A Bayesian latent class model to estimate the accuracy of pregnancy diagnosis by transrectal ultrasonography and laboratory detection of pregnancy-associated glycoproteins in dairy cows.

PubMed

Fosgate, G T; Motimele, B; Ganswindt, A; Irons, P C

2017-09-15

Accurate diagnosis of pregnancy is an essential component of an effective reproductive management plan for dairy cattle. Indirect methods of pregnancy detection can be performed soon after breeding and offer an advantage over traditional direct methods in not requiring an experienced veterinarian and having potential for automation. The objective of this study was to estimate the sensitivity and specificity of pregnancy-associated glycoprotein (PAG) detection ELISA and transrectal ultrasound (TRUS) in dairy cows of South Africa using a Bayesian latent class approach. Commercial dairy cattle from the five important dairy regions in South Africa were enrolled in a short-term prospective cohort study. Cattle were examined at 28-35days after artificial insemination (AI) and then followed up 14days later. At both sampling times, TRUS was performed to detect pregnancy and commercially available PAG detection ELISAs were performed on collected serum and milk. A total of 1236 cows were sampled and 1006 had complete test information for use in the Bayesian latent class model. The estimated sensitivity (95% probability interval) and specificity for PAG detection serum ELISA were 99.4% (98.5, 99.9) and 97.4% (94.7, 99.2), respectively. The estimated sensitivity and specificity for PAG detection milk ELISA were 99.2% (98.2, 99.8) and 93.4% (89.7, 96.1), respectively. Sensitivity of veterinarian performed TRUS at 28-35days post-AI varied between 77.8% and 90.5% and specificity varied between 94.7% and 99.8%. In summary, indirect detection of pregnancy using PAG ELISA is an accurate method for use in dairy cattle. The method is descriptively more sensitive than veterinarian-performed TRUS and therefore could be an economically viable addition to a reproductive management plan. Copyright © 2017 Elsevier B.V. All rights reserved.

Bayesian Adaptive Lasso for Ordinal Regression with Latent Variables

ERIC Educational Resources Information Center

Feng, Xiang-Nan; Wu, Hao-Tian; Song, Xin-Yuan

2017-01-01

We consider an ordinal regression model with latent variables to investigate the effects of observable and latent explanatory variables on the ordinal responses of interest. Each latent variable is characterized by correlated observed variables through a confirmatory factor analysis model. We develop a Bayesian adaptive lasso procedure to conduct…

Least Squares Distance Method of Cognitive Validation and Analysis for Binary Items Using Their Item Response Theory Parameters

ERIC Educational Resources Information Center

Dimitrov, Dimiter M.

2007-01-01

The validation of cognitive attributes required for correct answers on binary test items or tasks has been addressed in previous research through the integration of cognitive psychology and psychometric models using parametric or nonparametric item response theory, latent class modeling, and Bayesian modeling. All previous models, each with their…

On the Estimation of Disease Prevalence by Latent Class Models for Screening Studies Using Two Screening Tests with Categorical Disease Status Verified in Test Positives Only

PubMed Central

Chu, Haitao; Zhou, Yijie; Cole, Stephen R.; Ibrahim, Joseph G.

2010-01-01

Summary To evaluate the probabilities of a disease state, ideally all subjects in a study should be diagnosed by a definitive diagnostic or gold standard test. However, since definitive diagnostic tests are often invasive and expensive, it is generally unethical to apply them to subjects whose screening tests are negative. In this article, we consider latent class models for screening studies with two imperfect binary diagnostic tests and a definitive categorical disease status measured only for those with at least one positive screening test. Specifically, we discuss a conditional independent and three homogeneous conditional dependent latent class models and assess the impact of misspecification of the dependence structure on the estimation of disease category probabilities using frequentist and Bayesian approaches. Interestingly, the three homogeneous dependent models can provide identical goodness-of-fit but substantively different estimates for a given study. However, the parametric form of the assumed dependence structure itself is not “testable” from the data, and thus the dependence structure modeling considered here can only be viewed as a sensitivity analysis concerning a more complicated non-identifiable model potentially involving heterogeneous dependence structure. Furthermore, we discuss Bayesian model averaging together with its limitations as an alternative way to partially address this particularly challenging problem. The methods are applied to two cancer screening studies, and simulations are conducted to evaluate the performance of these methods. In summary, further research is needed to reduce the impact of model misspecification on the estimation of disease prevalence in such settings. PMID:20191614

Prediction of hemoglobin in blood donors using a latent class mixed-effects transition model.

PubMed

Nasserinejad, Kazem; van Rosmalen, Joost; de Kort, Wim; Rizopoulos, Dimitris; Lesaffre, Emmanuel

2016-02-20

Blood donors experience a temporary reduction in their hemoglobin (Hb) value after donation. At each visit, the Hb value is measured, and a too low Hb value leads to a deferral for donation. Because of the recovery process after each donation as well as state dependence and unobserved heterogeneity, longitudinal data of Hb values of blood donors provide unique statistical challenges. To estimate the shape and duration of the recovery process and to predict future Hb values, we employed three models for the Hb value: (i) a mixed-effects models; (ii) a latent-class mixed-effects model; and (iii) a latent-class mixed-effects transition model. In each model, a flexible function was used to model the recovery process after donation. The latent classes identify groups of donors with fast or slow recovery times and donors whose recovery time increases with the number of donations. The transition effect accounts for possible state dependence in the observed data. All models were estimated in a Bayesian way, using data of new entrant donors from the Donor InSight study. Informative priors were used for parameters of the recovery process that were not identified using the observed data, based on results from the clinical literature. The results show that the latent-class mixed-effects transition model fits the data best, which illustrates the importance of modeling state dependence, unobserved heterogeneity, and the recovery process after donation. The estimated recovery time is much longer than the current minimum interval between donations, suggesting that an increase of this interval may be warranted. Copyright © 2015 John Wiley & Sons, Ltd.

A Bayesian Model for the Estimation of Latent Interaction and Quadratic Effects When Latent Variables Are Non-Normally Distributed

ERIC Educational Resources Information Center

Kelava, Augustin; Nagengast, Benjamin

2012-01-01

Structural equation models with interaction and quadratic effects have become a standard tool for testing nonlinear hypotheses in the social sciences. Most of the current approaches assume normally distributed latent predictor variables. In this article, we present a Bayesian model for the estimation of latent nonlinear effects when the latent…

Bayesian latent class estimation of the incidence of chest radiograph-confirmed pneumonia in rural Thailand.

PubMed

Lu, Y; Baggett, H C; Rhodes, J; Thamthitiwat, S; Joseph, L; Gregory, C J

2016-10-01

Pneumonia is a leading cause of mortality and morbidity worldwide with radiographically confirmed pneumonia a key disease burden indicator. This is usually determined by a radiology panel which is assumed to be the best available standard; however, this assumption may introduce bias into pneumonia incidence estimates. To improve estimates of radiographic pneumonia incidence, we applied Bayesian latent class modelling (BLCM) to a large database of hospitalized patients with acute lower respiratory tract illness in Sa Kaeo and Nakhon Phanom provinces, Thailand from 2005 to 2010 with chest radiographs read by both a radiology panel and a clinician. We compared these estimates to those from conventional analysis. For children aged <5 years, estimated radiographically confirmed pneumonia incidence by BLCM was 2394/100 000 person-years (95% credible interval 2185-2574) vs. 1736/100 000 person-years (95% confidence interval 1706-1766) from conventional analysis. For persons aged ⩾5 years, estimated radiographically confirmed pneumonia incidence was similar between BLCM and conventional analysis (235 vs. 215/100 000 person-years). BLCM suggests the incidence of radiographically confirmed pneumonia in young children is substantially larger than estimated from the conventional approach using radiology panels as the reference standard.

Joint model-based clustering of nonlinear longitudinal trajectories and associated time-to-event data analysis, linked by latent class membership: with application to AIDS clinical studies.

PubMed

Huang, Yangxin; Lu, Xiaosun; Chen, Jiaqing; Liang, Juan; Zangmeister, Miriam

2017-10-27

Longitudinal and time-to-event data are often observed together. Finite mixture models are currently used to analyze nonlinear heterogeneous longitudinal data, which, by releasing the homogeneity restriction of nonlinear mixed-effects (NLME) models, can cluster individuals into one of the pre-specified classes with class membership probabilities. This clustering may have clinical significance, and be associated with clinically important time-to-event data. This article develops a joint modeling approach to a finite mixture of NLME models for longitudinal data and proportional hazard Cox model for time-to-event data, linked by individual latent class indicators, under a Bayesian framework. The proposed joint models and method are applied to a real AIDS clinical trial data set, followed by simulation studies to assess the performance of the proposed joint model and a naive two-step model, in which finite mixture model and Cox model are fitted separately.

Bayesian Latent Class Models in Malaria Diagnosis

PubMed Central

Gonçalves, Luzia; Subtil, Ana; de Oliveira, M. Rosário; do Rosário, Virgílio; Lee, Pei-Wen; Shaio, Men-Fang

2012-01-01

Aims The main focus of this study is to illustrate the importance of the statistical analysis in the evaluation of the accuracy of malaria diagnostic tests, without admitting a reference test, exploring a dataset (3317) collected in São Tomé and Príncipe. Methods Bayesian Latent Class Models (without and with constraints) are used to estimate the malaria infection prevalence, together with sensitivities, specificities, and predictive values of three diagnostic tests (RDT, Microscopy and PCR), in four subpopulations simultaneously based on a stratified analysis by age groups (, 5 years old) and fever status (febrile, afebrile). Results In the afebrile individuals with at least five years old, the posterior mean of the malaria infection prevalence is 3.2% with a highest posterior density interval of [2.3–4.1]. The other three subpopulations (febrile 5 years, afebrile or febrile children less than 5 years) present a higher prevalence around 10.3% [8.8–11.7]. In afebrile children under-five years old, the sensitivity of microscopy is 50.5% [37.7–63.2]. In children under-five, the estimated sensitivities/specificities of RDT are 95.4% [90.3–99.5]/93.8% [91.6–96.0] – afebrile – and 94.1% [87.5–99.4]/97.5% [95.5–99.3] – febrile. In individuals with at least five years old are 96.0% [91.5–99.7]/98.7% [98.1–99.2] – afebrile – and 97.9% [95.3–99.8]/97.7% [96.6–98.6] – febrile. The PCR yields the most reliable results in four subpopulations. Conclusions The utility of this RDT in the field seems to be relevant. However, in all subpopulations, data provide enough evidence to suggest caution with the positive predictive values of the RDT. Microscopy has poor sensitivity compared to the other tests, particularly, in the afebrile children less than 5 years. This type of findings reveals the danger of statistical analysis based on microscopy as a reference test. Bayesian Latent Class Models provide a powerful tool to evaluate malaria diagnostic tests, taking into account different groups of interest. PMID:22844405

A latent class distance association model for cross-classified data with a categorical response variable.

PubMed

Vera, José Fernando; de Rooij, Mark; Heiser, Willem J

2014-11-01

In this paper we propose a latent class distance association model for clustering in the predictor space of large contingency tables with a categorical response variable. The rows of such a table are characterized as profiles of a set of explanatory variables, while the columns represent a single outcome variable. In many cases such tables are sparse, with many zero entries, which makes traditional models problematic. By clustering the row profiles into a few specific classes and representing these together with the categories of the response variable in a low-dimensional Euclidean space using a distance association model, a parsimonious prediction model can be obtained. A generalized EM algorithm is proposed to estimate the model parameters and the adjusted Bayesian information criterion statistic is employed to test the number of mixture components and the dimensionality of the representation. An empirical example highlighting the advantages of the new approach and comparing it with traditional approaches is presented. © 2014 The British Psychological Society.

Bayesian latent structure modeling of walking behavior in a physical activity intervention

PubMed Central

Lawson, Andrew B; Ellerbe, Caitlyn; Carroll, Rachel; Alia, Kassandra; Coulon, Sandra; Wilson, Dawn K; VanHorn, M Lee; St George, Sara M

2017-01-01

The analysis of walking behavior in a physical activity intervention is considered. A Bayesian latent structure modeling approach is proposed whereby the ability and willingness of participants is modeled via latent effects. The dropout process is jointly modeled via a linked survival model. Computational issues are addressed via posterior sampling and a simulated evaluation of the longitudinal model’s ability to recover latent structure and predictor effects is considered. We evaluate the effect of a variety of socio-psychological and spatial neighborhood predictors on the propensity to walk and the estimation of latent ability and willingness in the full study. PMID:24741000

Test characteristics of milk amyloid A ELISA, somatic cell count, and bacteriological culture for detection of intramammary pathogens that cause subclinical mastitis.

PubMed

Jaeger, S; Virchow, F; Torgerson, P R; Bischoff, M; Biner, B; Hartnack, S; Rüegg, S R

2017-09-01

Bovine mastitis is an important disease in the dairy industry, causing economic losses as a result of withheld milk and treatment costs. Several studies have suggested milk amyloid A (MAA) as a promising biomarker in the diagnosis of mastitis. In the absence of a gold standard for diagnosis of subclinical mastitis, we estimated the diagnostic test accuracy of a commercial MAA-ELISA, somatic cell count (SCC), and bacteriological culture using Bayesian latent class modeling. We divided intramammary infections into 2 classes: those caused by major pathogens (e.g., Escherichia coli, Staphylococcus aureus, streptococci, and lacto-/enterococci) and those caused by all pathogens (major pathogens plus Corynebacterium bovis, coagulase-negative staphylococci, Bacillus spp., Streptomyces spp.). We applied the 3 diagnostic tests to all samples. Of 433 composite milk samples included in this study, 275 (63.5%) contained at least 1 colony of any bacterial species; of those, 56 contained major pathogens and 219 contained minor pathogens. The remaining 158 samples (36.5%) were sterile. We determined 2 different thresholds for the MAA-ELISA using Bayesian latent class modeling: 3.9 µg/mL to detect mastitis caused by major pathogens and 1.6 µg/mL to detect mastitis caused by all pathogens. The optimal SCC threshold for identification of subclinical mastitis was 150,000 cells/mL; this threshold led to higher specificity (Sp) than 100,000 cells/mL. Test accuracy for major-pathogen intramammary infections was as follows: SCC, sensitivity (Se) 92.6% and Sp 72.9%; MAA-ELISA, Se 81.4% and Sp 93.4%; bacteriological culture, Se 23.8% and Sp 95.2%. Test accuracy for all-pathogen intramammary infections was as follows: SCC, sensitivity 90.3% and Sp 71.8%; MAA-ELISA, Se 88.0% and Sp 65.2%; bacteriological culture, Se 83.8% and Sp 54.8%. We suggest the use of SCC and MAA-ELISA as a combined screening procedure for situations such as a Staphylococcus aureus control program. With Bayesian latent class analysis, we were able to identify a more differentiated use of the 3 diagnostic tools. The MAA-ELISA is a valuable addition to existing tools for the diagnosis of subclinical mastitis. The Authors. Published by the Federation of Animal Science Societies and Elsevier Inc. on behalf of the American Dairy Science Association®. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

Bayesian Analysis of Structural Equation Models with Nonlinear Covariates and Latent Variables

ERIC Educational Resources Information Center

Song, Xin-Yuan; Lee, Sik-Yum

2006-01-01

In this article, we formulate a nonlinear structural equation model (SEM) that can accommodate covariates in the measurement equation and nonlinear terms of covariates and exogenous latent variables in the structural equation. The covariates can come from continuous or discrete distributions. A Bayesian approach is developed to analyze the…

Patterns of Dating Violence Perpetration and Victimization in U.S. Young Adult Males and Females.

PubMed

Spencer, Rachael A; Renner, Lynette M; Clark, Cari Jo

2016-09-01

Dating violence (DV) is frequently reported by young adults in intimate relationships in the United States, but little is known about patterns of DV perpetration and victimization. In this study, we examined sexual and physical violence perpetration and victimization reported by young adults to determine how the violence patterns differ by sex and race/ethnicity. Data from non-Hispanic White, non-Hispanic Black, and Hispanic participants in Wave 3 of the National Longitudinal Study of Adolescent to Adult Health were analyzed. DV was assessed using responses to four questions focused on perpetration and four questions focused on victimization. The information on DV was taken from the most violent relationship reported by participants prior to Wave 3. Latent class analysis was first conducted separately by sex, adjusting for age, race/ethnicity, and financial stress, then by race/ethnicity, adjusting for age and financial stress. Relative model fit was established by comparing Bayesian Information Criteria (BIC), adjusted BIC, entropy, interpretability of latent classes, and certainty of latent class assignment for covariate-adjusted models. The results indicate that patterns of violence differed by sex and for females, by race/ethnicity. A three-class model was the best fit for males. For females, separate four-class models were parsimonious for White, Black, and Hispanic females. Financial stress was a significant predictor of violence classification for males and females and age predicted membership in White and Black female models. Variations in DV patterns by sex and race/ethnicity suggest the need for a more nuanced understanding of differences in DV. © The Author(s) 2015.

Identifying patterns of item missing survey data using latent groups: an observational study

PubMed Central

McElwee, Paul; Nathan, Andrea; Burton, Nicola W; Turrell, Gavin

2017-01-01

Objectives To examine whether respondents to a survey of health and physical activity and potential determinants could be grouped according to the questions they missed, known as ‘item missing’. Design Observational study of longitudinal data. Setting Residents of Brisbane, Australia. Participants 6901 people aged 40–65 years in 2007. Materials and methods We used a latent class model with a mixture of multinomial distributions and chose the number of classes using the Bayesian information criterion. We used logistic regression to examine if participants’ characteristics were associated with their modal latent class. We used logistic regression to examine whether the amount of item missing in a survey predicted wave missing in the following survey. Results Four per cent of participants missed almost one-fifth of the questions, and this group missed more questions in the middle of the survey. Eighty-three per cent of participants completed almost every question, but had a relatively high missing probability for a question on sleep time, a question which had an inconsistent presentation compared with the rest of the survey. Participants who completed almost every question were generally younger and more educated. Participants who completed more questions were less likely to miss the next longitudinal wave. Conclusions Examining patterns in item missing data has improved our understanding of how missing data were generated and has informed future survey design to help reduce missing data. PMID:29084795

Estimation of the latent mediated effect with ordinal data using the limited-information and Bayesian full-information approaches.

PubMed

Chen, Jinsong; Zhang, Dake; Choi, Jaehwa

2015-12-01

It is common to encounter latent variables with ordinal data in social or behavioral research. Although a mediated effect of latent variables (latent mediated effect, or LME) with ordinal data may appear to be a straightforward combination of LME with continuous data and latent variables with ordinal data, the methodological challenges to combine the two are not trivial. This research covers model structures as complex as LME and formulates both point and interval estimates of LME for ordinal data using the Bayesian full-information approach. We also combine weighted least squares (WLS) estimation with the bias-corrected bootstrapping (BCB; Efron Journal of the American Statistical Association, 82, 171-185, 1987) method or the traditional delta method as the limited-information approach. We evaluated the viability of these different approaches across various conditions through simulation studies, and provide an empirical example to illustrate the approaches. We found that the Bayesian approach with reasonably informative priors is preferred when both point and interval estimates are of interest and the sample size is 200 or above.

TENSOR DECOMPOSITIONS AND SPARSE LOG-LINEAR MODELS

PubMed Central

Johndrow, James E.; Bhattacharya, Anirban; Dunson, David B.

2017-01-01

Contingency table analysis routinely relies on log-linear models, with latent structure analysis providing a common alternative. Latent structure models lead to a reduced rank tensor factorization of the probability mass function for multivariate categorical data, while log-linear models achieve dimensionality reduction through sparsity. Little is known about the relationship between these notions of dimensionality reduction in the two paradigms. We derive several results relating the support of a log-linear model to nonnegative ranks of the associated probability tensor. Motivated by these findings, we propose a new collapsed Tucker class of tensor decompositions, which bridge existing PARAFAC and Tucker decompositions, providing a more flexible framework for parsimoniously characterizing multivariate categorical data. Taking a Bayesian approach to inference, we illustrate empirical advantages of the new decompositions. PMID:29332971

Using a bayesian latent class model to evaluate the utility of investigating persons with negative polymerase chain reaction results for pertussis.

PubMed

Tarr, Gillian A M; Eickhoff, Jens C; Koepke, Ruth; Hopfensperger, Daniel J; Davis, Jeffrey P; Conway, James H

2013-07-15

Pertussis remains difficult to control. Imperfect sensitivity of diagnostic tests and lack of specific guidance regarding interpretation of negative test results among patients with compatible symptoms may contribute to its spread. In this study, we examined whether additional pertussis cases could be identified if persons with negative pertussis test results were routinely investigated. We conducted interviews among 250 subjects aged ≤18 years with pertussis polymerase chain reaction (PCR) results reported from 2 reference laboratories in Wisconsin during July-September 2010 to determine whether their illnesses met the Centers for Disease Control and Prevention's clinical case definition (CCD) for pertussis. PCR validity measures were calculated using the CCD as the standard for pertussis disease. Two Bayesian latent class models were used to adjust the validity measures for pertussis detectable by 1) culture alone and 2) culture and/or more sensitive measures such as serology. Among 190 PCR-negative subjects, 54 (28%) had illnesses meeting the CCD. In adjusted analyses, PCR sensitivity and the negative predictive value were 1) 94% and 99% and 2) 43% and 87% in the 2 types of models, respectively. The models suggested that public health follow-up of reported pertussis patients with PCR-negative results leads to the detection of more true pertussis cases than follow-up of PCR-positive persons alone. The results also suggest a need for a more specific pertussis CCD.

