and 33 signals that reached P<5×10(-8) based on HapMap and 1000 Genomes imputation, respectively, and 9 and 11 that reached a stricter, likely conservative, threshold of P<5×10(-11) respectively. Imputation of 1000 Genomes genotype data modestly improved the strength of known associations. Of 20 associations detected at P<5×10(-8) in both analyses (17 of which represent well replicated signals in the NHGRI catalogue), six were captured by the same index SNP, five were nominally more strongly associated in 1000 Genomes imputed data and one was nominally more strongly associated in HapMap imputed data. We also detected an association between a low frequency variant and phenotype that was previously missed by HapMap based imputation approaches. An association between rs112635299 and alpha-1 globulin near the SERPINA gene represented the known association between rs28929474 (MAF = 0.007) and alpha1-antitrypsin that predisposes to emphysema (P = 2.5×10(-12)). Our data provide important proof of principle that 1000 Genomes imputation will detect novel, low frequency-large effect associations.

Imputation of Variants from the 1000 Genomes Project Modestly Improves Known Associations and Can Identify Low-frequency Variant - Phenotype Associations Undetected by HapMap Based Imputation

PubMed Central

Wood, Andrew R.; Perry, John R. B.; Tanaka, Toshiko; Hernandez, Dena G.; Zheng, Hou-Feng; Melzer, David; Gibbs, J. Raphael; Nalls, Michael A.; Weedon, Michael N.; Spector, Tim D.; Richards, J. Brent; Bandinelli, Stefania; Ferrucci, Luigi; Singleton, Andrew B.; Frayling, Timothy M.

2013-01-01

Genome-wide association (GWA) studies have been limited by the reliance on common variants present on microarrays or imputable from the HapMap Project data. More recently, the completion of the 1000 Genomes Project has provided variant and haplotype information for several million variants derived from sequencing over 1,000 individuals. To help understand the extent to which more variants (including low frequency (1% ≤ MAF <5%) and rare variants (<1%)) can enhance previously identified associations and identify novel loci, we selected 93 quantitative circulating factors where data was available from the InCHIANTI population study. These phenotypes included cytokines, binding proteins, hormones, vitamins and ions. We selected these phenotypes because many have known strong genetic associations and are potentially important to help understand disease processes. We performed a genome-wide scan for these 93 phenotypes in InCHIANTI. We identified 21 signals and 33 signals that reached P<5×10−8 based on HapMap and 1000 Genomes imputation, respectively, and 9 and 11 that reached a stricter, likely conservative, threshold of P<5×10−11 respectively. Imputation of 1000 Genomes genotype data modestly improved the strength of known associations. Of 20 associations detected at P<5×10−8 in both analyses (17 of which represent well replicated signals in the NHGRI catalogue), six were captured by the same index SNP, five were nominally more strongly associated in 1000 Genomes imputed data and one was nominally more strongly associated in HapMap imputed data. We also detected an association between a low frequency variant and phenotype that was previously missed by HapMap based imputation approaches. An association between rs112635299 and alpha-1 globulin near the SERPINA gene represented the known association between rs28929474 (MAF = 0.007) and alpha1-antitrypsin that predisposes to emphysema (P = 2.5×10−12). Our data provide important proof of principle that 1000 Genomes imputation will detect novel, low frequency-large effect associations. PMID:23696881

Accounting for dropout bias using mixed-effects models.

PubMed

Mallinckrodt, C H; Clark, W S; David, S R

2001-01-01

Treatment effects are often evaluated by comparing change over time in outcome measures. However, valid analyses of longitudinal data can be problematic when subjects discontinue (dropout) prior to completing the study. This study assessed the merits of likelihood-based repeated measures analyses (MMRM) compared with fixed-effects analysis of variance where missing values were imputed using the last observation carried forward approach (LOCF) in accounting for dropout bias. Comparisons were made in simulated data and in data from a randomized clinical trial. Subject dropout was introduced in the simulated data to generate ignorable and nonignorable missingness. Estimates of treatment group differences in mean change from baseline to endpoint from MMRM were, on average, markedly closer to the true value than estimates from LOCF in every scenario simulated. Standard errors and confidence intervals from MMRM accurately reflected the uncertainty of the estimates, whereas standard errors and confidence intervals from LOCF underestimated uncertainty.

[Analysis of economic cost of missed outpatient appointments].

PubMed

Jabalera Mesa, M L; Morales Asencio, J M; Rivas Ruiz, F; Porras González, M H

To estimate the economic costs of missed Outpatient appointments by the Costa del Sol Health Agency (ASCS). An analysis was performed on the costs arising from missed outpatient appointments (first appointment and examinations) of each of the specialities in the Centres belonging to the ASCS. A formula was used to determine the unit cost per appointment and per centre and speciality. This involved the direct imputation of the controllable costs and the indirect imputation of the service costs, together with an estimated cost of re-appointments based on a previous case-control study. The cost of missed appointments per centre in the Costa del Sol Hospital was €2,475,640, with a failure rate of 14.2% (256,377 appointments). In the Benalmádena High Resolution Hospital it was €515,936, with an absence rate of 12.2% (44,848 appointments), and in the Mijas High Resolution Centre, a cost of €395,342 with an absence rate of the 13.5% (99,536 appointments). The mean extra cost of a re-appointment was €12.95. The specialities with a higher medium cost were Digestive Diseases, Internal Medicine, and Rehabilitation. The economic cost of patients not turning up for scheduled appointments in the ASCS was greater than 3 million Euros for a non-attendance rate of the 13.8%, with Mijas High Resolution Centre being the centre that showed the lowest mean unitary cost per medical appointment. Copyright © 2017 SECA. Publicado por Elsevier España, S.L.U. All rights reserved.

Imputation method for lifetime exposure assessment in air pollution epidemiologic studies

PubMed Central

2013-01-01

Background Environmental epidemiology, when focused on the life course of exposure to a specific pollutant, requires historical exposure estimates that are difficult to obtain for the full time period due to gaps in the historical record, especially in earlier years. We show that these gaps can be filled by applying multiple imputation methods to a formal risk equation that incorporates lifetime exposure. We also address challenges that arise, including choice of imputation method, potential bias in regression coefficients, and uncertainty in age-at-exposure sensitivities. Methods During time periods when parameters needed in the risk equation are missing for an individual, the parameters are filled by an imputation model using group level information or interpolation. A random component is added to match the variance found in the estimates for study subjects not needing imputation. The process is repeated to obtain multiple data sets, whose regressions against health data can be combined statistically to develop confidence limits using Rubin’s rules to account for the uncertainty introduced by the imputations. To test for possible recall bias between cases and controls, which can occur when historical residence location is obtained by interview, and which can lead to misclassification of imputed exposure by disease status, we introduce an “incompleteness index,” equal to the percentage of dose imputed (PDI) for a subject. “Effective doses” can be computed using different functional dependencies of relative risk on age of exposure, allowing intercomparison of different risk models. To illustrate our approach, we quantify lifetime exposure (dose) from traffic air pollution in an established case–control study on Long Island, New York, where considerable in-migration occurred over a period of many decades. Results The major result is the described approach to imputation. The illustrative example revealed potential recall bias, suggesting that regressions against health data should be done as a function of PDI to check for consistency of results. The 1% of study subjects who lived for long durations near heavily trafficked intersections, had very high cumulative exposures. Thus, imputation methods must be designed to reproduce non-standard distributions. Conclusions Our approach meets a number of methodological challenges to extending historical exposure reconstruction over a lifetime and shows promise for environmental epidemiology. Application to assessment of breast cancer risks will be reported in a subsequent manuscript. PMID:23919666

Use of partial least squares regression to impute SNP genotypes in Italian cattle breeds.

PubMed

Dimauro, Corrado; Cellesi, Massimo; Gaspa, Giustino; Ajmone-Marsan, Paolo; Steri, Roberto; Marras, Gabriele; Macciotta, Nicolò P P

2013-06-05

The objective of the present study was to test the ability of the partial least squares regression technique to impute genotypes from low density single nucleotide polymorphisms (SNP) panels i.e. 3K or 7K to a high density panel with 50K SNP. No pedigree information was used. Data consisted of 2093 Holstein, 749 Brown Swiss and 479 Simmental bulls genotyped with the Illumina 50K Beadchip. First, a single-breed approach was applied by using only data from Holstein animals. Then, to enlarge the training population, data from the three breeds were combined and a multi-breed analysis was performed. Accuracies of genotypes imputed using the partial least squares regression method were compared with those obtained by using the Beagle software. The impact of genotype imputation on breeding value prediction was evaluated for milk yield, fat content and protein content. In the single-breed approach, the accuracy of imputation using partial least squares regression was around 90 and 94% for the 3K and 7K platforms, respectively; corresponding accuracies obtained with Beagle were around 85% and 90%. Moreover, computing time required by the partial least squares regression method was on average around 10 times lower than computing time required by Beagle. Using the partial least squares regression method in the multi-breed resulted in lower imputation accuracies than using single-breed data. The impact of the SNP-genotype imputation on the accuracy of direct genomic breeding values was small. The correlation between estimates of genetic merit obtained by using imputed versus actual genotypes was around 0.96 for the 7K chip. Results of the present work suggested that the partial least squares regression imputation method could be useful to impute SNP genotypes when pedigree information is not available.

Adult myeloid leukaemia and radon exposure: a Bayesian model for a case-control study with error in covariates.

PubMed

Toti, Simona; Biggeri, Annibale; Forastiere, Francesco

2005-06-30

The possible association between radon exposure in dwellings and adult myeloid leukaemia had been explored in an Italian province by a case-control study. A total of 44 cases and 211 controls were selected from death certificates file. No association had been found in the original study (OR = 0.58 for > 185 vs 80 < or = Bq/cm). Here we reanalyse the data taking into account the measurement error of radon concentration and the presence of missing data. A Bayesian hierarchical model with error in covariates is proposed which allows appropriate imputation of missing values. The general conclusion of no evidence of association with radon does not change, but a negative association is not observed anymore (OR = 0.99 for > 185 vs 80 < or = Bq/cm). After adjusting for residential house radon and gamma radiation, and for the multilevel data structure, geological features of the soil is associated with adult myeloid leukaemia risk (OR = 2.14, 95 per cent Cr.I. 1.0-5.5). Copyright 2005 John Wiley & Sons, Ltd.

Multi-test cervical cancer diagnosis with missing data estimation

NASA Astrophysics Data System (ADS)

Xu, Tao; Huang, Xiaolei; Kim, Edward; Long, L. Rodney; Antani, Sameer

2015-03-01

Cervical cancer is a leading most common type of cancer for women worldwide. Existing screening programs for cervical cancer suffer from low sensitivity. Using images of the cervix (cervigrams) as an aid in detecting pre-cancerous changes to the cervix has good potential to improve sensitivity and help reduce the number of cervical cancer cases. In this paper, we present a method that utilizes multi-modality information extracted from multiple tests of a patient's visit to classify the patient visit to be either low-risk or high-risk. Our algorithm integrates image features and text features to make a diagnosis. We also present two strategies to estimate the missing values in text features: Image Classifier Supervised Mean Imputation (ICSMI) and Image Classifier Supervised Linear Interpolation (ICSLI). We evaluate our method on a large medical dataset and compare it with several alternative approaches. The results show that the proposed method with ICSLI strategy achieves the best result of 83.03% specificity and 76.36% sensitivity. When higher specificity is desired, our method can achieve 90% specificity with 62.12% sensitivity.

A Temporal Mining Framework for Classifying Un-Evenly Spaced Clinical Data: An Approach for Building Effective Clinical Decision-Making System.

PubMed

Jane, Nancy Yesudhas; Nehemiah, Khanna Harichandran; Arputharaj, Kannan

2016-01-01

Clinical time-series data acquired from electronic health records (EHR) are liable to temporal complexities such as irregular observations, missing values and time constrained attributes that make the knowledge discovery process challenging. This paper presents a temporal rough set induced neuro-fuzzy (TRiNF) mining framework that handles these complexities and builds an effective clinical decision-making system. TRiNF provides two functionalities namely temporal data acquisition (TDA) and temporal classification. In TDA, a time-series forecasting model is constructed by adopting an improved double exponential smoothing method. The forecasting model is used in missing value imputation and temporal pattern extraction. The relevant attributes are selected using a temporal pattern based rough set approach. In temporal classification, a classification model is built with the selected attributes using a temporal pattern induced neuro-fuzzy classifier. For experimentation, this work uses two clinical time series dataset of hepatitis and thrombosis patients. The experimental result shows that with the proposed TRiNF framework, there is a significant reduction in the error rate, thereby obtaining the classification accuracy on an average of 92.59% for hepatitis and 91.69% for thrombosis dataset. The obtained classification results prove the efficiency of the proposed framework in terms of its improved classification accuracy.

Multivariate two-part statistics for analysis of correlated mass spectrometry data from multiple biological specimens.

PubMed

Taylor, Sandra L; Ruhaak, L Renee; Weiss, Robert H; Kelly, Karen; Kim, Kyoungmi

2017-01-01

High through-put mass spectrometry (MS) is now being used to profile small molecular compounds across multiple biological sample types from the same subjects with the goal of leveraging information across biospecimens. Multivariate statistical methods that combine information from all biospecimens could be more powerful than the usual univariate analyses. However, missing values are common in MS data and imputation can impact between-biospecimen correlation and multivariate analysis results. We propose two multivariate two-part statistics that accommodate missing values and combine data from all biospecimens to identify differentially regulated compounds. Statistical significance is determined using a multivariate permutation null distribution. Relative to univariate tests, the multivariate procedures detected more significant compounds in three biological datasets. In a simulation study, we showed that multi-biospecimen testing procedures were more powerful than single-biospecimen methods when compounds are differentially regulated in multiple biospecimens but univariate methods can be more powerful if compounds are differentially regulated in only one biospecimen. We provide R functions to implement and illustrate our method as supplementary information CONTACT: sltaylor@ucdavis.eduSupplementary information: Supplementary data are available at Bioinformatics online. © The Author 2016. Published by Oxford University Press. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com.

A phasing and imputation method for pedigreed populations that results in a single-stage genomic evaluation

PubMed Central

2012-01-01

Background Efficient, robust, and accurate genotype imputation algorithms make large-scale application of genomic selection cost effective. An algorithm that imputes alleles or allele probabilities for all animals in the pedigree and for all genotyped single nucleotide polymorphisms (SNP) provides a framework to combine all pedigree, genomic, and phenotypic information into a single-stage genomic evaluation. Methods An algorithm was developed for imputation of genotypes in pedigreed populations that allows imputation for completely ungenotyped animals and for low-density genotyped animals, accommodates a wide variety of pedigree structures for genotyped animals, imputes unmapped SNP, and works for large datasets. The method involves simple phasing rules, long-range phasing and haplotype library imputation and segregation analysis. Results Imputation accuracy was high and computational cost was feasible for datasets with pedigrees of up to 25 000 animals. The resulting single-stage genomic evaluation increased the accuracy of estimated genomic breeding values compared to a scenario in which phenotypes on relatives that were not genotyped were ignored. Conclusions The developed imputation algorithm and software and the resulting single-stage genomic evaluation method provide powerful new ways to exploit imputation and to obtain more accurate genetic evaluations. PMID:22462519

Use of missing data methods in longitudinal studies: the persistence of bad practices in developmental psychology.

PubMed

Jelicić, Helena; Phelps, Erin; Lerner, Richard M

2009-07-01

Developmental science rests on describing, explaining, and optimizing intraindividual changes and, hence, empirically requires longitudinal research. Problems of missing data arise in most longitudinal studies, thus creating challenges for interpreting the substance and structure of intraindividual change. Using a sample of reports of longitudinal studies obtained from three flagship developmental journals-Child Development, Developmental Psychology, and Journal of Research on Adolescence-we examined the number of longitudinal studies reporting missing data and the missing data techniques used. Of the 100 longitudinal studies sampled, 57 either reported having missing data or had discrepancies in sample sizes reported for different analyses. The majority of these studies (82%) used missing data techniques that are statistically problematic (either listwise deletion or pairwise deletion) and not among the methods recommended by statisticians (i.e., the direct maximum likelihood method and the multiple imputation method). Implications of these results for developmental theory and application, and the need for understanding the consequences of using statistically inappropriate missing data techniques with actual longitudinal data sets, are discussed.

Melancholic depression prediction by identifying representative features in metabolic and microarray profiles with missing values.

PubMed

Nie, Zhi; Yang, Tao; Liu, Yashu; Li, Qingyang; Narayan, Vaibhav A; Wittenberg, Gayle; Ye, Jieping

2015-01-01

Recent studies have revealed that melancholic depression, one major subtype of depression, is closely associated with the concentration of some metabolites and biological functions of certain genes and pathways. Meanwhile, recent advances in biotechnologies have allowed us to collect a large amount of genomic data, e.g., metabolites and microarray gene expression. With such a huge amount of information available, one approach that can give us new insights into the understanding of the fundamental biology underlying melancholic depression is to build disease status prediction models using classification or regression methods. However, the existence of strong empirical correlations, e.g., those exhibited by genes sharing the same biological pathway in microarray profiles, tremendously limits the performance of these methods. Furthermore, the occurrence of missing values which are ubiquitous in biomedical applications further complicates the problem. In this paper, we hypothesize that the problem of missing values might in some way benefit from the correlation between the variables and propose a method to learn a compressed set of representative features through an adapted version of sparse coding which is capable of identifying correlated variables and addressing the issue of missing values simultaneously. An efficient algorithm is also developed to solve the proposed formulation. We apply the proposed method on metabolic and microarray profiles collected from a group of subjects consisting of both patients with melancholic depression and healthy controls. Results show that the proposed method can not only produce meaningful clusters of variables but also generate a set of representative features that achieve superior classification performance over those generated by traditional clustering and data imputation techniques. In particular, on both datasets, we found that in comparison with the competing algorithms, the representative features learned by the proposed method give rise to significantly improved sensitivity scores, suggesting that the learned features allow prediction with high accuracy of disease status in those who are diagnosed with melancholic depression. To our best knowledge, this is the first work that applies sparse coding to deal with high feature correlations and missing values, which are common challenges in many biomedical applications. The proposed method can be readily adapted to other biomedical applications involving incomplete and high-dimensional data.

High dimensional land cover inference using remotely sensed modis data

NASA Astrophysics Data System (ADS)

Glanz, Hunter S.

Image segmentation persists as a major statistical problem, with the volume and complexity of data expanding alongside new technologies. Land cover classification, one of the most studied problems in Remote Sensing, provides an important example of image segmentation whose needs transcend the choice of a particular classification method. That is, the challenges associated with land cover classification pervade the analysis process from data pre-processing to estimation of a final land cover map. Many of the same challenges also plague the task of land cover change detection. Multispectral, multitemporal data with inherent spatial relationships have hardly received adequate treatment due to the large size of the data and the presence of missing values. In this work we propose a novel, concerted application of methods which provide a unified way to estimate model parameters, impute missing data, reduce dimensionality, classify land cover, and detect land cover changes. This comprehensive analysis adopts a Bayesian approach which incorporates prior knowledge to improve the interpretability, efficiency, and versatility of land cover classification and change detection. We explore a parsimonious, parametric model that allows for a natural application of principal components analysis to isolate important spectral characteristics while preserving temporal information. Moreover, it allows us to impute missing data and estimate parameters via expectation-maximization (EM). A significant byproduct of our framework includes a suite of training data assessment tools. To classify land cover, we employ a spanning tree approximation to a lattice Potts prior to incorporate spatial relationships in a judicious way and more efficiently access the posterior distribution of pixel labels. We then achieve exact inference of the labels via the centroid estimator. To detect land cover changes, we develop a new EM algorithm based on the same parametric model. We perform simulation studies to validate our models and methods, and conduct an extensive continental scale case study using MODIS data. The results show that we successfully classify land cover and recover the spatial patterns present in large scale data. Application of our change point method to an area in the Amazon successfully identifies the progression of deforestation through portions of the region.

Multiple imputation of rainfall missing data in the Iberian Mediterranean context

NASA Astrophysics Data System (ADS)

Miró, Juan Javier; Caselles, Vicente; Estrela, María José

2017-11-01

Given the increasing need for complete rainfall data networks, in recent years have been proposed diverse methods for filling gaps in observed precipitation series, progressively more advanced that traditional approaches to overcome the problem. The present study has consisted in validate 10 methods (6 linear, 2 non-linear and 2 hybrid) that allow multiple imputation, i.e., fill at the same time missing data of multiple incomplete series in a dense network of neighboring stations. These were applied for daily and monthly rainfall in two sectors in the Júcar River Basin Authority (east Iberian Peninsula), which is characterized by a high spatial irregularity and difficulty of rainfall estimation. A classification of precipitation according to their genetic origin was applied as pre-processing, and a quantile-mapping adjusting as post-processing technique. The results showed in general a better performance for the non-linear and hybrid methods, highlighting that the non-linear PCA (NLPCA) method outperforms considerably the Self Organizing Maps (SOM) method within non-linear approaches. On linear methods, the Regularized Expectation Maximization method (RegEM) was the best, but far from NLPCA. Applying EOF filtering as post-processing of NLPCA (hybrid approach) yielded the best results.

Statistical methods for change-point detection in surface temperature records

NASA Astrophysics Data System (ADS)

Pintar, A. L.; Possolo, A.; Zhang, N. F.

2013-09-01

We describe several statistical methods to detect possible change-points in a time series of values of surface temperature measured at a meteorological station, and to assess the statistical significance of such changes, taking into account the natural variability of the measured values, and the autocorrelations between them. These methods serve to determine whether the record may suffer from biases unrelated to the climate signal, hence whether there may be a need for adjustments as considered by M. J. Menne and C. N. Williams (2009) "Homogenization of Temperature Series via Pairwise Comparisons", Journal of Climate 22 (7), 1700-1717. We also review methods to characterize patterns of seasonality (seasonal decomposition using monthly medians or robust local regression), and explain the role they play in the imputation of missing values, and in enabling robust decompositions of the measured values into a seasonal component, a possible climate signal, and a station-specific remainder. The methods for change-point detection that we describe include statistical process control, wavelet multi-resolution analysis, adaptive weights smoothing, and a Bayesian procedure, all of which are applicable to single station records.

Comparison of HapMap and 1000 Genomes Reference Panels in a Large-Scale Genome-Wide Association Study.