Transrenal DNA-based diagnosis of Strongyloides stercoralis (Grassi, 1879) infection: Bayesian latent class modeling of test accuracy.

PubMed

Krolewiecki, Alejandro J; Koukounari, Artemis; Romano, Miryam; Caro, Reynaldo N; Scott, Alan L; Fleitas, Pedro; Cimino, Ruben; Shiff, Clive J

2018-06-01

For epidemiological work with soil transmitted helminths the recommended diagnostic approaches are to examine fecal samples for microscopic evidence of the parasite. In addition to several logistical and processing issues, traditional diagnostic approaches have been shown to lack the sensitivity required to reliably identify patients harboring low-level infections such as those associated with effective mass drug intervention programs. In this context, there is a need to rethink the approaches used for helminth diagnostics. Serological methods are now in use, however these tests are indirect and depend on individual immune responses, exposure patterns and the nature of the antigen. However, it has been demonstrated that cell-free DNA from pathogens and cancers can be readily detected in patient's urine which can be collected in the field, filtered in situ and processed later for analysis. In the work presented here, we employ three diagnostic procedures-stool examination, serology (NIE-ELISA) and PCR-based amplification of parasite transrenal DNA from urine-to determine their relative utility in the diagnosis of S. stercoralis infections from 359 field samples from an endemic area of Argentina. Bayesian Latent Class analysis was used to assess the relative performance of the three diagnostic procedures. The results underscore the low sensitivity of stool examination and support the idea that the use of serology combined with parasite transrenal DNA detection may be a useful strategy for sensitive and specific detection of low-level strongyloidiasis.

A Bayesian Approach to More Stable Estimates of Group-Level Effects in Contextual Studies.

PubMed

Zitzmann, Steffen; Lüdtke, Oliver; Robitzsch, Alexander

2015-01-01

Multilevel analyses are often used to estimate the effects of group-level constructs. However, when using aggregated individual data (e.g., student ratings) to assess a group-level construct (e.g., classroom climate), the observed group mean might not provide a reliable measure of the unobserved latent group mean. In the present article, we propose a Bayesian approach that can be used to estimate a multilevel latent covariate model, which corrects for the unreliable assessment of the latent group mean when estimating the group-level effect. A simulation study was conducted to evaluate the choice of different priors for the group-level variance of the predictor variable and to compare the Bayesian approach with the maximum likelihood approach implemented in the software Mplus. Results showed that, under problematic conditions (i.e., small number of groups, predictor variable with a small ICC), the Bayesian approach produced more accurate estimates of the group-level effect than the maximum likelihood approach did.

Data-driven classification of bipolar I disorder from longitudinal course of mood.

PubMed

Cochran, A L; McInnis, M G; Forger, D B

2016-10-11

The Diagnostic and Statistical Manual of Mental Disorder (DSM) classification of bipolar disorder defines categories to reflect common understanding of mood symptoms rather than scientific evidence. This work aimed to determine whether bipolar I can be objectively classified from longitudinal mood data and whether resulting classes have clinical associations. Bayesian nonparametric hierarchical models with latent classes and patient-specific models of mood are fit to data from Longitudinal Interval Follow-up Evaluations (LIFE) of bipolar I patients (N=209). Classes are tested for clinical associations. No classes are justified using the time course of DSM-IV mood states. Three classes are justified using the course of subsyndromal mood symptoms. Classes differed in attempted suicides (P=0.017), disability status (P=0.012) and chronicity of affective symptoms (P=0.009). Thus, bipolar I disorder can be objectively classified from mood course, and individuals in the resulting classes share clinical features. Data-driven classification from mood course could be used to enrich sample populations for pharmacological and etiological studies.

Identifying patterns of item missing survey data using latent groups: an observational study.

PubMed

Barnett, Adrian G; McElwee, Paul; Nathan, Andrea; Burton, Nicola W; Turrell, Gavin

2017-10-30

To examine whether respondents to a survey of health and physical activity and potential determinants could be grouped according to the questions they missed, known as 'item missing'. Observational study of longitudinal data. Residents of Brisbane, Australia. 6901 people aged 40-65 years in 2007. We used a latent class model with a mixture of multinomial distributions and chose the number of classes using the Bayesian information criterion. We used logistic regression to examine if participants' characteristics were associated with their modal latent class. We used logistic regression to examine whether the amount of item missing in a survey predicted wave missing in the following survey. Four per cent of participants missed almost one-fifth of the questions, and this group missed more questions in the middle of the survey. Eighty-three per cent of participants completed almost every question, but had a relatively high missing probability for a question on sleep time, a question which had an inconsistent presentation compared with the rest of the survey. Participants who completed almost every question were generally younger and more educated. Participants who completed more questions were less likely to miss the next longitudinal wave. Examining patterns in item missing data has improved our understanding of how missing data were generated and has informed future survey design to help reduce missing data. © Article author(s) (or their employer(s) unless otherwise stated in the text of the article) 2017. All rights reserved. No commercial use is permitted unless otherwise expressly granted.

Relating memory to functional performance in normal aging to dementia using hierarchical Bayesian cognitive processing models.

PubMed

Shankle, William R; Pooley, James P; Steyvers, Mark; Hara, Junko; Mangrola, Tushar; Reisberg, Barry; Lee, Michael D

2013-01-01

Determining how cognition affects functional abilities is important in Alzheimer disease and related disorders. A total of 280 patients (normal or Alzheimer disease and related disorders) received a total of 1514 assessments using the functional assessment staging test (FAST) procedure and the MCI Screen. A hierarchical Bayesian cognitive processing model was created by embedding a signal detection theory model of the MCI Screen-delayed recognition memory task into a hierarchical Bayesian framework. The signal detection theory model used latent parameters of discriminability (memory process) and response bias (executive function) to predict, simultaneously, recognition memory performance for each patient and each FAST severity group. The observed recognition memory data did not distinguish the 6 FAST severity stages, but the latent parameters completely separated them. The latent parameters were also used successfully to transform the ordinal FAST measure into a continuous measure reflecting the underlying continuum of functional severity. Hierarchical Bayesian cognitive processing models applied to recognition memory data from clinical practice settings accurately translated a latent measure of cognition into a continuous measure of functional severity for both individuals and FAST groups. Such a translation links 2 levels of brain information processing and may enable more accurate correlations with other levels, such as those characterized by biomarkers.

Bayesian Data-Model Fit Assessment for Structural Equation Modeling

ERIC Educational Resources Information Center

Levy, Roy

2011-01-01

Bayesian approaches to modeling are receiving an increasing amount of attention in the areas of model construction and estimation in factor analysis, structural equation modeling (SEM), and related latent variable models. However, model diagnostics and model criticism remain relatively understudied aspects of Bayesian SEM. This article describes…

A Bayesian Semiparametric Latent Variable Model for Mixed Responses

ERIC Educational Resources Information Center

Fahrmeir, Ludwig; Raach, Alexander

2007-01-01

In this paper we introduce a latent variable model (LVM) for mixed ordinal and continuous responses, where covariate effects on the continuous latent variables are modelled through a flexible semiparametric Gaussian regression model. We extend existing LVMs with the usual linear covariate effects by including nonparametric components for nonlinear…

Evaluation by latent class analysis of a magnetic capture based DNA extraction followed by real-time qPCR as a new diagnostic method for detection of Echinococcus multilocularis in definitive hosts.

PubMed

Maas, Miriam; van Roon, Annika; Dam-Deisz, Cecile; Opsteegh, Marieke; Massolo, Alessandro; Deksne, Gunita; Teunis, Peter; van der Giessen, Joke

2016-10-30

A new method, based on a magnetic capture based DNA extraction followed by qPCR, was developed for the detection of the zoonotic parasite Echinococcus multilocularis in definitive hosts. Latent class analysis was used to compare this new method with the currently used phenol-chloroform DNA extraction followed by single tube nested PCR. In total, 60 red foxes and coyotes from three different locations were tested with both molecular methods and the sedimentation and counting technique (SCT) or intestinal scraping technique (IST). Though based on a limited number of samples, it could be established that the magnetic capture based DNA extraction followed by qPCR showed similar sensitivity and specificity as the currently used phenol-chloroform DNA extraction followed by single tube nested PCR. All methods have a high specificity as shown by Bayesian latent class analysis. Both molecular assays have higher sensitivities than the combined SCT and IST, though the uncertainties in sensitivity estimates were wide for all assays tested. The magnetic capture based DNA extraction followed by qPCR has the advantage of not requiring hazardous chemicals like the phenol-chloroform DNA extraction followed by single tube nested PCR. This supports the replacement of the phenol-chloroform DNA extraction followed by single tube nested PCR by the magnetic capture based DNA extraction followed by qPCR for molecular detection of E. multilocularis in definitive hosts. Copyright © 2016 Elsevier B.V. All rights reserved.

Bayesian Analysis of Longitudinal Data Using Growth Curve Models

ERIC Educational Resources Information Center

Zhang, Zhiyong; Hamagami, Fumiaki; Wang, Lijuan Lijuan; Nesselroade, John R.; Grimm, Kevin J.

2007-01-01

Bayesian methods for analyzing longitudinal data in social and behavioral research are recommended for their ability to incorporate prior information in estimating simple and complex models. We first summarize the basics of Bayesian methods before presenting an empirical example in which we fit a latent basis growth curve model to achievement data…

Robust Bayesian clustering.

PubMed

Archambeau, Cédric; Verleysen, Michel

2007-01-01

A new variational Bayesian learning algorithm for Student-t mixture models is introduced. This algorithm leads to (i) robust density estimation, (ii) robust clustering and (iii) robust automatic model selection. Gaussian mixture models are learning machines which are based on a divide-and-conquer approach. They are commonly used for density estimation and clustering tasks, but are sensitive to outliers. The Student-t distribution has heavier tails than the Gaussian distribution and is therefore less sensitive to any departure of the empirical distribution from Gaussianity. As a consequence, the Student-t distribution is suitable for constructing robust mixture models. In this work, we formalize the Bayesian Student-t mixture model as a latent variable model in a different way from Svensén and Bishop [Svensén, M., & Bishop, C. M. (2005). Robust Bayesian mixture modelling. Neurocomputing, 64, 235-252]. The main difference resides in the fact that it is not necessary to assume a factorized approximation of the posterior distribution on the latent indicator variables and the latent scale variables in order to obtain a tractable solution. Not neglecting the correlations between these unobserved random variables leads to a Bayesian model having an increased robustness. Furthermore, it is expected that the lower bound on the log-evidence is tighter. Based on this bound, the model complexity, i.e. the number of components in the mixture, can be inferred with a higher confidence.

A Flexible Latent Class Approach to Estimating Test-Score Reliability

ERIC Educational Resources Information Center

van der Palm, Daniël W.; van der Ark, L. Andries; Sijtsma, Klaas

2014-01-01

The latent class reliability coefficient (LCRC) is improved by using the divisive latent class model instead of the unrestricted latent class model. This results in the divisive latent class reliability coefficient (DLCRC), which unlike LCRC avoids making subjective decisions about the best solution and thus avoids judgment error. A computational…

Directional data analysis under the general projected normal distribution

PubMed Central

Wang, Fangpo; Gelfand, Alan E.

2013-01-01

The projected normal distribution is an under-utilized model for explaining directional data. In particular, the general version provides flexibility, e.g., asymmetry and possible bimodality along with convenient regression specification. Here, we clarify the properties of this general class. We also develop fully Bayesian hierarchical models for analyzing circular data using this class. We show how they can be fit using MCMC methods with suitable latent variables. We show how posterior inference for distributional features such as the angular mean direction and concentration can be implemented as well as how prediction within the regression setting can be handled. With regard to model comparison, we argue for an out-of-sample approach using both a predictive likelihood scoring loss criterion and a cumulative rank probability score criterion. PMID:24046539

Latent class analysis of diagnostic science assessment data using Bayesian networks

NASA Astrophysics Data System (ADS)

Steedle, Jeffrey Thomas

2008-10-01

Diagnostic science assessments seek to draw inferences about student understanding by eliciting evidence about the mental models that underlie students' reasoning about physical systems. Measurement techniques for analyzing data from such assessments embody one of two contrasting assessment programs: learning progressions and facet-based assessments. Learning progressions assume that students have coherent theories that they apply systematically across different problem contexts. In contrast, the facet approach makes no such assumption, so students should not be expected to reason systematically across different problem contexts. A systematic comparison of these two approaches is of great practical value to assessment programs such as the National Assessment of Educational Progress as they seek to incorporate small clusters of related items in their tests for the purpose of measuring depth of understanding. This dissertation describes an investigation comparing learning progression and facet models. Data comprised student responses to small clusters of multiple-choice diagnostic science items focusing on narrow aspects of understanding of Newtonian mechanics. Latent class analysis was employed using Bayesian networks in order to model the relationship between students' science understanding and item responses. Separate models reflecting the assumptions of the learning progression and facet approaches were fit to the data. The technical qualities of inferences about student understanding resulting from the two models were compared in order to determine if either modeling approach was more appropriate. Specifically, models were compared on model-data fit, diagnostic reliability, diagnostic certainty, and predictive accuracy. In addition, the effects of test length were evaluated for both models in order to inform the number of items required to obtain adequately reliable latent class diagnoses. Lastly, changes in student understanding over time were studied with a longitudinal model in order to provide educators and curriculum developers with a sense of how students advance in understanding over the course of instruction. Results indicated that expected student response patterns rarely reflected the assumptions of the learning progression approach. That is, students tended not to systematically apply a coherent set of ideas across different problem contexts. Even those students expected to express scientifically-accurate understanding had substantial probabilities of reporting certain problematic ideas. The learning progression models failed to make as many substantively-meaningful distinctions among students as the facet models. In statistical comparisons, model-data fit was better for the facet model, but the models were quite comparable on all other statistical criteria. Studying the effects of test length revealed that approximately 8 items are needed to obtain adequate diagnostic certainty, but more items are needed to obtain adequate diagnostic reliability. The longitudinal analysis demonstrated that students either advance in their understanding (i.e., switch to the more advanced latent class) over a short period of instruction or stay at the same level. There was no significant relationship between the probability of changing latent classes and time between testing occasions. In all, this study is valuable because it provides evidence informing decisions about modeling and reporting on student understanding, it assesses the quality of measurement available from short clusters of diagnostic multiple-choice items, and it provides educators with knowledge of the paths that student may take as they advance from novice to expert understanding over the course of instruction.

A Bayesian hierarchical diffusion model decomposition of performance in Approach–Avoidance Tasks

PubMed Central

Krypotos, Angelos-Miltiadis; Beckers, Tom; Kindt, Merel; Wagenmakers, Eric-Jan

2015-01-01

Common methods for analysing response time (RT) tasks, frequently used across different disciplines of psychology, suffer from a number of limitations such as the failure to directly measure the underlying latent processes of interest and the inability to take into account the uncertainty associated with each individual's point estimate of performance. Here, we discuss a Bayesian hierarchical diffusion model and apply it to RT data. This model allows researchers to decompose performance into meaningful psychological processes and to account optimally for individual differences and commonalities, even with relatively sparse data. We highlight the advantages of the Bayesian hierarchical diffusion model decomposition by applying it to performance on Approach–Avoidance Tasks, widely used in the emotion and psychopathology literature. Model fits for two experimental data-sets demonstrate that the model performs well. The Bayesian hierarchical diffusion model overcomes important limitations of current analysis procedures and provides deeper insight in latent psychological processes of interest. PMID:25491372

Evaluation of performance of bacterial culture of feces and serum ELISA across stages of Johne's disease in cattle using a Bayesian latent class model.

PubMed

Espejo, L A; Zagmutt, F J; Groenendaal, H; Muñoz-Zanzi, C; Wells, S J

2015-11-01

The objective of this study was to evaluate the performance of bacterial culture of feces and serum ELISA to correctly identify cows with Mycobacterium avium ssp. paratuberculosis (MAP) at heavy, light, and non-fecal-shedding levels. A total of 29,785 parallel test results from bacterial culture of feces and serum ELISA were collected from 17 dairy herds in Minnesota, Pennsylvania, and Colorado. Samples were obtained from adult cows from dairy herds enrolled for up to 10 yr in the National Johne's Disease Demonstration Herd Project. A Bayesian latent class model was fitted to estimate the probabilities that bacterial culture of feces (using 72-h sedimentation or 30-min centrifugation methods) and serum ELISA results correctly identified cows as high positive, low positive, or negative given that cows were heavy, light, and non-shedders, respectively. The model assumed that no gold standard test was available and conditional independency existed between diagnostic tests. The estimated conditional probabilities that bacterial culture of feces correctly identified heavy shedders, light shedders, and non-shedders were 70.9, 32.0, and 98.5%, respectively. The same values for the serum ELISA were 60.6, 18.7, and 99.5%, respectively. Differences in diagnostic test performance were observed among states. These results improve the interpretation of results from bacterial culture of feces and serum ELISA for detection of MAP and MAP antibody (respectively), which can support on-farm infection control decisions and can be used to evaluate disease-testing strategies, taking into account the accuracy of these tests. Copyright © 2015 American Dairy Science Association. Published by Elsevier Inc. All rights reserved.

Rapid antigen detection tests for malaria diagnosis in severely ill Papua New Guinean children: a comparative study using Bayesian latent class models.

PubMed

Manning, Laurens; Laman, Moses; Rosanas-Urgell, Anna; Turlach, Berwin; Aipit, Susan; Bona, Cathy; Warrell, Jonathan; Siba, Peter; Mueller, Ivo; Davis, Timothy M E

2012-01-01

Although rapid diagnostic tests (RDTs) have practical advantages over light microscopy (LM) and good sensitivity in severe falciparum malaria in Africa, their utility where severe non-falciparum malaria occurs is unknown. LM, RDTs and polymerase chain reaction (PCR)-based methods have limitations, and thus conventional comparative malaria diagnostic studies employ imperfect gold standards. We assessed whether, using Bayesian latent class models (LCMs) which do not require a reference method, RDTs could safely direct initial anti-infective therapy in severe ill children from an area of hyperendemic transmission of both Plasmodium falciparum and P. vivax. We studied 797 Papua New Guinean children hospitalized with well-characterized severe illness for whom LM, RDT and nested PCR (nPCR) results were available. For any severe malaria, the estimated prevalence was 47.5% with RDTs exhibiting similar sensitivity and negative predictive value (NPV) to nPCR (≥96.0%). LM was the least sensitive test (87.4%) and had the lowest NPV (89.7%), but had the highest specificity (99.1%) and positive predictive value (98.9%). For severe falciparum malaria (prevalence 42.9%), the findings were similar. For non-falciparum severe malaria (prevalence 6.9%), no test had the WHO-recommended sensitivity and specificity of >95% and >90%, respectively. RDTs were the least sensitive (69.6%) and had the lowest NPV (96.7%). RDTs appear a valuable point-of-care test that is at least equivalent to LM in diagnosing severe falciparum malaria in this epidemiologic situation. None of the tests had the required sensitivity/specificity for severe non-falciparum malaria but the number of false-negative RDTs in this group was small.

Improved detection of Burkholderia pseudomallei from non-blood clinical specimens using enrichment culture and PCR: narrowing diagnostic gap in resource-constrained settings.

PubMed

Tellapragada, Chaitanya; Shaw, Tushar; D'Souza, Annet; Eshwara, Vandana Kalwaje; Mukhopadhyay, Chiranjay

2017-07-01

To evaluate the diagnostic utility of enrichment culture and PCR for improved case detection rates of non-bacteraemic form of melioidosis in limited resource settings. Clinical specimens (n = 525) obtained from patients presenting at a tertiary care hospital of South India with clinical symptoms suggestive of community-acquired pneumonia, lower respiratory tract infections, superficial or internal abscesses, chronic skin ulcers and bone or joint infections were tested for the presence of Burkholderia pseudomallei using conventional culture (CC), enrichment culture (EC) and PCR. Sensitivity, specificity, positive and negative predictive values of CC and PCR were initially deduced using EC as the gold standard method. Further, diagnostic accuracies of all the three methods were analysed using Bayesian latent class modelling (BLCM). Detection rates of B. pseudomallei using CC, EC and PCR were 3.8%, 5.3% and 6%, respectively. Diagnostic sensitivities and specificities of CC and PCR were 71.4, 98.4% and 100 and 99.4%, respectively in comparison with EC as the gold standard test. With Bayesian latent class modelling, EC and PCR demonstrated sensitivities of 98.7 and 99.3%, respectively, while CC showed a sensitivity of 70.3% for detection of B. pseudomallei. An increase of 1.6% (95% CI: 1.08-4.32%) in the case detection rate of melioidosis was observed in the study population when EC and/or PCR were used in adjunct to the conventional culture technique. Our study findings underscore the diagnostic superiority of enrichment culture and/or PCR over conventional microbiological culture for improved case detection of melioidosis from non-blood clinical specimens. © 2017 John Wiley & Sons Ltd.

Relating Memory To Functional Performance In Normal Aging to Dementia Using Hierarchical Bayesian Cognitive Processing Models

PubMed Central

Shankle, William R.; Pooley, James P.; Steyvers, Mark; Hara, Junko; Mangrola, Tushar; Reisberg, Barry; Lee, Michael D.

2012-01-01

Determining how cognition affects functional abilities is important in Alzheimer’s disease and related disorders (ADRD). 280 patients (normal or ADRD) received a total of 1,514 assessments using the Functional Assessment Staging Test (FAST) procedure and the MCI Screen (MCIS). A hierarchical Bayesian cognitive processing (HBCP) model was created by embedding a signal detection theory (SDT) model of the MCIS delayed recognition memory task into a hierarchical Bayesian framework. The SDT model used latent parameters of discriminability (memory process) and response bias (executive function) to predict, simultaneously, recognition memory performance for each patient and each FAST severity group. The observed recognition memory data did not distinguish the six FAST severity stages, but the latent parameters completely separated them. The latent parameters were also used successfully to transform the ordinal FAST measure into a continuous measure reflecting the underlying continuum of functional severity. HBCP models applied to recognition memory data from clinical practice settings accurately translated a latent measure of cognition to a continuous measure of functional severity for both individuals and FAST groups. Such a translation links two levels of brain information processing, and may enable more accurate correlations with other levels, such as those characterized by biomarkers. PMID:22407225

Diagnostic accuracy of clinical illness for bovine respiratory disease (BRD) diagnosis in beef cattle placed in feedlots: A systematic literature review and hierarchical Bayesian latent-class meta-analysis.

PubMed

Timsit, E; Dendukuri, N; Schiller, I; Buczinski, S

2016-12-01

Diagnosis of bovine respiratory disease (BRD) in beef cattle placed in feedlots is typically based on clinical illness (CI) detected by pen-checkers. Unfortunately, the accuracy of this diagnostic approach (namely, sensitivity [Se] and specificity [Sp]) remains poorly understood, in part due to the absence of a reference test for ante-mortem diagnosis of BRD. Our objective was to pool available estimates of CI's diagnostic accuracy for BRD diagnosis in feedlot beef cattle while adjusting for the inaccuracy in the reference test. The presence of lung lesions (LU) at slaughter was used as the reference test. A systematic review of the literature was conducted to identify research articles comparing CI detected by pen-checkers during the feeding period to LU at slaughter. A hierarchical Bayesian latent-class meta-analysis was used to model test accuracy. This approach accounted for imperfections of both tests as well as the within and between study variability in the accuracy of CI. Furthermore, it also predicted the Se CI and Sp CI for future studies. Conditional independence between CI and LU was assumed, as these two tests are not based on similar biological principles. Seven studies were included in the meta-analysis. Estimated pooled Se CI and Sp CI were 0.27 (95% Bayesian credible interval: 0.12-0.65) and 0.92 (0.72-0.98), respectively, whereas estimated pooled Se LU and Sp LU were 0.91 (0.82-0.99) and 0.67 (0.64-0.79). Predicted Se CI and Sp CI for future studies were 0.27 (0.01-0.96) and 0.92 (0.14-1.00), respectively. The wide credible intervals around predicted Se CI and Sp CI estimates indicated considerable heterogeneity among studies, which suggests that pooled Se CI and Sp CI are not generalizable to individual studies. In conclusion, CI appeared to have poor Se but high Sp for BRD diagnosis in feedlots. Furthermore, considerable heterogeneity among studies highlighted an urgent need to standardize BRD diagnosis in feedlots. Copyright © 2016 Elsevier B.V. All rights reserved.

A novel latent gaussian copula framework for modeling spatial correlation in quantized SAR imagery with applications to ATR

NASA Astrophysics Data System (ADS)

Thelen, Brian T.; Xique, Ismael J.; Burns, Joseph W.; Goley, G. Steven; Nolan, Adam R.; Benson, Jonathan W.

2017-04-01

With all of the new remote sensing modalities available, and with ever increasing capabilities and frequency of collection, there is a desire to fundamentally understand/quantify the information content in the collected image data relative to various exploitation goals, such as detection/classification. A fundamental approach for this is the framework of Bayesian decision theory, but a daunting challenge is to have significantly flexible and accurate multivariate models for the features and/or pixels that capture a wide assortment of distributions and dependen- cies. In addition, data can come in the form of both continuous and discrete representations, where the latter is often generated based on considerations of robustness to imaging conditions and occlusions/degradations. In this paper we propose a novel suite of "latent" models fundamentally based on multivariate Gaussian copula models that can be used for quantized data from SAR imagery. For this Latent Gaussian Copula (LGC) model, we derive an approximate, maximum-likelihood estimation algorithm and demonstrate very reasonable estimation performance even for the larger images with many pixels. However applying these LGC models to large dimen- sions/images within a Bayesian decision/classification theory is infeasible due to the computational/numerical issues in evaluating the true full likelihood, and we propose an alternative class of novel pseudo-likelihoood detection statistics that are computationally feasible. We show in a few simple examples that these statistics have the potential to provide very good and robust detection/classification performance. All of this framework is demonstrated on a simulated SLICY data set, and the results show the importance of modeling the dependencies, and of utilizing the pseudo-likelihood methods.

Prevalence of tuberculosis infection in healthcare workers of the public hospital network in Medellín, Colombia: a Bayesian approach.

PubMed

Ochoa, J; León, A L; Ramírez, I C; Lopera, C M; Bernal, E; Arbeláez, M P

2017-04-01

A latent tuberculosis infection (LTBI) prevalence survey was conducted using tuberculin skin test (TST) and Quantiferon test (QFT) in 1218 healthcare workers (HCWs) in Medellín, Colombia. In order to improve the prevalence estimates, a latent class model was built using a Bayesian approach with informative priors on the sensitivity and specificity of the TST. The proportion of concordant results (TST+,QFT+) was 41% and the discordant results contributed 27%. The marginal estimate of the prevalence P(LTBI+) was 62·1% [95% credible interval (CrI) 53·0-68·2]. The probability of LTBI+ given positive results for both tests was 99·6% (95% CrI 98·1-99·9). Sensitivity was 88·5 for TST and 74·3 for QFT, and specificity was 87·8 for TST and 97·6 for QFT. A high LTBI prevalence was found in HCWs with time-accumulated exposure in hospitals that lack control plans. In a context of intermediate tuberculosis (TB) incidence it is recommended to use only one test (either QFT or TST) in prevalence surveys or as pre-employment tests. Results will be useful to help implement TB infection control plans in hospitals where HCWs may be repeatedly exposed to unnoticed TB patients, and to inform the design of TB control policies.

Rasch Model Parameter Estimation in the Presence of a Nonnormal Latent Trait Using a Nonparametric Bayesian Approach

ERIC Educational Resources Information Center

Finch, Holmes; Edwards, Julianne M.

2016-01-01

Standard approaches for estimating item response theory (IRT) model parameters generally work under the assumption that the latent trait being measured by a set of items follows the normal distribution. Estimation of IRT parameters in the presence of nonnormal latent traits has been shown to generate biased person and item parameter estimates. A…

Accuracy of latent-variable estimation in Bayesian semi-supervised learning.

PubMed

Yamazaki, Keisuke

2015-09-01

Hierarchical probabilistic models, such as Gaussian mixture models, are widely used for unsupervised learning tasks. These models consist of observable and latent variables, which represent the observable data and the underlying data-generation process, respectively. Unsupervised learning tasks, such as cluster analysis, are regarded as estimations of latent variables based on the observable ones. The estimation of latent variables in semi-supervised learning, where some labels are observed, will be more precise than that in unsupervised, and one of the concerns is to clarify the effect of the labeled data. However, there has not been sufficient theoretical analysis of the accuracy of the estimation of latent variables. In a previous study, a distribution-based error function was formulated, and its asymptotic form was calculated for unsupervised learning with generative models. It has been shown that, for the estimation of latent variables, the Bayes method is more accurate than the maximum-likelihood method. The present paper reveals the asymptotic forms of the error function in Bayesian semi-supervised learning for both discriminative and generative models. The results show that the generative model, which uses all of the given data, performs better when the model is well specified. Copyright © 2015 Elsevier Ltd. All rights reserved.

Bayesian techniques for analyzing group differences in the Iowa Gambling Task: A case study of intuitive and deliberate decision-makers.

PubMed

Steingroever, Helen; Pachur, Thorsten; Šmíra, Martin; Lee, Michael D

2018-06-01

The Iowa Gambling Task (IGT) is one of the most popular experimental paradigms for comparing complex decision-making across groups. Most commonly, IGT behavior is analyzed using frequentist tests to compare performance across groups, and to compare inferred parameters of cognitive models developed for the IGT. Here, we present a Bayesian alternative based on Bayesian repeated-measures ANOVA for comparing performance, and a suite of three complementary model-based methods for assessing the cognitive processes underlying IGT performance. The three model-based methods involve Bayesian hierarchical parameter estimation, Bayes factor model comparison, and Bayesian latent-mixture modeling. We illustrate these Bayesian methods by applying them to test the extent to which differences in intuitive versus deliberate decision style are associated with differences in IGT performance. The results show that intuitive and deliberate decision-makers behave similarly on the IGT, and the modeling analyses consistently suggest that both groups of decision-makers rely on similar cognitive processes. Our results challenge the notion that individual differences in intuitive and deliberate decision styles have a broad impact on decision-making. They also highlight the advantages of Bayesian methods, especially their ability to quantify evidence in favor of the null hypothesis, and that they allow model-based analyses to incorporate hierarchical and latent-mixture structures.

Simple Estimators for the Simple Latent Class Mastery Testing Model. Twente Educational Memorandum No. 19.

ERIC Educational Resources Information Center

van der Linden, Wim J.

Latent class models for mastery testing differ from continuum models in that they do not postulate a latent mastery continuum but conceive mastery and non-mastery as two latent classes, each characterized by different probabilities of success. Several researchers use a simple latent class model that is basically a simultaneous application of the…

A Bayesian state-space approach for damage detection and classification

NASA Astrophysics Data System (ADS)

Dzunic, Zoran; Chen, Justin G.; Mobahi, Hossein; Büyüköztürk, Oral; Fisher, John W.

2017-11-01

The problem of automatic damage detection in civil structures is complex and requires a system that can interpret collected sensor data into meaningful information. We apply our recently developed switching Bayesian model for dependency analysis to the problems of damage detection and classification. The model relies on a state-space approach that accounts for noisy measurement processes and missing data, which also infers the statistical temporal dependency between measurement locations signifying the potential flow of information within the structure. A Gibbs sampling algorithm is used to simultaneously infer the latent states, parameters of the state dynamics, the dependence graph, and any changes in behavior. By employing a fully Bayesian approach, we are able to characterize uncertainty in these variables via their posterior distribution and provide probabilistic estimates of the occurrence of damage or a specific damage scenario. We also implement a single class classification method which is more realistic for most real world situations where training data for a damaged structure is not available. We demonstrate the methodology with experimental test data from a laboratory model structure and accelerometer data from a real world structure during different environmental and excitation conditions.

When do latent class models overstate accuracy for diagnostic and other classifiers in the absence of a gold standard?

PubMed

Spencer, Bruce D

2012-06-01

Latent class models are increasingly used to assess the accuracy of medical diagnostic tests and other classifications when no gold standard is available and the true state is unknown. When the latent class is treated as the true class, the latent class models provide measures of components of accuracy including specificity and sensitivity and their complements, type I and type II error rates. The error rates according to the latent class model differ from the true error rates, however, and empirical comparisons with a gold standard suggest the true error rates often are larger. We investigate conditions under which the true type I and type II error rates are larger than those provided by the latent class models. Results from Uebersax (1988, Psychological Bulletin 104, 405-416) are extended to accommodate random effects and covariates affecting the responses. The results are important for interpreting the results of latent class analyses. An error decomposition is presented that incorporates an error component from invalidity of the latent class model. © 2011, The International Biometric Society.

A Note on Cluster Effects in Latent Class Analysis

ERIC Educational Resources Information Center

Kaplan, David; Keller, Bryan

2011-01-01

This article examines the effects of clustering in latent class analysis. A comprehensive simulation study is conducted, which begins by specifying a true multilevel latent class model with varying within- and between-cluster sample sizes, varying latent class proportions, and varying intraclass correlations. These models are then estimated under…

Analysis of multinomial models with unknown index using data augmentation

USGS Publications Warehouse

Royle, J. Andrew; Dorazio, R.M.; Link, W.A.

2007-01-01

Multinomial models with unknown index ('sample size') arise in many practical settings. In practice, Bayesian analysis of such models has proved difficult because the dimension of the parameter space is not fixed, being in some cases a function of the unknown index. We describe a data augmentation approach to the analysis of this class of models that provides for a generic and efficient Bayesian implementation. Under this approach, the data are augmented with all-zero detection histories. The resulting augmented dataset is modeled as a zero-inflated version of the complete-data model where an estimable zero-inflation parameter takes the place of the unknown multinomial index. Interestingly, data augmentation can be justified as being equivalent to imposing a discrete uniform prior on the multinomial index. We provide three examples involving estimating the size of an animal population, estimating the number of diabetes cases in a population using the Rasch model, and the motivating example of estimating the number of species in an animal community with latent probabilities of species occurrence and detection.

Optimization-Based Model Fitting for Latent Class and Latent Profile Analyses

ERIC Educational Resources Information Center

Huang, Guan-Hua; Wang, Su-Mei; Hsu, Chung-Chu

2011-01-01

Statisticians typically estimate the parameters of latent class and latent profile models using the Expectation-Maximization algorithm. This paper proposes an alternative two-stage approach to model fitting. The first stage uses the modified k-means and hierarchical clustering algorithms to identify the latent classes that best satisfy the…

Latent Transition Analysis with a Mixture Item Response Theory Measurement Model

ERIC Educational Resources Information Center

Cho, Sun-Joo; Cohen, Allan S.; Kim, Seock-Ho; Bottge, Brian

2010-01-01

A latent transition analysis (LTA) model was described with a mixture Rasch model (MRM) as the measurement model. Unlike the LTA, which was developed with a latent class measurement model, the LTA-MRM permits within-class variability on the latent variable, making it more useful for measuring treatment effects within latent classes. A simulation…

Latent structure modeling underlying theophylline tablet formulations using a Bayesian network based on a self-organizing map clustering.

PubMed

Yasuda, Akihito; Onuki, Yoshinori; Obata, Yasuko; Takayama, Kozo

2015-01-01

The "quality by design" concept in pharmaceutical formulation development requires the establishment of a science-based rationale and design space. In this article, we integrate thin-plate spline (TPS) interpolation, Kohonen's self-organizing map (SOM) and a Bayesian network (BN) to visualize the latent structure underlying causal factors and pharmaceutical responses. As a model pharmaceutical product, theophylline tablets were prepared using a standard formulation. We measured the tensile strength and disintegration time as response variables and the compressibility, cohesion and dispersibility of the pretableting blend as latent variables. We predicted these variables quantitatively using nonlinear TPS, generated a large amount of data on pretableting blends and tablets and clustered these data into several clusters using a SOM. Our results show that we are able to predict the experimental values of the latent and response variables with a high degree of accuracy and are able to classify the tablet data into several distinct clusters. In addition, to visualize the latent structure between the causal and latent factors and the response variables, we applied a BN method to the SOM clustering results. We found that despite having inserted latent variables between the causal factors and response variables, their relation is equivalent to the results for the SOM clustering, and thus we are able to explain the underlying latent structure. Consequently, this technique provides a better understanding of the relationships between causal factors and pharmaceutical responses in theophylline tablet formulation.

Spectral Bayesian Knowledge Tracing

ERIC Educational Resources Information Center

Falakmasir, Mohammad; Yudelson, Michael; Ritter, Steve; Koedinger, Ken

2015-01-01

Bayesian Knowledge Tracing (BKT) has been in wide use for modeling student skill acquisition in Intelligent Tutoring Systems (ITS). BKT tracks and updates student's latent mastery of a skill as a probability distribution of a binary variable. BKT does so by accounting for observed student successes in applying the skill correctly, where success is…

Bayesian Estimation of Multivariate Latent Regression Models: Gauss versus Laplace

ERIC Educational Resources Information Center

Culpepper, Steven Andrew; Park, Trevor

2017-01-01

A latent multivariate regression model is developed that employs a generalized asymmetric Laplace (GAL) prior distribution for regression coefficients. The model is designed for high-dimensional applications where an approximate sparsity condition is satisfied, such that many regression coefficients are near zero after accounting for all the model…

A Latent Transition Model with Logistic Regression

ERIC Educational Resources Information Center

Chung, Hwan; Walls, Theodore A.; Park, Yousung

2007-01-01

Latent transition models increasingly include covariates that predict prevalence of latent classes at a given time or transition rates among classes over time. In many situations, the covariate of interest may be latent. This paper describes an approach for handling both manifest and latent covariates in a latent transition model. A Bayesian…

Paths to tobacco abstinence: A repeated-measures latent class analysis.

PubMed

McCarthy, Danielle E; Ebssa, Lemma; Witkiewitz, Katie; Shiffman, Saul

2015-08-01

Knowledge of smoking change processes may be enhanced by identifying pathways to stable abstinence. We sought to identify latent classes of smokers based on their day-to-day smoking status in the first weeks of a cessation attempt. We examined treatment effects on class membership and compared classes on baseline individual differences and 6-month abstinence rates. In this secondary analysis of a double-blind randomized placebo-controlled clinical trial (N = 1,433) of 5 smoking cessation pharmacotherapies (nicotine patch, nicotine lozenge, bupropion SR, patch and lozenge, or bupropion SR and lozenge), we conducted repeated-measures latent class analysis of daily smoking status (any smoking vs. none) for the first 27 days of a quit attempt. Treatment and covariate relations with latent class membership were examined. Distal outcome analysis compared confirmed 6-month abstinence rates among the latent classes. A 5-class solution was selected. Three-quarters of smokers were in stable smoking or abstinent classes, but 25% were in classes with unstable abstinence probabilities over time. Active treatment (compared to placebo), and particularly the patch and lozenge combination, promoted early quitting. Latent classes differed in 6-month abstinence rates and on several baseline variables, including nicotine dependence, quitting history, self-efficacy, sleep disturbance, and minority status. Repeated-measures latent class analysis identified latent classes of smoking change patterns affected by treatment, related to known risk factors, and predictive of distal outcomes. Tracking behavior early in a change attempt may identify prognostic patterns of change and facilitate adaptive treatment planning. (c) 2015 APA, all rights reserved).

Spatial Dependence and Heterogeneity in Bayesian Factor Analysis: A Cross-National Investigation of Schwartz Values

ERIC Educational Resources Information Center

Stakhovych, Stanislav; Bijmolt, Tammo H. A.; Wedel, Michel

2012-01-01

In this article, we present a Bayesian spatial factor analysis model. We extend previous work on confirmatory factor analysis by including geographically distributed latent variables and accounting for heterogeneity and spatial autocorrelation. The simulation study shows excellent recovery of the model parameters and demonstrates the consequences…

On the Performance Characteristics of Latent-Factor and Knowledge Tracing Models

ERIC Educational Resources Information Center

Klingler, Severin; Käser, Tanja; Solenthaler, Barbara; Gross, Markus

2015-01-01

Modeling student knowledge is a fundamental task of an intelligent tutoring system. A popular approach for modeling the acquisition of knowledge is Bayesian Knowledge Tracing (BKT). Various extensions to the original BKT model have been proposed, among them two novel models that unify BKT and Item Response Theory (IRT). Latent Factor Knowledge…

Examining Factor Score Distributions to Determine the Nature of Latent Spaces

ERIC Educational Resources Information Center

Steinley, Douglas; McDonald, Roderick P.

2007-01-01

Similarities between latent class models with K classes and linear factor models with K-1 factors are investigated. Specifically, the mathematical equivalence between the covariance structure of the two models is discussed, and a Monte Carlo simulation is performed using generated data that represents both latent factors and latent classes with…

Evidence of Associations between Cytokine Genes and Subjective Reports of Sleep Disturbance in Oncology Patients and Their Family Caregivers

PubMed Central

Miaskowski, Christine; Cooper, Bruce A.; Dhruva, Anand; Dunn, Laura B.; Langford, Dale J.; Cataldo, Janine K.; Baggott, Christina R.; Merriman, John D.; Dodd, Marylin; Lee, Kathryn; West, Claudia; Paul, Steven M.; Aouizerat, Bradley E.

2012-01-01

The purposes of this study were to identify distinct latent classes of individuals based on subjective reports of sleep disturbance; to examine differences in demographic, clinical, and symptom characteristics between the latent classes; and to evaluate for variations in pro- and anti-inflammatory cytokine genes between the latent classes. Among 167 oncology outpatients with breast, prostate, lung, or brain cancer and 85 of their FCs, growth mixture modeling (GMM) was used to identify latent classes of individuals based on General Sleep Disturbance Scale (GSDS) obtained prior to, during, and for four months following completion of radiation therapy. Single nucleotide polymorphisms (SNPs) and haplotypes in candidate cytokine genes were interrogated for differences between the two latent classes. Multiple logistic regression was used to assess the effect of phenotypic and genotypic characteristics on GSDS group membership. Two latent classes were identified: lower sleep disturbance (88.5%) and higher sleep disturbance (11.5%). Participants who were younger and had a lower Karnofsky Performance status score were more likely to be in the higher sleep disturbance class. Variation in two cytokine genes (i.e., IL6, NFKB) predicted latent class membership. Evidence was found for latent classes with distinct sleep disturbance trajectories. Unique genetic markers in cytokine genes may partially explain the interindividual heterogeneity characterizing these trajectories. PMID:22844404

A general class of multinomial mixture models for anuran calling survey data

USGS Publications Warehouse

Royle, J. Andrew; Link, W.A.