PubMed

de Vries, Paul S; Sabater-Lleal, Maria; Chasman, Daniel I; Trompet, Stella; Ahluwalia, Tarunveer S; Teumer, Alexander; Kleber, Marcus E; Chen, Ming-Huei; Wang, Jie Jin; Attia, John R; Marioni, Riccardo E; Steri, Maristella; Weng, Lu-Chen; Pool, Rene; Grossmann, Vera; Brody, Jennifer A; Venturini, Cristina; Tanaka, Toshiko; Rose, Lynda M; Oldmeadow, Christopher; Mazur, Johanna; Basu, Saonli; Frånberg, Mattias; Yang, Qiong; Ligthart, Symen; Hottenga, Jouke J; Rumley, Ann; Mulas, Antonella; de Craen, Anton J M; Grotevendt, Anne; Taylor, Kent D; Delgado, Graciela E; Kifley, Annette; Lopez, Lorna M; Berentzen, Tina L; Mangino, Massimo; Bandinelli, Stefania; Morrison, Alanna C; Hamsten, Anders; Tofler, Geoffrey; de Maat, Moniek P M; Draisma, Harmen H M; Lowe, Gordon D; Zoledziewska, Magdalena; Sattar, Naveed; Lackner, Karl J; Völker, Uwe; McKnight, Barbara; Huang, Jie; Holliday, Elizabeth G; McEvoy, Mark A; Starr, John M; Hysi, Pirro G; Hernandez, Dena G; Guan, Weihua; Rivadeneira, Fernando; McArdle, Wendy L; Slagboom, P Eline; Zeller, Tanja; Psaty, Bruce M; Uitterlinden, André G; de Geus, Eco J C; Stott, David J; Binder, Harald; Hofman, Albert; Franco, Oscar H; Rotter, Jerome I; Ferrucci, Luigi; Spector, Tim D; Deary, Ian J; März, Winfried; Greinacher, Andreas; Wild, Philipp S; Cucca, Francesco; Boomsma, Dorret I; Watkins, Hugh; Tang, Weihong; Ridker, Paul M; Jukema, Jan W; Scott, Rodney J; Mitchell, Paul; Hansen, Torben; O'Donnell, Christopher J; Smith, Nicholas L; Strachan, David P; Dehghan, Abbas

2017-01-01

An increasing number of genome-wide association (GWA) studies are now using the higher resolution 1000 Genomes Project reference panel (1000G) for imputation, with the expectation that 1000G imputation will lead to the discovery of additional associated loci when compared to HapMap imputation. In order to assess the improvement of 1000G over HapMap imputation in identifying associated loci, we compared the results of GWA studies of circulating fibrinogen based on the two reference panels. Using both HapMap and 1000G imputation we performed a meta-analysis of 22 studies comprising the same 91,953 individuals. We identified six additional signals using 1000G imputation, while 29 loci were associated using both HapMap and 1000G imputation. One locus identified using HapMap imputation was not significant using 1000G imputation. The genome-wide significance threshold of 5×10-8 is based on the number of independent statistical tests using HapMap imputation, and 1000G imputation may lead to further independent tests that should be corrected for. When using a stricter Bonferroni correction for the 1000G GWA study (P-value < 2.5×10-8), the number of loci significant only using HapMap imputation increased to 4 while the number of loci significant only using 1000G decreased to 5. In conclusion, 1000G imputation enabled the identification of 20% more loci than HapMap imputation, although the advantage of 1000G imputation became less clear when a stricter Bonferroni correction was used. More generally, our results provide insights that are applicable to the implementation of other dense reference panels that are under development.

Comparison of HapMap and 1000 Genomes Reference Panels in a Large-Scale Genome-Wide Association Study

PubMed Central

de Vries, Paul S.; Sabater-Lleal, Maria; Chasman, Daniel I.; Trompet, Stella; Kleber, Marcus E.; Chen, Ming-Huei; Wang, Jie Jin; Attia, John R.; Marioni, Riccardo E.; Weng, Lu-Chen; Grossmann, Vera; Brody, Jennifer A.; Venturini, Cristina; Tanaka, Toshiko; Rose, Lynda M.; Oldmeadow, Christopher; Mazur, Johanna; Basu, Saonli; Yang, Qiong; Ligthart, Symen; Hottenga, Jouke J.; Rumley, Ann; Mulas, Antonella; de Craen, Anton J. M.; Grotevendt, Anne; Taylor, Kent D.; Delgado, Graciela E.; Kifley, Annette; Lopez, Lorna M.; Berentzen, Tina L.; Mangino, Massimo; Bandinelli, Stefania; Morrison, Alanna C.; Hamsten, Anders; Tofler, Geoffrey; de Maat, Moniek P. M.; Draisma, Harmen H. M.; Lowe, Gordon D.; Zoledziewska, Magdalena; Sattar, Naveed; Lackner, Karl J.; Völker, Uwe; McKnight, Barbara; Huang, Jie; Holliday, Elizabeth G.; McEvoy, Mark A.; Starr, John M.; Hysi, Pirro G.; Hernandez, Dena G.; Guan, Weihua; Rivadeneira, Fernando; McArdle, Wendy L.; Slagboom, P. Eline; Zeller, Tanja; Psaty, Bruce M.; Uitterlinden, André G.; de Geus, Eco J. C.; Stott, David J.; Binder, Harald; Hofman, Albert; Franco, Oscar H.; Rotter, Jerome I.; Ferrucci, Luigi; Spector, Tim D.; Deary, Ian J.; März, Winfried; Greinacher, Andreas; Wild, Philipp S.; Cucca, Francesco; Boomsma, Dorret I.; Watkins, Hugh; Tang, Weihong; Ridker, Paul M.; Jukema, Jan W.; Scott, Rodney J.; Mitchell, Paul; Hansen, Torben; O'Donnell, Christopher J.; Smith, Nicholas L.; Strachan, David P.

2017-01-01

An increasing number of genome-wide association (GWA) studies are now using the higher resolution 1000 Genomes Project reference panel (1000G) for imputation, with the expectation that 1000G imputation will lead to the discovery of additional associated loci when compared to HapMap imputation. In order to assess the improvement of 1000G over HapMap imputation in identifying associated loci, we compared the results of GWA studies of circulating fibrinogen based on the two reference panels. Using both HapMap and 1000G imputation we performed a meta-analysis of 22 studies comprising the same 91,953 individuals. We identified six additional signals using 1000G imputation, while 29 loci were associated using both HapMap and 1000G imputation. One locus identified using HapMap imputation was not significant using 1000G imputation. The genome-wide significance threshold of 5×10−8 is based on the number of independent statistical tests using HapMap imputation, and 1000G imputation may lead to further independent tests that should be corrected for. When using a stricter Bonferroni correction for the 1000G GWA study (P-value < 2.5×10−8), the number of loci significant only using HapMap imputation increased to 4 while the number of loci significant only using 1000G decreased to 5. In conclusion, 1000G imputation enabled the identification of 20% more loci than HapMap imputation, although the advantage of 1000G imputation became less clear when a stricter Bonferroni correction was used. More generally, our results provide insights that are applicable to the implementation of other dense reference panels that are under development. PMID:28107422

The impact of loss to follow-up on hypothesis tests of the treatment effect for several statistical methods in substance abuse clinical trials.

PubMed

Hedden, Sarra L; Woolson, Robert F; Carter, Rickey E; Palesch, Yuko; Upadhyaya, Himanshu P; Malcolm, Robert J

2009-07-01

"Loss to follow-up" can be substantial in substance abuse clinical trials. When extensive losses to follow-up occur, one must cautiously analyze and interpret the findings of a research study. Aims of this project were to introduce the types of missing data mechanisms and describe several methods for analyzing data with loss to follow-up. Furthermore, a simulation study compared Type I error and power of several methods when missing data amount and mechanism varies. Methods compared were the following: Last observation carried forward (LOCF), multiple imputation (MI), modified stratified summary statistics (SSS), and mixed effects models. Results demonstrated nominal Type I error for all methods; power was high for all methods except LOCF. Mixed effect model, modified SSS, and MI are generally recommended for use; however, many methods require that the data are missing at random or missing completely at random (i.e., "ignorable"). If the missing data are presumed to be nonignorable, a sensitivity analysis is recommended.

Tackling missing radiographic progression data: multiple imputation technique compared with inverse probability weights and complete case analysis.

PubMed

Descalzo, Miguel Á; Garcia, Virginia Villaverde; González-Alvaro, Isidoro; Carbonell, Jordi; Balsa, Alejandro; Sanmartí, Raimon; Lisbona, Pilar; Hernandez-Barrera, Valentín; Jiménez-Garcia, Rodrigo; Carmona, Loreto

2013-02-01

To describe the results of different statistical ways of addressing radiographic outcome affected by missing data--multiple imputation technique, inverse probability weights and complete case analysis--using data from an observational study. A random sample of 96 RA patients was selected for a follow-up study in which radiographs of hands and feet were scored. Radiographic progression was tested by comparing the change in the total Sharp-van der Heijde radiographic score (TSS) and the joint erosion score (JES) from baseline to the end of the second year of follow-up. MI technique, inverse probability weights in weighted estimating equation (WEE) and CC analysis were used to fit a negative binomial regression. Major predictors of radiographic progression were JES and joint space narrowing (JSN) at baseline, together with baseline disease activity measured by DAS28 for TSS and MTX use for JES. Results from CC analysis show larger coefficients and s.e.s compared with MI and weighted techniques. The results from the WEE model were quite in line with those of MI. If it seems plausible that CC or MI analysis may be valid, then MI should be preferred because of its greater efficiency. CC analysis resulted in inefficient estimates or, translated into non-statistical terminology, could guide us into inaccurate results and unwise conclusions. The methods discussed here will contribute to the use of alternative approaches for tackling missing data in observational studies.

Estimates of adherence and error analysis of physical activity data collected via accelerometry in a large study of free-living adults.

PubMed

Paul, David R; Kramer, Matthew; Stote, Kim S; Spears, Karen E; Moshfegh, Alanna J; Baer, David J; Rumpler, William V

2008-06-09

Activity monitors (AM) are small, electronic devices used to quantify the amount and intensity of physical activity (PA). Unfortunately, it has been demonstrated that data loss that occurs when AMs are not worn by subjects (removals during sleeping and waking hours) tend to result in biased estimates of PA and total energy expenditure (TEE). No study has reported the degree of data loss in a large study of adults, and/or the degree to which the estimates of PA and TEE are affected. Also, no study in adults has proposed a methodology to minimize the effects of AM removals. Adherence estimates were generated from a pool of 524 women and men that wore AMs for 13 - 15 consecutive days. To simulate the effect of data loss due to AM removal, a reference dataset was first compiled from a subset consisting of 35 highly adherent subjects (24 HR; minimum of 20 hrs/day for seven consecutive days). AM removals were then simulated during sleep and between one and ten waking hours using this 24 HR dataset. Differences in the mean values for PA and TEE between the 24 HR reference dataset and the different simulations were compared using paired t-tests and/or coefficients of variation. The estimated average adherence of the pool of 524 subjects was 15.8 +/- 3.4 hrs/day for approximately 11.7 +/- 2.0 days. Simulated data loss due to AM removals during sleeping hours in the 24 HR database (n = 35), resulted in biased estimates of PA (p < 0.05), but not TEE. Losing as little as one hour of data from the 24 HR dataset during waking hours results in significant biases (p < 0.0001) and variability (coefficients of variation between 7 and 21%) in the estimates of PA. Inserting a constant value for sleep and imputing estimates for missing data during waking hours significantly improved the estimates of PA. Although estimated adherence was good, measurements of PA can be improved by relatively simple imputation of missing AM data.

Linear regression analysis of survival data with missing censoring indicators.

PubMed

Wang, Qihua; Dinse, Gregg E

2011-04-01

Linear regression analysis has been studied extensively in a random censorship setting, but typically all of the censoring indicators are assumed to be observed. In this paper, we develop synthetic data methods for estimating regression parameters in a linear model when some censoring indicators are missing. We define estimators based on regression calibration, imputation, and inverse probability weighting techniques, and we prove all three estimators are asymptotically normal. The finite-sample performance of each estimator is evaluated via simulation. We illustrate our methods by assessing the effects of sex and age on the time to non-ambulatory progression for patients in a brain cancer clinical trial.

Dealing with missing standard deviation and mean values in meta-analysis of continuous outcomes: a systematic review.

PubMed

Weir, Christopher J; Butcher, Isabella; Assi, Valentina; Lewis, Stephanie C; Murray, Gordon D; Langhorne, Peter; Brady, Marian C

2018-03-07

Rigorous, informative meta-analyses rely on availability of appropriate summary statistics or individual participant data. For continuous outcomes, especially those with naturally skewed distributions, summary information on the mean or variability often goes unreported. While full reporting of original trial data is the ideal, we sought to identify methods for handling unreported mean or variability summary statistics in meta-analysis. We undertook two systematic literature reviews to identify methodological approaches used to deal with missing mean or variability summary statistics. Five electronic databases were searched, in addition to the Cochrane Colloquium abstract books and the Cochrane Statistics Methods Group mailing list archive. We also conducted cited reference searching and emailed topic experts to identify recent methodological developments. Details recorded included the description of the method, the information required to implement the method, any underlying assumptions and whether the method could be readily applied in standard statistical software. We provided a summary description of the methods identified, illustrating selected methods in example meta-analysis scenarios. For missing standard deviations (SDs), following screening of 503 articles, fifteen methods were identified in addition to those reported in a previous review. These included Bayesian hierarchical modelling at the meta-analysis level; summary statistic level imputation based on observed SD values from other trials in the meta-analysis; a practical approximation based on the range; and algebraic estimation of the SD based on other summary statistics. Following screening of 1124 articles for methods estimating the mean, one approximate Bayesian computation approach and three papers based on alternative summary statistics were identified. Illustrative meta-analyses showed that when replacing a missing SD the approximation using the range minimised loss of precision and generally performed better than omitting trials. When estimating missing means, a formula using the median, lower quartile and upper quartile performed best in preserving the precision of the meta-analysis findings, although in some scenarios, omitting trials gave superior results. Methods based on summary statistics (minimum, maximum, lower quartile, upper quartile, median) reported in the literature facilitate more comprehensive inclusion of randomised controlled trials with missing mean or variability summary statistics within meta-analyses.

Practical implementation of cost-effective genomic selection in commercial pig breeding using imputation.

PubMed

Cleveland, M A; Hickey, J M

2013-08-01

Genomic selection can be implemented in pig breeding at a reduced cost using genotype imputation. Accuracy of imputation and the impact on resulting genomic breeding values (gEBV) was investigated. High-density genotype data was available for 4,763 animals from a single pig line. Three low-density genotype panels were constructed with SNP densities of 450 (L450), 3,071 (L3k) and 5,963 (L6k). Accuracy of imputation was determined using 184 test individuals with no genotyped descendants in the data but with parents and grandparents genotyped using the Illumina PorcineSNP60 Beadchip. Alternative genotyping scenarios were created in which parents, grandparents, and individuals that were not direct ancestors of test animals (Other) were genotyped at high density (S1), grandparents were not genotyped (S2), dams and granddams were not genotyped (S3), and dams and granddams were genotyped at low density (S4). Four additional scenarios were created by excluding Other animal genotypes. Test individuals were always genotyped at low density. Imputation was performed with AlphaImpute. Genomic breeding values were calculated using the single-step genomic evaluation. Test animals were evaluated for the information retained in the gEBV, calculated as the correlation between gEBV using imputed genotypes and gEBV using true genotypes. Accuracy of imputation was high for all scenarios but decreased with fewer SNP on the low-density panel (0.995 to 0.965 for S1) and with reduced genotyping of ancestors, where the largest changes were for L450 (0.965 in S1 to 0.914 in S3). Exclusion of genotypes for Other animals resulted in only small accuracy decreases. Imputation accuracy was not consistent across the genome. Information retained in the gEBV was related to genotyping scenario and thus to imputation accuracy. Reducing the number of SNP on the low-density panel reduced the information retained in the gEBV, with the largest decrease observed from L3k to L450. Excluding Other animal genotypes had little impact on imputation accuracy but caused large decreases in the information retained in the gEBV. These results indicate that accuracy of gEBV from imputed genotypes depends on the level of genotyping in close relatives and the size of the genotyped dataset. Fewer high-density genotyped individuals are needed to obtain accurate imputation than are needed to obtain accurate gEBV. Strategies to optimize development of low-density panels can improve both imputation and gEBV accuracy.

Estimation of habitual iodine intake in Japanese adults using 16 d diet records over four seasons with a newly developed food composition database for iodine.

PubMed

Katagiri, Ryoko; Asakura, Keiko; Sasaki, Satoshi; Hirota, Naoko; Notsu, Akiko; Miura, Ayako; Todoriki, Hidemi; Fukui, Mitsuru; Date, Chigusa

2015-08-28

Although habitual seaweed consumption in Japan would suggest that iodine intake in Japanese is exceptionally high, intake data from diet records are limited. In the present study, we developed a composition database of iodine and estimated the habitual intake of iodine among Japanese adults. Missing values for iodine content in the existing composition table were imputed based on established criteria. 16 d diet records (4 d over four seasons) from adults (120 women aged 30-69 years and 120 men aged 30-76 years) living in Japan were collected, and iodine intake was estimated. Habitual intake was estimated with the Best-power method. Totally, 995 food items were imputed. The distribution of iodine intake in 24 h was highly skewed, and approximately 55 % of 24 h values were < 300 μg/d. The median iodine intake in 24 h was 229 μg/d for women and 273 μg/d for men. All subjects consumed iodine-rich foods (kelp or soup stock) on one or more days of the sixteen survey days. The mean (median) habitual iodine intake was 1414 (857) μg/d for women and 1572 (1031) μg/d for men. Older participants had higher intake than younger participants. The major contributors to iodine intake were kelp (60 %) and soup stock (30 %). Habitual iodine intake among Japanese was sufficient or higher than the tolerable upper intake level, particularly in older generations. The association between high iodine intake as that observed in the present study and thyroid disease requires further study.

GACT: a Genome build and Allele definition Conversion Tool for SNP imputation and meta-analysis in genetic association studies.

PubMed

Sulovari, Arvis; Li, Dawei

2014-07-19

Genome-wide association studies (GWAS) have successfully identified genes associated with complex human diseases. Although much of the heritability remains unexplained, combining single nucleotide polymorphism (SNP) genotypes from multiple studies for meta-analysis will increase the statistical power to identify new disease-associated variants. Meta-analysis requires same allele definition (nomenclature) and genome build among individual studies. Similarly, imputation, commonly-used prior to meta-analysis, requires the same consistency. However, the genotypes from various GWAS are generated using different genotyping platforms, arrays or SNP-calling approaches, resulting in use of different genome builds and allele definitions. Incorrect assumptions of identical allele definition among combined GWAS lead to a large portion of discarded genotypes or incorrect association findings. There is no published tool that predicts and converts among all major allele definitions. In this study, we have developed a tool, GACT, which stands for Genome build and Allele definition Conversion Tool, that predicts and inter-converts between any of the common SNP allele definitions and between the major genome builds. In addition, we assessed several factors that may affect imputation quality, and our results indicated that inclusion of singletons in the reference had detrimental effects while ambiguous SNPs had no measurable effect. Unexpectedly, exclusion of genotypes with missing rate > 0.001 (40% of study SNPs) showed no significant decrease of imputation quality (even significantly higher when compared to the imputation with singletons in the reference), especially for rare SNPs. GACT is a new, powerful, and user-friendly tool with both command-line and interactive online versions that can accurately predict, and convert between any of the common allele definitions and between genome builds for genome-wide meta-analysis and imputation of genotypes from SNP-arrays or deep-sequencing, particularly for data from the dbGaP and other public databases. http://www.uvm.edu/genomics/software/gact.

Risk-Stratified Imputation in Survival Analysis

PubMed Central

Kennedy, Richard E.; Adragni, Kofi P.; Tiwari, Hemant K.; Voeks, Jenifer H.; Brott, Thomas G.; Howard, George

2013-01-01

Background Censoring that is dependent on covariates associated with survival can arise in randomized trials due to changes in recruitment and eligibility criteria to minimize withdrawals, potentially leading to biased treatment effect estimates. Imputation approaches have been proposed to address censoring in survival analysis; and while these approaches may provide unbiased estimates of treatment effects, imputation of a large number of outcomes may over- or underestimate the associated variance based on the imputation pool selected. Purpose We propose an improved method, risk-stratified imputation, as an alternative to address withdrawal related to the risk of events in the context of time-to-event analyses. Methods Our algorithm performs imputation from a pool of replacement subjects with similar values of both treatment and covariate(s) of interest, that is, from a risk-stratified sample. This stratification prior to imputation addresses the requirement of time-to-event analysis that censored observations are representative of all other observations in the risk group with similar exposure variables. We compared our risk-stratified imputation to case deletion and bootstrap imputation in a simulated dataset in which the covariate of interest (study withdrawal) was related to treatment. A motivating example from a recent clinical trial is also presented to demonstrate the utility of our method. Results In our simulations, risk-stratified imputation gives estimates of treatment effect comparable to bootstrap and auxiliary variable imputation while avoiding inaccuracies of the latter two in estimating the associated variance. Similar results were obtained in analysis of clinical trial data. Limitations Risk-stratified imputation has little advantage over other imputation methods when covariates of interest are not related to treatment, although its performance is superior when covariates are related to treatment. Risk-stratified imputation is intended for categorical covariates, and may be sensitive to the width of the matching window if continuous covariates are used. Conclusions The use of the risk-stratified imputation should facilitate the analysis of many clinical trials, in which one group has a higher withdrawal rate that is related to treatment. PMID:23818434

Prevalence of Autism Spectrum Disorder in Nurseries in Lebanon: A Cross Sectional Study.

PubMed

Chaaya, Monique; Saab, Dahlia; Maalouf, Fadi T; Boustany, Rose-Mary

2016-02-01

In Lebanon, no estimate for autism prevalence exists. This cross-sectional study examines the prevalence of Autism spectrum disorder (ASD) in toddlers in nurseries in Beirut and Mount-Lebanon. The final sample included 998 toddlers (16-48 months) from 177 nurseries. We sent parents the Modified Checklist for Autism in Toddlers (M-CHAT) for screening, and a self-administered questionnaire (associated factors). We imputed missing M-CHAT data with successful answers. Since there were no follow-up interviews for the M-CHAT, we used the positive predictive value (0.058) from a large study for prevalence estimates. ASD prevalence was 1 in 66 children (comparable to US). Ratios were: male/female: 1.05; Beirut/Mount-Lebanon: 1.2. Using a more representative sample and ascertaining results are needed for better prevalence estimates in Lebanon.

Linear Regression with a Randomly Censored Covariate: Application to an Alzheimer's Study.

PubMed

Atem, Folefac D; Qian, Jing; Maye, Jacqueline E; Johnson, Keith A; Betensky, Rebecca A

2017-01-01

The association between maternal age of onset of dementia and amyloid deposition (measured by in vivo positron emission tomography (PET) imaging) in cognitively normal older offspring is of interest. In a regression model for amyloid, special methods are required due to the random right censoring of the covariate of maternal age of onset of dementia. Prior literature has proposed methods to address the problem of censoring due to assay limit of detection, but not random censoring. We propose imputation methods and a survival regression method that do not require parametric assumptions about the distribution of the censored covariate. Existing imputation methods address missing covariates, but not right censored covariates. In simulation studies, we compare these methods to the simple, but inefficient complete case analysis, and to thresholding approaches. We apply the methods to the Alzheimer's study.

Estimating hazard ratios in cohort data with missing disease information due to death.

PubMed

Binder, Nadine; Herrnböck, Anne-Sophie; Schumacher, Martin

2017-03-01

In clinical and epidemiological studies information on the primary outcome of interest, that is, the disease status, is usually collected at a limited number of follow-up visits. The disease status can often only be retrieved retrospectively in individuals who are alive at follow-up, but will be missing for those who died before. Right-censoring the death cases at the last visit (ad-hoc analysis) yields biased hazard ratio estimates of a potential risk factor, and the bias can be substantial and occur in either direction. In this work, we investigate three different approaches that use the same likelihood contributions derived from an illness-death multistate model in order to more adequately estimate the hazard ratio by including the death cases into the analysis: a parametric approach, a penalized likelihood approach, and an imputation-based approach. We investigate to which extent these approaches allow for an unbiased regression analysis by evaluating their performance in simulation studies and on a real data example. In doing so, we use the full cohort with complete illness-death data as reference and artificially induce missing information due to death by setting discrete follow-up visits. Compared to an ad-hoc analysis, all considered approaches provide less biased or even unbiased results, depending on the situation studied. In the real data example, the parametric approach is seen to be too restrictive, whereas the imputation-based approach could almost reconstruct the original event history information. © 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

Longitudinal association of short-term, metronome-paced heart rate variability and echocardiographically assessed cardiac structure at a 4-year follow-up: results from the prospective, population-based CARLA cohort.