2005-01-01

We propose a general framework for modeling anuran abundance using data collected from commonly used calling surveys. The data generated from calling surveys are indices of calling intensity (vocalization of males) that do not have a precise link to actual population size and are sensitive to factors that influence anuran behavior. We formulate a model for calling-index data in terms of the maximum potential calling index that could be observed at a site (the 'latent abundance class'), given its underlying breeding population, and we focus attention on estimating the distribution of this latent abundance class. A critical consideration in estimating the latent structure is imperfect detection, which causes the observed abundance index to be less than or equal to the latent abundance class. We specify a multinomial sampling model for the observed abundance index that is conditional on the latent abundance class. Estimation of the latent abundance class distribution is based on the marginal likelihood of the index data, having integrated over the latent class distribution. We apply the proposed modeling framework to data collected as part of the North American Amphibian Monitoring Program (NAAMP).

Diagnostic accuracy of a bayesian latent group analysis for the detection of malingering-related poor effort.

PubMed

Ortega, Alonso; Labrenz, Stephan; Markowitsch, Hans J; Piefke, Martina

2013-01-01

In the last decade, different statistical techniques have been introduced to improve assessment of malingering-related poor effort. In this context, we have recently shown preliminary evidence that a Bayesian latent group model may help to optimize classification accuracy using a simulation research design. In the present study, we conducted two analyses. Firstly, we evaluated how accurately this Bayesian approach can distinguish between participants answering in an honest way (honest response group) and participants feigning cognitive impairment (experimental malingering group). Secondly, we tested the accuracy of our model in the differentiation between patients who had real cognitive deficits (cognitively impaired group) and participants who belonged to the experimental malingering group. All Bayesian analyses were conducted using the raw scores of a visual recognition forced-choice task (2AFC), the Test of Memory Malingering (TOMM, Trial 2), and the Word Memory Test (WMT, primary effort subtests). The first analysis showed 100% accuracy for the Bayesian model in distinguishing participants of both groups with all effort measures. The second analysis showed outstanding overall accuracy of the Bayesian model when estimates were obtained from the 2AFC and the TOMM raw scores. Diagnostic accuracy of the Bayesian model diminished when using the WMT total raw scores. Despite, overall diagnostic accuracy can still be considered excellent. The most plausible explanation for this decrement is the low performance in verbal recognition and fluency tasks of some patients of the cognitively impaired group. Additionally, the Bayesian model provides individual estimates, p(zi |D), of examinees' effort levels. In conclusion, both high classification accuracy levels and Bayesian individual estimates of effort may be very useful for clinicians when assessing for effort in medico-legal settings.

Comparison of the Performance of the TPTest, Tubex, Typhidot and Widal Immunodiagnostic Assays and Blood Cultures in Detecting Patients with Typhoid Fever in Bangladesh, Including Using a Bayesian Latent Class Modeling Approach.

PubMed

Islam, Kamrul; Sayeed, Md Abu; Hossen, Emran; Khanam, Farhana; Charles, Richelle C; Andrews, Jason; Ryan, Edward T; Qadri, Firdausi

2016-04-01

There is an urgent need for an improved diagnostic assay for typhoid fever. In this current study, we compared the recently developed TPTest (Typhoid and Paratyphoid Test) with the Widal test, blood culture, and two commonly used commercially available kits, Tubex and Typhidot. For analysis, we categorized 92 Bangladeshi patients with suspected enteric fever into four groups: S. Typhi bacteremic patients (n = 28); patients with a fourfold change in Widal test from day 0 to convalescent period (n = 7); patients with Widal titer ≥1:320 (n = 13) at either acute or convalescent stage of disease; and patients suspected with enteric fever, but with a negative blood culture and Widal titer (n = 44). We also tested healthy endemic zone controls (n = 20) and Bangladeshi patients with other febrile illnesses (n = 15). Sample size was based on convenience to facilitate preliminary analysis. Of 28 S. Typhi bacteremic patients, 28 (100%), 21 (75%) and 18 (64%) patients were positive by TPTest, Tubex and Typhidot, respectively. In healthy endemic zone controls, the TPTest method was negative in all, whereas Tubex and Typhidot were positive in 3 (15%) and 5 (25%), respectively. We then estimated sensitivity and specificity of all diagnostic tests using Bayesian latent class modeling. The sensitivity of TPTest, Tubex and Typhidot were estimated at 96.0% (95% CI: 87.1%-99.8%), 60.2% (95% CI: 49.3%-71.2%), and 59.6% (95% CI: 50.1%-69.3%), respectively. Specificity was estimated at 96.6% (90.7%-99.2%) for TPTest, 89.9% (79.6%-96.8%) for Tubex, and 80.0% (67.7%-89.7%) for Typhidot. These results suggest that the TPTest is highly sensitive and specific in diagnosing individuals with typhoid fever in a typhoid endemic setting, outperforming currently available and commonly used alternatives.

Comparison of Diagnostic Accuracy of Periprosthetic Tissue Culture in Blood Culture Bottles to That of Prosthesis Sonication Fluid Culture for Diagnosis of Prosthetic Joint Infection (PJI) by Use of Bayesian Latent Class Modeling and IDSA PJI Criteria for Classification.

PubMed

Yan, Qun; Karau, Melissa J; Greenwood-Quaintance, Kerryl E; Mandrekar, Jayawant N; Osmon, Douglas R; Abdel, Matthew P; Patel, Robin

2018-06-01

We have previously demonstrated that culturing periprosthetic tissue in blood culture bottles (BCBs) improves sensitivity compared to conventional agar and broth culture methods for diagnosis of prosthetic joint infection (PJI). We have also shown that prosthesis sonication culture improves sensitivity compared to periprosthetic tissue culture using conventional agar and broth methods. The purpose of this study was to compare the diagnostic accuracy of tissue culture in BCBs (subsequently referred to as tissue culture) to prosthesis sonication culture (subsequently referred to as sonicate fluid culture). We studied 229 subjects who underwent arthroplasty revision or resection surgery between March 2016 and October 2017 at Mayo Clinic in Rochester, Minnesota. Using the Infectious Diseases Society of America (IDSA) PJI diagnostic criteria (omitting culture criteria) as the gold standard, the sensitivity of tissue culture was similar to that of the sonicate fluid culture (66.4% versus 73.1%, P = 0.07) but was significantly lower than that of the two tests combined (66.4% versus 76.9%, P < 0.001). Using Bayesian latent class modeling, which assumes no gold standard for PJI diagnosis, the sensitivity of tissue culture was slightly lower than that of sonicate fluid culture (86.3% versus 88.7%) and much lower than that of the two tests combined (86.3% versus 99.1%). In conclusion, tissue culture in BCBs reached sensitivity similar to that of prosthesis sonicate fluid culture for diagnosis of PJI, but the two tests combined had the highest sensitivity without compromising specificity. The combination of tissue culture in BCBs and sonicate fluid culture is recommended to achieve the highest level of microbiological diagnosis of PJI. Copyright © 2018 American Society for Microbiology.

Latent Class Analysis of Differential Item Functioning on the Peabody Picture Vocabulary Test-III

ERIC Educational Resources Information Center

Webb, Mi-young Lee; Cohen, Allan S.; Schwanenflugel, Paula J.

2008-01-01

This study investigated the use of latent class analysis for the detection of differences in item functioning on the Peabody Picture Vocabulary Test-Third Edition (PPVT-III). A two-class solution for a latent class model appeared to be defined in part by ability because Class 1 was lower in ability than Class 2 on both the PPVT-III and the…

Rejoinder to MacCallum, Edwards, and Cai (2012) and Rindskopf (2012): Mastering a New Method

ERIC Educational Resources Information Center

Muthen, Bengt; Asparouhov, Tihomir

2012-01-01

This rejoinder discusses the general comments on how to use Bayesian structural equation modeling (BSEM) wisely and how to get more people better trained in using Bayesian methods. Responses to specific comments cover how to handle sign switching, nonconvergence and nonidentification, and prior choices in latent variable models. Two new…

Developmental Profiles of Eczema, Wheeze, and Rhinitis: Two Population-Based Birth Cohort Studies

PubMed Central

2014-01-01

Background The term “atopic march” has been used to imply a natural progression of a cascade of symptoms from eczema to asthma and rhinitis through childhood. We hypothesize that this expression does not adequately describe the natural history of eczema, wheeze, and rhinitis during childhood. We propose that this paradigm arose from cross-sectional analyses of longitudinal studies, and may reflect a population pattern that may not predominate at the individual level. Methods and Findings Data from 9,801 children in two population-based birth cohorts were used to determine individual profiles of eczema, wheeze, and rhinitis and whether the manifestations of these symptoms followed an atopic march pattern. Children were assessed at ages 1, 3, 5, 8, and 11 y. We used Bayesian machine learning methods to identify distinct latent classes based on individual profiles of eczema, wheeze, and rhinitis. This approach allowed us to identify groups of children with similar patterns of eczema, wheeze, and rhinitis over time. Using a latent disease profile model, the data were best described by eight latent classes: no disease (51.3%), atopic march (3.1%), persistent eczema and wheeze (2.7%), persistent eczema with later-onset rhinitis (4.7%), persistent wheeze with later-onset rhinitis (5.7%), transient wheeze (7.7%), eczema only (15.3%), and rhinitis only (9.6%). When latent variable modelling was carried out separately for the two cohorts, similar results were obtained. Highly concordant patterns of sensitisation were associated with different profiles of eczema, rhinitis, and wheeze. The main limitation of this study was the difference in wording of the questions used to ascertain the presence of eczema, wheeze, and rhinitis in the two cohorts. Conclusions The developmental profiles of eczema, wheeze, and rhinitis are heterogeneous; only a small proportion of children (∼7% of those with symptoms) follow trajectory profiles resembling the atopic march. Please see later in the article for the Editors' Summary PMID:25335105

Exploring Latent Class Based on Growth Rates in Number Sense Ability

ERIC Educational Resources Information Center

Kim, Dongil; Shin, Jaehyun; Lee, Kijyung

2013-01-01

The purpose of this study was to explore latent class based on growth rates in number sense ability by using latent growth class modeling (LGCM). LGCM is one of the noteworthy methods for identifying growth patterns of the progress monitoring within the response to intervention framework in that it enables us to analyze latent sub-groups based not…

The Impact of Ignoring the Level of Nesting Structure in Nonparametric Multilevel Latent Class Models

ERIC Educational Resources Information Center

Park, Jungkyu; Yu, Hsiu-Ting

2016-01-01

The multilevel latent class model (MLCM) is a multilevel extension of a latent class model (LCM) that is used to analyze nested structure data structure. The nonparametric version of an MLCM assumes a discrete latent variable at a higher-level nesting structure to account for the dependency among observations nested within a higher-level unit. In…

On Local Homogeneity and Stochastically Ordered Mixed Rasch Models

ERIC Educational Resources Information Center

Kreiner, Svend; Hansen, Mogens; Hansen, Carsten Rosenberg

2006-01-01

Mixed Rasch models add latent classes to conventional Rasch models, assuming that the Rasch model applies within each class and that relative difficulties of items are different in two or more latent classes. This article considers a family of stochastically ordered mixed Rasch models, with ordinal latent classes characterized by increasing total…

Comparing hierarchical models via the marginalized deviance information criterion.

PubMed

Quintero, Adrian; Lesaffre, Emmanuel

2018-07-20

Hierarchical models are extensively used in pharmacokinetics and longitudinal studies. When the estimation is performed from a Bayesian approach, model comparison is often based on the deviance information criterion (DIC). In hierarchical models with latent variables, there are several versions of this statistic: the conditional DIC (cDIC) that incorporates the latent variables in the focus of the analysis and the marginalized DIC (mDIC) that integrates them out. Regardless of the asymptotic and coherency difficulties of cDIC, this alternative is usually used in Markov chain Monte Carlo (MCMC) methods for hierarchical models because of practical convenience. The mDIC criterion is more appropriate in most cases but requires integration of the likelihood, which is computationally demanding and not implemented in Bayesian software. Therefore, we consider a method to compute mDIC by generating replicate samples of the latent variables that need to be integrated out. This alternative can be easily conducted from the MCMC output of Bayesian packages and is widely applicable to hierarchical models in general. Additionally, we propose some approximations in order to reduce the computational complexity for large-sample situations. The method is illustrated with simulated data sets and 2 medical studies, evidencing that cDIC may be misleading whilst mDIC appears pertinent. Copyright © 2018 John Wiley & Sons, Ltd.

A Multidimensional Item Response Model: Constrained Latent Class Analysis Using the Gibbs Sampler and Posterior Predictive Checks.

ERIC Educational Resources Information Center

Hoijtink, Herbert; Molenaar, Ivo W.

1997-01-01

This paper shows that a certain class of constrained latent class models may be interpreted as a special case of nonparametric multidimensional item response models. Parameters of this latent class model are estimated using an application of the Gibbs sampler, and model fit is investigated using posterior predictive checks. (SLD)

Model Selection and Psychological Theory: A Discussion of the Differences between the Akaike Information Criterion (AIC) and the Bayesian Information Criterion (BIC)

ERIC Educational Resources Information Center

Vrieze, Scott I.

2012-01-01

This article reviews the Akaike information criterion (AIC) and the Bayesian information criterion (BIC) in model selection and the appraisal of psychological theory. The focus is on latent variable models, given their growing use in theory testing and construction. Theoretical statistical results in regression are discussed, and more important…

Investigating the Theoretical Structure of the DAS-II Core Battery at School Age Using Bayesian Structural Equation Modeling

ERIC Educational Resources Information Center

Dombrowski, Stefan C.; Golay, Philippe; McGill, Ryan J.; Canivez, Gary L.

2018-01-01

Bayesian structural equation modeling (BSEM) was used to investigate the latent structure of the Differential Ability Scales-Second Edition core battery using the standardization sample normative data for ages 7-17. Results revealed plausibility of a three-factor model, consistent with publisher theory, expressed as either a higher-order (HO) or a…

A Latent Class Approach to Fitting the Weighted Euclidean Model, CLASCAL.

ERIC Educational Resources Information Center

Winsberg, Suzanne; De Soete, Geert

1993-01-01

A weighted Euclidean distance model is proposed that incorporates a latent class approach (CLASCAL). The contribution to the distance function between two stimuli is per dimension weighted identically by all subjects in the same latent class. A model selection strategy is proposed and illustrated. (SLD)

A Latent Class Unfolding Model for Analyzing Single Stimulus Preference Ratings.

ERIC Educational Resources Information Center

De Soete, Geert; Heiser, Willem J.

1993-01-01

A latent class unfolding model is developed for single stimulus preference ratings. One advantage is the possibility of testing the spatial unfolding model against the unconstrained latent class model for rating data. The model is applied to data about party preferences of members of the Dutch parliament. (SLD)

Managerial performance and cost efficiency of Japanese local public hospitals: a latent class stochastic frontier model.

PubMed

Besstremyannaya, Galina

2011-09-01

The paper explores the link between managerial performance and cost efficiency of 617 Japanese general local public hospitals in 1999-2007. Treating managerial performance as unobservable heterogeneity, the paper employs a panel data stochastic cost frontier model with latent classes. Financial parameters associated with better managerial performance are found to be positively significant in explaining the probability of belonging to the more efficient latent class. The analysis of latent class membership was consistent with the conjecture that unobservable technological heterogeneity reflected in the existence of the latent classes is related to managerial performance. The findings may support the cause for raising efficiency of Japanese local public hospitals by enhancing the quality of management. Copyright © 2011 John Wiley & Sons, Ltd.

Latent class instrumental variables: A clinical and biostatistical perspective

PubMed Central

Baker, Stuart G.; Kramer, Barnett S.; Lindeman, Karen S.

2015-01-01

In some two-arm randomized trials, some participants receive the treatment assigned to the other arm as a result of technical problems, refusal of a treatment invitation, or a choice of treatment in an encouragement design. In some before-and-after studies, the availability of a new treatment changes from one time period to this next. Under assumptions that are often reasonable, the latent class instrumental variable (IV) method estimates the effect of treatment received in the aforementioned scenarios involving all-or-none compliance and all-or-none availability. Key aspects are four initial latent classes (sometimes called principal strata) based on treatment received if in each randomization group or time period, the exclusion restriction assumption (in which randomization group or time period is an instrumental variable), the monotonicity assumption (which drops an implausible latent class from the analysis), and the estimated effect of receiving treatment in one latent class (sometimes called efficacy, the local average treatment effect, or the complier average causal effect). Since its independent formulations in the biostatistics and econometrics literatures, the latent class IV method (which has no well-established name) has gained increasing popularity. We review the latent class IV method from a clinical and biostatistical perspective, focusing on underlying assumptions, methodological extensions, and applications in our fields of obstetrics and cancer research. PMID:26239275

Comparing the Performance of Improved Classify-Analyze Approaches For Distal Outcomes in Latent Profile Analysis

PubMed Central

Dziak, John J.; Bray, Bethany C.; Zhang, Jieting; Zhang, Minqiang; Lanza, Stephanie T.

2016-01-01

Several approaches are available for estimating the relationship of latent class membership to distal outcomes in latent profile analysis (LPA). A three-step approach is commonly used, but has problems with estimation bias and confidence interval coverage. Proposed improvements include the correction method of Bolck, Croon, and Hagenaars (BCH; 2004), Vermunt’s (2010) maximum likelihood (ML) approach, and the inclusive three-step approach of Bray, Lanza, & Tan (2015). These methods have been studied in the related case of latent class analysis (LCA) with categorical indicators, but not as well studied for LPA with continuous indicators. We investigated the performance of these approaches in LPA with normally distributed indicators, under different conditions of distal outcome distribution, class measurement quality, relative latent class size, and strength of association between latent class and the distal outcome. The modified BCH implemented in Latent GOLD had excellent performance. The maximum likelihood and inclusive approaches were not robust to violations of distributional assumptions. These findings broadly agree with and extend the results presented by Bakk and Vermunt (2016) in the context of LCA with categorical indicators. PMID:28630602

Applying Latent Class Analysis to Risk Stratification for Perioperative Mortality in Patients Undergoing Intraabdominal General Surgery.

PubMed

Kim, Minjae; Wall, Melanie M; Li, Guohua

2016-07-01

Perioperative risk stratification is often performed using individual risk factors without consideration of the syndemic of these risk factors. We used latent class analysis (LCA) to identify the classes of comorbidities and risk factors associated with perioperative mortality in patients presenting for intraabdominal general surgery. The 2005 to 2010 American College of Surgeons National Surgical Quality Improvement Program was used to obtain a cohort of patients undergoing intraabdominal general surgery. Risk factors and comorbidities were entered into LCA models to identify the latent classes, and individuals were assigned to a class based on the highest posterior probability of class membership. Relative risk regression was used to determine the associations between the latent classes and 30-day mortality, with adjustments for procedure. A 9-class model was fit using LCA on 466,177 observations. After combining classes with similar adjusted mortality risks, 5 risk classes were obtained. Compared with the class with average mortality risk (class 4), the risk ratios (95% confidence interval) ranged from 0.020 (0.014-0.027) in the lowest risk class (class 1) to 6.75 (6.46-7.02) in the highest risk class. After adjusting for procedure and ASA physical status, the latent classes remained significantly associated with 30-day mortality. The addition of the risk class variable to a model containing ASA physical status and surgical procedure demonstrated a significant increase in the area under the receiver operator characteristic curve (0.892 vs 0.915; P < 0.0001). Latent classes of risk factors and comorbidities in patients undergoing intraabdominal surgery are predictive of 30-day mortality independent of the ASA physical status and improve risk prediction with the ASA physical status.

A Latent Class Approach to Estimating Test-Score Reliability

ERIC Educational Resources Information Center

van der Ark, L. Andries; van der Palm, Daniel W.; Sijtsma, Klaas

2011-01-01

This study presents a general framework for single-administration reliability methods, such as Cronbach's alpha, Guttman's lambda-2, and method MS. This general framework was used to derive a new approach to estimating test-score reliability by means of the unrestricted latent class model. This new approach is the latent class reliability…

Introduction to Latent Class Analysis with Applications

ERIC Educational Resources Information Center

Porcu, Mariano; Giambona, Francesca

2017-01-01

Latent class analysis (LCA) is a statistical method used to group individuals (cases, units) into classes (categories) of an unobserved (latent) variable on the basis of the responses made on a set of nominal, ordinal, or continuous observed variables. In this article, we introduce LCA in order to demonstrate its usefulness to early adolescence…

A Latent Class Regression Analysis of Men's Conformity to Masculine Norms and Psychological Distress

ERIC Educational Resources Information Center

Wong, Y. Joel; Owen, Jesse; Shea, Munyi

2012-01-01

How are specific dimensions of masculinity related to psychological distress in specific groups of men? To address this question, the authors used latent class regression to assess the optimal number of latent classes that explained differential relationships between conformity to masculine norms and psychological distress in a racially diverse…

Regression mixture models: Does modeling the covariance between independent variables and latent classes improve the results?

PubMed Central

Lamont, Andrea E.; Vermunt, Jeroen K.; Van Horn, M. Lee

2016-01-01

Regression mixture models are increasingly used as an exploratory approach to identify heterogeneity in the effects of a predictor on an outcome. In this simulation study, we test the effects of violating an implicit assumption often made in these models – i.e., independent variables in the model are not directly related to latent classes. Results indicated that the major risk of failing to model the relationship between predictor and latent class was an increase in the probability of selecting additional latent classes and biased class proportions. Additionally, this study tests whether regression mixture models can detect a piecewise relationship between a predictor and outcome. Results suggest that these models are able to detect piecewise relations, but only when the relationship between the latent class and the predictor is included in model estimation. We illustrate the implications of making this assumption through a re-analysis of applied data examining heterogeneity in the effects of family resources on academic achievement. We compare previous results (which assumed no relation between independent variables and latent class) to the model where this assumption is lifted. Implications and analytic suggestions for conducting regression mixture based on these findings are noted. PMID:26881956

The job content questionnaire in various occupational contexts: applying a latent class model

PubMed Central

Santos, Kionna Oliveira Bernardes; de Araújo, Tânia Maria; Karasek, Robert

2017-01-01

Objective To evaluate Job Content Questionnaire(JCQ) performance using the latent class model. Methods We analysed cross-sectional studies conducted in Brazil and examined three occupational categories: petroleum industry workers (n=489), teachers (n=4392) and primary healthcare workers (3078)and 1552 urban workers from a representative sample of the city of Feira de Santana in Bahia, Brazil. An appropriate number of latent classes was extracted and described each occupational category using latent class analysis, a multivariate method that evaluates constructs and takes into account the latent characteristics underlying the structure of measurement scales. The conditional probabilities of workers belonging to each class were then analysed graphically. Results Initially, the latent class analysis extracted four classes corresponding to the four job types (active, passive, low strain and high strain) proposed by the Job-Strain model (JSM) and operationalised by the JCQ. However, after taking into consideration the adequacy criteria to evaluate the number of extracted classes, three classes (active, low strain and high strain) were extracted from the studies of urban workers and teachers and four classes (active, passive, low strain and high strain) from the study of primary healthcare and petroleum industry workers. Conclusion The four job types proposed by the JSM were identified among primary healthcare and petroleum industry workers—groups with relatively high levels of skill discretion and decision authority. Three job types were identified for teachers and urban workers; however, passive job situations were not found within these groups. The latent class analysis enabled us to describe the conditional standard responses of the job types proposed by the model, particularly in relation to active jobs and high and low strain situations. PMID:28515185

Bayesian CP Factorization of Incomplete Tensors with Automatic Rank Determination.

PubMed

Zhao, Qibin; Zhang, Liqing; Cichocki, Andrzej

2015-09-01

CANDECOMP/PARAFAC (CP) tensor factorization of incomplete data is a powerful technique for tensor completion through explicitly capturing the multilinear latent factors. The existing CP algorithms require the tensor rank to be manually specified, however, the determination of tensor rank remains a challenging problem especially for CP rank . In addition, existing approaches do not take into account uncertainty information of latent factors, as well as missing entries. To address these issues, we formulate CP factorization using a hierarchical probabilistic model and employ a fully Bayesian treatment by incorporating a sparsity-inducing prior over multiple latent factors and the appropriate hyperpriors over all hyperparameters, resulting in automatic rank determination. To learn the model, we develop an efficient deterministic Bayesian inference algorithm, which scales linearly with data size. Our method is characterized as a tuning parameter-free approach, which can effectively infer underlying multilinear factors with a low-rank constraint, while also providing predictive distributions over missing entries. Extensive simulations on synthetic data illustrate the intrinsic capability of our method to recover the ground-truth of CP rank and prevent the overfitting problem, even when a large amount of entries are missing. Moreover, the results from real-world applications, including image inpainting and facial image synthesis, demonstrate that our method outperforms state-of-the-art approaches for both tensor factorization and tensor completion in terms of predictive performance.

Bayesian Nonparametric Ordination for the Analysis of Microbial Communities.

PubMed

Ren, Boyu; Bacallado, Sergio; Favaro, Stefano; Holmes, Susan; Trippa, Lorenzo

2017-01-01

Human microbiome studies use sequencing technologies to measure the abundance of bacterial species or Operational Taxonomic Units (OTUs) in samples of biological material. Typically the data are organized in contingency tables with OTU counts across heterogeneous biological samples. In the microbial ecology community, ordination methods are frequently used to investigate latent factors or clusters that capture and describe variations of OTU counts across biological samples. It remains important to evaluate how uncertainty in estimates of each biological sample's microbial distribution propagates to ordination analyses, including visualization of clusters and projections of biological samples on low dimensional spaces. We propose a Bayesian analysis for dependent distributions to endow frequently used ordinations with estimates of uncertainty. A Bayesian nonparametric prior for dependent normalized random measures is constructed, which is marginally equivalent to the normalized generalized Gamma process, a well-known prior for nonparametric analyses. In our prior, the dependence and similarity between microbial distributions is represented by latent factors that concentrate in a low dimensional space. We use a shrinkage prior to tune the dimensionality of the latent factors. The resulting posterior samples of model parameters can be used to evaluate uncertainty in analyses routinely applied in microbiome studies. Specifically, by combining them with multivariate data analysis techniques we can visualize credible regions in ecological ordination plots. The characteristics of the proposed model are illustrated through a simulation study and applications in two microbiome datasets.

Using a web-based application to define the accuracy of diagnostic tests when the gold standard is imperfect.

PubMed

Lim, Cherry; Wannapinij, Prapass; White, Lisa; Day, Nicholas P J; Cooper, Ben S; Peacock, Sharon J; Limmathurotsakul, Direk

2013-01-01

Estimates of the sensitivity and specificity for new diagnostic tests based on evaluation against a known gold standard are imprecise when the accuracy of the gold standard is imperfect. Bayesian latent class models (LCMs) can be helpful under these circumstances, but the necessary analysis requires expertise in computational programming. Here, we describe open-access web-based applications that allow non-experts to apply Bayesian LCMs to their own data sets via a user-friendly interface. Applications for Bayesian LCMs were constructed on a web server using R and WinBUGS programs. The models provided (http://mice.tropmedres.ac) include two Bayesian LCMs: the two-tests in two-population model (Hui and Walter model) and the three-tests in one-population model (Walter and Irwig model). Both models are available with simplified and advanced interfaces. In the former, all settings for Bayesian statistics are fixed as defaults. Users input their data set into a table provided on the webpage. Disease prevalence and accuracy of diagnostic tests are then estimated using the Bayesian LCM, and provided on the web page within a few minutes. With the advanced interfaces, experienced researchers can modify all settings in the models as needed. These settings include correlation among diagnostic test results and prior distributions for all unknown parameters. The web pages provide worked examples with both models using the original data sets presented by Hui and Walter in 1980, and by Walter and Irwig in 1988. We also illustrate the utility of the advanced interface using the Walter and Irwig model on a data set from a recent melioidosis study. The results obtained from the web-based applications were comparable to those published previously. The newly developed web-based applications are open-access and provide an important new resource for researchers worldwide to evaluate new diagnostic tests.

Latent class instrumental variables: a clinical and biostatistical perspective.

PubMed

Baker, Stuart G; Kramer, Barnett S; Lindeman, Karen S

2016-01-15

In some two-arm randomized trials, some participants receive the treatment assigned to the other arm as a result of technical problems, refusal of a treatment invitation, or a choice of treatment in an encouragement design. In some before-and-after studies, the availability of a new treatment changes from one time period to this next. Under assumptions that are often reasonable, the latent class instrumental variable (IV) method estimates the effect of treatment received in the aforementioned scenarios involving all-or-none compliance and all-or-none availability. Key aspects are four initial latent classes (sometimes called principal strata) based on treatment received if in each randomization group or time period, the exclusion restriction assumption (in which randomization group or time period is an instrumental variable), the monotonicity assumption (which drops an implausible latent class from the analysis), and the estimated effect of receiving treatment in one latent class (sometimes called efficacy, the local average treatment effect, or the complier average causal effect). Since its independent formulations in the biostatistics and econometrics literatures, the latent class IV method (which has no well-established name) has gained increasing popularity. We review the latent class IV method from a clinical and biostatistical perspective, focusing on underlying assumptions, methodological extensions, and applications in our fields of obstetrics and cancer research. Copyright © 2015 John Wiley & Sons, Ltd.

Combined Patterns of Risk for Problem and Obesogenic Behaviors in Adolescents: A Latent Class Analysis Approach

ERIC Educational Resources Information Center

Fleary, Sasha A.

2017-01-01

Background: Several studies have used latent class analyses to explore obesogenic behaviors and substance use in adolescents independently. We explored a variety of health risks jointly to identify distinct patterns of risk behaviors among adolescents. Methods: Latent class models were estimated using Youth Risk Behavior Surveillance System…

Predicting the heat of vaporization of iron at high temperatures using time-resolved laser-induced incandescence and Bayesian model selection

NASA Astrophysics Data System (ADS)

Sipkens, Timothy A.; Hadwin, Paul J.; Grauer, Samuel J.; Daun, Kyle J.

2018-03-01

Competing theories have been proposed to account for how the latent heat of vaporization of liquid iron varies with temperature, but experimental confirmation remains elusive, particularly at high temperatures. We propose time-resolved laser-induced incandescence measurements on iron nanoparticles combined with Bayesian model plausibility, as a novel method for evaluating these relationships. Our approach scores the explanatory power of candidate models, accounting for parameter uncertainty, model complexity, measurement noise, and goodness-of-fit. The approach is first validated with simulated data and then applied to experimental data for iron nanoparticles in argon. Our results justify the use of Román's equation to account for the temperature dependence of the latent heat of vaporization of liquid iron.

Accuracy of Herdsmen Reporting versus Serologic Testing for Estimating Foot-and-Mouth Disease Prevalence

PubMed Central

Handel, Ian G.; Tanya, Vincent N.; Hamman, Saidou M.; Nfon, Charles; Bergman, Ingrid E.; Malirat, Viviana; Sorensen, Karl J.; Bronsvoort, Barend M. de C.

2014-01-01

Herdsman-reported disease prevalence is widely used in veterinary epidemiologic studies, especially for diseases with visible external lesions; however, the accuracy of such reports is rarely validated. Thus, we used latent class analysis in a Bayesian framework to compare sensitivity and specificity of herdsman reporting with virus neutralization testing and use of 3 nonstructural protein ELISAs for estimates of foot-and-mouth disease (FMD) prevalence on the Adamawa plateau of Cameroon in 2000. Herdsman-reported estimates in this FMD-endemic area were comparable to those obtained from serologic testing. To harness to this cost-effective resource of monitoring emerging infectious diseases, we suggest that estimates of the sensitivity and specificity of herdsmen reporting should be done in parallel with serologic surveys of other animal diseases. PMID:25417556

Latent-Trait Latent-Class Analysis of Self-Disclosure in the Work Environment

ERIC Educational Resources Information Center

Maij-de Meij, Annette M.; Kelderman, Henk; van der Flier, Henk

2005-01-01

Based on the literature about self-disclosure, it was hypothesized that different groups of subjects differ in their pattern of self-disclosure with respect to different areas of social interaction. An extended latent-trait latent-class model was proposed to describe these general patterns of self-disclosure. The model was used to analyze the data…

Mixture IRT Model with a Higher-Order Structure for Latent Traits

ERIC Educational Resources Information Center

Huang, Hung-Yu

2017-01-01

Mixture item response theory (IRT) models have been suggested as an efficient method of detecting the different response patterns derived from latent classes when developing a test. In testing situations, multiple latent traits measured by a battery of tests can exhibit a higher-order structure, and mixtures of latent classes may occur on…

Modeling Heterogeneity in Relationships between Initial Status and Rates of Change: Latent Variable Regression in a Three-Level Hierarchical Model. CSE Report 647

ERIC Educational Resources Information Center

Choi, Kilchan; Seltzer, Michael

2005-01-01

In studies of change in education and numerous other fields, interest often centers on how differences in the status of individuals at the start of a time period of substantive interest relate to differences in subsequent change. This report presents a fully Bayesian approach to estimating three-level hierarchical models in which latent variable…

Bayesian Estimation in the One-Parameter Latent Trait Model.

DTIC Science & Technology

1980-03-01

Journal of Mathematical and Statistical Psychology , 1973, 26, 31-44. (a) Andersen, E. B. A goodness of fit test for the Rasch model. Psychometrika, 1973, 28...technique for estimating latent trait mental test parameters. Educational and Psychological Measurement, 1976, 36, 705-715. Lindley, D. V. The...Lord, F. M. An analysis of verbal Scholastic Aptitude Test using Birnbaum’s three-parameter logistic model. Educational and Psychological

Exploring heterogeneity in clinical trials with latent class analysis

PubMed Central

Abarda, Abdallah; Contractor, Ateka A.; Wang, Juan; Dayton, C. Mitchell

2018-01-01

Case-mix is common in clinical trials and treatment effect can vary across different subgroups. Conventionally, a subgroup analysis is performed by dividing the overall study population by one or two grouping variables. It is usually impossible to explore complex high-order intersections among confounding variables. Latent class analysis (LCA) provides a framework to identify latent classes by observed manifest variables. Distal clinical outcomes and treatment effect can be different across these classes. This paper provides a step-by-step tutorial on how to perform LCA with R. A simulated dataset is generated to illustrate the process. In the example, the classify-analyze approach is employed to explore the differential treatment effects on distal outcomes across latent classes. PMID:29955579

A latent transition analysis of bullying and victimization in Chinese primary school students

PubMed Central

Lau, Puiyi; Luo, Fang

2017-01-01

Bullying is a social phenomenon that impacts a large number of children and young people, worldwide. This study aimed to longitudinally examine the development of bullying and victimization in Chinese students in grades 4, 5, and 6. We used latent class analysis to empirically identify groups of youth with different bullying and victimization patterns, and then used latent transition analysis to explore the movement of children between these latent classes over time. Results showed that: (1) across the three time points, students could be classified into four classes: bullies, victims, bully-victims, and non-involved children; and (2) students in the non-involved class tended to remain in that class when moving to higher grades, students in the bully and victims classes tended to transition to the non-involved class, while students in the bully-victims class tended to transition to the bullies class. Thus, future intervention should be implemented to prevent bully-victims from bullying behaviors. PMID:28837571

Growth Modeling with Non-Ignorable Dropout: Alternative Analyses of the STAR*D Antidepressant Trial

PubMed Central

Muthén, Bengt; Asparouhov, Tihomir; Hunter, Aimee; Leuchter, Andrew

2011-01-01

This paper uses a general latent variable framework to study a series of models for non-ignorable missingness due to dropout. Non-ignorable missing data modeling acknowledges that missingness may depend on not only covariates and observed outcomes at previous time points as with the standard missing at random (MAR) assumption, but also on latent variables such as values that would have been observed (missing outcomes), developmental trends (growth factors), and qualitatively different types of development (latent trajectory classes). These alternative predictors of missing data can be explored in a general latent variable framework using the Mplus program. A flexible new model uses an extended pattern-mixture approach where missingness is a function of latent dropout classes in combination with growth mixture modeling using latent trajectory classes. A new selection model allows not only an influence of the outcomes on missingness, but allows this influence to vary across latent trajectory classes. Recommendations are given for choosing models. The missing data models are applied to longitudinal data from STAR*D, the largest antidepressant clinical trial in the U.S. to date. Despite the importance of this trial, STAR*D growth model analyses using non-ignorable missing data techniques have not been explored until now. The STAR*D data are shown to feature distinct trajectory classes, including a low class corresponding to substantial improvement in depression, a minority class with a U-shaped curve corresponding to transient improvement, and a high class corresponding to no improvement. The analyses provide a new way to assess drug efficiency in the presence of dropout. PMID:21381817

A Latent Class Analysis of Dyadic Perfectionism in a College Sample

ERIC Educational Resources Information Center

Lopez, Frederick G.; Fons-Scheyd, Alia; Bush-King, Imelda; McDermott, Ryon C.

2011-01-01

A latent class analysis of dyadic perfectionism scores within a college sample (N = 369) identified four classes of participants. Controlling for gender and current dating status, class membership was associated with significant differences on several measures of relationship attitudes. Gender and class membership also significantly interacted in…

Sexual Behavior Latent Classes Among Men Who Have Sex With Men: Associations With Sexually Transmitted Infections.

PubMed

Rice, Cara E; Norris Turner, Abigail; Lanza, Stephanie T

2017-01-01

Men who have sex with men (MSM) are at disproportionate risk of acquisition of sexually transmitted infections (STIs). We used latent class analysis (LCA) to examine patterns of sexual behavior among MSM and how those patterns are related to STIs. We examined patterns of sexual behavior using behavioral and clinical data from a cross-sectional study of 235 MSM who presented to an urban sexual health clinic for STI testing. Analyzed data were collected using a combination of interviewer- and self-administered surveys and electronic health records. We used LCA to identify underlying subgroups of men based on their sexual behavior, described the demographics of the latent classes, and examined the association between the latent classes and STI status. We identified three latent classes of sexual behavior: Unprotected Anal Intercourse (UAI) Only (67%), Partner Seekers (14%), and Multiple Behaviors (19%). Men in the Multiple Behaviors class had a 67% probability of being STI positive, followed by men in the UAI Only class (27%) and men in the Partner Seekers class (22%). Examining the intersection of a variety of sexual practices indicates particular subgroups of MSM have the highest probability of being STI positive.

Bayesian Regression with Network Prior: Optimal Bayesian Filtering Perspective

PubMed Central

Qian, Xiaoning; Dougherty, Edward R.

2017-01-01

The recently introduced intrinsically Bayesian robust filter (IBRF) provides fully optimal filtering relative to a prior distribution over an uncertainty class ofjoint random process models, whereas formerly the theory was limited to model-constrained Bayesian robust filters, for which optimization was limited to the filters that are optimal for models in the uncertainty class. This paper extends the IBRF theory to the situation where there are both a prior on the uncertainty class and sample data. The result is optimal Bayesian filtering (OBF), where optimality is relative to the posterior distribution derived from the prior and the data. The IBRF theories for effective characteristics and canonical expansions extend to the OBF setting. A salient focus of the present work is to demonstrate the advantages of Bayesian regression within the OBF setting over the classical Bayesian approach in the context otlinear Gaussian models. PMID:28824268

Economic considerations and patients' preferences affect treatment selection for patients with rheumatoid arthritis: a discrete choice experiment among European rheumatologists.

PubMed

Hifinger, M; Hiligsmann, M; Ramiro, S; Watson, V; Severens, J L; Fautrel, B; Uhlig, T; van Vollenhoven, R; Jacques, P; Detert, J; Canas da Silva, J; Scirè, C A; Berghea, F; Carmona, L; Péntek, M; Keat, A; Boonen, A

2017-01-01

To compare the value that rheumatologists across Europe attach to patients' preferences and economic aspects when choosing treatments for patients with rheumatoid arthritis. In a discrete choice experiment, European rheumatologists chose between two hypothetical drug treatments for a patient with moderate disease activity. Treatments differed in five attributes: efficacy (improvement and achieved state on disease activity), safety (probability of serious adverse events), patient's preference (level of agreement), medication costs and cost-effectiveness (incremental cost-effectiveness ratio (ICER)). A Bayesian efficient design defined 14 choice sets, and a random parameter logit model was used to estimate relative preferences for rheumatologists across countries. Cluster analyses and latent class models were applied to understand preference patterns across countries and among individual rheumatologists. Responses of 559 rheumatologists from 12 European countries were included in the analysis (49% females, mean age 48 years). In all countries, efficacy dominated treatment decisions followed by economic considerations and patients' preferences. Across countries, rheumatologists avoided selecting a treatment that patients disliked. Latent class models revealed four respondent profiles: one traded off all attributes except safety, and the remaining three classes disregarded ICER. Among individual rheumatologists, 57% disregarded ICER and these were more likely from Italy, Romania, Portugal or France, whereas 43% disregarded uncommon/rare side effects and were more likely from Belgium, Germany, Hungary, the Netherlands, Norway, Spain, Sweden or UK. Overall, European rheumatologists are willing to trade between treatment efficacy, patients' treatment preferences and economic considerations. However, the degree of trade-off differs between countries and among individuals. Published by the BMJ Publishing Group Limited. For permission to use (where not already granted under a licence) please go to http://www.bmj.com/company/products-services/rights-and-licensing/.

Bayesian estimation of the sensitivity and specificity of individual fecal culture and Paralisa to detect Mycobacterium avium subspecies paratuberculosis infection in young farmed deer.

PubMed

Stringer, Lesley A; Jones, Geoff; Jewell, Chris P; Noble, Alasdair D; Heuer, Cord; Wilson, Peter R; Johnson, Wesley O

2013-11-01

A Bayesian latent class model was used to estimate the sensitivity and specificity of an immunoglobulin G1 serum enzyme-linked immunosorbent assay (Paralisa) and individual fecal culture to detect young deer infected with Mycobacterium avium subsp. paratuberculosis. Paired fecal and serum samples were collected, between July 2009 and April 2010, from 20 individual yearling (12-24-month-old) deer in each of 20 South Island and 18 North Island herds in New Zealand and subjected to culture and Paralisa, respectively. Two fecal samples and 16 serum samples from 356 North Island deer, and 55 fecal and 37 serum samples from 401 South Island deer, were positive. The estimate of individual fecal culture sensitivity was 77% (95% credible interval [CI] = 61-92%) with specificity of 99% (95% CI = 98-99.7%). The Paralisa sensitivity estimate was 19% (95% CI = 10-30%), with specificity of 94% (95% CI = 93-96%). All estimates were robust to variation of priors and assumptions tested in a sensitivity analysis. The data informs the use of the tests in determining infection status at the individual and herd level.

Evaluation of fecal culture and fecal RT-PCR to detect Mycobacterium avium ssp. paratuberculosis fecal shedding in dairy goats and dairy sheep using latent class Bayesian modeling.