PubMed

Medenwald, Daniel; Swenne, Cees A; Frantz, Stefan; Nuding, Sebastian; Kors, Jan A; Pietzner, Diana; Tiller, Daniel; Greiser, Karin H; Kluttig, Alexander; Haerting, Johannes

2017-12-01

To assess the value of cardiac structure/function in predicting heart rate variability (HRV) and the possibly predictive value of HRV on cardiac parameters. Baseline and 4-year follow-up data from the population-based CARLA cohort were used (790 men, 646 women, aged 45-83 years at baseline and 50-87 years at follow-up). Echocardiographic and HRV recordings were performed at baseline and at follow-up. Linear regression models with a quadratic term were used. Crude and covariate adjusted estimates were calculated. Missing values were imputed by means of multiple imputation. Heart rate variability measures taken into account consisted of linear time and frequency domain [standard deviation of normal-to-normal intervals (SDNN), high-frequency power (HF), low-frequency power (LF), LF/HF ratio] and non-linear measures [detrended fluctuation analysis (DFA1), SD1, SD2, SD1/SD2 ratio]. Echocardiographic parameters considered were ventricular mass index, diastolic interventricular septum thickness, left ventricular diastolic dimension, left atrial dimension systolic (LADS), and ejection fraction (Teichholz). A negative quadratic relation between baseline LADS and change in SDNN and HF was observed. The maximum HF and SDNN change (an increase of roughly 0.02%) was predicted at LADS of 3.72 and 3.57 cm, respectively, while the majority of subjects experienced a decrease in HRV. There was no association between further echocardiographic parameters and change in HRV, and there was no evidence of a predictive value of HRV in the prediction of changes in cardiac structure. In the general population, LADS predicts 4-year alteration in SDNN and HF non-linearly. Because of the novelty of the result, analyses should be replicated in other populations. Published on behalf of the European Society of Cardiology. All rights reserved. © The Author 2017. For permissions please email: journals.permissions@oup.com.

Extent of linkage disequilibrium, consistency of gametic phase, and imputation accuracy within and across Canadian dairy breeds.

PubMed

Larmer, S G; Sargolzaei, M; Schenkel, F S

2014-05-01

Genomic selection requires a large reference population to accurately estimate single nucleotide polymorphism (SNP) effects. In some Canadian dairy breeds, the available reference populations are not large enough for accurate estimation of SNP effects for traits of interest. If marker phase is highly consistent across multiple breeds, it is theoretically possible to increase the accuracy of genomic prediction for one or all breeds by pooling several breeds into a common reference population. This study investigated the extent of linkage disequilibrium (LD) in 5 major dairy breeds using a 50,000 (50K) SNP panel and 3 of the same breeds using the 777,000 (777K) SNP panel. Correlation of pair-wise SNP phase was also investigated on both panels. The level of LD was measured using the squared correlation of alleles at 2 loci (r(2)), and the consistency of SNP gametic phases was correlated using the signed square root of these values. Because of the high cost of the 777K panel, the accuracy of imputation from lower density marker panels [6,000 (6K) or 50K] was examined both within breed and using a multi-breed reference population in Holstein, Ayrshire, and Guernsey. Imputation was carried out using FImpute V2.2 and Beagle 3.3.2 software. Imputation accuracies were then calculated as both the proportion of correct SNP filled in (concordance rate) and allelic R(2). Computation time was also explored to determine the efficiency of the different algorithms for imputation. Analysis showed that LD values >0.2 were found in all breeds at distances at or shorter than the average adjacent pair-wise distance between SNP on the 50K panel. Correlations of r-values, however, did not reach high levels (<0.9) at these distances. High correlation values of SNP phase between breeds were observed (>0.94) when the average pair-wise distances using the 777K SNP panel were examined. High concordance rate (0.968-0.995) and allelic R(2) (0.946-0.991) were found for all breeds when imputation was carried out with FImpute from 50K to 777K. Imputation accuracy for Guernsey and Ayrshire was slightly lower when using the imputation method in Beagle. Computing time was significantly greater when using Beagle software, with all comparable procedures being 9 to 13 times less efficient, in terms of time, compared with FImpute. These findings suggest that use of a multi-breed reference population might increase prediction accuracy using the 777K SNP panel and that 777K genotypes can be efficiently and effectively imputed using the lower density 50K SNP panel. Copyright © 2014 American Dairy Science Association. Published by Elsevier Inc. All rights reserved.

Screening strategies for tubal factor subfertility.

PubMed

den Hartog, J E; Lardenoije, C M J G; Severens, J L; Land, J A; Evers, J L H; Kessels, A G H

2008-08-01

Different screening strategies exist to estimate the risk of tubal factor subfertility, preceding laparoscopy. Three screening strategies, comprising Chlamydia trachomatis IgG antibody testing (CAT), high-sensitivity C-reactive protein (hs-CRP) testing and hysterosalpingography (HSG), were explored using laparoscopy as reference standard and the occurrence of a spontaneous pregnancy as a surrogate marker for the absence of tubal pathology. In this observational study, 642 subfertile women, who underwent tubal testing, participated. Data on serological testing, HSG, laparoscopy and interval conception were collected. Multiple imputations were used to compensate for missing data. Strategy A (HSG) has limited value in estimating the risk of tubal pathology. Strategy B (CAT-->HSG) shows that CAT significantly discerns patients with a high versus low risk of tubal pathology, whereas HSG following CAT has no additional value. Strategy C (CAT-->hs-CRP-->HSG) demonstrates that hs-CRP may be valuable in CAT-positive patients only and HSG has no additional value. CAT is proposed as first screening test for tubal factor subfertility. In CAT-negative women, HSG may be performed because of its high specificity and fertility-enhancing effect. In CAT-positive women, hs-CRP seems promising, whereas HSG has no additional value. The position and timing of laparoscopy deserves critical reappraisal.

Missing data in trial-based cost-effectiveness analysis: An incomplete journey.

PubMed

Leurent, Baptiste; Gomes, Manuel; Carpenter, James R

2018-06-01

Cost-effectiveness analyses (CEA) conducted alongside randomised trials provide key evidence for informing healthcare decision making, but missing data pose substantive challenges. Recently, there have been a number of developments in methods and guidelines addressing missing data in trials. However, it is unclear whether these developments have permeated CEA practice. This paper critically reviews the extent of and methods used to address missing data in recently published trial-based CEA. Issues of the Health Technology Assessment journal from 2013 to 2015 were searched. Fifty-two eligible studies were identified. Missing data were very common; the median proportion of trial participants with complete cost-effectiveness data was 63% (interquartile range: 47%-81%). The most common approach for the primary analysis was to restrict analysis to those with complete data (43%), followed by multiple imputation (30%). Half of the studies conducted some sort of sensitivity analyses, but only 2 (4%) considered possible departures from the missing-at-random assumption. Further improvements are needed to address missing data in cost-effectiveness analyses conducted alongside randomised trials. These should focus on limiting the extent of missing data, choosing an appropriate method for the primary analysis that is valid under contextually plausible assumptions, and conducting sensitivity analyses to departures from the missing-at-random assumption. © 2018 The Authors Health Economics published by John Wiley & Sons Ltd.

A Validation Method to Determine Missing Years of Birth in a Cohort Study of Shipyard Workers Using Social Security Number.

PubMed

Cabasag, Citadel J; Ziogas, Argyrios; Shehata, Merna; Anton-Culver, Hoda

2016-06-01

The objective of the study is to evaluate the generalizability and feasibility of a prediction method developed by Block et al (1983) to estimate missing age of workers by using their Social Security Numbers (SSNs). Our study used a retrospective occupational cohort of about 15,000 workers from the Long Beach Naval Shipyard (1978 to 1985). Our results showed an 89.2% agreement (P < 0.0001) between the predicted and observed years of birth using Block's method. The correlation increased to 93.1% (P < 0.0001) after our modifications by removing workers with SSNs issued in U.S. territories. The method performed better in shipyard workers who obtained their SSNs after 1950. Although the method is useful in imputing missing age, the accuracy is dependent on the year, age, and location of SSN issuance.

Filling gaps in notification data: a model-based approach applied to travel related campylobacteriosis cases in New Zealand.

PubMed

Amene, E; Horn, B; Pirie, R; Lake, R; Döpfer, D

2016-09-06

Data containing notified cases of disease are often compromised by incomplete or partial information related to individual cases. In an effort to enhance the value of information from enteric disease notifications in New Zealand, this study explored the use of Bayesian and Multiple Imputation (MI) models to fill risk factor data gaps. As a test case, overseas travel as a risk factor for infection with campylobacteriosis has been examined. Two methods, namely Bayesian Specification (BAS) and Multiple Imputation (MI), were compared regarding predictive performance for various levels of artificially induced missingness of overseas travel status in campylobacteriosis notification data. Predictive performance of the models was assessed through the Brier Score, the Area Under the ROC Curve and the Percent Bias of regression coefficients. Finally, the best model was selected and applied to predict missing overseas travel status of campylobacteriosis notifications. While no difference was observed in the predictive performance of the BAS and MI methods at a lower rate of missingness (<10 %), but the BAS approach performed better than MI at a higher rate of missingness (50 %, 65 %, 80 %). The estimated proportion (95 % Credibility Intervals) of travel related cases was greatest in highly urban District Health Boards (DHBs) in Counties Manukau, Auckland and Waitemata, at 0.37 (0.12, 0.57), 0.33 (0.13, 0.55) and 0.28 (0.10, 0.49), whereas the lowest proportion was estimated for more rural West Coast, Northland and Tairawhiti DHBs at 0.02 (0.01, 0.05), 0.03 (0.01, 0.08) and 0.04 (0.01, 0.06), respectively. The national rate of travel related campylobacteriosis cases was estimated at 0.16 (0.02, 0.48). The use of BAS offers a flexible approach to data augmentation particularly when the missing rate is very high and when the Missing At Random (MAR) assumption holds. High rates of travel associated cases in urban regions of New Zealand predicted by this approach are plausible given the high rate of travel in these regions, including destinations with higher risk of infection. The added advantage of using a Bayesian approach is that the model's prediction can be improved whenever new information becomes available.

Missing data in trial‐based cost‐effectiveness analysis: An incomplete journey

PubMed Central

Gomes, Manuel; Carpenter, James R.

2018-01-01

SUMMARY Cost‐effectiveness analyses (CEA) conducted alongside randomised trials provide key evidence for informing healthcare decision making, but missing data pose substantive challenges. Recently, there have been a number of developments in methods and guidelines addressing missing data in trials. However, it is unclear whether these developments have permeated CEA practice. This paper critically reviews the extent of and methods used to address missing data in recently published trial‐based CEA. Issues of the Health Technology Assessment journal from 2013 to 2015 were searched. Fifty‐two eligible studies were identified. Missing data were very common; the median proportion of trial participants with complete cost‐effectiveness data was 63% (interquartile range: 47%–81%). The most common approach for the primary analysis was to restrict analysis to those with complete data (43%), followed by multiple imputation (30%). Half of the studies conducted some sort of sensitivity analyses, but only 2 (4%) considered possible departures from the missing‐at‐random assumption. Further improvements are needed to address missing data in cost‐effectiveness analyses conducted alongside randomised trials. These should focus on limiting the extent of missing data, choosing an appropriate method for the primary analysis that is valid under contextually plausible assumptions, and conducting sensitivity analyses to departures from the missing‐at‐random assumption. PMID:29573044

Causal Structure of Brain Physiology after Brain Injury from Subarachnoid Hemorrhage.

PubMed

Claassen, Jan; Rahman, Shah Atiqur; Huang, Yuxiao; Frey, Hans-Peter; Schmidt, J Michael; Albers, David; Falo, Cristina Maria; Park, Soojin; Agarwal, Sachin; Connolly, E Sander; Kleinberg, Samantha

2016-01-01

High frequency physiologic data are routinely generated for intensive care patients. While massive amounts of data make it difficult for clinicians to extract meaningful signals, these data could provide insight into the state of critically ill patients and guide interventions. We develop uniquely customized computational methods to uncover the causal structure within systemic and brain physiologic measures recorded in a neurological intensive care unit after subarachnoid hemorrhage. While the data have many missing values, poor signal-to-noise ratio, and are composed from a heterogeneous patient population, our advanced imputation and causal inference techniques enable physiologic models to be learned for individuals. Our analyses confirm that complex physiologic relationships including demand and supply of oxygen underlie brain oxygen measurements and that mechanisms for brain swelling early after injury may differ from those that develop in a delayed fashion. These inference methods will enable wider use of ICU data to understand patient physiology.

PANDA-view: An easy-to-use tool for statistical analysis and visualization of quantitative proteomics data.

PubMed

Chang, Cheng; Xu, Kaikun; Guo, Chaoping; Wang, Jinxia; Yan, Qi; Zhang, Jian; He, Fuchu; Zhu, Yunping

2018-05-22

Compared with the numerous software tools developed for identification and quantification of -omics data, there remains a lack of suitable tools for both downstream analysis and data visualization. To help researchers better understand the biological meanings in their -omics data, we present an easy-to-use tool, named PANDA-view, for both statistical analysis and visualization of quantitative proteomics data and other -omics data. PANDA-view contains various kinds of analysis methods such as normalization, missing value imputation, statistical tests, clustering and principal component analysis, as well as the most commonly-used data visualization methods including an interactive volcano plot. Additionally, it provides user-friendly interfaces for protein-peptide-spectrum representation of the quantitative proteomics data. PANDA-view is freely available at https://sourceforge.net/projects/panda-view/. 1987ccpacer@163.com and zhuyunping@gmail.com. Supplementary data are available at Bioinformatics online.

Genotype imputation from various low-density SNP panels and its impact on accuracy of genomic breeding values in pigs.

PubMed

Grossi, D A; Brito, L F; Jafarikia, M; Schenkel, F S; Feng, Z

2018-04-30

The uptake of genomic selection (GS) by the swine industry is still limited by the costs of genotyping. A feasible alternative to overcome this challenge is to genotype animals using an affordable low-density (LD) single nucleotide polymorphism (SNP) chip panel followed by accurate imputation to a high-density panel. Therefore, the main objective of this study was to screen incremental densities of LD panels in order to systematically identify one that balances the tradeoffs among imputation accuracy, prediction accuracy of genomic estimated breeding values (GEBVs), and genotype density (directly associated with genotyping costs). Genotypes using the Illumina Porcine60K BeadChip were available for 1378 Duroc (DU), 2361 Landrace (LA) and 3192 Yorkshire (YO) pigs. In addition, pseudo-phenotypes (de-regressed estimated breeding values) for five economically important traits were provided for the analysis. The reference population for genotyping imputation consisted of 931 DU, 1631 LA and 2103 YO animals and the remainder individuals were included in the validation population of each breed. A LD panel of 3000 evenly spaced SNPs (LD3K) yielded high imputation accuracy rates: 93.78% (DU), 97.07% (LA) and 97.00% (YO) and high correlations (>0.97) between the predicted GEBVs using the actual 60 K SNP genotypes and the imputed 60 K SNP genotypes for all traits and breeds. The imputation accuracy was influenced by the reference population size as well as the amount of parental genotype information available in the reference population. However, parental genotype information became less important when the LD panel had at least 3000 SNPs. The correlation of the GEBVs directly increased with an increase in imputation accuracy. When genotype information for both parents was available, a panel of 300 SNPs (imputed to 60 K) yielded GEBV predictions highly correlated (⩾0.90) with genomic predictions obtained based on the true 60 K panel, for all traits and breeds. For a small reference population size with no parents on reference population, it is recommended the use of a panel at least as dense as the LD3K and, when there are two parents in the reference population, a panel as small as the LD300 might be a feasible option. These findings are of great importance for the development of LD panels for swine in order to reduce genotyping costs, increase the uptake of GS and, therefore, optimize the profitability of the swine industry.

Improved imputation of low-frequency and rare variants using the UK10K haplotype reference panel.

PubMed

Huang, Jie; Howie, Bryan; McCarthy, Shane; Memari, Yasin; Walter, Klaudia; Min, Josine L; Danecek, Petr; Malerba, Giovanni; Trabetti, Elisabetta; Zheng, Hou-Feng; Gambaro, Giovanni; Richards, J Brent; Durbin, Richard; Timpson, Nicholas J; Marchini, Jonathan; Soranzo, Nicole

2015-09-14

Imputing genotypes from reference panels created by whole-genome sequencing (WGS) provides a cost-effective strategy for augmenting the single-nucleotide polymorphism (SNP) content of genome-wide arrays. The UK10K Cohorts project has generated a data set of 3,781 whole genomes sequenced at low depth (average 7x), aiming to exhaustively characterize genetic variation down to 0.1% minor allele frequency in the British population. Here we demonstrate the value of this resource for improving imputation accuracy at rare and low-frequency variants in both a UK and an Italian population. We show that large increases in imputation accuracy can be achieved by re-phasing WGS reference panels after initial genotype calling. We also present a method for combining WGS panels to improve variant coverage and downstream imputation accuracy, which we illustrate by integrating 7,562 WGS haplotypes from the UK10K project with 2,184 haplotypes from the 1000 Genomes Project. Finally, we introduce a novel approximation that maintains speed without sacrificing imputation accuracy for rare variants.

Reporting and handling missing outcome data in mental health: a systematic review of Cochrane systematic reviews and meta-analyses.

PubMed

Spineli, Loukia M; Pandis, Nikolaos; Salanti, Georgia

2015-06-01

The purpose of the study was to provide empirical evidence about the reporting of methodology to address missing outcome data and the acknowledgement of their impact in Cochrane systematic reviews in the mental health field. Systematic reviews published in the Cochrane Database of Systematic Reviews after January 1, 2009 by three Cochrane Review Groups relating to mental health were included. One hundred ninety systematic reviews were considered. Missing outcome data were present in at least one included study in 175 systematic reviews. Of these 175 systematic reviews, 147 (84%) accounted for missing outcome data by considering a relevant primary or secondary outcome (e.g., dropout). Missing outcome data implications were reported only in 61 (35%) systematic reviews and primarily in the discussion section by commenting on the amount of the missing outcome data. One hundred forty eligible meta-analyses with missing data were scrutinized. Seventy-nine (56%) of them had studies with total dropout rate between 10 and 30%. One hundred nine (78%) meta-analyses reported to have performed intention-to-treat analysis by including trials with imputed outcome data. Sensitivity analysis for incomplete outcome data was implemented in less than 20% of the meta-analyses. Reporting of the techniques for handling missing outcome data and their implications in the findings of the systematic reviews are suboptimal. Copyright © 2014 John Wiley & Sons, Ltd.

Inference in randomized trials with death and missingness.

PubMed

Wang, Chenguang; Scharfstein, Daniel O; Colantuoni, Elizabeth; Girard, Timothy D; Yan, Ying

2017-06-01

In randomized studies involving severely ill patients, functional outcomes are often unobserved due to missed clinic visits, premature withdrawal, or death. It is well known that if these unobserved functional outcomes are not handled properly, biased treatment comparisons can be produced. In this article, we propose a procedure for comparing treatments that is based on a composite endpoint that combines information on both the functional outcome and survival. We further propose a missing data imputation scheme and sensitivity analysis strategy to handle the unobserved functional outcomes not due to death. Illustrations of the proposed method are given by analyzing data from a recent non-small cell lung cancer clinical trial and a recent trial of sedation interruption among mechanically ventilated patients. © 2016, The International Biometric Society.

Addressing the Challenges of Obtaining Functional Outcomes in Traumatic Brain Injury Research: Missing Data Patterns, Timing of Follow-Up, and Three Prognostic Models

PubMed Central

Morrison, Laurie J.; Devlin, Sean M.; Bulger, Eileen M.; Brasel, Karen J.; Sheehan, Kellie; Minei, Joseph P.; Kerby, Jeffrey D.; Tisherman, Samuel A.; Rizoli, Sandro; Karmy-Jones, Riyad; van Heest, Rardi; Newgard, Craig D.

2014-01-01

Abstract Traumatic brain injury (TBI) is common and debilitating. Randomized trials of interventions for TBI ideally assess effectiveness by using long-term functional neurological outcomes, but such outcomes are difficult to obtain and costly. If there is little change between functional status at hospital discharge versus 6 months, then shorter-term outcomes may be adequate for use in future clinical trials. Using data from a previously published multi-center, randomized, placebo-controlled TBI clinical trial, we evaluated patterns of missing outcome data, changes in functional status between hospital discharge and 6 months, and three prognostic models to predict long-term functional outcome from covariates available at hospital discharge (functional measures, demographics, and injury characteristics). The Resuscitation Outcomes Consortium Hypertonic Saline trial enrolled 1282 TBI patients, obtaining the primary outcome of 6-month Glasgow Outcome Score Extended (GOSE) for 85% of patients, but missing the primary outcome for the remaining 15%. Patients with missing outcomes had less-severe injuries, higher neurological function at discharge (GOSE), and shorter hospital stays than patients whose GOSE was obtained. Of 1066 (83%) patients whose GOSE was obtained both at hospital discharge and at 6-months, 71% of patients had the same dichotomized functional status (severe disability/death vs. moderate/no disability) after 6 months as at discharge, 28% had an improved functional status, and 1% had worsened. Performance was excellent (C-statistic between 0.88 and 0.91) for all three prognostic models and calibration adequate for two models (p values, 0.22 and 0.85). Our results suggest that multiple imputation of the standard 6-month GOSE may be reasonable in TBI research when the primary outcome cannot be obtained through other means. PMID:24552494

Addressing the challenges of obtaining functional outcomes in traumatic brain injury research: missing data patterns, timing of follow-up, and three prognostic models.

PubMed

Zelnick, Leila R; Morrison, Laurie J; Devlin, Sean M; Bulger, Eileen M; Brasel, Karen J; Sheehan, Kellie; Minei, Joseph P; Kerby, Jeffrey D; Tisherman, Samuel A; Rizoli, Sandro; Karmy-Jones, Riyad; van Heest, Rardi; Newgard, Craig D

2014-06-01

Traumatic brain injury (TBI) is common and debilitating. Randomized trials of interventions for TBI ideally assess effectiveness by using long-term functional neurological outcomes, but such outcomes are difficult to obtain and costly. If there is little change between functional status at hospital discharge versus 6 months, then shorter-term outcomes may be adequate for use in future clinical trials. Using data from a previously published multi-center, randomized, placebo-controlled TBI clinical trial, we evaluated patterns of missing outcome data, changes in functional status between hospital discharge and 6 months, and three prognostic models to predict long-term functional outcome from covariates available at hospital discharge (functional measures, demographics, and injury characteristics). The Resuscitation Outcomes Consortium Hypertonic Saline trial enrolled 1282 TBI patients, obtaining the primary outcome of 6-month Glasgow Outcome Score Extended (GOSE) for 85% of patients, but missing the primary outcome for the remaining 15%. Patients with missing outcomes had less-severe injuries, higher neurological function at discharge (GOSE), and shorter hospital stays than patients whose GOSE was obtained. Of 1066 (83%) patients whose GOSE was obtained both at hospital discharge and at 6-months, 71% of patients had the same dichotomized functional status (severe disability/death vs. moderate/no disability) after 6 months as at discharge, 28% had an improved functional status, and 1% had worsened. Performance was excellent (C-statistic between 0.88 and 0.91) for all three prognostic models and calibration adequate for two models (p values, 0.22 and 0.85). Our results suggest that multiple imputation of the standard 6-month GOSE may be reasonable in TBI research when the primary outcome cannot be obtained through other means.