PubMed

Bauman, Cathy A; Jones-Bitton, Andria; Jansen, Jocelyn; Kelton, David; Menzies, Paula

2016-09-20

The study's objective was to evaluate the ability of fecal culture (FCUL) and fecal PCR (FPCR) to identify dairy goat and dairy sheep shedding Mycobacterium avium ssp. paratuberculosis. A cross-sectional study of the small ruminant populations was performed in Ontario, Canada between October 2010 and August 2011. Twenty-nine dairy goat herds and 21 dairy sheep flocks were visited, and 20 lactating females > two years of age were randomly selected from each farm resulting in 580 goats and 397 sheep participating in the study. Feces were collected per rectum and cultured using the BD BACTEC™ MGIT™ 960 system using a standard (49 days) and an extended (240 days) incubation time, and underwent RT-PCR based on the hsp-X gene (Tetracore®). Statistical analysis was performed using a 2-test latent class Bayesian hierarchical model for each species fitted in WinBUGS. Extending the fecal culture incubation time statistically improved FCUL sensitivity from 23.1 % (95 % PI: 15.9-34.1) to 42.7 % (95 % PI: 33.0-54.5) in dairy goats and from 5.8 % (95 % PI: 2.3-12.4) to 19.0 % (95 % PI: 11.9-28.9) in dairy sheep. FPCR demonstrated statistically higher sensitivity than FCUL (49 day incubation) with a sensitivity of 31.9 % (95 % PI: 22.4-43.1) in goats and 42.6 % (95 % PI: 28.8-63.3) in sheep. Fecal culture demonstrates such low sensitivity at the standard incubation time it cannot be recommended as a screening test to detect shedding of MAP in either goats or sheep. Extending the incubation time resulted in improved sensitivity; however, it is still disappointingly low for screening purposes. Fecal PCR should be the screening test of choice in both species; however, it is important to recognize that control programs should not be based on testing alone when they demonstrate such low sensitivity.

Comparison of the performance of three PCR assays for the detection and differentiation of Theileria orientalis genotypes.

PubMed

Perera, Piyumali K; Gasser, Robin B; Pulford, David J; Stevenson, Mark A; Firestone, Simon M; McFadden, Andrew M J; Jabbar, Abdul

2015-03-31

Oriental theileriosis is a tick-borne disease of bovines caused by the members of the Theileria orientalis complex. Recently, we developed a multiplexed tandem (MT) PCR to detect, differentiate and quantitate four genotypes (i.e., buffeli, chitose, ikeda and type 5) of T. orientalis. In this study, we used MT PCR to assess the prevalence and infection intensity of four T. orientalis genotypes in selected cattle herds that experienced oriental theileriosis outbreaks in New Zealand, and compared the sensitivities and specificities of MT PCR, PCR-high resolution melting (PCR-HRM) and a TaqMan qPCR. MT PCR, PCR-HRM analysis for T. orientalis and a TaqMan qPCR assay for ikeda genotype were employed to test 154 and 88 cattle blood samples from North (where oriental theileriosis outbreaks had occurred; designated as Group 1) and South (where no outbreaks had been reported; Group 2) Islands of New Zealand, respectively. Quantitative data from MT PCR assay were analyzed using generalized linear model and paired-sample t-test. The diagnostic specificity and sensitivity of the assays were estimated using a Bayesian latent class modeling approach. In Group 1, 99.4% (153/154) of cattle were test-positive for T. orientalis in both the MT PCR and PCR-HRM assays. The apparent prevalences of genotype ikeda in Group 1 were 87.6% (134/153) and 87.7% (135/154) using the MT PCR and Ikeda TaqMan qPCR assays, respectively. Using the MT PCR test, all four genotypes of T. orientalis were detected. The infection intensity estimated for genotype ikeda was significantly higher (P = 0.009) in severely anaemic cattle than in those without anaemia, and this intensity was significantly higher than that of buffeli (P < 0.001) in the former cattle. Bayesian latent class analysis showed that the diagnostic sensitivities (97.1-98.9%) and specificities (96.5-98.9%) of the three PCR assays were very comparable. The present findings show the advantages of using the MT PCR assay as a useful tool for in-depth epidemiological and transmission studies of T. orientalis worldwide.

The job content questionnaire in various occupational contexts: applying a latent class model.

PubMed

Santos, Kionna Oliveira Bernardes; Araújo, Tânia Maria de; Carvalho, Fernando Martins; Karasek, Robert

2017-05-17

To evaluate Job Content Questionnaire(JCQ) performance using the latent class model. We analysed cross-sectional studies conducted in Brazil and examined three occupational categories: petroleum industry workers (n=489), teachers (n=4392) and primary healthcare workers (3078)and 1552 urban workers from a representative sample of the city of Feira de Santana in Bahia, Brazil. An appropriate number of latent classes was extracted and described each occupational category using latent class analysis, a multivariate method that evaluates constructs and takes into accountthe latent characteristics underlying the structure of measurement scales. The conditional probabilities of workers belonging to each class were then analysed graphically. Initially, the latent class analysis extracted four classes corresponding to the four job types (active, passive, low strain and high strain) proposed by the Job-Strain model (JSM) and operationalised by the JCQ. However, after taking into consideration the adequacy criteria to evaluate the number of extracted classes, three classes (active, low strain and high strain) were extracted from the studies of urban workers and teachers and four classes (active, passive, low strain and high strain) from the study of primary healthcare and petroleum industry workers. The four job types proposed by the JSM were identified among primary healthcare and petroleum industry workers-groups with relatively high levels of skill discretion and decision authority. Three job types were identified for teachers and urban workers; however, passive job situations were not found within these groups. The latent class analysis enabled us to describe the conditional standard responses of the job types proposed by the model, particularly in relation to active jobs and high and low strain situations. © Article author(s) (or their employer(s) unless otherwise stated in the text of the article) 2017. All rights reserved. No commercial use is permitted unless otherwise expressly granted.

Multimethod latent class analysis

PubMed Central

Nussbeck, Fridtjof W.; Eid, Michael

2015-01-01

Correct and, hence, valid classifications of individuals are of high importance in the social sciences as these classifications are the basis for diagnoses and/or the assignment to a treatment. The via regia to inspect the validity of psychological ratings is the multitrait-multimethod (MTMM) approach. First, a latent variable model for the analysis of rater agreement (latent rater agreement model) will be presented that allows for the analysis of convergent validity between different measurement approaches (e.g., raters). Models of rater agreement are transferred to the level of latent variables. Second, the latent rater agreement model will be extended to a more informative MTMM latent class model. This model allows for estimating (i) the convergence of ratings, (ii) method biases in terms of differential latent distributions of raters and differential associations of categorizations within raters (specific rater bias), and (iii) the distinguishability of categories indicating if categories are satisfyingly distinct from each other. Finally, an empirical application is presented to exemplify the interpretation of the MTMM latent class model. PMID:26441714

Validation of Diagnostic Measures Based on Latent Class Analysis: A Step Forward in Response Bias Research

ERIC Educational Resources Information Center

Thomas, Michael L.; Lanyon, Richard I.; Millsap, Roger E.

2009-01-01

The use of criterion group validation is hindered by the difficulty of classifying individuals on latent constructs. Latent class analysis (LCA) is a method that can be used for determining the validity of scales meant to assess latent constructs without such a priori classifications. The authors used this method to examine the ability of the L…

Caregiver profiles in dementia related to quality of life, depression and perseverance time in the European Actifcare study: the importance of social health.

PubMed

Janssen, Eveline P C J; de Vugt, Marjolein; Köhler, Sebastian; Wolfs, Claire; Kerpershoek, Liselot; Handels, Ron L H; Orrell, Martin; Woods, Bob; Jelley, Hannah; Stephan, Astrid; Bieber, Anja; Meyer, Gabriele; Engedal, Knut; Selbaek, Geir; Wimo, Anders; Irving, Kate; Hopper, Louise; Gonçalves-Pereira, Manuel; Portolani, Elisa; Zanetti, Orazio; Verhey, Frans R

2017-01-01

To identify caregiver profiles of persons with mild to moderate dementia and to investigate differences between identified caregiver profiles, using baseline data of the international prospective cohort study Actifcare. A latent class analysis was used to discover different caregiver profiles based on disease related characteristics of 453 persons with dementia and their 453 informal caregivers. These profiles were compared with regard to quality of life (CarerQoL score), depressive symptoms (HADS-D score) and perseverance time. A 5-class model was identified, with the best Bayesian Information Criterion value, significant likelihood ratio test (p < 0.001), high entropy score (0.88) and substantive interpretability. The classes could be differentiated on two axes: (i) caregivers' age, relationship with persons with dementia, severity of dementia, and (ii) tendency towards stress and difficulty adapting to stress. Classes showed significant differences with all dependent variables, and were labelled 'older low strain', 'older intermediate strain', 'older high strain', 'younger low strain' and 'younger high strain'. Differences exist between types of caregivers that explain variability in quality of life, depressive symptoms and perseverance time. Our findings may give direction for tailored interventions for caregivers of persons with dementia, which may improve social health and reduce health care costs.

Studying Psychosocial Barriers to Drug Treatment Among Chinese Methamphetamine Users Using A 3-Step Latent Class Analysis.

PubMed

Wang, Jichuan; Kelly, Brian C; Liu, Tieqiao; Hao, Wei

2016-03-01

Given the growth in methamphetamine use in China during the 21st century, we assessed perceived psychosocial barriers to drug treatment among this population. Using a sample of 303 methamphetamine users recruited via Respondent Driven Sampling, we use Latent Class Analysis (LCA) to identify possible distinct latent groups among Chinese methamphetamine users on the basis of their perceptions of psychosocial barriers to drug treatment. After covariates were included to predict latent class membership, the 3-step modeling approach was applied. Our findings indicate that the Chinese methamphetamine using population was heterogeneous on perceptions of drug treatment barriers; four distinct latent classes (subpopulations) were identified--Unsupported Deniers, Deniers, Privacy Anxious, and Low Barriers--and individual characteristics shaped the probability of class membership. Efforts to link Chinese methamphetamine users to treatment may require a multi-faceted approach that attends to differing perceptions about impediments to drug treatment. Copyright © 2015. Published by Elsevier Inc.

A Latent Class Analysis of Maternal Depressive Symptoms over 12 Years and Offspring Adjustment in Adolescence

PubMed Central

Campbell, Susan B.; Morgan-Lopez, Antonio A.; Cox, Martha J.; McLoyd, Vonnie C.

2009-01-01

We used data from the NICHD Study of Early Child Care and Youth Development and latent class analysis to model patterns of maternal depressive symptoms from infant age 1 month to the transition to adolescence (age 12), and then examined adolescent adjustment at age 15 as a function of the course and severity of maternal symptoms. We identified five latent classes of symptoms in these 1357 women while also taking into account sociodemographic measures: never depressed; stable subclinical; early-decreasing; moderately elevated; chronic. Women with few symptoms were more likely to be married, better educated, and in better physical health than women with more elevated symptoms. Family size and whether the pregnancy was planned also differentiated among classes. At age 15, adolescents whose mothers were in the chronic, elevated, and stable subclinical latent classes reported more internalizing and externalizing problems and acknowledged engaging in more risky behavior than did children of never-depressed mothers. Latent class differences in self-reported loneliness and dysphoria were also found. Finally, several significant interactions between sex and latent class suggested that girls whose mothers reported elevated symptoms of depression over time experienced more internalizing distress and dysphoric mood relative to their male counterparts. Discussion focuses on adolescent adjustment, especially among offspring whose mothers report stable symptoms of depression across their childhoods. PMID:19685946

Class Evolution Tree: A Graphical Tool to Support Decisions on the Number of Classes in Exploratory Categorical Latent Variable Modeling for Rehabilitation Research

ERIC Educational Resources Information Center

Kriston, Levente; Melchior, Hanne; Hergert, Anika; Bergelt, Corinna; Watzke, Birgit; Schulz, Holger; von Wolff, Alessa

2011-01-01

The aim of our study was to develop a graphical tool that can be used in addition to standard statistical criteria to support decisions on the number of classes in explorative categorical latent variable modeling for rehabilitation research. Data from two rehabilitation research projects were used. In the first study, a latent profile analysis was…

Latent class analysis of real time qPCR and bacteriological culturing for the diagnosis of Streptococcus agalactiae in cow composite milk samples.

PubMed

Holmøy, Ingrid H; Toft, Nils; Jørgensen, Hannah J; Mørk, Tormod; Sølverød, Liv; Nødtvedt, Ane

2018-06-01

Streptococcus agalactiae (S. agalactiae) has re-emerged as a mastitis pathogen among Norwegian dairy cows. The Norwegian cattle health services recommend that infected herds implement measures to eradicate S. agalactiae, this includes a screening of milk samples from all lactating cows. The performance of the qPCR-test currently in use for this purpose has not been evaluated under field conditions. The objective of this study was to estimate the sensitivity and specificity of the real-time qPCR assay in use in Norway (Mastitis 4 qPCR, DNA Diagnostics A/S, Risskov, Denmark) and compare it to conventional bacteriological culturing for detection of S. agalactiae in milk samples. Because none of these tests are considered a perfect reference test, the evaluation was performed using latent class models in a Bayesian analysis. Aseptically collected cow-composite milk samples from 578 cows belonging to 6 herds were cultured and tested by qPCR. While 37 (6.4%) samples were positive for S. agalactiae by bacteriological culture, 66 (11.4%) samples were positive by qPCR. The within-herd prevalence in the six herds, as estimated by the latent class models ranged from 7.7 to 50.8%. At the recommended cut-off (cycle threshold 37), the sensitivity of the qPCR was significantly higher at 95.3 (95% posterior probability interval [PPI] [84.2; 99.6]) than that of bacteriological culture at 58.2 (95% PPI [43.8; 74.4]). However, bacterial culture had a higher specificity of 99.7 (95% PPI [98.5; 100.0]) compared to the qPCR at 98.5 (95% PPI [94.6; 99.9]). The median estimated negative predictive values of qPCR was consistently higher than those of the BC at all estimated prevalences, and the superiority of the qPCR increased with increasing within-herd prevalence. The median positive predictive values of BC was in general higher than the estimates for the qPCR, however, at the highest prevalence the predictive ability of both tests were similar. Copyright © 2018 Elsevier B.V. All rights reserved.

A latent transition model of the effects of a teen dating violence prevention initiative.

PubMed

Williams, Jason; Miller, Shari; Cutbush, Stacey; Gibbs, Deborah; Clinton-Sherrod, Monique; Jones, Sarah

2015-02-01

Patterns of physical and psychological teen dating violence (TDV) perpetration, victimization, and related behaviors were examined with data from the evaluation of the Start Strong: Building Healthy Teen Relationships initiative, a dating violence primary prevention program targeting middle school students. Latent class and latent transition models were used to estimate distinct patterns of TDV and related behaviors of bullying and sexual harassment in seventh grade students at baseline and to estimate transition probabilities from one pattern of behavior to another at the 1-year follow-up. Intervention effects were estimated by conditioning transitions on exposure to Start Strong. Latent class analyses suggested four classes best captured patterns of these interrelated behaviors. Classes were characterized by elevated perpetration and victimization on most behaviors (the multiproblem class), bullying perpetration/victimization and sexual harassment victimization (the bully-harassment victimization class), bullying perpetration/victimization and psychological TDV victimization (bully-psychological victimization), and experience of bully victimization (bully victimization). Latent transition models indicated greater stability of class membership in the comparison group. Intervention students were less likely to transition to the most problematic pattern and more likely to transition to the least problem class. Although Start Strong has not been found to significantly change TDV, alternative evaluation models may find important differences. Latent transition analysis models suggest positive intervention impact, especially for the transitions at the most and the least positive end of the spectrum. Copyright © 2015. Published by Elsevier Inc.

Substance Use Profiles of Urban American Indian Adolescents: A Latent Class Analysis.

PubMed

Kulis, Stephen S; Jager, Justin; Ayers, Stephanie L; Lateef, Husain; Kiehne, Elizabeth

2016-07-28

A growing majority of American Indian adolescents now live in cities and are at high risk of early and problematic substance use and its negative health effects. This study used latent class analysis to empirically derive heterogeneous patterns of substance use among urban American Indian adolescents, examined demographic correlates of the resulting latent classes, and tested for differences among the latent classes in other risk behavior and prosocial outcomes. The study employed a representative sample of 8th, 10th, and 12th grade American Indian adolescents (n = 2,407) in public or charter schools in metropolitan areas of Arizona in 2012. Latent class analysis examined eight types of last 30 day substance use. Four latent classes emerged: a large group of "nonusers" (69%); a substantial minority using alcohol, tobacco, and/or marijuana [ATM] (17%); a smaller group of polysubstance users consuming, alcohol, tobacco, marijuana, other illicit drugs, and prescription or OTC drugs in combination (6%); and a "not alcohol" group reporting combinations of tobacco, marijuana, and prescription drug use, but rarely alcohol use (4%). The latent classes varied by age and grade level, but not by other demographic characteristics, and aligned in highly consistent patterns on other non-substance use outcomes. Polysubstance users reported the most problematic and nonusers the least problematic outcomes, with ATM and "not alcohol" users in the middle. Urban AI adolescent substance use occurs in three somewhat distinctive patterns of combinations of recent alcohol and drug consumption, covarying in systematic ways with other problematic risk behaviors and attitudes.

Discordant inflammation and pain in early and established rheumatoid arthritis: Latent Class Analysis of Early Rheumatoid Arthritis Network and British Society for Rheumatology Biologics Register data.

PubMed

McWilliams, Daniel F; Ferguson, Eamonn; Young, Adam; Kiely, Patrick D W; Walsh, David A

2016-12-13

Rheumatoid arthritis (RA) disease activity is often measured using the 28-joint Disease Activity Score (DAS28). We aimed to identify and independently verify subgroups of people with RA that may be discordant with respect to self-reported and objective disease state, with potentially different clinical needs. Data were derived from three cohorts: (1) the Early Rheumatoid Arthritis Network (ERAN) and the British Society for Rheumatology Biologics Register (BSRBR), (2) those commencing tumour necrosis factor (TNF)-α inhibitors and (3) those using non-biologic drugs. In latent class analysis, we used variables related to pain, central pain mechanisms or inflammation (pain, vitality, mental health, erythrocyte sedimentation rate, swollen joint count, tender joint count, visual analogue scale of general health). Clinically relevant outcomes were examined. Five, four and four latent classes were found in the ERAN, BSRBR TNF inhibitor and non-biologic cohorts, respectively. The proportions of people assigned with >80% probability into latent classes were 76%, 58% and 72% in the ERAN, TNF inhibitor and non-biologic cohorts, respectively. The latent classes displayed either concordance between measures indicative of mild, moderate or severe disease activity; discordantly worse patient-reported measures despite less markedly elevated inflammation; or discordantly less severe patient-reported measures despite elevated inflammation. Latent classes with discordantly worse patient-reported measures represented 12%, 40% and 21% of the ERAN, TNF inhibitor and non-biologic cohorts, respectively; contained more females; and showed worse function. In those latent classes with worse scores at baseline, DAS28 and function improved over 1 year (p < 0.001 for all comparisons), and scores differed less at follow-up than at baseline. Discordant latent classes can be identified in people with RA, and these findings are robust across three cohorts with varying disease duration and activity. These findings could be used to identify a sizeable subgroup of people with RA who might gain added benefit from pain management strategies.

The Latent Class Structure of Chinese Patients with Eating Disorders in Shanghai.

PubMed

Zheng, Yuchen; Kang, Qing; Huang, Jiabin; Jiang, Wenhui; Liu, Qiang; Chen, Han; Fan, Qing; Wang, Zhen; Chen, Jue; Xiao, Zeping

2017-08-25

Eating disorder is culture related, and the clinical symptoms are different between eastern and western patients. So the validity of feeding and eating disorders in the upcoming ICD-11 guide for Chinese patients is unclear. To explore the latent class structure of Chinese patients with eating disorder and the cross-cultural validity of the eating disorder section of the new ICD-11 guide in China. A total of 379 patients with eating disorders at Shanghai Mental Health Center were evaluated using the EDI questionnaire and a questionnaire developed by researchers from 2010 to 2016. SPSS 20.0 was used to enter data and analyze demographic data, and Latent GOLD was employed to conduct latent profile analysis. According to the results of latent profile analysis, patients with eating disorder were divided into five classes: low-weight fasting class (23.1%), non-fat-phobic binge/purge class (21.54%), low-fat-phobic binge class (19.27%), fat-phobic binge class (19.27%), and non-fat-phobic low-weight class (16.76%). Among the clinical symptoms extracted, there were significant differences in Body Mass Index (BMI), binge eating behavior, self-induced vomiting, laxative use and fat-phobic opinion; while there was no significant difference in restrictive food intake. Based on the clinical symptoms, there are five latent classes in Chinese patients with eating disorder, which is in accordance with the diagnostic categories of feeding and eating disorder in ICD-11. However, further work is needed in improving the fat-phobic opinion of patients with eating disorder and clarifying the BMI standard of thinness in the Chinese population.

Spurious Latent Classes in the Mixture Rasch Model

ERIC Educational Resources Information Center

Alexeev, Natalia; Templin, Jonathan; Cohen, Allan S.

2011-01-01

Mixture Rasch models have been used to study a number of psychometric issues such as goodness of fit, response strategy differences, strategy shifts, and multidimensionality. Although these models offer the potential for improving understanding of the latent variables being measured, under some conditions overextraction of latent classes may…

Cross-Informant Agreement on Child and Adolescent Withdrawn Behavior: A Latent Class Approach

ERIC Educational Resources Information Center

Rubin, David H.; Althoff, Robert R.; Walkup, John T.; Hudziak, James J.