Reconstruction of spatially detailed global map of NH4+ and NO3- application in synthetic nitrogen fertilizer

NASA Astrophysics Data System (ADS)

Nishina, Kazuya; Ito, Akihiko; Hanasaki, Naota; Hayashi, Seiji

2017-02-01

Currently, available historical global N fertilizer map as an input data to global biogeochemical model is still limited and existing maps were not considered NH4+ and NO3- in the fertilizer application rates. This paper provides a method for constructing a new historical global nitrogen fertilizer application map (0.5° × 0.5° resolution) for the period 1961-2010 based on country-specific information from Food and Agriculture Organization statistics (FAOSTAT) and various global datasets. This new map incorporates the fraction of NH4+ (and NO3-) in N fertilizer inputs by utilizing fertilizer species information in FAOSTAT, in which species can be categorized as NH4+- and/or NO3--forming N fertilizers. During data processing, we applied a statistical data imputation method for the missing data (19 % of national N fertilizer consumption) in FAOSTAT. The multiple imputation method enabled us to fill gaps in the time-series data using plausible values using covariates information (year, population, GDP, and crop area). After the imputation, we downscaled the national consumption data to a gridded cropland map. Also, we applied the multiple imputation method to the available chemical fertilizer species consumption, allowing for the estimation of the NH4+ / NO3- ratio in national fertilizer consumption. In this study, the synthetic N fertilizer inputs in 2000 showed a general consistency with the existing N fertilizer map (Potter et al., 2010) in relation to the ranges of N fertilizer inputs. Globally, the estimated N fertilizer inputs based on the sum of filled data increased from 15 to 110 Tg-N during 1961-2010. On the other hand, the global NO3- input started to decline after the late 1980s and the fraction of NO3- in global N fertilizer decreased consistently from 35 to 13 % over a 50-year period. NH4+-forming fertilizers are dominant in most countries; however, the NH4+ / NO3- ratio in N fertilizer inputs shows clear differences temporally and geographically. This new map can be utilized as input data to global model studies and bring new insights for the assessment of historical terrestrial N cycling changes. Datasets available at doi:10.1594/PANGAEA.861203.

Effectiveness of a Web-Based Screening and Fully Automated Brief Motivational Intervention for Adolescent Substance Use: A Randomized Controlled Trial.

PubMed

Arnaud, Nicolas; Baldus, Christiane; Elgán, Tobias H; De Paepe, Nina; Tønnesen, Hanne; Csémy, Ladislav; Thomasius, Rainer

2016-05-24

Mid-to-late adolescence is a critical period for initiation of alcohol and drug problems, which can be reduced by targeted brief motivational interventions. Web-based brief interventions have advantages in terms of acceptability and accessibility and have shown significant reductions of substance use among college students. However, the evidence is sparse among adolescents with at-risk use of alcohol and other drugs. This study evaluated the effectiveness of a targeted and fully automated Web-based brief motivational intervention with no face-to-face components on substance use among adolescents screened for at-risk substance use in four European countries. In an open-access, purely Web-based randomized controlled trial, a convenience sample of adolescents aged 16-18 years from Sweden, Germany, Belgium, and the Czech Republic was recruited using online and offline methods and screened online for at-risk substance use using the CRAFFT (Car, Relax, Alone, Forget, Friends, Trouble) screening instrument. Participants were randomized to a single session brief motivational intervention group or an assessment-only control group but not blinded. Primary outcome was differences in past month drinking measured by a self-reported AUDIT-C-based index score for drinking frequency, quantity, and frequency of binge drinking with measures collected online at baseline and after 3 months. Secondary outcomes were the AUDIT-C-based separate drinking indicators, illegal drug use, and polydrug use. All outcome analyses were conducted with and without Expectation Maximization (EM) imputation of missing follow-up data. In total, 2673 adolescents were screened and 1449 (54.2%) participants were randomized to the intervention or control group. After 3 months, 211 adolescents (14.5%) provided follow-up data. Compared to the control group, results from linear mixed models revealed significant reductions in self-reported past-month drinking in favor of the intervention group in both the non-imputed (P=.010) and the EM-imputed sample (P=.022). Secondary analyses revealed a significant effect on drinking frequency (P=.037) and frequency of binge drinking (P=.044) in the non-imputation-based analyses and drinking quantity (P=.021) when missing data were imputed. Analyses for illegal drug use and polydrug use revealed no significant differences between the study groups (Ps>.05). Although the study is limited by a large drop-out, significant between-group effects for alcohol use indicate that targeted brief motivational intervention in a fully automated Web-based format can be effective to reduce drinking and lessen existing substance use service barriers for at-risk drinking European adolescents. International Standard Randomized Controlled Trial Registry: ISRCTN95538913; http://www.isrctn.com/ISRCTN95538913 (Archived by WebCite at http://www.webcitation.org/6XkuUEwBx).

Approaches in Characterizing Genetic Structure and Mapping in a Rice Multiparental Population.

PubMed

Raghavan, Chitra; Mauleon, Ramil; Lacorte, Vanica; Jubay, Monalisa; Zaw, Hein; Bonifacio, Justine; Singh, Rakesh Kumar; Huang, B Emma; Leung, Hei

2017-06-07

Multi-parent Advanced Generation Intercross (MAGIC) populations are fast becoming mainstream tools for research and breeding, along with the technology and tools for analysis. This paper demonstrates the analysis of a rice MAGIC population from data filtering to imputation and processing of genetic data to characterizing genomic structure, and finally quantitative trait loci (QTL) mapping. In this study, 1316 S6:8 indica MAGIC (MI) lines and the eight founders were sequenced using Genotyping by Sequencing (GBS). As the GBS approach often includes missing data, the first step was to impute the missing SNPs. The observable number of recombinations in the population was then explored. Based on this case study, a general outline of procedures for a MAGIC analysis workflow is provided, as well as for QTL mapping of agronomic traits and biotic and abiotic stress, using the results from both association and interval mapping approaches. QTL for agronomic traits (yield, flowering time, and plant height), physical (grain length and grain width) and cooking properties (amylose content) of the rice grain, abiotic stress (submergence tolerance), and biotic stress (brown spot disease) were mapped. Through presenting this extensive analysis in the MI population in rice, we highlight important considerations when choosing analytical approaches. The methods and results reported in this paper will provide a guide to future genetic analysis methods applied to multi-parent populations. Copyright © 2017 Raghavan et al.

Fusing Data Mining, Machine Learning and Traditional Statistics to Detect Biomarkers Associated with Depression

PubMed Central

Dipnall, Joanna F.

2016-01-01

Background Atheoretical large-scale data mining techniques using machine learning algorithms have promise in the analysis of large epidemiological datasets. This study illustrates the use of a hybrid methodology for variable selection that took account of missing data and complex survey design to identify key biomarkers associated with depression from a large epidemiological study. Methods The study used a three-step methodology amalgamating multiple imputation, a machine learning boosted regression algorithm and logistic regression, to identify key biomarkers associated with depression in the National Health and Nutrition Examination Study (2009–2010). Depression was measured using the Patient Health Questionnaire-9 and 67 biomarkers were analysed. Covariates in this study included gender, age, race, smoking, food security, Poverty Income Ratio, Body Mass Index, physical activity, alcohol use, medical conditions and medications. The final imputed weighted multiple logistic regression model included possible confounders and moderators. Results After the creation of 20 imputation data sets from multiple chained regression sequences, machine learning boosted regression initially identified 21 biomarkers associated with depression. Using traditional logistic regression methods, including controlling for possible confounders and moderators, a final set of three biomarkers were selected. The final three biomarkers from the novel hybrid variable selection methodology were red cell distribution width (OR 1.15; 95% CI 1.01, 1.30), serum glucose (OR 1.01; 95% CI 1.00, 1.01) and total bilirubin (OR 0.12; 95% CI 0.05, 0.28). Significant interactions were found between total bilirubin with Mexican American/Hispanic group (p = 0.016), and current smokers (p<0.001). Conclusion The systematic use of a hybrid methodology for variable selection, fusing data mining techniques using a machine learning algorithm with traditional statistical modelling, accounted for missing data and complex survey sampling methodology and was demonstrated to be a useful tool for detecting three biomarkers associated with depression for future hypothesis generation: red cell distribution width, serum glucose and total bilirubin. PMID:26848571

Fusing Data Mining, Machine Learning and Traditional Statistics to Detect Biomarkers Associated with Depression.

PubMed

Dipnall, Joanna F; Pasco, Julie A; Berk, Michael; Williams, Lana J; Dodd, Seetal; Jacka, Felice N; Meyer, Denny

2016-01-01

Atheoretical large-scale data mining techniques using machine learning algorithms have promise in the analysis of large epidemiological datasets. This study illustrates the use of a hybrid methodology for variable selection that took account of missing data and complex survey design to identify key biomarkers associated with depression from a large epidemiological study. The study used a three-step methodology amalgamating multiple imputation, a machine learning boosted regression algorithm and logistic regression, to identify key biomarkers associated with depression in the National Health and Nutrition Examination Study (2009-2010). Depression was measured using the Patient Health Questionnaire-9 and 67 biomarkers were analysed. Covariates in this study included gender, age, race, smoking, food security, Poverty Income Ratio, Body Mass Index, physical activity, alcohol use, medical conditions and medications. The final imputed weighted multiple logistic regression model included possible confounders and moderators. After the creation of 20 imputation data sets from multiple chained regression sequences, machine learning boosted regression initially identified 21 biomarkers associated with depression. Using traditional logistic regression methods, including controlling for possible confounders and moderators, a final set of three biomarkers were selected. The final three biomarkers from the novel hybrid variable selection methodology were red cell distribution width (OR 1.15; 95% CI 1.01, 1.30), serum glucose (OR 1.01; 95% CI 1.00, 1.01) and total bilirubin (OR 0.12; 95% CI 0.05, 0.28). Significant interactions were found between total bilirubin with Mexican American/Hispanic group (p = 0.016), and current smokers (p<0.001). The systematic use of a hybrid methodology for variable selection, fusing data mining techniques using a machine learning algorithm with traditional statistical modelling, accounted for missing data and complex survey sampling methodology and was demonstrated to be a useful tool for detecting three biomarkers associated with depression for future hypothesis generation: red cell distribution width, serum glucose and total bilirubin.

Effects of cloud cover and meteorology in estimating ground-level air pollution using MAIAC AOD in the Yangtze River Delta

NASA Astrophysics Data System (ADS)

Xiao, Q.; Liu, Y.

2017-12-01

Satellite aerosol optical depth (AOD) has been used to assess fine particulate matter (PM2.5) pollution worldwide. However, non-random missing AOD due to cloud cover or high surface reflectance can cause up to 80% data loss and bias model-estimated spatial and temporal trends of PM2.5. Previous studies filled the data gap largely by spatial smoothing which ignored the impact of cloud cover and meteorology on aerosol loadings and has been shown to exhibit poor performance when monitoring stations are sparse or when there is seasonal large-scale missingness. Using the Yangtze River Delta of China as an example, we present a flexible Multiple Imputation (MI) method that combines cloud fraction, elevation, humidity, temperature, and spatiotemporal trends to impute the missing AOD. A two-stage statistical model driven by gap-filled AOD, meteorology and land use information was then fitted to estimate daily ground PM2.5 concentrations in 2013 and 2014 at 1 km resolution with complete coverage in space and time. The daily MI models have an average R2 of 0.77, with an inter-quartile range of 0.71 to 0.82 across days. The overall model 10-fold cross-validation R2 were 0.81 and 0.73 (for year 2013 and 2014, respectively. Predictions with only observational AOD or only imputed AOD showed similar accuracy. This method provides reliable PM2.5 predictions with complete coverage at high resolution. By including all the pixels of all days into model development, this method corrected the sampling bias in satellite-driven air pollution modelling due to non-random missingness in AOD. Comparing with previously reported gap-filling methods, the MI method has the strength of not relying on ground PM2.5 measurements, therefore allows the prediction of historical PM2.5 levels prior to the establishment of regular ground monitoring networks.

On using summary statistics from an external calibration sample to correct for covariate measurement error.

PubMed

Guo, Ying; Little, Roderick J; McConnell, Daniel S

2012-01-01

Covariate measurement error is common in epidemiologic studies. Current methods for correcting measurement error with information from external calibration samples are insufficient to provide valid adjusted inferences. We consider the problem of estimating the regression of an outcome Y on covariates X and Z, where Y and Z are observed, X is unobserved, but a variable W that measures X with error is observed. Information about measurement error is provided in an external calibration sample where data on X and W (but not Y and Z) are recorded. We describe a method that uses summary statistics from the calibration sample to create multiple imputations of the missing values of X in the regression sample, so that the regression coefficients of Y on X and Z and associated standard errors can be estimated using simple multiple imputation combining rules, yielding valid statistical inferences under the assumption of a multivariate normal distribution. The proposed method is shown by simulation to provide better inferences than existing methods, namely the naive method, classical calibration, and regression calibration, particularly for correction for bias and achieving nominal confidence levels. We also illustrate our method with an example using linear regression to examine the relation between serum reproductive hormone concentrations and bone mineral density loss in midlife women in the Michigan Bone Health and Metabolism Study. Existing methods fail to adjust appropriately for bias due to measurement error in the regression setting, particularly when measurement error is substantial. The proposed method corrects this deficiency.

Daily home gardening improved survival for older people with mobility limitations: an 11-year follow-up study in Taiwan.

PubMed

Lêng, Chhian Hūi; Wang, Jung-Der

2016-01-01

To test the hypothesis that gardening is beneficial for survival after taking time-dependent comorbidities, mobility, and depression into account in a longitudinal middle-aged (50-64 years) and older (≥65 years) cohort in Taiwan. The cohort contained 5,058 nationally sampled adults ≥50 years old from the Taiwan Longitudinal Study on Aging (1996-2007). Gardening was defined as growing flowers, gardening, or cultivating potted plants for pleasure with five different frequencies. We calculated hazard ratios for the mortality risks of gardening and adjusted the analysis for socioeconomic status, health behaviors and conditions, depression, mobility limitations, and comorbidities. Survival models also examined time-dependent effects and risks in each stratum contingent upon baseline mobility and depression. Sensitivity analyses used imputation methods for missing values. Daily home gardening was associated with a high survival rate (hazard ratio: 0.82; 95% confidence interval: 0.71-0.94). The benefits were robust for those with mobility limitations, but without depression at baseline (hazard ratio: 0.64, 95% confidence interval: 0.48-0.87) when adjusted for time-dependent comorbidities, mobility limitations, and depression. Chronic or relapsed depression weakened the protection of gardening. For those without mobility limitations and not depressed at baseline, gardening had no effect. Sensitivity analyses using different imputation methods yielded similar results and corroborated the hypothesis. Daily gardening for pleasure was associated with reduced mortality for Taiwanese >50 years old with mobility limitations but without depression.

Matrix and Tensor Completion on a Human Activity Recognition Framework.

PubMed

Savvaki, Sofia; Tsagkatakis, Grigorios; Panousopoulou, Athanasia; Tsakalides, Panagiotis

2017-11-01

Sensor-based activity recognition is encountered in innumerable applications of the arena of pervasive healthcare and plays a crucial role in biomedical research. Nonetheless, the frequent situation of unobserved measurements impairs the ability of machine learning algorithms to efficiently extract context from raw streams of data. In this paper, we study the problem of accurate estimation of missing multimodal inertial data and we propose a classification framework that considers the reconstruction of subsampled data during the test phase. We introduce the concept of forming the available data streams into low-rank two-dimensional (2-D) and 3-D Hankel structures, and we exploit data redundancies using sophisticated imputation techniques, namely matrix and tensor completion. Moreover, we examine the impact of reconstruction on the classification performance by experimenting with several state-of-the-art classifiers. The system is evaluated with respect to different data structuring scenarios, the volume of data available for reconstruction, and various levels of missing values per device. Finally, the tradeoff between subsampling accuracy and energy conservation in wearable platforms is examined. Our analysis relies on two public datasets containing inertial data, which extend to numerous activities, multiple sensing parameters, and body locations. The results highlight that robust classification accuracy can be achieved through recovery, even for extremely subsampled data streams.

A Statistical Method for Estimating Missing GHG Emissions in Bottom-Up Inventories: The Case of Fossil Fuel Combustion in Industry in the Bogota Region, Colombia

NASA Astrophysics Data System (ADS)

Jimenez-Pizarro, R.; Rojas, A. M.; Pulido-Guio, A. D.

2012-12-01

The development of environmentally, socially and financially suitable greenhouse gas (GHG) mitigation portfolios requires detailed disaggregation of emissions by activity sector, preferably at the regional level. Bottom-up (BU) emission inventories are intrinsically disaggregated, but although detailed, they are frequently incomplete. Missing and erroneous activity data are rather common in emission inventories of GHG, criteria and toxic pollutants, even in developed countries. The fraction of missing and erroneous data can be rather large in developing country inventories. In addition, the cost and time for obtaining or correcting this information can be prohibitive or can delay the inventory development. This is particularly true for regional BU inventories in the developing world. Moreover, a rather common practice is to disregard or to arbitrarily impute low default activity or emission values to missing data, which typically leads to significant underestimation of the total emissions. Our investigation focuses on GHG emissions by fossil fuel combustion in industry in the Bogota Region, composed by Bogota and its adjacent, semi-rural area of influence, the Province of Cundinamarca. We found that the BU inventories for this sub-category substantially underestimate emissions when compared to top-down (TD) estimations based on sub-sector specific national fuel consumption data and regional energy intensities. Although both BU inventories have a substantial number of missing and evidently erroneous entries, i.e. information on fuel consumption per combustion unit per company, the validated energy use and emission data display clear and smooth frequency distributions, which can be adequately fitted to bimodal log-normal distributions. This is not unexpected as industrial plant sizes are typically log-normally distributed. Moreover, our statistical tests suggest that industrial sub-sectors, as classified by the International Standard Industrial Classification (ISIC), are also well represented by log-normal distributions. Using the validated data, we tested several missing data estimation procedures, including Montecarlo sampling of the real and fitted distributions, and a per ISIC estimation based on bootstrap-calculated mean values. These results will be presented and discussed in detail. Our results suggest that the accuracy of sub-sector BU emission inventories, particularly in developing regions, could be significantly improved if they are designed and carried out to be representative sub-samples (surveys) of the actual universe of emitters. A large fraction the missing data could be subsequently estimated by robust statistical procedures provided that most of the emitters were accounted by number and ISIC.

Application of pattern mixture models to address missing data in longitudinal data analysis using SPSS.

PubMed

Son, Heesook; Friedmann, Erika; Thomas, Sue A

2012-01-01

Longitudinal studies are used in nursing research to examine changes over time in health indicators. Traditional approaches to longitudinal analysis of means, such as analysis of variance with repeated measures, are limited to analyzing complete cases. This limitation can lead to biased results due to withdrawal or data omission bias or to imputation of missing data, which can lead to bias toward the null if data are not missing completely at random. Pattern mixture models are useful to evaluate the informativeness of missing data and to adjust linear mixed model (LMM) analyses if missing data are informative. The aim of this study was to provide an example of statistical procedures for applying a pattern mixture model to evaluate the informativeness of missing data and conduct analyses of data with informative missingness in longitudinal studies using SPSS. The data set from the Patients' and Families' Psychological Response to Home Automated External Defibrillator Trial was used as an example to examine informativeness of missing data with pattern mixture models and to use a missing data pattern in analysis of longitudinal data. Prevention of withdrawal bias, omitted data bias, and bias toward the null in longitudinal LMMs requires the assessment of the informativeness of the occurrence of missing data. Missing data patterns can be incorporated as fixed effects into LMMs to evaluate the contribution of the presence of informative missingness to and control for the effects of missingness on outcomes. Pattern mixture models are a useful method to address the presence and effect of informative missingness in longitudinal studies.

Missing Data in Alcohol Clinical Trials with Binary Outcomes

PubMed Central

Hallgren, Kevin A.; Witkiewitz, Katie; Kranzler, Henry R.; Falk, Daniel E.; Litten, Raye Z.; O’Malley, Stephanie S.; Anton, Raymond F.

2017-01-01

Background Missing data are common in alcohol clinical trials for both continuous and binary endpoints. Approaches to handle missing data have been explored for continuous outcomes, yet no studies have compared missing data approaches for binary outcomes (e.g., abstinence, no heavy drinking days). The present study compares approaches to modeling binary outcomes with missing data in the COMBINE study. Method We included participants in the COMBINE Study who had complete drinking data during treatment and who were assigned to active medication or placebo conditions (N=1146). Using simulation methods, missing data were introduced under common scenarios with varying sample sizes and amounts of missing data. Logistic regression was used to estimate the effect of naltrexone (vs. placebo) in predicting any drinking and any heavy drinking outcomes at the end of treatment using four analytic approaches: complete case analysis (CCA), last observation carried forward (LOCF), the worst-case scenario of missing equals any drinking or heavy drinking (WCS), and multiple imputation (MI). In separate analyses, these approaches were compared when drinking data were manually deleted for those participants who discontinued treatment but continued to provide drinking data. Results WCS produced the greatest amount of bias in treatment effect estimates. MI usually yielded less biased estimates than WCS and CCA in the simulated data, and performed considerably better than LOCF when estimating treatment effects among individuals who discontinued treatment. Conclusions Missing data can introduce bias in treatment effect estimates in alcohol clinical trials. Researchers should utilize modern missing data methods, including MI, and avoid WCS and CCA when analyzing binary alcohol clinical trial outcomes. PMID:27254113

Missing Data in Alcohol Clinical Trials with Binary Outcomes.

PubMed

Hallgren, Kevin A; Witkiewitz, Katie; Kranzler, Henry R; Falk, Daniel E; Litten, Raye Z; O'Malley, Stephanie S; Anton, Raymond F

2016-07-01

Missing data are common in alcohol clinical trials for both continuous and binary end points. Approaches to handle missing data have been explored for continuous outcomes, yet no studies have compared missing data approaches for binary outcomes (e.g., abstinence, no heavy drinking days). This study compares approaches to modeling binary outcomes with missing data in the COMBINE study. We included participants in the COMBINE study who had complete drinking data during treatment and who were assigned to active medication or placebo conditions (N = 1,146). Using simulation methods, missing data were introduced under common scenarios with varying sample sizes and amounts of missing data. Logistic regression was used to estimate the effect of naltrexone (vs. placebo) in predicting any drinking and any heavy drinking outcomes at the end of treatment using 4 analytic approaches: complete case analysis (CCA), last observation carried forward (LOCF), the worst case scenario (WCS) of missing equals any drinking or heavy drinking, and multiple imputation (MI). In separate analyses, these approaches were compared when drinking data were manually deleted for those participants who discontinued treatment but continued to provide drinking data. WCS produced the greatest amount of bias in treatment effect estimates. MI usually yielded less biased estimates than WCS and CCA in the simulated data and performed considerably better than LOCF when estimating treatment effects among individuals who discontinued treatment. Missing data can introduce bias in treatment effect estimates in alcohol clinical trials. Researchers should utilize modern missing data methods, including MI, and avoid WCS and CCA when analyzing binary alcohol clinical trial outcomes. Copyright © 2016 by the Research Society on Alcoholism.

Dissolved organic nitrogen dynamics in the North Sea: A time series analysis (1995-2005)

NASA Astrophysics Data System (ADS)

Van Engeland, T.; Soetaert, K.; Knuijt, A.; Laane, R. W. P. M.; Middelburg, J. J.

2010-09-01

Dissolved organic nitrogen (DON) dynamics in the North Sea was explored by means of long-term time series of nitrogen parameters from the Dutch national monitoring program. Generally, the data quality was good with little missing data points. Different imputation methods were used to verify the robustness of the patterns against these missing data. No long-term trends in DON concentrations were found over the sampling period (1995-2005). Inter-annual variability in the different time series showed both common and station-specific behavior. The stations could be divided into two regions, based on absolute concentrations and the dominant times scales of variability. Average DON concentrations were 11 μmol l -1 in the coastal region and 5 μmol l -1 in the open sea. Organic fractions of total dissolved nitrogen (TDN) averaged 38 and 71% in the coastal zone and open sea, respectively, but increased over time due to decreasing dissolved inorganic nitrogen (DIN) concentrations. In both regions intra-annual variability dominated over inter-annual variability, but DON variation in the open sea was markedly shifted towards shorter time scales relative to coastal stations. In the coastal zone a consistent seasonal DON cycle existed with high values in spring-summer and low values in autumn-winter. In the open sea seasonality was weak. A marked shift in the seasonality was found at the Dogger Bank, with DON accumulation towards summer and low values in winter prior to 1999, and accumulation in spring and decline throughout summer after 1999. This study clearly shows that DON is a dynamic actor in the North Sea and should be monitored systematically to enable us to understand fully the functioning of this ecosystem.

MaCH-Admix: Genotype Imputation for Admixed Populations

PubMed Central

Liu, Eric Yi; Li, Mingyao; Wang, Wei; Li, Yun

2012-01-01

Imputation in admixed populations is an important problem but challenging due to the complex linkage disequilibrium (LD) pattern. The emergence of large reference panels such as that from the 1,000 Genomes Project enables more accurate imputation in general, and in particular for admixed populations and for uncommon variants. To efficiently benefit from these large reference panels, one key issue to consider in modern genotype imputation framework is the selection of effective reference panels. In this work, we consider a number of methods for effective reference panel construction inside a hidden Markov model and specific to each target individual. These methods fall into two categories: identity-by-state (IBS) based and ancestry-weighted approach. We evaluated the performance on individuals from recently admixed populations. Our target samples include 8,421 African Americans and 3,587 Hispanic Americans from the Women’s Health Initiative, which allow assessment of imputation quality for uncommon variants. Our experiments include both large and small reference panels; large, medium, and small target samples; and in genome regions of varying levels of LD. We also include BEAGLE and IMPUTE2 for comparison. Experiment results with large reference panel suggest that our novel piecewise IBS method yields consistently higher imputation quality than other methods/software. The advantage is particularly noteworthy among uncommon variants where we observe up to 5.1% information gain with the difference being highly significant (Wilcoxon signed rank test P-value < 0.0001). Our work is the first that considers various sensible approaches for imputation in admixed populations and presents a comprehensive comparison. PMID:23074066

Mapping change of older forest with nearest-neighbor imputation and Landsat time-series

Treesearch

Janet L. Ohmann; Matthew J. Gregory; Heather M. Roberts; Warren B. Cohen; Robert E. Kennedy; Zhiqiang Yang

2012-01-01

The Northwest Forest Plan (NWFP), which aims to conserve late-successional and old-growth forests (older forests) and associated species, established new policies on federal lands in the Pacific Northwest USA. As part of monitoring for the NWFP, we tested nearest-neighbor imputation for mapping change in older forest, defined by threshold values for forest attributes...

Imputation of Baseline LDL Cholesterol Concentration in Patients with Familial Hypercholesterolemia on Statins or Ezetimibe.

PubMed

Ruel, Isabelle; Aljenedil, Sumayah; Sadri, Iman; de Varennes, Émilie; Hegele, Robert A; Couture, Patrick; Bergeron, Jean; Wanneh, Eric; Baass, Alexis; Dufour, Robert; Gaudet, Daniel; Brisson, Diane; Brunham, Liam R; Francis, Gordon A; Cermakova, Lubomira; Brophy, James M; Ryomoto, Arnold; Mancini, G B John; Genest, Jacques

2018-02-01

Familial hypercholesterolemia (FH) is the most frequent genetic disorder seen clinically and is characterized by increased LDL cholesterol (LDL-C) (>95th percentile), family history of increased LDL-C, premature atherosclerotic cardiovascular disease (ASCVD) in the patient or in first-degree relatives, presence of tendinous xanthomas or premature corneal arcus, or presence of a pathogenic mutation in the LDLR , PCSK9 , or APOB genes. A diagnosis of FH has important clinical implications with respect to lifelong risk of ASCVD and requirement for intensive pharmacological therapy. The concentration of baseline LDL-C (untreated) is essential for the diagnosis of FH but is often not available because the individual is already on statin therapy. To validate a new algorithm to impute baseline LDL-C, we examined 1297 patients. The baseline LDL-C was compared with the imputed baseline obtained within 18 months of the initiation of therapy. We compared the percent reduction in LDL-C on treatment from baseline with the published percent reductions. After eliminating individuals with missing data, nonstandard doses of statins, or medications other than statins or ezetimibe, we provide data on 951 patients. The mean ± SE baseline LDL-C was 243.0 (2.2) mg/dL [6.28 (0.06) mmol/L], and the mean ± SE imputed baseline LDL-C was 244.2 (2.6) mg/dL [6.31 (0.07) mmol/L] ( P = 0.48). There was no difference in response according to the patient's sex or in percent reduction between observed and expected for individual doses or types of statin or ezetimibe. We provide a validated estimation of baseline LDL-C for patients with FH that may help clinicians in making a diagnosis. © 2017 American Association for Clinical Chemistry.

Biomarkers to monitor safety in people on art and risk of mortality.

PubMed

Bansi, Loveleen; Gazzard, Brian; Post, Frank; Phillips, Andrew; Johnson, Margaret; Hill, Teresa; Gilson, Richard; Leen, Clifford; Walsh, John; Fisher, Martin; Sabin, Caroline

2012-05-01

Though patients with HIV now have near normal life expectancies as a result of antiretroviral treatment, long-term adverse effects are of growing concern. Using time-updated laboratory measurements, we use several methods to derive a score that can be used to identify individuals at high risk of mortality. Patients who started highly active antiretroviral therapy after 2000 and had ≥1 CD4 count, viral load, and laboratory marker recorded after the date of starting highly active antiretroviral therapy were included in the analyses. Laboratory markers were stratified into quintiles and associations between each marker and mortality was assessed using Poisson regression. The estimates of the final model were used to construct a score for predicting short-term mortality. Several methods, including multiple imputation, were used for analyzing records with missing measurements. Of the 7232 patients included in this analysis, 247 died over 24,796 person-years of follow-up, giving an overall mortality rate of 1.00 (95% confidence interval: 0.87 to 1.12) per 100 person-years. Regardless of which method was used to deal with missing data, albumin, alkaline phosphatase, and hemoglobin were independently associated with mortality. Alanine transaminase was independently associated with mortality when patients with missing measurements were assumed to have measurements within the normal range. The C-statistics for all models ranged from 0.76 to 0.78. Measures of alanine transaminase, albumin, alkaline phosphatase, and hemoglobin in the normal range were predictive of mortality, and hence we suggest using a scoring system to predict mortality which relies on the raw values of these 4 laboratory markers.

The impact of different strategies to handle missing data on both precision and bias in a drug safety study: a multidatabase multinational population-based cohort study

PubMed Central

Martín-Merino, Elisa; Calderón-Larrañaga, Amaia; Hawley, Samuel; Poblador-Plou, Beatriz; Llorente-García, Ana; Petersen, Irene; Prieto-Alhambra, Daniel

2018-01-01

Background Missing data are often an issue in electronic medical records (EMRs) research. However, there are many ways that people deal with missing data in drug safety studies. Aim To compare the risk estimates resulting from different strategies for the handling of missing data in the study of venous thromboembolism (VTE) risk associated with antiosteoporotic medications (AOM). Methods New users of AOM (alendronic acid, other bisphosphonates, strontium ranelate, selective estrogen receptor modulators, teriparatide, or denosumab) aged ≥50 years during 1998–2014 were identified in two Spanish (the Base de datos para la Investigación Farmacoepidemiológica en Atención Primaria [BIFAP] and EpiChron cohort) and one UK (Clinical Practice Research Datalink [CPRD]) EMR. Hazard ratios (HRs) according to AOM (with alendronic acid as reference) were calculated adjusting for VTE risk factors, body mass index (that was missing in 61% of patients included in the three databases), and smoking (that was missing in 23% of patients) in the year of AOM therapy initiation. HRs and standard errors obtained using cross-sectional multiple imputation (MI) (reference method) were compared to complete case (CC) analysis – using only patients with complete data – and longitudinal MI – adding to the cross-sectional MI model the body mass index/smoking values as recorded in the year before and after therapy initiation. Results Overall, 422/95,057 (0.4%), 19/12,688 (0.1%), and 2,051/161,202 (1.3%) VTE cases/participants were seen in BIFAP, EpiChron, and CPRD, respectively. HRs moved from 100.00% underestimation to 40.31% overestimation in CC compared with cross-sectional MI, while longitudinal MI methods provided similar risk estimates compared with cross-sectional MI. Precision for HR improved in cross-sectional MI versus CC by up to 160.28%, while longitudinal MI improved precision (compared with cross-sectional) only minimally (up to 0.80%). Conclusion CC may substantially affect relative risk estimation in EMR-based drug safety studies, since missing data are not often completely at random. Little improvement was seen in these data in terms of power with the inclusion of longitudinal MI compared with cross-sectional MI. The strategy for handling missing data in drug safety studies can have a large impact on both risk estimates and precision.

The impact of different strategies to handle missing data on both precision and bias in a drug safety study: a multidatabase multinational population-based cohort study.

PubMed

Martín-Merino, Elisa; Calderón-Larrañaga, Amaia; Hawley, Samuel; Poblador-Plou, Beatriz; Llorente-García, Ana; Petersen, Irene; Prieto-Alhambra, Daniel

2018-01-01

Missing data are often an issue in electronic medical records (EMRs) research. However, there are many ways that people deal with missing data in drug safety studies. To compare the risk estimates resulting from different strategies for the handling of missing data in the study of venous thromboembolism (VTE) risk associated with antiosteoporotic medications (AOM). New users of AOM (alendronic acid, other bisphosphonates, strontium ranelate, selective estrogen receptor modulators, teriparatide, or denosumab) aged ≥50 years during 1998-2014 were identified in two Spanish (the Base de datos para la Investigación Farmacoepidemiológica en Atención Primaria [BIFAP] and EpiChron cohort) and one UK (Clinical Practice Research Datalink [CPRD]) EMR. Hazard ratios (HRs) according to AOM (with alendronic acid as reference) were calculated adjusting for VTE risk factors, body mass index (that was missing in 61% of patients included in the three databases), and smoking (that was missing in 23% of patients) in the year of AOM therapy initiation. HRs and standard errors obtained using cross-sectional multiple imputation (MI) (reference method) were compared to complete case (CC) analysis - using only patients with complete data - and longitudinal MI - adding to the cross-sectional MI model the body mass index/smoking values as recorded in the year before and after therapy initiation. Overall, 422/95,057 (0.4%), 19/12,688 (0.1%), and 2,051/161,202 (1.3%) VTE cases/participants were seen in BIFAP, EpiChron, and CPRD, respectively. HRs moved from 100.00% underestimation to 40.31% overestimation in CC compared with cross-sectional MI, while longitudinal MI methods provided similar risk estimates compared with cross-sectional MI. Precision for HR improved in cross-sectional MI versus CC by up to 160.28%, while longitudinal MI improved precision (compared with cross-sectional) only minimally (up to 0.80%). CC may substantially affect relative risk estimation in EMR-based drug safety studies, since missing data are not often completely at random. Little improvement was seen in these data in terms of power with the inclusion of longitudinal MI compared with cross-sectional MI. The strategy for handling missing data in drug safety studies can have a large impact on both risk estimates and precision.

A comparative study: classification vs. user-based collaborative filtering for clinical prediction.

PubMed

Hao, Fang; Blair, Rachael Hageman

2016-12-08

Recommender systems have shown tremendous value for the prediction of personalized item recommendations for individuals in a variety of settings (e.g., marketing, e-commerce, etc.). User-based collaborative filtering is a popular recommender system, which leverages an individuals' prior satisfaction with items, as well as the satisfaction of individuals that are "similar". Recently, there have been applications of collaborative filtering based recommender systems for clinical risk prediction. In these applications, individuals represent patients, and items represent clinical data, which includes an outcome. Application of recommender systems to a problem of this type requires the recasting a supervised learning problem as unsupervised. The rationale is that patients with similar clinical features carry a similar disease risk. As the "Big Data" era progresses, it is likely that approaches of this type will be reached for as biomedical data continues to grow in both size and complexity (e.g., electronic health records). In the present study, we set out to understand and assess the performance of recommender systems in a controlled yet realistic setting. User-based collaborative filtering recommender systems are compared to logistic regression and random forests with different types of imputation and varying amounts of missingness on four different publicly available medical data sets: National Health and Nutrition Examination Survey (NHANES, 2011-2012 on Obesity), Study to Understand Prognoses Preferences Outcomes and Risks of Treatment (SUPPORT), chronic kidney disease, and dermatology data. We also examined performance using simulated data with observations that are Missing At Random (MAR) or Missing Completely At Random (MCAR) under various degrees of missingness and levels of class imbalance in the response variable. Our results demonstrate that user-based collaborative filtering is consistently inferior to logistic regression and random forests with different imputations on real and simulated data. The results warrant caution for the collaborative filtering for the purpose of clinical risk prediction when traditional classification is feasible and practical. CF may not be desirable in datasets where classification is an acceptable alternative. We describe some natural applications related to "Big Data" where CF would be preferred and conclude with some insights as to why caution may be warranted in this context.

The utility of low-density genotyping for imputation in the Thoroughbred horse

PubMed Central

2014-01-01

Background Despite the dramatic reduction in the cost of high-density genotyping that has occurred over the last decade, it remains one of the limiting factors for obtaining the large datasets required for genomic studies of disease in the horse. In this study, we investigated the potential for low-density genotyping and subsequent imputation to address this problem. Results Using the haplotype phasing and imputation program, BEAGLE, it is possible to impute genotypes from low- to high-density (50K) in the Thoroughbred horse with reasonable to high accuracy. Analysis of the sources of variation in imputation accuracy revealed dependence both on the minor allele frequency of the single nucleotide polymorphisms (SNPs) being imputed and on the underlying linkage disequilibrium structure. Whereas equidistant spacing of the SNPs on the low-density panel worked well, optimising SNP selection to increase their minor allele frequency was advantageous, even when the panel was subsequently used in a population of different geographical origin. Replacing base pair position with linkage disequilibrium map distance reduced the variation in imputation accuracy across SNPs. Whereas a 1K SNP panel was generally sufficient to ensure that more than 80% of genotypes were correctly imputed, other studies suggest that a 2K to 3K panel is more efficient to minimize the subsequent loss of accuracy in genomic prediction analyses. The relationship between accuracy and genotyping costs for the different low-density panels, suggests that a 2K SNP panel would represent good value for money. Conclusions Low-density genotyping with a 2K SNP panel followed by imputation provides a compromise between cost and accuracy that could promote more widespread genotyping, and hence the use of genomic information in horses. In addition to offering a low cost alternative to high-density genotyping, imputation provides a means to combine datasets from different genotyping platforms, which is becoming necessary since researchers are starting to use the recently developed equine 70K SNP chip. However, more work is needed to evaluate the impact of between-breed differences on imputation accuracy. PMID:24495673

Multiple hot-deck imputation for network inference from RNA sequencing data.

PubMed

Imbert, Alyssa; Valsesia, Armand; Le Gall, Caroline; Armenise, Claudia; Lefebvre, Gregory; Gourraud, Pierre-Antoine; Viguerie, Nathalie; Villa-Vialaneix, Nathalie

2018-05-15

Network inference provides a global view of the relations existing between gene expression in a given transcriptomic experiment (often only for a restricted list of chosen genes). However, it is still a challenging problem: even if the cost of sequencing techniques has decreased over the last years, the number of samples in a given experiment is still (very) small compared to the number of genes. We propose a method to increase the reliability of the inference when RNA-seq expression data have been measured together with an auxiliary dataset that can provide external information on gene expression similarity between samples. Our statistical approach, hd-MI, is based on imputation for samples without available RNA-seq data that are considered as missing data but are observed on the secondary dataset. hd-MI can improve the reliability of the inference for missing rates up to 30% and provides more stable networks with a smaller number of false positive edges. On a biological point of view, hd-MI was also found relevant to infer networks from RNA-seq data acquired in adipose tissue during a nutritional intervention in obese individuals. In these networks, novel links between genes were highlighted, as well as an improved comparability between the two steps of the nutritional intervention. Software and sample data are available as an R package, RNAseqNet, that can be downloaded from the Comprehensive R Archive Network (CRAN). alyssa.imbert@inra.fr or nathalie.villa-vialaneix@inra.fr. Supplementary data are available at Bioinformatics online.

Adapt-Mix: learning local genetic correlation structure improves summary statistics-based analyses

PubMed Central

Park, Danny S.; Brown, Brielin; Eng, Celeste; Huntsman, Scott; Hu, Donglei; Torgerson, Dara G.; Burchard, Esteban G.; Zaitlen, Noah

2015-01-01

Motivation: Approaches to identifying new risk loci, training risk prediction models, imputing untyped variants and fine-mapping causal variants from summary statistics of genome-wide association studies are playing an increasingly important role in the human genetics community. Current summary statistics-based methods rely on global ‘best guess’ reference panels to model the genetic correlation structure of the dataset being studied. This approach, especially in admixed populations, has the potential to produce misleading results, ignores variation in local structure and is not feasible when appropriate reference panels are missing or small. Here, we develop a method, Adapt-Mix, that combines information across all available reference panels to produce estimates of local genetic correlation structure for summary statistics-based methods in arbitrary populations. Results: We applied Adapt-Mix to estimate the genetic correlation structure of both admixed and non-admixed individuals using simulated and real data. We evaluated our method by measuring the performance of two summary statistics-based methods: imputation and joint-testing. When using our method as opposed to the current standard of ‘best guess’ reference panels, we observed a 28% decrease in mean-squared error for imputation and a 73.7% decrease in mean-squared error for joint-testing. Availability and implementation: Our method is publicly available in a software package called ADAPT-Mix available at https://github.com/dpark27/adapt_mix. Contact: noah.zaitlen@ucsf.edu PMID:26072481

Making an unknown unknown a known unknown: Missing data in longitudinal neuroimaging studies.

PubMed

Matta, Tyler H; Flournoy, John C; Byrne, Michelle L

2017-10-28

The analysis of longitudinal neuroimaging data within the massively univariate framework provides the opportunity to study empirical questions about neurodevelopment. Missing outcome data are an all-to-common feature of any longitudinal study, a feature that, if handled improperly, can reduce statistical power and lead to biased parameter estimates. The goal of this paper is to provide conceptual clarity of the issues and non-issues that arise from analyzing incomplete data in longitudinal studies with particular focus on neuroimaging data. This paper begins with a review of the hierarchy of missing data mechanisms and their relationship to likelihood-based methods, a review that is necessary not just for likelihood-based methods, but also for multiple-imputation methods. Next, the paper provides a series of simulation studies with designs common in longitudinal neuroimaging studies to help illustrate missing data concepts regardless of interpretation. Finally, two applied examples are used to demonstrate the sensitivity of inferences under different missing data assumptions and how this may change the substantive interpretation. The paper concludes with a set of guidelines for analyzing incomplete longitudinal data that can improve the validity of research findings in developmental neuroimaging research. Copyright © 2017 The Authors. Published by Elsevier Ltd.. All rights reserved.

Detecting Anomalies from End-to-End Internet Performance Measurements (PingER) Using Cluster Based Local Outlier Factor

DOE Office of Scientific and Technical Information (OSTI.GOV)

Ali, Saqib; Wang, Guojun; Cottrell, Roger Leslie

PingER (Ping End-to-End Reporting) is a worldwide end-to-end Internet performance measurement framework. It was developed by the SLAC National Accelerator Laboratory, Stanford, USA and running from the last 20 years. It has more than 700 monitoring agents and remote sites which monitor the performance of Internet links around 170 countries of the world. At present, the size of the compressed PingER data set is about 60 GB comprising of 100,000 flat files. The data is publicly available for valuable Internet performance analyses. However, the data sets suffer from missing values and anomalies due to congestion, bottleneck links, queuing overflow, networkmore » software misconfiguration, hardware failure, cable cuts, and social upheavals. Therefore, the objective of this paper is to detect such performance drops or spikes labeled as anomalies or outliers for the PingER data set. In the proposed approach, the raw text files of the data set are transformed into a PingER dimensional model. The missing values are imputed using the k-NN algorithm. The data is partitioned into similar instances using the k-means clustering algorithm. Afterward, clustering is integrated with the Local Outlier Factor (LOF) using the Cluster Based Local Outlier Factor (CBLOF) algorithm to detect the anomalies or outliers from the PingER data. Lastly, anomalies are further analyzed to identify the time frame and location of the hosts generating the major percentage of the anomalies in the PingER data set ranging from 1998 to 2016.« less

CMIP5 based downscaled temperature over Western Himalayan region

NASA Astrophysics Data System (ADS)

Dutta, M.; Das, L.; Meher, J. K.

2016-12-01

Limited numbers of reliable temperature data is available for assessing warming over the Western Himalayan Region (WHR) of India. India meteorological Department provided many stations having more than 30% missing values. Stations having <30% missing values, were replaced using the Multiple Imputation Chained Equation (MICE) technique. Finally 16 stations having continuous records during 1969-2009 were considered as the "reference stations" for assessing the trends in addition to evaluate the Coupled Model Intercomparison, phase 5 (CMIP5) Global Circulation Model(GCMs). Station data indicates higher and rapid (1.41oC) winter warming than the other seasons and least warming was observed in the post monsoon (0.31oC) season. Mean annual warming is 0.84 oC during 1969-2009 indicating the warming over the WHR is more than double the global warming (0.85oC during 1880-2012). The performance of 34 CMIP5 models was evaluated through three different approaches namely comparison of: i) mean seasonal cycle ii) temporal trends and iii) spatial correlation and a rank was assigned to each GCM. How the better performing GCMs able to reproduce the observed spatial details were verified the ERA-interim reanalysis data. Finally station level future downscaled winter temperature has constructed using Empirical Statistical Downscaling (ESD) technique where 2 meter air temperature (T2m) is considered as predictor and station temperature as predictant. Future range of downscaled temperature change for the stations Dheradun, Manali and Gulmarg are 1.3-6.1OC, 1.1-5.8OC and 0.5-5.8OC respectively at the end of 21st century.

Detecting Anomalies from End-to-End Internet Performance Measurements (PingER) Using Cluster Based Local Outlier Factor

DOE PAGES

Ali, Saqib; Wang, Guojun; Cottrell, Roger Leslie; ...

2018-05-28

PingER (Ping End-to-End Reporting) is a worldwide end-to-end Internet performance measurement framework. It was developed by the SLAC National Accelerator Laboratory, Stanford, USA and running from the last 20 years. It has more than 700 monitoring agents and remote sites which monitor the performance of Internet links around 170 countries of the world. At present, the size of the compressed PingER data set is about 60 GB comprising of 100,000 flat files. The data is publicly available for valuable Internet performance analyses. However, the data sets suffer from missing values and anomalies due to congestion, bottleneck links, queuing overflow, networkmore » software misconfiguration, hardware failure, cable cuts, and social upheavals. Therefore, the objective of this paper is to detect such performance drops or spikes labeled as anomalies or outliers for the PingER data set. In the proposed approach, the raw text files of the data set are transformed into a PingER dimensional model. The missing values are imputed using the k-NN algorithm. The data is partitioned into similar instances using the k-means clustering algorithm. Afterward, clustering is integrated with the Local Outlier Factor (LOF) using the Cluster Based Local Outlier Factor (CBLOF) algorithm to detect the anomalies or outliers from the PingER data. Lastly, anomalies are further analyzed to identify the time frame and location of the hosts generating the major percentage of the anomalies in the PingER data set ranging from 1998 to 2016.« less

Using generalized estimating equations and extensions in randomized trials with missing longitudinal patient reported outcome data.