2013-01-01

Withdrawn behavior (WB) relates to many developmental outcomes, including pervasive developmental disorders, anxiety, depression, psychosis, personality disorders and suicide. No study has compared the latent profiles of different informants' reports on WB. This study uses multi-informant latent class analyses (LCA) of the child behavior checklist…

The effects of rurality on substance use disorder diagnosis: A multiple-groups latent class analysis.

PubMed

Brooks, Billy; McBee, Matthew; Pack, Robert; Alamian, Arsham

2017-05-01

Rates of accidental overdose mortality from substance use disorder (SUD) have risen dramatically in the United States since 1990. Between 1999 and 2004 alone rates increased 62% nationwide, with rural overdose mortality increasing at a rate 3 times that seen in urban populations. Cultural differences between rural and urban populations (e.g., educational attainment, unemployment rates, social characteristics, etc.) affect the nature of SUD, leading to disparate risk of overdose across these communities. Multiple-groups latent class analysis with covariates was applied to data from the 2011 and 2012 National Survey on Drug Use and Health (n=12.140) to examine potential differences in latent classifications of SUD between rural and urban adult (aged 18years and older) populations. Nine drug categories were used to identify latent classes of SUD defined by probability of diagnosis within these categories. Once the class structures were established for rural and urban samples, posterior membership probabilities were entered into a multinomial regression analysis of socio-demographic predictors' association with the likelihood of SUD latent class membership. Latent class structures differed across the sub-groups, with the rural sample fitting a 3-class structure (Bootstrap Likelihood Ratio Test P value=0.03) and the urban fitting a 6-class model (Bootstrap Likelihood Ratio Test P value<0.0001). Overall the rural class structure exhibited less diversity in class structure and lower prevalence of SUD in multiple drug categories (e.g. cocaine, hallucinogens, and stimulants). This result supports the hypothesis that different underlying elements exist in the two populations that affect SUD patterns, and thus can inform the development of surveillance instruments, clinical services, and prevention programming tailored to specific communities. Copyright © 2017 Elsevier Ltd. All rights reserved.

Understanding the Heterogeneity of BPD Symptoms through Latent Class Analysis: Initial Results and Clinical Correlates among Inner-City Substance Users

ERIC Educational Resources Information Center

Bornovalova, Marina A.; Levy, Roy; Gratz, Kim L.; Lejuez, C. W.

2010-01-01

The current study investigated the heterogeneity of borderline personality disorder (BPD) symptoms in a sample of 382 inner-city, predominantly African American male substance users through the use of latent class analysis. A 4-class model was statistically preferred, with 1 class interpreted to be a baseline class, 1 class interpreted to be a…

Using Bayesian statistics for modeling PTSD through Latent Growth Mixture Modeling: implementation and discussion.

PubMed

Depaoli, Sarah; van de Schoot, Rens; van Loey, Nancy; Sijbrandij, Marit

2015-01-01

After traumatic events, such as disaster, war trauma, and injuries including burns (which is the focus here), the risk to develop posttraumatic stress disorder (PTSD) is approximately 10% (Breslau & Davis, 1992). Latent Growth Mixture Modeling can be used to classify individuals into distinct groups exhibiting different patterns of PTSD (Galatzer-Levy, 2015). Currently, empirical evidence points to four distinct trajectories of PTSD patterns in those who have experienced burn trauma. These trajectories are labeled as: resilient, recovery, chronic, and delayed onset trajectories (e.g., Bonanno, 2004; Bonanno, Brewin, Kaniasty, & Greca, 2010; Maercker, Gäbler, O'Neil, Schützwohl, & Müller, 2013; Pietrzak et al., 2013). The delayed onset trajectory affects only a small group of individuals, that is, about 4-5% (O'Donnell, Elliott, Lau, & Creamer, 2007). In addition to its low frequency, the later onset of this trajectory may contribute to the fact that these individuals can be easily overlooked by professionals. In this special symposium on Estimating PTSD trajectories (Van de Schoot, 2015a), we illustrate how to properly identify this small group of individuals through the Bayesian estimation framework using previous knowledge through priors (see, e.g., Depaoli & Boyajian, 2014; Van de Schoot, Broere, Perryck, Zondervan-Zwijnenburg, & Van Loey, 2015). We used latent growth mixture modeling (LGMM) (Van de Schoot, 2015b) to estimate PTSD trajectories across 4 years that followed a traumatic burn. We demonstrate and compare results from traditional (maximum likelihood) and Bayesian estimation using priors (see, Depaoli, 2012, 2013). Further, we discuss where priors come from and how to define them in the estimation process. We demonstrate that only the Bayesian approach results in the desired theory-driven solution of PTSD trajectories. Since the priors are chosen subjectively, we also present a sensitivity analysis of the Bayesian results to illustrate how to check the impact of the prior knowledge integrated into the model. We conclude with recommendations and guidelines for researchers looking to implement theory-driven LGMM, and we tailor this discussion to the context of PTSD research.

Precipitation and Latent Heating Distributions from Satellite Passive Microwave Radiometry. Part 1; Improved Method and Uncertainties

NASA Technical Reports Server (NTRS)

Olson, William S.; Kummerow, Christian D.; Yang, Song; Petty, Grant W.; Tao, Wei-Kuo; Bell, Thomas L.; Braun, Scott A.; Wang, Yansen; Lang, Stephen E.; Johnson, Daniel E.;

A Tutorial Introduction to Bayesian Models of Cognitive Development

DTIC Science & Technology

2011-01-01

typewriter with an infinite amount of paper. There is a space of documents that it is capable of producing, which includes things like The Tempest and does...not include, say, a Vermeer painting or a poem written in Russian. This typewriter represents a means of generating the hypothesis space for a Bayesian...learner: each possible document that can be typed on it is a hypothesis, the infinite set of documents producible by the typewriter is the latent

Discontinuous Patterns of Cigarette Smoking From Ages 18 to 50 in the United States: A Repeated-Measures Latent Class Analysis.

PubMed

Terry-McElrath, Yvonne M; O'Malley, Patrick M; Johnston, Lloyd D

2017-12-13

Effective cigarette smoking prevention and intervention programming is enhanced by accurate understanding of developmental smoking pathways across the life span. This study investigated within-person patterns of cigarette smoking from ages 18 to 50 among a US national sample of high school graduates, focusing on identifying ages of particular importance for smoking involvement change. Using data from approximately 15,000 individuals participating in the longitudinal Monitoring the Future study, trichotomous measures of past 30-day smoking obtained at 11 time points were modeled using repeated-measures latent class analyses. Sex differences in latent class structure and membership were examined. Twelve latent classes were identified: three characterized by consistent smoking patterns across age (no smoking; smoking < pack per day; smoking pack + per day); three showing uptake to a higher category of smoking across age; four reflecting successful quit behavior by age 50; and two defined by discontinuous shifts between smoking categories. The same latent class structure was found for both males and females, but membership probabilities differed between sexes. Although evidence of increases or decreases in smoking behavior was observed at virtually all ages through 35, 21/22 and 29/30 appeared to be particularly key for smoking category change within class. This examination of latent classes of cigarette smoking among a national US longitudinal sample of high school graduates from ages 18 to 50 identified unique patterns and critical ages of susceptibility to change in smoking category within class. Such information may be of particular use in developing effective smoking prevention and intervention programming. This study examined cigarette smoking among a national longitudinal US sample of high school graduates from ages 18 to 50 and identified distinct latent classes characterized by patterns of movement between no cigarette use, light-to-moderate smoking, and the conventional definition of heavy smoking at 11 time points via repeated-measures latent class analysis. Membership probabilities for each smoking class were estimated, and critical ages of susceptibility to change in smoking behaviors were identified. © The Author 2017. Published by Oxford University Press on behalf of the Society for Research on Nicotine and Tobacco. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Applying the Mixed Rasch Model to the Runco Ideational Behavior Scale

ERIC Educational Resources Information Center

Sen, Sedat

2016-01-01

Previous research using creativity assessments has used latent class models and identified multiple classes (a 3-class solution) associated with various domains. This study explored the latent class structure of the Runco Ideational Behavior Scale, which was designed to quantify ideational capacity. A robust state-of the-art technique called the…

Use of Latent Class Analysis to define groups based on validity, cognition, and emotional functioning.

PubMed

Morin, Ruth T; Axelrod, Bradley N

Latent Class Analysis (LCA) was used to classify a heterogeneous sample of neuropsychology data. In particular, we used measures of performance validity, symptom validity, cognition, and emotional functioning to assess and describe latent groups of functioning in these areas. A data-set of 680 neuropsychological evaluation protocols was analyzed using a LCA. Data were collected from evaluations performed for clinical purposes at an urban medical center. A four-class model emerged as the best fitting model of latent classes. The resulting classes were distinct based on measures of performance validity and symptom validity. Class A performed poorly on both performance and symptom validity measures. Class B had intact performance validity and heightened symptom reporting. The remaining two Classes performed adequately on both performance and symptom validity measures, differing only in cognitive and emotional functioning. In general, performance invalidity was associated with worse cognitive performance, while symptom invalidity was associated with elevated emotional distress. LCA appears useful in identifying groups within a heterogeneous sample with distinct performance patterns. Further, the orthogonal nature of performance and symptom validities is supported.

A Framework for Estimating Causal Effects in Latent Class Analysis: Is There a Causal Link Between Early Sex and Subsequent Profiles of Delinquency?

PubMed Central

Lanza, Stephanie T.; Coffman, Donna L.

2013-01-01

Prevention scientists use latent class analysis (LCA) with increasing frequency to characterize complex behavior patterns and profiles of risk. Often, the most important research questions in these studies involve establishing characteristics that predict membership in the latent classes, thus describing the composition of the subgroups and suggesting possible points of intervention. More recently, prevention scientists have begun to adopt modern methods for drawing causal inference from observational data because of the bias that can be introduced by confounders. This same issue of confounding exists in any analysis of observational data, including prediction of latent class membership. This study demonstrates a straightforward approach to causal inference in LCA that builds on propensity score methods. We demonstrate this approach by examining the causal effect of early sex on subsequent delinquency latent classes using data from 1,890 adolescents in 11th and 12th grade from wave I of the National Longitudinal Study of Adolescent Health. Prior to the statistical adjustment for potential confounders, early sex was significantly associated with delinquency latent class membership for both genders (p=0.02). However, the propensity score adjusted analysis indicated no evidence for a causal effect of early sex on delinquency class membership (p=0.76) for either gender. Sample R and SAS code is included in an Appendix in the ESM so that prevention scientists may adopt this approach to causal inference in LCA in their own work. PMID:23839479

A framework for estimating causal effects in latent class analysis: is there a causal link between early sex and subsequent profiles of delinquency?

PubMed

Butera, Nicole M; Lanza, Stephanie T; Coffman, Donna L

2014-06-01

Prevention scientists use latent class analysis (LCA) with increasing frequency to characterize complex behavior patterns and profiles of risk. Often, the most important research questions in these studies involve establishing characteristics that predict membership in the latent classes, thus describing the composition of the subgroups and suggesting possible points of intervention. More recently, prevention scientists have begun to adopt modern methods for drawing causal inference from observational data because of the bias that can be introduced by confounders. This same issue of confounding exists in any analysis of observational data, including prediction of latent class membership. This study demonstrates a straightforward approach to causal inference in LCA that builds on propensity score methods. We demonstrate this approach by examining the causal effect of early sex on subsequent delinquency latent classes using data from 1,890 adolescents in 11th and 12th grade from wave I of the National Longitudinal Study of Adolescent Health. Prior to the statistical adjustment for potential confounders, early sex was significantly associated with delinquency latent class membership for both genders (p = 0.02). However, the propensity score adjusted analysis indicated no evidence for a causal effect of early sex on delinquency class membership (p = 0.76) for either gender. Sample R and SAS code is included in an Appendix in the ESM so that prevention scientists may adopt this approach to causal inference in LCA in their own work.

Application of Latent Class Analysis to Identify Behavioral Patterns of Response to Behavioral Lifestyle Interventions in Overweight and Obese Adults

PubMed Central

Fitzpatrick, Stephanie L.; Coughlin, Janelle W.; Appel, Lawrence J.; Tyson, Crystal; Stevens, Victor J.; Jerome, Gerald J.; Dalcin, Arlene; Brantley, Phillip J.; Hill-Briggs, Felicia

2016-01-01

Background Examining responders and non-responders to behavioral lifestyle interventions among overweight/obese adults with additional comorbidities may aid in refining and tailoring obesity treatment. Purpose The purpose of this study is to demonstrate the use of latent class analysis to identify patterns of response to behavioral lifestyle interventions based on adherence to diet and exercise recommendations. Method Repeated measures latent class analysis was applied to two clinical trial datasets, combination of two active interventions in the PREMIER Trial (n=501) and phase 1 of the Weight Loss Maintenance Trial (WLM; n=1685), to identify patterns of response to behavioral lifestyle interventions. Treatment response was based on adherence to daily recommendations for fruit/vegetable, fat, saturated fat, sodium, and exercise at baseline and 6 months. Results In PREMIER, three distinct latent classes emerged: responders (45.9 %), non-responders (23.6 %), and early adherers (30.5 %). Responders and Early Adherers had greater weight loss at 6 and 18 months and were more likely to meet behavioral recommendations at 18 months than Non-responders. For WLM, there were four latent classes: partial responders (16 %), non-responders (40 %), early adherers (2 %), and fruit/veggie only responders (41 %). Non-responders in WLM had significantly less weight loss at 6 months compared to that of the other three latent classes. Conclusion Latent class analysis is a useful method to apply to clinical trial data to identify distinct patterns of response to behavioral interventions. Overweight/ obese participants who respond to behavioral lifestyle treatment (i.e., meet behavioral recommendations) have significantly greater weight loss than that of participants who do not make behavioral changes. PMID:25331853

Application of Latent Class Analysis to Identify Behavioral Patterns of Response to Behavioral Lifestyle Interventions in Overweight and Obese Adults.

PubMed

Fitzpatrick, Stephanie L; Coughlin, Janelle W; Appel, Lawrence J; Tyson, Crystal; Stevens, Victor J; Jerome, Gerald J; Dalcin, Arlene; Brantley, Phillip J; Hill-Briggs, Felicia

2015-08-01

Examining responders and non-responders to behavioral lifestyle interventions among overweight/obese adults with additional comorbidities may aid in refining and tailoring obesity treatment. The purpose of this study is to demonstrate the use of latent class analysis to identify patterns of response to behavioral lifestyle interventions based on adherence to diet and exercise recommendations. Repeated measures latent class analysis was applied to two clinical trial datasets, combination of two active interventions in the PREMIER Trial (n = 501) and phase 1 of the Weight Loss Maintenance Trial (WLM; n = 1685), to identify patterns of response to behavioral lifestyle interventions. Treatment response was based on adherence to daily recommendations for fruit/vegetable, fat, saturated fat, sodium, and exercise at baseline and 6 months. In PREMIER, three distinct latent classes emerged: responders (45.9%), non-responders (23.6%), and early adherers (30.5%). Responders and Early Adherers had greater weight loss at 6 and 18 months and were more likely to meet behavioral recommendations at 18 months than Non-responders. For WLM, there were four latent classes: partial responders (16%), non-responders (40%), early adherers (2%), and fruit/veggie only responders (41%). Non-responders in WLM had significantly less weight loss at 6 months compared to that of the other three latent classes. Latent class analysis is a useful method to apply to clinical trial data to identify distinct patterns of response to behavioral interventions. Overweight/ obese participants who respond to behavioral lifestyle treatment (i.e., meet behavioral recommendations) have significantly greater weight loss than that of participants who do not make behavioral changes.

Do recognizable lifetime eating disorder phenotypes naturally occur in a culturally asian population? A combined latent profile and taxometric approach.

PubMed

Thomas, Jennifer J; Eddy, Kamryn T; Ruscio, John; Ng, King Lam; Casale, Kristen E; Becker, Anne E; Lee, Sing

2015-05-01

We examined whether empirically derived eating disorder (ED) categories in Hong Kong Chinese patients (N = 454) would be consistent with recognizable lifetime ED phenotypes derived from latent structure models of European and American samples. We performed latent profile analysis (LPA) using indicator variables from data collected during routine assessment, and then applied taxometric analysis to determine whether latent classes were qualitatively versus quantitatively distinct. Latent profile analysis identified four classes: (i) binge/purge (47%); (ii) non-fat-phobic low-weight (34%); (iii) fat-phobic low-weight (12%); and (iv) overweight disordered eating (6%). Taxometric analysis identified qualitative (categorical) distinctions between the binge/purge and non-fat-phobic low-weight classes, and also between the fat-phobic and non-fat-phobic low-weight classes. Distinctions between the fat-phobic low-weight and binge/purge classes were indeterminate. Empirically derived categories in Hong Kong showed recognizable correspondence with recognizable lifetime ED phenotypes. Although taxometric findings support two distinct classes of low weight EDs, LPA findings also support heterogeneity among non-fat-phobic individuals. Copyright © 2015 John Wiley & Sons, Ltd and Eating Disorders Association.

Short-term memory development: differences in serial position curves between age groups and latent classes.

PubMed

Koppenol-Gonzalez, Gabriela V; Bouwmeester, Samantha; Vermunt, Jeroen K

2014-10-01

In studies on the development of cognitive processes, children are often grouped based on their ages before analyzing the data. After the analysis, the differences between age groups are interpreted as developmental differences. We argue that this approach is problematic because the variance in cognitive performance within an age group is considered to be measurement error. However, if a part of this variance is systematic, it can provide very useful information about the cognitive processes used by some children of a certain age but not others. In the current study, we presented 210 children aged 5 to 12 years with serial order short-term memory tasks. First we analyze our data according to the approach using age groups, and then we apply latent class analysis to form latent classes of children based on their performance instead of their ages. We display the results of the age groups and the latent classes in terms of serial position curves, and we discuss the differences in results. Our findings show that there are considerable differences in performance between the age groups and the latent classes. We interpret our findings as indicating that the latent class analysis yielded a much more meaningful way of grouping children in terms of cognitive processes than the a priori grouping of children based on their ages. Copyright © 2014 Elsevier Inc. All rights reserved.

Variational learning and bits-back coding: an information-theoretic view to Bayesian learning.

PubMed

Honkela, Antti; Valpola, Harri

2004-07-01

The bits-back coding first introduced by Wallace in 1990 and later by Hinton and van Camp in 1993 provides an interesting link between Bayesian learning and information-theoretic minimum-description-length (MDL) learning approaches. The bits-back coding allows interpreting the cost function used in the variational Bayesian method called ensemble learning as a code length in addition to the Bayesian view of misfit of the posterior approximation and a lower bound of model evidence. Combining these two viewpoints provides interesting insights to the learning process and the functions of different parts of the model. In this paper, the problem of variational Bayesian learning of hierarchical latent variable models is used to demonstrate the benefits of the two views. The code-length interpretation provides new views to many parts of the problem such as model comparison and pruning and helps explain many phenomena occurring in learning.

A New Monte Carlo Method for Estimating Marginal Likelihoods.

PubMed

Wang, Yu-Bo; Chen, Ming-Hui; Kuo, Lynn; Lewis, Paul O

2018-06-01

Evaluating the marginal likelihood in Bayesian analysis is essential for model selection. Estimators based on a single Markov chain Monte Carlo sample from the posterior distribution include the harmonic mean estimator and the inflated density ratio estimator. We propose a new class of Monte Carlo estimators based on this single Markov chain Monte Carlo sample. This class can be thought of as a generalization of the harmonic mean and inflated density ratio estimators using a partition weighted kernel (likelihood times prior). We show that our estimator is consistent and has better theoretical properties than the harmonic mean and inflated density ratio estimators. In addition, we provide guidelines on choosing optimal weights. Simulation studies were conducted to examine the empirical performance of the proposed estimator. We further demonstrate the desirable features of the proposed estimator with two real data sets: one is from a prostate cancer study using an ordinal probit regression model with latent variables; the other is for the power prior construction from two Eastern Cooperative Oncology Group phase III clinical trials using the cure rate survival model with similar objectives.

On selecting a prior for the precision parameter of Dirichlet process mixture models

USGS Publications Warehouse

Dorazio, R.M.

2009-01-01

In hierarchical mixture models the Dirichlet process is used to specify latent patterns of heterogeneity, particularly when the distribution of latent parameters is thought to be clustered (multimodal). The parameters of a Dirichlet process include a precision parameter ?? and a base probability measure G0. In problems where ?? is unknown and must be estimated, inferences about the level of clustering can be sensitive to the choice of prior assumed for ??. In this paper an approach is developed for computing a prior for the precision parameter ?? that can be used in the presence or absence of prior information about the level of clustering. This approach is illustrated in an analysis of counts of stream fishes. The results of this fully Bayesian analysis are compared with an empirical Bayes analysis of the same data and with a Bayesian analysis based on an alternative commonly used prior.

An introduction to Bayesian statistics in health psychology.

PubMed

Depaoli, Sarah; Rus, Holly M; Clifton, James P; van de Schoot, Rens; Tiemensma, Jitske

2017-09-01

The aim of the current article is to provide a brief introduction to Bayesian statistics within the field of health psychology. Bayesian methods are increasing in prevalence in applied fields, and they have been shown in simulation research to improve the estimation accuracy of structural equation models, latent growth curve (and mixture) models, and hierarchical linear models. Likewise, Bayesian methods can be used with small sample sizes since they do not rely on large sample theory. In this article, we discuss several important components of Bayesian statistics as they relate to health-based inquiries. We discuss the incorporation and impact of prior knowledge into the estimation process and the different components of the analysis that should be reported in an article. We present an example implementing Bayesian estimation in the context of blood pressure changes after participants experienced an acute stressor. We conclude with final thoughts on the implementation of Bayesian statistics in health psychology, including suggestions for reviewing Bayesian manuscripts and grant proposals. We have also included an extensive amount of online supplementary material to complement the content presented here, including Bayesian examples using many different software programmes and an extensive sensitivity analysis examining the impact of priors.