PubMed

Bell, Melanie L; Horton, Nicholas J; Dhillon, Haryana M; Bray, Victoria J; Vardy, Janette

2018-05-26

Patient reported outcomes (PROs) are important in oncology research; however, missing data can pose a threat to the validity of results. Psycho-oncology researchers should be aware of the statistical options for handling missing data robustly. One rarely used set of methods, which includes extensions for handling missing data, is generalized estimating equations (GEEs). Our objective was to demonstrate use of GEEs to analyze PROs with missing data in randomized trials with assessments at fixed time points. We introduce GEEs and show, with a worked example, how to use GEEs that account for missing data: inverse probability weighted GEEs and multiple imputation with GEE. We use data from an RCT evaluating a web-based brain training for cancer survivors reporting cognitive symptoms after chemotherapy treatment. The primary outcome for this demonstration is the binary outcome of cognitive impairment. Several methods are used, and results are compared. We demonstrate that estimates can vary depending on the choice of analytical approach, with odds ratios for no cognitive impairment ranging from 2.04 to 5.74. While most of these estimates were statistically significant (P < 0.05), a few were not. Researchers using PROs should use statistical methods that handle missing data in a way as to result in unbiased estimates. GEE extensions are analytic options for handling dropouts in longitudinal RCTs, particularly if the outcome is not continuous. Copyright © 2018 John Wiley & Sons, Ltd.

Daily home gardening improved survival for older people with mobility limitations: an 11-year follow-up study in Taiwan

PubMed Central

Lêng, Chhian Hūi; Wang, Jung-Der

2016-01-01

Aims To test the hypothesis that gardening is beneficial for survival after taking time-dependent comorbidities, mobility, and depression into account in a longitudinal middle-aged (50–64 years) and older (≥65 years) cohort in Taiwan. Methods The cohort contained 5,058 nationally sampled adults ≥50 years old from the Taiwan Longitudinal Study on Aging (1996–2007). Gardening was defined as growing flowers, gardening, or cultivating potted plants for pleasure with five different frequencies. We calculated hazard ratios for the mortality risks of gardening and adjusted the analysis for socioeconomic status, health behaviors and conditions, depression, mobility limitations, and comorbidities. Survival models also examined time-dependent effects and risks in each stratum contingent upon baseline mobility and depression. Sensitivity analyses used imputation methods for missing values. Results Daily home gardening was associated with a high survival rate (hazard ratio: 0.82; 95% confidence interval: 0.71–0.94). The benefits were robust for those with mobility limitations, but without depression at baseline (hazard ratio: 0.64, 95% confidence interval: 0.48–0.87) when adjusted for time-dependent comorbidities, mobility limitations, and depression. Chronic or relapsed depression weakened the protection of gardening. For those without mobility limitations and not depressed at baseline, gardening had no effect. Sensitivity analyses using different imputation methods yielded similar results and corroborated the hypothesis. Conclusion Daily gardening for pleasure was associated with reduced mortality for Taiwanese >50 years old with mobility limitations but without depression. PMID:27486315

Routine hospital management of self-harm and risk of further self-harm: propensity score analysis using record-based cohort data.

PubMed

Steeg, S; Emsley, R; Carr, M; Cooper, J; Kapur, N

2018-01-01

The care received by people presenting to hospital following self-harm varies and it is unclear how different types of treatment affect risk of further self-harm. Observational cohort data from the Manchester Self-Harm Project, UK, included 16 456 individuals presenting to an Emergency Department with self-harm between 2003 and 2011. Individuals were followed up for 12 months. We also used data from a smaller cohort of individuals presenting to 31 hospitals in England during a 3-month period in 2010/2011, followed up for 6 months. Propensity score (PS) methods were used to address observed confounding. Missing data were imputed using multiple imputation. Following PS stratification, those who received a psychosocial assessment had a lower risk of repeat hospital attendance for self-harm than those who were not assessed [RR 0.87, 95% confidence interval (CI) 0.80-0.95]. The risk was reduced most among people less likely to be assessed. Following PS matching, we found no associations between risks of repeat self-harm and admission to a medical bed, referral to outpatient psychiatry or admission to a psychiatric bed. We did not find a relationship between psychosocial assessment and repeat self-harm in the 31 centre cohort. This study shows the potential value of using novel statistical techniques in large mental health datasets to estimate treatment effects. We found that specialist psychosocial assessment may reduce the risk of repeat self-harm. This type of routine care should be provided for all individuals who present to hospital after self-harm, regardless of perceived risk.

Comparison of Three Methods Estimating Baseline Creatinine For Acute Kidney Injury in Hospitalized Patients: a Multicentre Survey in Third-Level Urban Hospitals of China.

PubMed

Lang, Xia-Bing; Yang, Yi; Yang, Ju-Rong; Wan, Jian-Xin; Yu, Sheng-Qiang; Cui, Jiong; Tang, Xiao-Jing; Chen, Jianghua

2018-01-01

A lack of baseline serum creatinine (SCr) data leads to underestimation of the burden caused by acute kidney injury (AKI) in developing countries. The goal of this study was to investigate the effects of various baseline SCr analysis methods on the current diagnosis of AKI in hospitalized patients. Patients with at least one SCr value during their hospital stay between January 1, 2011 and December 31, 2012 were retrospectively included in the study. The baseline SCr was determined either by the minimum SCr (SCrMIN) or the estimated SCr using the MDRD formula (SCrGFR-75). We also used the dynamic baseline SCr (SCrdynamic) in accordance with the 7 day/48 hour time window. AKI was defined based on the KDIGO SCr criteria. Of 562,733 hospitalized patients, 350,458 (62.3%) had at least one SCr determination, and 146,185 (26.0%) had repeat SCr tests. AKI was diagnosed in 13,883 (2.5%) patients using the SCrMIN, 21,281 (3.8%) using the SCrGFR-75 and 9,288 (1.7%) using the SCrdynamic. Compared with the non-AKI patients, AKI patients had a higher in-hospital mortality rate regardless of the baseline SCr analysis method. Because of the scarcity of SCr data, imputation of the baseline SCr is necessary to remedy the missing data. The detection rate of AKI varies depending on the different imputation methods. SCrGFR-75 can identify more AKI cases than the other two methods. © 2018 The Author(s). Published by S. Karger AG, Basel.

Latent Patient Cluster Discovery for Robust Future Forecasting and New-Patient Generalization

PubMed Central

Masino, Aaron J.

2016-01-01

Commonly referred to as predictive modeling, the use of machine learning and statistical methods to improve healthcare outcomes has recently gained traction in biomedical informatics research. Given the vast opportunities enabled by large Electronic Health Records (EHR) data and powerful resources for conducting predictive modeling, we argue that it is yet crucial to first carefully examine the prediction task and then choose predictive methods accordingly. Specifically, we argue that there are at least three distinct prediction tasks that are often conflated in biomedical research: 1) data imputation, where a model fills in the missing values in a dataset, 2) future forecasting, where a model projects the development of a medical condition for a known patient based on existing observations, and 3) new-patient generalization, where a model transfers the knowledge learned from previously observed patients to newly encountered ones. Importantly, the latter two tasks—future forecasting and new-patient generalizations—tend to be more difficult than data imputation as they require predictions to be made on potentially out-of-sample data (i.e., data following a different predictable pattern from what has been learned by the model). Using hearing loss progression as an example, we investigate three regression models and show that the modeling of latent clusters is a robust method for addressing the more challenging prediction scenarios. Overall, our findings suggest that there exist significant differences between various kinds of prediction tasks and that it is important to evaluate the merits of a predictive model relative to the specific purpose of a prediction task. PMID:27636203

Latent Patient Cluster Discovery for Robust Future Forecasting and New-Patient Generalization.

PubMed

Qian, Ting; Masino, Aaron J

2016-01-01

Commonly referred to as predictive modeling, the use of machine learning and statistical methods to improve healthcare outcomes has recently gained traction in biomedical informatics research. Given the vast opportunities enabled by large Electronic Health Records (EHR) data and powerful resources for conducting predictive modeling, we argue that it is yet crucial to first carefully examine the prediction task and then choose predictive methods accordingly. Specifically, we argue that there are at least three distinct prediction tasks that are often conflated in biomedical research: 1) data imputation, where a model fills in the missing values in a dataset, 2) future forecasting, where a model projects the development of a medical condition for a known patient based on existing observations, and 3) new-patient generalization, where a model transfers the knowledge learned from previously observed patients to newly encountered ones. Importantly, the latter two tasks-future forecasting and new-patient generalizations-tend to be more difficult than data imputation as they require predictions to be made on potentially out-of-sample data (i.e., data following a different predictable pattern from what has been learned by the model). Using hearing loss progression as an example, we investigate three regression models and show that the modeling of latent clusters is a robust method for addressing the more challenging prediction scenarios. Overall, our findings suggest that there exist significant differences between various kinds of prediction tasks and that it is important to evaluate the merits of a predictive model relative to the specific purpose of a prediction task.

Who’s not driving among U.S. high school seniors: A closer look at race/ethnicity, socioeconomic factors, and driving status

PubMed Central

Shults, Ruth A.; Banerjee, Tanima; Perry, Timothy

2017-01-01

Objectives We examined associations among race/ethnicity, socioeconomic factors, and driving status in a nationally representative sample of >26,000 U.S. high school seniors. Methods Weighted data from the 2012 and 2013 Monitoring the Future surveys were combined and analyzed. We imputed missing values using fully conditional specification multiple imputation methods. Multivariate logistic regression modeling was conducted to explore associations among race/ethnicity, socioeconomic factors, and driving status, while accounting for selected student behaviors and location. Lastly, odds ratios were converted to prevalence ratios. Results 23% of high school seniors did not drive during an average week; 14% of white students were nondrivers compared to 40% of black students. Multivariate analysis revealed that minority students were 1.8 to 2.5 times more likely to be nondrivers than their white counterparts, and students who had no earned income were 2.8 times more likely to be nondrivers than those earning an average of ≥$36 a week. Driving status also varied considerably by student academic performance, number of parents in the household, parental education, census region, and urbanicity. Conclusions Our findings suggest that resources—both financial and time—influence when or whether a teen will learn to drive. Many young people from minority or lower socioeconomic families who learn to drive may be doing so after their 18th birthday and therefore would not take advantage of the safety benefits provided by graduated driver licensing. Innovative approaches may be needed to improve safety for these young novice drivers. PMID:27064697

A Statistical Framework for Protein Quantitation in Bottom-Up MS-Based Proteomics

DOE Office of Scientific and Technical Information (OSTI.GOV)

Karpievitch, Yuliya; Stanley, Jeffrey R.; Taverner, Thomas

2009-08-15

Motivation: Quantitative mass spectrometry-based proteomics requires protein-level estimates and associated confidence measures. Challenges include the presence of low quality or incorrectly identified peptides and informative missingness. Furthermore, models are required for rolling peptide-level information up to the protein level. Results: We present a statistical model that carefully accounts for informative missingness in peak intensities and allows unbiased, model-based, protein-level estimation and inference. The model is applicable to both label-based and label-free quantitation experiments. We also provide automated, model-based, algorithms for filtering of proteins and peptides as well as imputation of missing values. Two LC/MS datasets are used to illustrate themore » methods. In simulation studies, our methods are shown to achieve substantially more discoveries than standard alternatives. Availability: The software has been made available in the opensource proteomics platform DAnTE (http://omics.pnl.gov/software/). Contact: adabney@stat.tamu.edu Supplementary information: Supplementary data are available at Bioinformatics online.« less

On the way toward systems biology of Aspergillus fumigatus infection.

PubMed

Albrecht, Daniela; Kniemeyer, Olaf; Mech, Franziska; Gunzer, Matthias; Brakhage, Axel; Guthke, Reinhard

2011-06-01

Pathogenicity of Aspergillus fumigatus is multifactorial. Thus, global studies are essential for the understanding of the infection process. Therefore, a data warehouse was established where genome sequence, transcriptome and proteome data are stored. These data are analyzed for the elucidation of virulence determinants. The data analysis workflow starts with pre-processing including imputing of missing values and normalization. Last step is the identification of differentially expressed genes/proteins as interesting candidates for further analysis, in particular for functional categorization and correlation studies. Sequence data and other prior knowledge extracted from databases are integrated to support the inference of gene regulatory networks associated with pathogenicity. This knowledge-assisted data analysis aims at establishing mathematical models with predictive strength to assist further experimental work. Recently, first steps were done to extend the integrative data analysis and computational modeling by evaluating spatio-temporal data (movies) that monitor interactions of A. fumigatus morphotypes (e.g. conidia) with host immune cells. Copyright © 2011 Elsevier GmbH. All rights reserved.

Hazard Function Estimation with Cause-of-Death Data Missing at Random.

PubMed

Wang, Qihua; Dinse, Gregg E; Liu, Chunling

2012-04-01

Hazard function estimation is an important part of survival analysis. Interest often centers on estimating the hazard function associated with a particular cause of death. We propose three nonparametric kernel estimators for the hazard function, all of which are appropriate when death times are subject to random censorship and censoring indicators can be missing at random. Specifically, we present a regression surrogate estimator, an imputation estimator, and an inverse probability weighted estimator. All three estimators are uniformly strongly consistent and asymptotically normal. We derive asymptotic representations of the mean squared error and the mean integrated squared error for these estimators and we discuss a data-driven bandwidth selection method. A simulation study, conducted to assess finite sample behavior, demonstrates that the proposed hazard estimators perform relatively well. We illustrate our methods with an analysis of some vascular disease data.

The use of imputed sibling genotypes in sibship-based association analysis: on modeling alternatives, power and model misspecification.

PubMed

Minică, Camelia C; Dolan, Conor V; Hottenga, Jouke-Jan; Willemsen, Gonneke; Vink, Jacqueline M; Boomsma, Dorret I

2013-05-01

When phenotypic, but no genotypic data are available for relatives of participants in genetic association studies, previous research has shown that family-based imputed genotypes can boost the statistical power when included in such studies. Here, using simulations, we compared the performance of two statistical approaches suitable to model imputed genotype data: the mixture approach, which involves the full distribution of the imputed genotypes and the dosage approach, where the mean of the conditional distribution features as the imputed genotype. Simulations were run by varying sibship size, size of the phenotypic correlations among siblings, imputation accuracy and minor allele frequency of the causal SNP. Furthermore, as imputing sibling data and extending the model to include sibships of size two or greater requires modeling the familial covariance matrix, we inquired whether model misspecification affects power. Finally, the results obtained via simulations were empirically verified in two datasets with continuous phenotype data (height) and with a dichotomous phenotype (smoking initiation). Across the settings considered, the mixture and the dosage approach are equally powerful and both produce unbiased parameter estimates. In addition, the likelihood-ratio test in the linear mixed model appears to be robust to the considered misspecification in the background covariance structure, given low to moderate phenotypic correlations among siblings. Empirical results show that the inclusion in association analysis of imputed sibling genotypes does not always result in larger test statistic. The actual test statistic may drop in value due to small effect sizes. That is, if the power benefit is small, that the change in distribution of the test statistic under the alternative is relatively small, the probability is greater of obtaining a smaller test statistic. As the genetic effects are typically hypothesized to be small, in practice, the decision on whether family-based imputation could be used as a means to increase power should be informed by prior power calculations and by the consideration of the background correlation.

Trends in study design and the statistical methods employed in a leading general medicine journal.

PubMed

Gosho, M; Sato, Y; Nagashima, K; Takahashi, S

2018-02-01

Study design and statistical methods have become core components of medical research, and the methodology has become more multifaceted and complicated over time. The study of the comprehensive details and current trends of study design and statistical methods is required to support the future implementation of well-planned clinical studies providing information about evidence-based medicine. Our purpose was to illustrate study design and statistical methods employed in recent medical literature. This was an extension study of Sato et al. (N Engl J Med 2017; 376: 1086-1087), which reviewed 238 articles published in 2015 in the New England Journal of Medicine (NEJM) and briefly summarized the statistical methods employed in NEJM. Using the same database, we performed a new investigation of the detailed trends in study design and individual statistical methods that were not reported in the Sato study. Due to the CONSORT statement, prespecification and justification of sample size are obligatory in planning intervention studies. Although standard survival methods (eg Kaplan-Meier estimator and Cox regression model) were most frequently applied, the Gray test and Fine-Gray proportional hazard model for considering competing risks were sometimes used for a more valid statistical inference. With respect to handling missing data, model-based methods, which are valid for missing-at-random data, were more frequently used than single imputation methods. These methods are not recommended as a primary analysis, but they have been applied in many clinical trials. Group sequential design with interim analyses was one of the standard designs, and novel design, such as adaptive dose selection and sample size re-estimation, was sometimes employed in NEJM. Model-based approaches for handling missing data should replace single imputation methods for primary analysis in the light of the information found in some publications. Use of adaptive design with interim analyses is increasing after the presentation of the FDA guidance for adaptive design. © 2017 John Wiley & Sons Ltd.

Estimates of cancer incidence, mortality and survival in aboriginal people from NSW, Australia

PubMed Central

2012-01-01

Background Aboriginal status has been unreliably and incompletely recorded in health and vital registration data collections for the most populous areas of Australia, including NSW where 29% of Australian Aboriginal people reside. This paper reports an assessment of Aboriginal status recording in NSW cancer registrations and estimates incidence, mortality and survival from cancer in NSW Aboriginal people using multiple imputation of missing Aboriginal status in NSW Central Cancer Registry (CCR) records. Methods Logistic regression modelling and multiple imputation were used to assign Aboriginal status to those records of cancer diagnosed from 1999 to 2008 with missing Aboriginality (affecting 12-18% of NSW cancers registered in this period). Estimates of incidence, mortality and survival from cancer in NSW Aboriginal people were compared with the NSW total population, as standardised incidence and mortality ratios, and with the non-Aboriginal population. Results Following imputation, 146 (12.2%) extra cancers in Aboriginal males and 140 (12.5%) in Aboriginal females were found for 1999-2007. Mean annual cancer incidence in NSW Aboriginal people was estimated to be 660 per 100,000 and 462 per 100,000, 9% and 6% higher than all NSW males and females respectively. Mean annual cancer mortality in NSW Aboriginal people was estimated to be 373 per 100,000 in males and 240 per 100,000 in females, 68% and 73% higher than for all NSW males and females respectively. Despite similar incidence of localised cancer, mortality from localised cancer in Aboriginal people is significantly higher than in non-Aboriginal people, as is mortality from cancers with regional, distant and unknown degree of spread at diagnosis. Cancer survival in Aboriginal people is significantly lower: 51% of males and 43% of females had died of the cancer by 5 years following diagnosis, compared to 36% and 33% of non-Aboriginal males and females respectively. Conclusion The present study is the first to produce valid and reliable estimates of cancer incidence, survival and mortality in Australian Aboriginal people from NSW. Despite somewhat higher cancer incidence in Aboriginal than in non-Aboriginal people, substantially higher mortality and lower survival in Aboriginal people is only partly explained by more advanced cancer at diagnosis. PMID:22559220

Evaluation of an imputed pitch velocity model of the auditory kappa effect.

PubMed

Henry, Molly J; McAuley, J Devin

2009-04-01

Three experiments evaluated an imputed pitch velocity model of the auditory kappa effect. Listeners heard 3-tone sequences and judged the timing of the middle (target) tone relative to the timing of the 1st and 3rd (bounding) tones. Experiment 1 held pitch constant but varied the time (T) interval between bounding tones (T = 728, 1,000, or 1,600 ms) in order to establish baseline performance levels for the 3 values of T. Experiments 2 and 3 combined the values of T tested in Experiment 1 with a pitch manipulation in order to create fast (8 semitones/728 ms), medium (8 semitones/1,000 ms), and slow (8 semitones/1,600 ms) velocity conditions. Consistent with an auditory motion hypothesis, distortions in perceived timing were larger for fast than for slow velocity conditions for both ascending sequences (Experiment 2) and descending sequences (Experiment 3). Overall, results supported the proposed imputed pitch velocity model of the auditory kappa effect. (c) 2009 APA, all rights reserved.

Underestimation of Variance of Predicted Health Utilities Derived from Multiattribute Utility Instruments.

PubMed

Chan, Kelvin K W; Xie, Feng; Willan, Andrew R; Pullenayegum, Eleanor M

2017-04-01

Parameter uncertainty in value sets of multiattribute utility-based instruments (MAUIs) has received little attention previously. This false precision leads to underestimation of the uncertainty of the results of cost-effectiveness analyses. The aim of this study is to examine the use of multiple imputation as a method to account for this uncertainty of MAUI scoring algorithms. We fitted a Bayesian model with random effects for respondents and health states to the data from the original US EQ-5D-3L valuation study, thereby estimating the uncertainty in the EQ-5D-3L scoring algorithm. We applied these results to EQ-5D-3L data from the Commonwealth Fund (CWF) Survey for Sick Adults ( n = 3958), comparing the standard error of the estimated mean utility in the CWF population using the predictive distribution from the Bayesian mixed-effect model (i.e., incorporating parameter uncertainty in the value set) with the standard error of the estimated mean utilities based on multiple imputation and the standard error using the conventional approach of using MAUI (i.e., ignoring uncertainty in the value set). The mean utility in the CWF population based on the predictive distribution of the Bayesian model was 0.827 with a standard error (SE) of 0.011. When utilities were derived using the conventional approach, the estimated mean utility was 0.827 with an SE of 0.003, which is only 25% of the SE based on the full predictive distribution of the mixed-effect model. Using multiple imputation with 20 imputed sets, the mean utility was 0.828 with an SE of 0.011, which is similar to the SE based on the full predictive distribution. Ignoring uncertainty of the predicted health utilities derived from MAUIs could lead to substantial underestimation of the variance of mean utilities. Multiple imputation corrects for this underestimation so that the results of cost-effectiveness analyses using MAUIs can report the correct degree of uncertainty.

Implications of the Trauma Quality Improvement Project inclusion of nonsurvivable injuries in performance benchmarking.

PubMed

Heaney, Jiselle Bock; Schroll, Rebecca; Turney, Jennifer; Stuke, Lance; Marr, Alan B; Greiffenstein, Patrick; Robledo, Rosemarie; Theriot, Amanda; Duchesne, Juan; Hunt, John

2017-10-01

The Trauma Quality Improvement Project (TQIP) uses an injury prediction model for performance benchmarking. We hypothesize that at a Level I high-volume penetrating trauma center, performance outcomes will be biased due to inclusion of patients with nonsurvivable injuries. Retrospective chart review was conducted for all patients included in the institutional TQIP analysis from 2013 to 2014 with length of stay (LOS) less than 1 day to determine survivability of the injuries. Observed (O)/expected (E) mortality ratios were calculated before and after exclusion of these patients. Completeness of data reported to TQIP was examined. Eight hundred twenty-six patients were reported to TQIP including 119 deaths. Nonsurvivable injuries accounted 90.9% of the deaths in patients with an LOS of 1 day or less. The O/E mortality ratio for all patients was 1.061, and the O/E ratio after excluding all patients with LOS less than 1 day found to have nonsurvivable injuries was 0.895. Data for key variables were missing in 63.3% of patients who died in the emergency department, 50% of those taken to the operating room and 0% of those admitted to the intensive care unit. Charts for patients who died with LOS less than 1 day were significantly more likely than those who lived to be missing crucial. This study shows TQIP inclusion of patients with nonsurvivable injuries biases outcomes at an urban trauma center. Missing data results in imputation of values, increasing inaccuracy. Further investigation is needed to determine if these findings exist at other institutions, and whether the current TQIP model needs revision to accurately identify and exclude patients with nonsurvivable injuries. Prognostic and epidemiological, level III.

Computing tools for implementing standards for single-case designs.

PubMed

Chen, Li-Ting; Peng, Chao-Ying Joanne; Chen, Ming-E

2015-11-01

In the single-case design (SCD) literature, five sets of standards have been formulated and distinguished: design standards, assessment standards, analysis standards, reporting standards, and research synthesis standards. This article reviews computing tools that can assist researchers and practitioners in meeting the analysis standards recommended by the What Works Clearinghouse: Procedures and Standards Handbook-the WWC standards. These tools consist of specialized web-based calculators or downloadable software for SCD data, and algorithms or programs written in Excel, SAS procedures, SPSS commands/Macros, or the R programming language. We aligned these tools with the WWC standards and evaluated them for accuracy and treatment of missing data, using two published data sets. All tools were tested to be accurate. When missing data were present, most tools either gave an error message or conducted analysis based on the available data. Only one program used a single imputation method. This article concludes with suggestions for an inclusive computing tool or environment, additional research on the treatment of missing data, and reasonable and flexible interpretations of the WWC standards. © The Author(s) 2015.

Statistical methods for incomplete data: Some results on model misspecification.

PubMed

McIsaac, Michael; Cook, R J

2017-02-01

Inverse probability weighted estimating equations and multiple imputation are two of the most studied frameworks for dealing with incomplete data in clinical and epidemiological research. We examine the limiting behaviour of estimators arising from inverse probability weighted estimating equations, augmented inverse probability weighted estimating equations and multiple imputation when the requisite auxiliary models are misspecified. We compute limiting values for settings involving binary responses and covariates and illustrate the effects of model misspecification using simulations based on data from a breast cancer clinical trial. We demonstrate that, even when both auxiliary models are misspecified, the asymptotic biases of double-robust augmented inverse probability weighted estimators are often smaller than the asymptotic biases of estimators arising from complete-case analyses, inverse probability weighting or multiple imputation. We further demonstrate that use of inverse probability weighting or multiple imputation with slightly misspecified auxiliary models can actually result in greater asymptotic bias than the use of naïve, complete case analyses. These asymptotic results are shown to be consistent with empirical results from simulation studies.

Moderation analysis with missing data in the predictors.

PubMed

Zhang, Qian; Wang, Lijuan

2017-12-01

The most widely used statistical model for conducting moderation analysis is the moderated multiple regression (MMR) model. In MMR modeling, missing data could pose a challenge, mainly because the interaction term is a product of two or more variables and thus is a nonlinear function of the involved variables. In this study, we consider a simple MMR model, where the effect of the focal predictor X on the outcome Y is moderated by a moderator U. The primary interest is to find ways of estimating and testing the moderation effect with the existence of missing data in X. We mainly focus on cases when X is missing completely at random (MCAR) and missing at random (MAR). Three methods are compared: (a) Normal-distribution-based maximum likelihood estimation (NML); (b) Normal-distribution-based multiple imputation (NMI); and (c) Bayesian estimation (BE). Via simulations, we found that NML and NMI could lead to biased estimates of moderation effects under MAR missingness mechanism. The BE method outperformed NMI and NML for MMR modeling with missing data in the focal predictor, missingness depending on the moderator and/or auxiliary variables, and correctly specified distributions for the focal predictor. In addition, more robust BE methods are needed in terms of the distribution mis-specification problem of the focal predictor. An empirical example was used to illustrate the applications of the methods with a simple sensitivity analysis. (PsycINFO Database Record (c) 2017 APA, all rights reserved).

An efficient method to transcription factor binding sites imputation via simultaneous completion of multiple matrices with positional consistency.

PubMed

Guo, Wei-Li; Huang, De-Shuang

2017-08-22

Transcription factors (TFs) are DNA-binding proteins that have a central role in regulating gene expression. Identification of DNA-binding sites of TFs is a key task in understanding transcriptional regulation, cellular processes and disease. Chromatin immunoprecipitation followed by high-throughput sequencing (ChIP-seq) enables genome-wide identification of in vivo TF binding sites. However, it is still difficult to map every TF in every cell line owing to cost and biological material availability, which poses an enormous obstacle for integrated analysis of gene regulation. To address this problem, we propose a novel computational approach, TFBSImpute, for predicting additional TF binding profiles by leveraging information from available ChIP-seq TF binding data. TFBSImpute fuses the dataset to a 3-mode tensor and imputes missing TF binding signals via simultaneous completion of multiple TF binding matrices with positional consistency. We show that signals predicted by our method achieve overall similarity with experimental data and that TFBSImpute significantly outperforms baseline approaches, by assessing the performance of imputation methods against observed ChIP-seq TF binding profiles. Besides, motif analysis shows that TFBSImpute preforms better in capturing binding motifs enriched in observed data compared with baselines, indicating that the higher performance of TFBSImpute is not simply due to averaging related samples. We anticipate that our approach will constitute a useful complement to experimental mapping of TF binding, which is beneficial for further study of regulation mechanisms and disease.

Evaluation of methods to estimate missing days' supply within pharmacy data of the Clinical Practice Research Datalink (CPRD) and The Health Improvement Network (THIN).

PubMed

Lum, Kirsten J; Newcomb, Craig W; Roy, Jason A; Carbonari, Dena M; Saine, M Elle; Cardillo, Serena; Bhullar, Harshvinder; Gallagher, Arlene M; Lo Re, Vincent

2017-01-01

The extent to which days' supply data are missing in pharmacoepidemiologic databases and effective methods for estimation is unknown. We determined the percentage of missing days' supply on prescription and patient levels for oral anti-diabetic drugs (OADs) and evaluated three methods for estimating days' supply within the Clinical Practice Research Datalink (CPRD) and The Health Improvement Network (THIN). We estimated the percentage of OAD prescriptions and patients with missing days' supply in each database from 2009 to 2013. Within a random sample of prescriptions with known days' supply, we measured the accuracy of three methods to estimate missing days' supply by imputing the following: (1) 28 days' supply, (2) mode number of tablets/day by drug strength and number of tablets/prescription, and (3) number of tablets/day via a machine learning algorithm. We determined incidence rates (IRs) of acute myocardial infarction (AMI) using each method to evaluate the impact on ascertainment of exposure time and outcomes. Days' supply was missing for 24 % of OAD prescriptions in CPRD and 33 % in THIN (affecting 48 and 57 % of patients, respectively). Methods 2 and 3 were very accurate in estimating days' supply for OADs prescribed at a consistent number of tablets/day. Method 3 was more accurate for OADs prescribed at varying number of tablets/day. IRs of AMI were similar across methods for most OADs. Missing days' supply is a substantial problem in both databases. Method 2 is easy and very accurate for most OADs and results in IRs comparable to those from method 3.

The African Genome Variation Project shapes medical genetics in Africa

NASA Astrophysics Data System (ADS)

Gurdasani, Deepti; Carstensen, Tommy; Tekola-Ayele, Fasil; Pagani, Luca; Tachmazidou, Ioanna; Hatzikotoulas, Konstantinos; Karthikeyan, Savita; Iles, Louise; Pollard, Martin O.; Choudhury, Ananyo; Ritchie, Graham R. S.; Xue, Yali; Asimit, Jennifer; Nsubuga, Rebecca N.; Young, Elizabeth H.; Pomilla, Cristina; Kivinen, Katja; Rockett, Kirk; Kamali, Anatoli; Doumatey, Ayo P.; Asiki, Gershim; Seeley, Janet; Sisay-Joof, Fatoumatta; Jallow, Muminatou; Tollman, Stephen; Mekonnen, Ephrem; Ekong, Rosemary; Oljira, Tamiru; Bradman, Neil; Bojang, Kalifa; Ramsay, Michele; Adeyemo, Adebowale; Bekele, Endashaw; Motala, Ayesha; Norris, Shane A.; Pirie, Fraser; Kaleebu, Pontiano; Kwiatkowski, Dominic; Tyler-Smith, Chris; Rotimi, Charles; Zeggini, Eleftheria; Sandhu, Manjinder S.

2015-01-01

Given the importance of Africa to studies of human origins and disease susceptibility, detailed characterization of African genetic diversity is needed. The African Genome Variation Project provides a resource with which to design, implement and interpret genomic studies in sub-Saharan Africa and worldwide. The African Genome Variation Project represents dense genotypes from 1,481 individuals and whole-genome sequences from 320 individuals across sub-Saharan Africa. Using this resource, we find novel evidence of complex, regionally distinct hunter-gatherer and Eurasian admixture across sub-Saharan Africa. We identify new loci under selection, including loci related to malaria susceptibility and hypertension. We show that modern imputation panels (sets of reference genotypes from which unobserved or missing genotypes in study sets can be inferred) can identify association signals at highly differentiated loci across populations in sub-Saharan Africa. Using whole-genome sequencing, we demonstrate further improvements in imputation accuracy, strengthening the case for large-scale sequencing efforts of diverse African haplotypes. Finally, we present an efficient genotype array design capturing common genetic variation in Africa.

Family structure as a predictor of screen time among youth.

PubMed

McMillan, Rachel; McIsaac, Michael; Janssen, Ian

2015-01-01

The family plays a central role in the development of health-related behaviors among youth. The objective of this study was to determine whether non-traditional parental structure and shared custody arrangements predict how much time youth spend watching television, using a computer recreationally, and playing video games. Participants were a nationally representative sample of Canadian youth (N = 26,068) in grades 6-10 who participated in the 2009/10 Health Behaviour in School-aged Children Survey. Screen time in youth from single parent and reconstituted families, with or without regular visitation with their non-residential parent, was compared to that of youth from traditional dual-parent families. Multiple imputation was used to account for missing data. After multiple imputation, the relative odds of being in the highest television, computer use, video game, and total screen time quartiles were not different in boys and girls from non-traditional families by comparison to boys and girls from traditional dual-parent families. In conclusion, parental structure and child custody arrangements did not have a meaningful impact on screen time among youth.

Loneliness in senior housing communities.

PubMed

Taylor, Harry Owen; Wang, Yi; Morrow-Howell, Nancy

2018-05-23

There are many studies on loneliness among community-dwelling older adults; however, there is limited research examining the extent and correlates of loneliness among older adults who reside in senior housing communities. This study examines the extent and correlates of loneliness in three public senior housing communities in the St. Louis area. Data for this project was collected with survey questionnaires with a total sample size of 148 respondents. Loneliness was measured using the Hughes 3-item loneliness scale. Additionally, the questionnaire contained measures on socio-demographics, health/mental health, social engagement, and social support. Missing data for the hierarchical multivariate regression models were imputed using multiple imputation methods. Results showed approximately 30.8% of the sample was not lonely, 42.7% was moderately lonely, and 26.6% was severely lonely. In the multivariate analyses, loneliness was primarily associated with depressive symptoms. Contrary to popular opinion, our study found the prevalence of loneliness was high in senior housing communities. Nevertheless, senior housing communities could be ideal locations for reducing loneliness among older adults. Interventions should focus on concomitantly addressing both an individual's loneliness and mental health.

Multiple imputation to evaluate the impact of an assay change in national surveys

PubMed Central

Sternberg, Maya

2017-01-01

National health surveys, such as the National Health and Nutrition Examination Survey, are used to monitor trends of nutritional biomarkers. These surveys try to maintain the same biomarker assay over time, but there are a variety of reasons why the assay may change. In these cases, it is important to evaluate the potential impact of a change so that any observed fluctuations in concentrations over time are not confounded by changes in the assay. To this end, a subset of stored specimens previously analyzed with the old assay is retested using the new assay. These paired data are used to estimate an adjustment equation, which is then used to ‘adjust’ all the old assay results and convert them into ‘equivalent’ units of the new assay. In this paper, we present a new way of approaching this problem using modern statistical methods designed for missing data. Using simulations, we compare the proposed multiple imputation approach with the adjustment equation approach currently in use. We also compare these approaches using real National Health and Nutrition Examination Survey data for 25-hydroxyvitamin D. PMID:28419523

The African Genome Variation Project shapes medical genetics in Africa.

PubMed

Gurdasani, Deepti; Carstensen, Tommy; Tekola-Ayele, Fasil; Pagani, Luca; Tachmazidou, Ioanna; Hatzikotoulas, Konstantinos; Karthikeyan, Savita; Iles, Louise; Pollard, Martin O; Choudhury, Ananyo; Ritchie, Graham R S; Xue, Yali; Asimit, Jennifer; Nsubuga, Rebecca N; Young, Elizabeth H; Pomilla, Cristina; Kivinen, Katja; Rockett, Kirk; Kamali, Anatoli; Doumatey, Ayo P; Asiki, Gershim; Seeley, Janet; Sisay-Joof, Fatoumatta; Jallow, Muminatou; Tollman, Stephen; Mekonnen, Ephrem; Ekong, Rosemary; Oljira, Tamiru; Bradman, Neil; Bojang, Kalifa; Ramsay, Michele; Adeyemo, Adebowale; Bekele, Endashaw; Motala, Ayesha; Norris, Shane A; Pirie, Fraser; Kaleebu, Pontiano; Kwiatkowski, Dominic; Tyler-Smith, Chris; Rotimi, Charles; Zeggini, Eleftheria; Sandhu, Manjinder S

2015-01-15

Given the importance of Africa to studies of human origins and disease susceptibility, detailed characterization of African genetic diversity is needed. The African Genome Variation Project provides a resource with which to design, implement and interpret genomic studies in sub-Saharan Africa and worldwide. The African Genome Variation Project represents dense genotypes from 1,481 individuals and whole-genome sequences from 320 individuals across sub-Saharan Africa. Using this resource, we find novel evidence of complex, regionally distinct hunter-gatherer and Eurasian admixture across sub-Saharan Africa. We identify new loci under selection, including loci related to malaria susceptibility and hypertension. We show that modern imputation panels (sets of reference genotypes from which unobserved or missing genotypes in study sets can be inferred) can identify association signals at highly differentiated loci across populations in sub-Saharan Africa. Using whole-genome sequencing, we demonstrate further improvements in imputation accuracy, strengthening the case for large-scale sequencing efforts of diverse African haplotypes. Finally, we present an efficient genotype array design capturing common genetic variation in Africa.

Very low-depth sequencing in a founder population identifies a cardioprotective APOC3 signal missed by genome-wide imputation.

PubMed

Gilly, Arthur; Ritchie, Graham Rs; Southam, Lorraine; Farmaki, Aliki-Eleni; Tsafantakis, Emmanouil; Dedoussis, George; Zeggini, Eleftheria

2016-06-01

Cohort-wide very low-depth whole-genome sequencing (WGS) can comprehensively capture low-frequency sequence variation for the cost of a dense genome-wide genotyping array. Here, we analyse 1x sequence data across the APOC3 gene in a founder population from the island of Crete in Greece (n = 1239) and find significant evidence for association with blood triglyceride levels with the previously reported R19X cardioprotective null mutation (β = -1.09,σ = 0.163, P = 8.2 × 10 -11 ) and a second loss of function mutation, rs138326449 (β = -1.17,σ = 0.188, P = 1.14 × 10 -9 ). The signal cannot be recapitulated by imputing genome-wide genotype data on a large reference panel of 5122 individuals including 249 with 4x WGS data from the same population. Gene-level meta-analysis with other studies reporting burden signals at APOC3 provides robust evidence for a replicable cardioprotective rare variant aggregation (P = 3.2 × 10 -31 , n = 13 480). © The Author 2016. Published by Oxford University Press.

Very low-depth sequencing in a founder population identifies a cardioprotective APOC3 signal missed by genome-wide imputation

PubMed Central

Gilly, Arthur; Ritchie, Graham Rs; Southam, Lorraine; Farmaki, Aliki-Eleni; Tsafantakis, Emmanouil; Dedoussis, George; Zeggini, Eleftheria

2016-01-01

Cohort-wide very low-depth whole-genome sequencing (WGS) can comprehensively capture low-frequency sequence variation for the cost of a dense genome-wide genotyping array. Here, we analyse 1x sequence data across the APOC3 gene in a founder population from the island of Crete in Greece (n = 1239) and find significant evidence for association with blood triglyceride levels with the previously reported R19X cardioprotective null mutation (β = −1.09,σ = 0.163, P = 8.2 × 10−11) and a second loss of function mutation, rs138326449 (β = −1.17,σ = 0.188, P = 1.14 × 10−9). The signal cannot be recapitulated by imputing genome-wide genotype data on a large reference panel of 5122 individuals including 249 with 4x WGS data from the same population. Gene-level meta-analysis with other studies reporting burden signals at APOC3 provides robust evidence for a replicable cardioprotective rare variant aggregation (P = 3.2 × 10−31, n = 13 480). PMID:27146844

Hazard Function Estimation with Cause-of-Death Data Missing at Random

PubMed Central

Wang, Qihua; Dinse, Gregg E.; Liu, Chunling

2010-01-01

Hazard function estimation is an important part of survival analysis. Interest often centers on estimating the hazard function associated with a particular cause of death. We propose three nonparametric kernel estimators for the hazard function, all of which are appropriate when death times are subject to random censorship and censoring indicators can be missing at random. Specifically, we present a regression surrogate estimator, an imputation estimator, and an inverse probability weighted estimator. All three estimators are uniformly strongly consistent and asymptotically normal. We derive asymptotic representations of the mean squared error and the mean integrated squared error for these estimators and we discuss a data-driven bandwidth selection method. A simulation study, conducted to assess finite sample behavior, demonstrates that the proposed hazard estimators perform relatively well. We illustrate our methods with an analysis of some vascular disease data. PMID:22267874

Missing data in trauma registries: A systematic review.

PubMed

Shivasabesan, Gowri; Mitra, Biswadev; O'Reilly, Gerard M

2018-03-30

Trauma registries play an integral role in trauma systems but their valid use hinges on data quality. The aim of this study was to determine, among contemporary publications using trauma registry data, the level of reporting of data completeness and the methods used to deal with missing data. A systematic review was conducted of all trauma registry-based manuscripts published from 01 January 2015 to current date (17 March 2017). Studies were identified by searching MEDLINE, EMBASE, and CINAHL using relevant subject headings and keywords. Included manuscripts were evaluated based on previously published recommendations regarding the reporting and discussion of missing data. Manuscripts were graded on their degree of characterization of such observations. In addition, the methods used to manage missing data were examined. There were 539 manuscripts that met inclusion criteria. Among these, 208 (38.6%) manuscripts did not mention data completeness and 88 (16.3%) mentioned missing data but did not quantify the extent. Only a handful (n = 26; 4.8%) quantified the 'missingness' of all variables. Most articles (n = 477; 88.5%) contained no details such as a comparison between patient characteristics in cohorts with and without missing data. Of the 331 articles which made at least some mention of data completeness, the method of managing missing data was unknown in 34 (10.3%). When method(s) to handle missing data were identified, 234 (78.8%) manuscripts used complete case analysis only, 18 (6.1%) used multiple imputation only and 34 (11.4%) used a combination of these. Most manuscripts using trauma registry data did not quantify the extent of missing data for any variables and contained minimal discussion regarding missingness. Out of the studies which identified a method of managing missing data, most used complete case analysis, a method that may bias results. The lack of standardization in the reporting and management of missing data questions the validity of conclusions from research based on trauma registry data. Copyright © 2018 Elsevier Ltd. All rights reserved.

Effects of breast stimulation for spontaneous onset of labor on salivary oxytocin levels in low-risk pregnant women: A feasibility study

PubMed Central

Tadokoro, Yuriko; Shuo, Takuya; Sawano, Erika; Shinohara, Kazuyuki

2018-01-01

Objectives This preliminary study aimed to 1) determine changes in the salivary oxytocin (OT) level during breast stimulation for promoting the spontaneous onset of labor in low-risk term pregnancies, and 2) clarify the feasibility of the breast stimulation intervention protocol in terms of practicality and acceptability. Methods We used a single arm trial design. Sixteen low-risk pregnant women between 38 and 40 weeks of gestation with cephalic presentation participated. They performed breast stimulation for 3 days with an attendant midwife in a single maternity hospital. Each breast was stimulated for 15 minutes for a total of 1 hour per day. Saliva was collected 10 minutes before the intervention and 15, 30, 60, 75, and 90 minutes after the intervention, yielding 18 samples per woman. Results Among a total of 282 saliva samples from the 16 participants, OT level was measured in 142 samples (missing rate: 49.6%). The median OT level showed the highest values on day 3 of the breast stimulation, with a marked increase 30 min after the intervention. In the mixed models after multiple imputation for missing data, the OT level on the first day of intervention was significantly lower than that on the third day of intervention. Fatigue from breast stimulation decreased on subsequent days, and most of the women (75%) felt no discomfort with the protocol. Uterine hyperstimulation was not observed. Conclusion Following a 3-day breast stimulation protocol for spontaneous onset of labor, the mean OT level showed the highest values on day 3. The breast stimulation intervention protocol showed good feasibility in terms of practicality and acceptability among the pregnant women. Additional large-scale studies are warranted to confirm the protocol’s effectiveness. PMID:29447299

Effects of breast stimulation for spontaneous onset of labor on salivary oxytocin levels in low-risk pregnant women: A feasibility study.

PubMed

Takahata, Kaori; Horiuchi, Shigeko; Tadokoro, Yuriko; Shuo, Takuya; Sawano, Erika; Shinohara, Kazuyuki

2018-01-01

This preliminary study aimed to 1) determine changes in the salivary oxytocin (OT) level during breast stimulation for promoting the spontaneous onset of labor in low-risk term pregnancies, and 2) clarify the feasibility of the breast stimulation intervention protocol in terms of practicality and acceptability. We used a single arm trial design. Sixteen low-risk pregnant women between 38 and 40 weeks of gestation with cephalic presentation participated. They performed breast stimulation for 3 days with an attendant midwife in a single maternity hospital. Each breast was stimulated for 15 minutes for a total of 1 hour per day. Saliva was collected 10 minutes before the intervention and 15, 30, 60, 75, and 90 minutes after the intervention, yielding 18 samples per woman. Among a total of 282 saliva samples from the 16 participants, OT level was measured in 142 samples (missing rate: 49.6%). The median OT level showed the highest values on day 3 of the breast stimulation, with a marked increase 30 min after the intervention. In the mixed models after multiple imputation for missing data, the OT level on the first day of intervention was significantly lower than that on the third day of intervention. Fatigue from breast stimulation decreased on subsequent days, and most of the women (75%) felt no discomfort with the protocol. Uterine hyperstimulation was not observed. Following a 3-day breast stimulation protocol for spontaneous onset of labor, the mean OT level showed the highest values on day 3. The breast stimulation intervention protocol showed good feasibility in terms of practicality and acceptability among the pregnant women. Additional large-scale studies are warranted to confirm the protocol's effectiveness.

The Association Between Electronic Bullying and School Absenteeism Among High School Students in the United States.

PubMed

Grinshteyn, Erin; Yang, Y T

2017-02-01

We examined the relationship between exposure to electronic bullying and absenteeism as a result of being afraid. This multivariate, multinomial regression analysis of the 2013 Youth Risk Behavior Survey data assessed the association between experiencing electronic bullying in the past year and how often students were absent in the last month due to feeling unsafe at/in transit to school. The model controlled for other predictors of school absence including demographics, physical/behavioral health, and risk factors. Missing data were multiply imputed. Electronic bullying was significantly associated with absences. Controlling for model covariates, the relative risk of missing 1 day of school was 1.77 times higher, the relative risk of missing 2 to 3 days of school per month increased by a factor of 2.08, and the relative risk of missing 4 or more days of school per month increased by a factor of 1.77 for those who experienced electronic bullying in the past year compared with those who were not electronically bullied. Electronic bullying's association with absenteeism places it among already recognized negative influences such as depression and binge drinking, necessitating schools to implement policies to mediate the resulting harmful effects. © 2017, American School Health Association.

Identification and characterization of novel associations in the CASP8/ALS2CR12 region on chromosome 2 with breast cancer risk

PubMed Central

Lin, Wei-Yu; Camp, Nicola J.; Ghoussaini, Maya; Beesley, Jonathan; Michailidou, Kyriaki; Hopper, John L.; Apicella, Carmel; Southey, Melissa C.; Stone, Jennifer; Schmidt, Marjanka K.; Broeks, Annegien; Van't Veer, Laura J.; Th Rutgers, Emiel J.; Muir, Kenneth; Lophatananon, Artitaya; Stewart-Brown, Sarah; Siriwanarangsan, Pornthep; Fasching, Peter A.; Haeberle, Lothar; Ekici, Arif B.; Beckmann, Matthias W.; Peto, Julian; Dos-Santos-Silva, Isabel; Fletcher, Olivia; Johnson, Nichola; Bolla, Manjeet K.; Wang, Qin; Dennis, Joe; Sawyer, Elinor J.; Cheng, Timothy; Tomlinson, Ian; Kerin, Michael J.; Miller, Nicola; Marmé, Frederik; Surowy, Harald M.; Burwinkel, Barbara; Guénel, Pascal; Truong, Thérèse; Menegaux, Florence; Mulot, Claire; Bojesen, Stig E.; Nordestgaard, Børge G.; Nielsen, Sune F.; Flyger, Henrik; Benitez, Javier; Zamora, M. Pilar; Arias Perez, Jose Ignacio; Menéndez, Primitiva; González-Neira, Anna; Pita, Guillermo; Alonso, M. Rosario; Álvarez, Nuria; Herrero, Daniel; Anton-Culver, Hoda; Brenner, Hermann; Dieffenbach, Aida Karina; Arndt, Volker; Stegmaier, Christa; Meindl, Alfons; Lichtner, Peter; Schmutzler, Rita K.; Müller-Myhsok, Bertram; Brauch, Hiltrud; Brüning, Thomas; Ko, Yon-Dschun; Tessier, Daniel C.; Vincent, Daniel; Bacot, Francois; Nevanlinna, Heli; Aittomäki, Kristiina; Blomqvist, Carl; Khan, Sofia; Matsuo, Keitaro; Ito, Hidemi; Iwata, Hiroji; Horio, Akiyo; Bogdanova, Natalia V.; Antonenkova, Natalia N.; Dörk, Thilo; Lindblom, Annika; Margolin, Sara; Mannermaa, Arto; Kataja, Vesa; Kosma, Veli-Matti; Hartikainen, Jaana M.; Wu, Anna H.; Tseng, Chiu-Chen; Van Den Berg, David; Stram, Daniel O.; Neven, Patrick; Wauters, Els; Wildiers, Hans; Lambrechts, Diether; Chang-Claude, Jenny; Rudolph, Anja; Seibold, Petra; Flesch-Janys, Dieter; Radice, Paolo; Peterlongo, Paolo; Manoukian, Siranoush; Bonanni, Bernardo; Couch, Fergus J.; Wang, Xianshu; Vachon, Celine; Purrington, Kristen; Giles, Graham G.; Milne, Roger L.; Mclean, Catriona; Haiman, Christopher A.; Henderson, Brian E.; Schumacher, Fredrick; Le Marchand, Loic; Simard, Jacques; Goldberg, Mark S.; Labrèche, France; Dumont, Martine; Teo, Soo Hwang; Yip, Cheng Har; Hassan, Norhashimah; Vithana, Eranga Nishanthie; Kristensen, Vessela; Zheng, Wei; Deming-Halverson, Sandra; Shrubsole, Martha J.; Long, Jirong; Winqvist, Robert; Pylkäs, Katri; Jukkola-Vuorinen, Arja; Kauppila, Saila; Andrulis, Irene L.; Knight, Julia A.; Glendon, Gord; Tchatchou, Sandrine; Devilee, Peter; Tollenaar, Robert A.E.M.; Seynaeve, Caroline; Van Asperen, Christi J.; García-Closas, Montserrat; Figueroa, Jonine; Lissowska, Jolanta; Brinton, Louise; Czene, Kamila; Darabi, Hatef; Eriksson, Mikael; Brand, Judith S.; Hooning, Maartje J.; Hollestelle, Antoinette; Van Den Ouweland, Ans M.W.; Jager, Agnes; Li, Jingmei; Liu, Jianjun; Humphreys, Keith; Shu, Xiao-Ou; Lu, Wei; Gao, Yu-Tang; Cai, Hui; Cross, Simon S.; Reed, Malcolm W. R.; Blot, William; Signorello, Lisa B.; Cai, Qiuyin; Pharoah, Paul D.P.; Perkins, Barbara; Shah, Mitul; Blows, Fiona M.; Kang, Daehee; Yoo, Keun-Young; Noh, Dong-Young; Hartman, Mikael; Miao, Hui; Chia, Kee Seng; Putti, Thomas Choudary; Hamann, Ute; Luccarini, Craig; Baynes, Caroline; Ahmed, Shahana; Maranian, Mel; Healey, Catherine S.; Jakubowska, Anna; Lubinski, Jan; Jaworska-Bieniek, Katarzyna; Durda, Katarzyna; Sangrajrang, Suleeporn; Gaborieau, Valerie; Brennan, Paul; Mckay, James; Slager, Susan; Toland, Amanda E.; Yannoukakos, Drakoulis; Shen, Chen-Yang; Hsiung, Chia-Ni; Wu, Pei-Ei; Ding, Shian-ling; Ashworth, Alan; Jones, Michael; Orr, Nick; Swerdlow, Anthony J; Tsimiklis, Helen; Makalic, Enes; Schmidt, Daniel F.; Bui, Quang M.; Chanock, Stephen J.; Hunter, David J.; Hein, Rebecca; Dahmen, Norbert; Beckmann, Lars; Aaltonen, Kirsimari; Muranen, Taru A.; Heikkinen, Tuomas; Irwanto, Astrid; Rahman, Nazneen; Turnbull, Clare A.; Waisfisz, Quinten; Meijers-Heijboer, Hanne E. J.; Adank, Muriel A.; Van Der Luijt, Rob B.; Hall, Per; Chenevix-Trench, Georgia; Dunning, Alison; Easton, Douglas F.; Cox, Angela

2015-01-01

Previous studies have suggested that polymorphisms in CASP8 on chromosome 2 are associated with breast cancer risk. To clarify the role of CASP8 in breast cancer susceptibility, we carried out dense genotyping of this region in the Breast Cancer Association Consortium (BCAC). Single-nucleotide polymorphisms (SNPs) spanning a 1 Mb region around CASP8 were genotyped in 46 450 breast cancer cases and 42 600 controls of European origin from 41 studies participating in the BCAC as part of a custom genotyping array experiment (iCOGS). Missing genotypes and SNPs were imputed and, after quality exclusions, 501 typed and 1232 imputed SNPs were included in logistic regression models adjusting for study and ancestry principal components. The SNPs retained in the final model were investigated further in data from nine genome-wide association studies (GWAS) comprising in total 10 052 case and 12 575 control subjects. The most significant association signal observed in European subjects was for the imputed intronic SNP rs1830298 in ALS2CR12 (telomeric to CASP8), with per allele odds ratio and 95% confidence interval [OR (95% confidence interval, CI)] for the minor allele of 1.05 (1.03–1.07), P = 1 × 10−5. Three additional independent signals from intronic SNPs were identified, in CASP8 (rs36043647), ALS2CR11 (rs59278883) and CFLAR (rs7558475). The association with rs1830298 was replicated in the imputed results from the combined GWAS (P = 3 × 10−6), yielding a combined OR (95% CI) of 1.06 (1.04–1.08), P = 1 × 10−9. Analyses of gene expression associations in peripheral blood and normal breast tissue indicate that CASP8 might be the target gene, suggesting a mechanism involving apoptosis. PMID:25168388

Identification and characterization of novel associations in the CASP8/ALS2CR12 region on chromosome 2 with breast cancer risk.

PubMed

Lin, Wei-Yu; Camp, Nicola J; Ghoussaini, Maya; Beesley, Jonathan; Michailidou, Kyriaki; Hopper, John L; Apicella, Carmel; Southey, Melissa C; Stone, Jennifer; Schmidt, Marjanka K; Broeks, Annegien; Van't Veer, Laura J; Th Rutgers, Emiel J; Muir, Kenneth; Lophatananon, Artitaya; Stewart-Brown, Sarah; Siriwanarangsan, Pornthep; Fasching, Peter A; Haeberle, Lothar; Ekici, Arif B; Beckmann, Matthias W; Peto, Julian; Dos-Santos-Silva, Isabel; Fletcher, Olivia; Johnson, Nichola; Bolla, Manjeet K; Wang, Qin; Dennis, Joe; Sawyer, Elinor J; Cheng, Timothy; Tomlinson, Ian; Kerin, Michael J; Miller, Nicola; Marmé, Frederik; Surowy, Harald M; Burwinkel, Barbara; Guénel, Pascal; Truong, Thérèse; Menegaux, Florence; Mulot, Claire; Bojesen, Stig E; Nordestgaard, Børge G; Nielsen, Sune F; Flyger, Henrik; Benitez, Javier; Zamora, M Pilar; Arias Perez, Jose Ignacio; Menéndez, Primitiva; González-Neira, Anna; Pita, Guillermo; Alonso, M Rosario; Alvarez, Nuria; Herrero, Daniel; Anton-Culver, Hoda; Brenner, Hermann; Dieffenbach, Aida Karina; Arndt, Volker; Stegmaier, Christa; Meindl, Alfons; Lichtner, Peter; Schmutzler, Rita K; Müller-Myhsok, Bertram; Brauch, Hiltrud; Brüning, Thomas; Ko, Yon-Dschun; Tessier, Daniel C; Vincent, Daniel; Bacot, Francois; Nevanlinna, Heli; Aittomäki, Kristiina; Blomqvist, Carl; Khan, Sofia; Matsuo, Keitaro; Ito, Hidemi; Iwata, Hiroji; Horio, Akiyo; Bogdanova, Natalia V; Antonenkova, Natalia N; Dörk, Thilo; Lindblom, Annika; Margolin, Sara; Mannermaa, Arto; Kataja, Vesa; Kosma, Veli-Matti; Hartikainen, Jaana M; Wu, Anna H; Tseng, Chiu-Chen; Van Den Berg, David; Stram, Daniel O; Neven, Patrick; Wauters, Els; Wildiers, Hans; Lambrechts, Diether; Chang-Claude, Jenny; Rudolph, Anja; Seibold, Petra; Flesch-Janys, Dieter; Radice, Paolo; Peterlongo, Paolo; Manoukian, Siranoush; Bonanni, Bernardo; Couch, Fergus J; Wang, Xianshu; Vachon, Celine; Purrington, Kristen; Giles, Graham G; Milne, Roger L; Mclean, Catriona; Haiman, Christopher A; Henderson, Brian E; Schumacher, Fredrick; Le Marchand, Loic; Simard, Jacques; Goldberg, Mark S; Labrèche, France; Dumont, Martine; Teo, Soo Hwang; Yip, Cheng Har; Hassan, Norhashimah; Vithana, Eranga Nishanthie; Kristensen, Vessela; Zheng, Wei; Deming-Halverson, Sandra; Shrubsole, Martha J; Long, Jirong; Winqvist, Robert; Pylkäs, Katri; Jukkola-Vuorinen, Arja; Kauppila, Saila; Andrulis, Irene L; Knight, Julia A; Glendon, Gord; Tchatchou, Sandrine; Devilee, Peter; Tollenaar, Robert A E M; Seynaeve, Caroline; Van Asperen, Christi J; García-Closas, Montserrat; Figueroa, Jonine; Lissowska, Jolanta; Brinton, Louise; Czene, Kamila; Darabi, Hatef; Eriksson, Mikael; Brand, Judith S; Hooning, Maartje J; Hollestelle, Antoinette; Van Den Ouweland, Ans M W; Jager, Agnes; Li, Jingmei; Liu, Jianjun; Humphreys, Keith; Shu, Xiao-Ou; Lu, Wei; Gao, Yu-Tang; Cai, Hui; Cross, Simon S; Reed, Malcolm W R; Blot, William; Signorello, Lisa B; Cai, Qiuyin; Pharoah, Paul D P; Perkins, Barbara; Shah, Mitul; Blows, Fiona M; Kang, Daehee; Yoo, Keun-Young; Noh, Dong-Young; Hartman, Mikael; Miao, Hui; Chia, Kee Seng; Putti, Thomas Choudary; Hamann, Ute; Luccarini, Craig; Baynes, Caroline; Ahmed, Shahana; Maranian, Mel; Healey, Catherine S; Jakubowska, Anna; Lubinski, Jan; Jaworska-Bieniek, Katarzyna; Durda, Katarzyna; Sangrajrang, Suleeporn; Gaborieau, Valerie; Brennan, Paul; Mckay, James; Slager, Susan; Toland, Amanda E; Yannoukakos, Drakoulis; Shen, Chen-Yang; Hsiung, Chia-Ni; Wu, Pei-Ei; Ding, Shian-Ling; Ashworth, Alan; Jones, Michael; Orr, Nick; Swerdlow, Anthony J; Tsimiklis, Helen; Makalic, Enes; Schmidt, Daniel F; Bui, Quang M; Chanock, Stephen J; Hunter, David J; Hein, Rebecca; Dahmen, Norbert; Beckmann, Lars; Aaltonen, Kirsimari; Muranen, Taru A; Heikkinen, Tuomas; Irwanto, Astrid; Rahman, Nazneen; Turnbull, Clare A; Waisfisz, Quinten; Meijers-Heijboer, Hanne E J; Adank, Muriel A; Van Der Luijt, Rob B; Hall, Per; Chenevix-Trench, Georgia; Dunning, Alison; Easton, Douglas F; Cox, Angela

2015-01-01

Previous studies have suggested that polymorphisms in CASP8 on chromosome 2 are associated with breast cancer risk. To clarify the role of CASP8 in breast cancer susceptibility, we carried out dense genotyping of this region in the Breast Cancer Association Consortium (BCAC). Single-nucleotide polymorphisms (SNPs) spanning a 1 Mb region around CASP8 were genotyped in 46 450 breast cancer cases and 42 600 controls of European origin from 41 studies participating in the BCAC as part of a custom genotyping array experiment (iCOGS). Missing genotypes and SNPs were imputed and, after quality exclusions, 501 typed and 1232 imputed SNPs were included in logistic regression models adjusting for study and ancestry principal components. The SNPs retained in the final model were investigated further in data from nine genome-wide association studies (GWAS) comprising in total 10 052 case and 12 575 control subjects. The most significant association signal observed in European subjects was for the imputed intronic SNP rs1830298 in ALS2CR12 (telomeric to CASP8), with per allele odds ratio and 95% confidence interval [OR (95% confidence interval, CI)] for the minor allele of 1.05 (1.03-1.07), P = 1 × 10(-5). Three additional independent signals from intronic SNPs were identified, in CASP8 (rs36043647), ALS2CR11 (rs59278883) and CFLAR (rs7558475). The association with rs1830298 was replicated in the imputed results from the combined GWAS (P = 3 × 10(-6)), yielding a combined OR (95% CI) of 1.06 (1.04-1.08), P = 1 × 10(-9). Analyses of gene expression associations in peripheral blood and normal breast tissue indicate that CASP8 might be the target gene, suggesting a mechanism involving apoptosis. © The Author 2014. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com.

Early variations of laboratory parameters predicting shunt-dependent hydrocephalus after subarachnoid hemorrhage.

PubMed

Na, Min Kyun; Won, Yu Deok; Kim, Choong Hyun; Kim, Jae Min; Cheong, Jin Hwan; Ryu, Je Il; Han, Myung-Hoon

2017-01-01

Hydrocephalus is a frequent complication following subarachnoid hemorrhage. Few studies investigated the association between laboratory parameters and shunt-dependent hydrocephalus. This study aimed to investigate the variations of laboratory parameters after subarachnoid hemorrhage. We also attempted to identify predictive laboratory parameters for shunt-dependent hydrocephalus. Multiple imputation was performed to fill the missing laboratory data using Bayesian methods in SPSS. We used univariate and multivariate Cox regression analyses to calculate hazard ratios for shunt-dependent hydrocephalus based on clinical and laboratory factors. The area under the receiver operating characteristic curve was used to determine the laboratory risk values predicting shunt-dependent hydrocephalus. We included 181 participants with a mean age of 54.4 years. Higher sodium (hazard ratio, 1.53; 95% confidence interval, 1.13-2.07; p = 0.005), lower potassium, and higher glucose levels were associated with higher shunt-dependent hydrocephalus. The receiver operating characteristic curve analysis showed that the areas under the curve of sodium, potassium, and glucose were 0.649 (cutoff value, 142.75 mEq/L), 0.609 (cutoff value, 3.04 mmol/L), and 0.664 (cutoff value, 140.51 mg/dL), respectively. Despite the exploratory nature of this study, we found that higher sodium, lower potassium, and higher glucose levels were predictive values for shunt-dependent hydrocephalus from postoperative day (POD) 1 to POD 12-16 after subarachnoid hemorrhage. Strict correction of electrolyte imbalance seems necessary to reduce shunt-dependent hydrocephalus. Further large studies are warranted to confirm our findings.

yaImpute: An R package for kNN imputation

Treesearch

Nicholas L. Crookston; Andrew O. Finley

2008-01-01

This article introduces yaImpute, an R package for nearest neighbor search and imputation. Although nearest neighbor imputation is used in a host of disciplines, the methods implemented in the yaImpute package are tailored to imputation-based forest attribute estimation and mapping. The impetus to writing the yaImpute is a growing interest in nearest neighbor...

Applications of temporal kernel canonical correlation analysis in adherence studies.

PubMed

John, Majnu; Lencz, Todd; Ferbinteanu, Janina; Gallego, Juan A; Robinson, Delbert G

2017-10-01

Adherence to medication is often measured as a continuous outcome but analyzed as a dichotomous outcome due to lack of appropriate tools. In this paper, we illustrate the use of the temporal kernel canonical correlation analysis (tkCCA) as a method to analyze adherence measurements and symptom levels on a continuous scale. The tkCCA is a novel method developed for studying the relationship between neural signals and hemodynamic response detected by functional MRI during spontaneous activity. Although the tkCCA is a powerful tool, it has not been utilized outside the application that it was originally developed for. In this paper, we simulate time series of symptoms and adherence levels for patients with a hypothetical brain disorder and show how the tkCCA can be used to understand the relationship between them. We also examine, via simulations, the behavior of the tkCCA under various missing value mechanisms and imputation methods. Finally, we apply the tkCCA to a real data example of psychotic symptoms and adherence levels obtained from a study based on subjects with a first episode of schizophrenia, schizophreniform or schizoaffective disorder.

Naproxen or Estradiol for Bleeding and Spotting with the Levonorgestrel Intrauterine System: A Randomized Controlled Trial

PubMed Central

MADDEN, Tessa; PROEHL, Sarah; ALLSWORTH, Jenifer E.; SECURA, Gina M.; PEIPERT, Jeffrey F.

2011-01-01

Objective To evaluate whether oral naproxen or transdermal estradiol decreases bleeding and spotting in women initiating the levonorgestrel-releasing intrauterine system (LNG-IUS). Study Design We conducted a randomized controlled trial of naproxen, estradiol, or placebo administered over the first 12 weeks of LNG-IUS use. Participants completed a written bleeding diary. We imputed missing values and performed an intention-to-treat analysis. Results There were 129 women randomized to naproxen (n=42), estradiol (n=44), or placebo (n=43). The naproxen group was more likely to be in the lowest quartile of bleeding and spotting days compared to placebo, 42.9% versus 16.3% (p=0.03). In the multivariable analysis, the naproxen group had a 10% reduction in bleeding and spotting days (RRadj 0.90, 95%CI 0.84–0.97) compared to placebo. More frequent bleeding and spotting was observed in the estradiol group (RRadj 1.25, 95%CI 1.17–1.34). Conclusions Administration of naproxen resulted in a reduction in bleeding and spotting days compared to placebo. (150 words) PMID:22055339

Bi-level Multi-Source Learning for Heterogeneous Block-wise Missing Data

PubMed Central

Xiang, Shuo; Yuan, Lei; Fan, Wei; Wang, Yalin; Thompson, Paul M.; Ye, Jieping

2013-01-01

Bio-imaging technologies allow scientists to collect large amounts of high-dimensional data from multiple heterogeneous sources for many biomedical applications. In the study of Alzheimer's Disease (AD), neuroimaging data, gene/protein expression data, etc., are often analyzed together to improve predictive power. Joint learning from multiple complementary data sources is advantageous, but feature-pruning and data source selection are critical to learn interpretable models from high-dimensional data. Often, the data collected has block-wise missing entries. In the Alzheimer’s Disease Neuroimaging Initiative (ADNI), most subjects have MRI and genetic information, but only half have cerebrospinal fluid (CSF) measures, a different half has FDG-PET; only some have proteomic data. Here we propose how to effectively integrate information from multiple heterogeneous data sources when data is block-wise missing. We present a unified “bi-level” learning model for complete multi-source data, and extend it to incomplete data. Our major contributions are: (1) our proposed models unify feature-level and source-level analysis, including several existing feature learning approaches as special cases; (2) the model for incomplete data avoids imputing missing data and offers superior performance; it generalizes to other applications with block-wise missing data sources; (3) we present efficient optimization algorithms for modeling complete and incomplete data. We comprehensively evaluate the proposed models including all ADNI subjects with at least one of four data types at baseline: MRI, FDG-PET, CSF and proteomics. Our proposed models compare favorably with existing approaches. PMID:23988272

Bi-level multi-source learning for heterogeneous block-wise missing data.

PubMed

Xiang, Shuo; Yuan, Lei; Fan, Wei; Wang, Yalin; Thompson, Paul M; Ye, Jieping

2014-11-15

Bio-imaging technologies allow scientists to collect large amounts of high-dimensional data from multiple heterogeneous sources for many biomedical applications. In the study of Alzheimer's Disease (AD), neuroimaging data, gene/protein expression data, etc., are often analyzed together to improve predictive power. Joint learning from multiple complementary data sources is advantageous, but feature-pruning and data source selection are critical to learn interpretable models from high-dimensional data. Often, the data collected has block-wise missing entries. In the Alzheimer's Disease Neuroimaging Initiative (ADNI), most subjects have MRI and genetic information, but only half have cerebrospinal fluid (CSF) measures, a different half has FDG-PET; only some have proteomic data. Here we propose how to effectively integrate information from multiple heterogeneous data sources when data is block-wise missing. We present a unified "bi-level" learning model for complete multi-source data, and extend it to incomplete data. Our major contributions are: (1) our proposed models unify feature-level and source-level analysis, including several existing feature learning approaches as special cases; (2) the model for incomplete data avoids imputing missing data and offers superior performance; it generalizes to other applications with block-wise missing data sources; (3) we present efficient optimization algorithms for modeling complete and incomplete data. We comprehensively evaluate the proposed models including all ADNI subjects with at least one of four data types at baseline: MRI, FDG-PET, CSF and proteomics. Our proposed models compare favorably with existing approaches. © 2013 Elsevier Inc. All rights reserved.