A latent class regression analysis of men's conformity to masculine norms and psychological distress.

PubMed

Wong, Y Joel; Owen, Jesse; Shea, Munyi

2012-01-01

How are specific dimensions of masculinity related to psychological distress in specific groups of men? To address this question, the authors used latent class regression to assess the optimal number of latent classes that explained differential relationships between conformity to masculine norms and psychological distress in a racially diverse sample of 223 men. The authors identified a 2-class solution. Both latent classes demonstrated very different associations between conformity to masculine norms and psychological distress. In Class 1 (labeled risk avoiders; n = 133), conformity to the masculine norm of risk-taking was negatively related to psychological distress. In Class 2 (labeled detached risk-takers; n = 90), conformity to the masculine norms of playboy, self-reliance, and risk-taking was positively related to psychological distress, whereas conformity to the masculine norm of violence was negatively related to psychological distress. A post hoc analysis revealed that younger men and Asian American men (compared with Latino and White American men) had significantly greater odds of being in Class 2 versus Class 1. The implications of these findings for future research and clinical practice are examined. (c) 2012 APA, all rights reserved.

Prevalence Estimation and Validation of New Instruments in Psychiatric Research: An Application of Latent Class Analysis and Sensitivity Analysis

ERIC Educational Resources Information Center

Pence, Brian Wells; Miller, William C.; Gaynes, Bradley N.

2009-01-01

Prevalence and validation studies rely on imperfect reference standard (RS) diagnostic instruments that can bias prevalence and test characteristic estimates. The authors illustrate 2 methods to account for RS misclassification. Latent class analysis (LCA) combines information from multiple imperfect measures of an unmeasurable latent condition to…

A Latent Transition Analysis of Academic Intrinsic Motivation from Childhood through Adolescence

ERIC Educational Resources Information Center

Marcoulides, George A.; Gottfried, Adele Eskeles; Gottfried, Allen W.; Oliver, Pamella H.

2008-01-01

A longitudinal modeling approach was utilized to determine the existence of latent classes with regard to academic intrinsic motivation and the points of stability and transition of individuals between and within classes. A special type of latent Markov Chain model using "Mplus" was fit to data from the Fullerton Longitudinal Study, with…

Mixture Item Response Theory-MIMIC Model: Simultaneous Estimation of Differential Item Functioning for Manifest Groups and Latent Classes

ERIC Educational Resources Information Center

Bilir, Mustafa Kuzey

2009-01-01

This study uses a new psychometric model (mixture item response theory-MIMIC model) that simultaneously estimates differential item functioning (DIF) across manifest groups and latent classes. Current DIF detection methods investigate DIF from only one side, either across manifest groups (e.g., gender, ethnicity, etc.), or across latent classes…

Empirically derived lifespan polytraumatization typologies: A systematic review.

PubMed

Contractor, Ateka A; Caldas, Stephanie; Fletcher, Shelley; Shea, M Tracie; Armour, Cherie

2018-07-01

Polytraumatization classes based on trauma endorsement patterns relate to distinct clinical outcomes. Person-centered approaches robustly evaluate the nature, and construct validity of polytraumatization classes. Our review examined evidence for the nature and construct validity of lifespan polytraumatization typologies. In September 2016, we searched Pubmed, PSYCINFO, PSYC ARTICLES, Academic Search Complete, PILPTS, Web of Science, CINAHL, Medline, PsycEXTRA, and PBSC. Search terms included "latent profile," "latent class," "latent analysis," "person-centered," "polytrauma," "polyvictimization," "traumatization," "lifetime," "cooccurring," "complex," "typology," "multidimensional," "sequential," "multiple," "subtype," "(re)victimization," "cumulative," "maltreatment," "abuse," and "stressor." Inclusionary criteria included: peer-reviewed; latent class/latent profile analyses (LCA/LPA) of lifespan polytrauma classes; adult samples of size greater than 200; only trauma types as LCA/LPA indicators; mental health correlates of typologies; and individual-level trauma assessment. Of 1,397 articles, nine met inclusion criteria. Following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, research assistants completed a secondary reference search, and independently extracted data with standardized coding forms. Three-class (n = 5) or four-class (n = 4) solutions were found. Seven studies found a class characterized by higher trauma endorsement (high-trauma). All studies found a class characterized by lower trauma endorsement (low-trauma), and predominance of specific traumas (specific-trauma; e.g., childhood maltreatment). High-trauma versus low-trauma classes and specific-trauma versus low-trauma classes differed on mental health correlates. Evidence supports the prevalence of a high-trauma class experiencing poorer mental health, and the detrimental impact of aggregated interpersonal and other traumas. We highlight the clinical importance of addressing polytraumatization classes, and comprehensively assessing the impact of all traumas. © 2018 Wiley Periodicals, Inc.

Patterns of Chronic Conditions and Their Associations With Behaviors and Quality of Life, 2010

PubMed Central

Mitchell, Sandra A.; Thompson, William W.; Zack, Matthew M.; Reeve, Bryce B.; Cella, David; Smith, Ashley Wilder

2015-01-01

Introduction Co-occurring chronic health conditions elevate the risk of poor health outcomes such as death and disability, are associated with poor quality of life, and magnify the complexities of self-management, care coordination, and treatment planning. This study assessed patterns of both singular and multiple chronic conditions, behavioral risk factors, and quality of life in a population-based sample. Methods In a national survey, adults (n = 4,184) answered questions about the presence of 27 chronic conditions. We used latent class analysis to identify patterns of chronic conditions and to explore associations of latent class membership with sociodemographic characteristics, behavioral risk factors, and health. Results Latent class analyses indicated 4 morbidity profiles: a healthy class (class 1), a class with predominantly physical health conditions (class 2), a class with predominantly mental health conditions (class 3), and a class with both physical and mental health conditions (class 4). Class 4 respondents reported significantly worse physical health and well-being and more days of activity limitation than those in the other latent classes. Class 4 respondents were also more likely to be obese and sedentary, and those with predominantly mental health conditions were most likely to be current smokers. Conclusions Subgroups with distinct patterns of chronic conditions can provide direction for screening and surveillance, guideline development, and the delivery of complex care services. PMID:26679491

Latent class evaluation of a milk test, a urine test, and the fat-to-protein percentage ratio in milk to diagnose ketosis in dairy cows.

PubMed

Krogh, M A; Toft, N; Enevoldsen, C

2011-05-01

In this study, 3 commonly used tests to diagnose ketosis were evaluated with a latent class model to avoid the assumption of an available perfect test. The 3 tests were the KetoLac BHB (Sanwa Kagaku Kenkyusho Co. Ltd., Nagoya, Japan) test strip that tests milk for β-hydroxybutyrate, the KetoStix (Bayer Diagnostics Europe Ltd., Dublin, Ireland) test strip that tests urine for acetoacetate, and the fat-to-protein percentage ratio (FPR) in milk. A total of 8,902 cows were included in the analysis. The cows were considered to be a random sample from the population of Danish dairy cattle under intensive management, thus representing a natural spectrum of ketosis as a disease. All cows had a recorded FPR between 7 and 21 d postpartum. The KetoLac BHB recordings were available from 2,257 cows and 6,645 cows had a KetoStix recording. The recordings were analyzed with a modified Hui-Walter model, in a Bayesian framework. The specificity of the KetoLac BHB test and the KetoStix test were both high [0.99 (0.97-0.99)], whereas the specificity of FPR was somewhat lower [0.79 (0.77-0.81)]. The best sensitivity was for the KetoStix test [0.78 (0.55-0.98)], followed by the FPR [0.63 (0.58-0.71)] and KetoLac BHB test [0.58 (0.35-0.93)]. Copyright © 2011 American Dairy Science Association. Published by Elsevier Inc. All rights reserved.

Class Extraction and Classification Accuracy in Latent Class Models

ERIC Educational Resources Information Center

Wu, Qiong

2009-01-01

Despite the increasing popularity of latent class models (LCM) in educational research, methodological studies have not yet accumulated much information on the appropriate application of this modeling technique, especially with regard to requirement on sample size and number of indicators. This dissertation study represented an initial attempt to…

An Assessment of Character and Leadership Development Latent Factor Structures through Confirmatory Factor, Item Response Theory, and Latent Class Analyses

ERIC Educational Resources Information Center

Higginbotham, David L.

2013-01-01

This study leveraged the complementary nature of confirmatory factor (CFA), item response theory (IRT), and latent class (LCA) analyses to strengthen the rigor and sophistication of evaluation of two new measures of the Air Force Academy's "leader of character" definition--the Character Mosaic Virtues (CMV) and the Leadership Mosaic…

Longitudinal Physical Activity Patterns Among Older Adults: A Latent Transition Analysis.

PubMed

Mooney, Stephen J; Joshi, Spruha; Cerdá, Magdalena; Kennedy, Gary J; Beard, John R; Rundle, Andrew G

2018-05-14

Most epidemiologic studies of physical activity measure either total energy expenditure or engagement in a single activity type, such as walking. These approaches may gloss over important nuances in activity patterns. We performed a latent transition analysis to identify patterns of activity types as well as neighborhood and individual determinants of changes in those activity patterns over two years in a cohort of 2,023 older adult residents of New York City, NY, surveyed between 2011 and 2013. We identified seven latent classes: 1) Mostly Inactive, 2) Walking, 3) Exercise, 4) Household Activities and Walking, 5) Household Activities and Exercise, 6) Gardening and Household Activities, and 7) Gardening, Household Activities, and Exercise. The majority of subjects retained the same activity patterns between waves (54% unchanged between waves 1 and 2, 66% unchanged between waves 2 and 3).Most latent class transitions were between classes distinguished only by one form of activity, and only neighborhood unemployment was consistently associated with changing between activity latent classes. Future latent transition analyses of physical activity would benefit from larger cohorts and longer follow-up periods to assess predictors of and long-term impacts of changes in activity patterns.

Heterosexual Casual Sex and STI Diagnosis: A Latent Class Analysis

PubMed Central

Ann Lyons, Heidi

2017-01-01

Casual sex is common during the emerging adult life course stage, but little research has taken a person-centered approach to investigate if casual sexual behavior influences STI rates. Using a nationally representative sample and latent class analysis, results showed three distinctive latent classes. Abstainers were the least likely to have an STI, followed by the casual sex experienced, and then the casual sex risk-takers. Once other covariates were included in the model, there was no significant difference between the abstainers and casual sex experienced classes. These results highlight the need for future research to include diverse samples of emerging adults. PMID:29276549

Generative Topographic Mapping (GTM): Universal Tool for Data Visualization, Structure-Activity Modeling and Dataset Comparison.

PubMed

Kireeva, N; Baskin, I I; Gaspar, H A; Horvath, D; Marcou, G; Varnek, A

2012-04-01

Here, the utility of Generative Topographic Maps (GTM) for data visualization, structure-activity modeling and database comparison is evaluated, on hand of subsets of the Database of Useful Decoys (DUD). Unlike other popular dimensionality reduction approaches like Principal Component Analysis, Sammon Mapping or Self-Organizing Maps, the great advantage of GTMs is providing data probability distribution functions (PDF), both in the high-dimensional space defined by molecular descriptors and in 2D latent space. PDFs for the molecules of different activity classes were successfully used to build classification models in the framework of the Bayesian approach. Because PDFs are represented by a mixture of Gaussian functions, the Bhattacharyya kernel has been proposed as a measure of the overlap of datasets, which leads to an elegant method of global comparison of chemical libraries. Copyright © 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

Detecting Math Anxiety with a Mixture Partial Credit Model

ERIC Educational Resources Information Center

Ölmez, Ibrahim Burak; Cohen, Allan S.

2017-01-01

The purpose of this study was to investigate a new methodology for detection of differences in middle grades students' math anxiety. A mixture partial credit model analysis revealed two distinct latent classes based on homogeneities in response patterns within each latent class. Students in Class 1 had less anxiety about apprehension of math…

Classes in the Balance: Latent Class Analysis and the Balance Scale Task

ERIC Educational Resources Information Center

Boom, Jan; ter Laak, Jan

2007-01-01

Latent class analysis (LCA) has been successfully applied to tasks measuring higher cognitive functioning, suggesting the existence of distinct strategies used in such tasks. With LCA it became possible to classify post hoc. This important step forward in modeling and analyzing cognitive strategies is relevant to the overlapping waves model for…

Trajectories of Substance Use Disorders in Youth: Identifying and Predicting Group Memberships

ERIC Educational Resources Information Center

Lee, Chih-Yuan S.; Winters, Ken C.; Wall, Melanie M.

2010-01-01

This study used latent class regression to identify latent trajectory classes based on individuals' diagnostic course of substance use disorders (SUDs) from late adolescence to early adulthood as well as to examine whether several psychosocial risk factors predicted the trajectory class membership. The study sample consisted of 310 individuals…

Examining Variation in Adolescent Bystanders' Responses to Bullying

ERIC Educational Resources Information Center

Waasdorp, Tracy Evian; Bradshaw, Catherine P.

2018-01-01

Latent class analysis was used to examine whether patterns of bystander responses varied as a function of both student- and school-level characteristics. Data from 18,863 high school students from 58 schools who "ever witnessed bullying" were used to identify five latent classes of bystander behavior. Three of the classes identified…

Latent Class Analysis of Peer Conformity: Who Is Yielding to Pressure and Why?

ERIC Educational Resources Information Center

Kosten, Paul A.; Scheier, Lawrence M.; Grenard, Jerry L.

2013-01-01

This study used latent class analysis to examine typologies of peer conformity in a community sample of middle school students. Students responded to 31 items assessing diverse facets of conformity dispositions. The most parsimonious model produced three qualitatively distinct classes that differed on the basis of conformity to recreational…

A Latent Class Analysis of Weight-Related Health Behaviors among 2-and 4-Year College Students and Associated Risk of Obesity

ERIC Educational Resources Information Center

Mathur, Charu; Stigler, Melissa; Lust, Katherine; Laska, Melissa

2014-01-01

Little is known about the complex patterning of weight-related health behaviors in 2-and 4-year college students. The objective of this study was to identify and describe unique classes of weight-related health behaviors among college students. Latent class analysis was used to identify homogenous, mutually exclusive classes of nine health…

Tracking Children Who Fly below the Radar: Latent Transition Modeling of Students with Late-Emerging Reading Disability

ERIC Educational Resources Information Center

Compton, Donald L.; Fuchs, Douglas; Fuchs, Lynn S.; Elleman, Amy M.; Gilbert, Jennifer K.

2008-01-01

The purpose of this study was to examine (1) the stability of latent classes associated with reading disability (RD) and typical development (TD) across time, (2) the importance of speeded word recognition as a latent class indicator of RD and TD, and (3) possible early indicators of students with late-emerging RD. Analyses were based on a…

Indentifying Latent Classes and Testing Their Determinants in Early Adolescents' Use of Computers and Internet for Learning

ERIC Educational Resources Information Center

Heo, Gyun

2013-01-01

The purpose of the present study was to identify latent classes resting on early adolescents' change trajectory patterns in using computers and the Internet for learning and to test the effects of gender, self-control, self-esteem, and game use in South Korea. Latent growth mixture modeling (LGMM) was used to identify subpopulations in the Korea…

Data-driven subtypes of major depressive disorder: a systematic review

PubMed Central

2012-01-01

Background According to current classification systems, patients with major depressive disorder (MDD) may have very different combinations of symptoms. This symptomatic diversity hinders the progress of research into the causal mechanisms and treatment allocation. Theoretically founded subtypes of depression such as atypical, psychotic, and melancholic depression have limited clinical applicability. Data-driven analyses of symptom dimensions or subtypes of depression are scarce. In this systematic review, we examine the evidence for the existence of data-driven symptomatic subtypes of depression. Methods We undertook a systematic literature search of MEDLINE, PsycINFO and Embase in May 2012. We included studies analyzing the depression criteria of the Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV) of adults with MDD in latent variable analyses. Results In total, 1176 articles were retrieved, of which 20 satisfied the inclusion criteria. These reports described a total of 34 latent variable analyses: 6 confirmatory factor analyses, 6 exploratory factor analyses, 12 principal component analyses, and 10 latent class analyses. The latent class techniques distinguished 2 to 5 classes, which mainly reflected subgroups with different overall severity: 62 of 71 significant differences on symptom level were congruent with a latent class solution reflecting severity. The latent class techniques did not consistently identify specific symptom clusters. Latent factor techniques mostly found a factor explaining the variance in the symptoms depressed mood and interest loss (11 of 13 analyses), often complemented by psychomotor retardation or fatigue (8 of 11 analyses). However, differences in found factors and classes were substantial. Conclusions The studies performed to date do not provide conclusive evidence for the existence of depressive symptom dimensions or symptomatic subtypes. The wide diversity of identified factors and classes might result either from the absence of patterns to be found, or from the theoretical and modeling choices preceding analysis. PMID:23210727

Dissociative Experiences are Associated with Obsessive-Compulsive Symptoms in a Non-clinical Sample: A Latent Profile Analysis

PubMed Central

BOYSAN, Murat

2014-01-01

Introduction There has been a burgeoning literature considering the significant associations between obsessive-compulsive symptoms and dissociative experiences. In this study, the relationsips between dissociative symtomotology and dimensions of obsessive-compulsive symptoms were examined in homogeneous sub-groups obtained with latent class algorithm in an undergraduate Turkish sample. Method Latent profile analysis, a recently developed classification method based on latent class analysis, was applied to the Dissociative Experiences Scale (DES) item-response data from 2976 undergraduates. Differences in severity of obsessive-compulsive symptoms, anxiety and depression across groups were evaluated by running multinomial logistic regression analyses. Associations between latent class probabilities and psychological variables in terms of obsessive-compulsive sub-types, anxiety, and depression were assessed by computing Pearson’s product-moment correlation coefficients. Results The findings of the latent profile analysis supported further evidence for discontinuity model of dissociative experiences. The analysis empirically justified the distinction among three sub-groups based on the DES items. A marked proportion of the sample (42%) was assigned to the high dissociative class. In the further analyses, all sub-types of obsessive-compulsive symptoms significantly differed across latent classes. Regarding the relationships between obsessive-compulsive symptoms and dissociative symptomatology, low dissociation appeared to be a buffering factor dealing with obsessive-compulsive symptoms; whereas high dissociation appeared to be significantly associated with high levels of obsessive-compulsive symptoms. Conclusion It is concluded that the concept of dissociation can be best understood in a typological approach that dissociative symptomatology not only exacerbates obsessive-compulsive symptoms but also serves as an adaptive coping mechanism. PMID:28360635

Dissociative Experiences are Associated with Obsessive-Compulsive Symptoms in a Non-clinical Sample: A Latent Profile Analysis.

PubMed

Boysan, Murat

2014-09-01

There has been a burgeoning literature considering the significant associations between obsessive-compulsive symptoms and dissociative experiences. In this study, the relationsips between dissociative symtomotology and dimensions of obsessive-compulsive symptoms were examined in homogeneous sub-groups obtained with latent class algorithm in an undergraduate Turkish sample. Latent profile analysis, a recently developed classification method based on latent class analysis, was applied to the Dissociative Experiences Scale (DES) item-response data from 2976 undergraduates. Differences in severity of obsessive-compulsive symptoms, anxiety and depression across groups were evaluated by running multinomial logistic regression analyses. Associations between latent class probabilities and psychological variables in terms of obsessive-compulsive sub-types, anxiety, and depression were assessed by computing Pearson's product-moment correlation coefficients. The findings of the latent profile analysis supported further evidence for discontinuity model of dissociative experiences. The analysis empirically justified the distinction among three sub-groups based on the DES items. A marked proportion of the sample (42%) was assigned to the high dissociative class. In the further analyses, all sub-types of obsessive-compulsive symptoms significantly differed across latent classes. Regarding the relationships between obsessive-compulsive symptoms and dissociative symptomatology, low dissociation appeared to be a buffering factor dealing with obsessive-compulsive symptoms; whereas high dissociation appeared to be significantly associated with high levels of obsessive-compulsive symptoms. It is concluded that the concept of dissociation can be best understood in a typological approach that dissociative symptomatology not only exacerbates obsessive-compulsive symptoms but also serves as an adaptive coping mechanism.

Precipitation and Latent Heating Distributions from Satellite Passive Microwave Radiometry. Part 1; Method and Uncertainties

NASA Technical Reports Server (NTRS)

Olson, William S.; Kummerow, Christian D.; Yang, Song; Petty, Grant W.; Tao, Wei-Kuo; Bell, Thomas L.; Braun, Scott A.; Wang, Yansen; Lang, Stephen E.; Johnson, Daniel E.

2004-01-01

A revised Bayesian algorithm for estimating surface rain rate, convective rain proportion, and latent heating/drying profiles from satellite-borne passive microwave radiometer observations over ocean backgrounds is described. The algorithm searches a large database of cloud-radiative model simulations to find cloud profiles that are radiatively consistent with a given set of microwave radiance measurements. The properties of these radiatively consistent profiles are then composited to obtain best estimates of the observed properties. The revised algorithm is supported by an expanded and more physically consistent database of cloud-radiative model simulations. The algorithm also features a better quantification of the convective and non-convective contributions to total rainfall, a new geographic database, and an improved representation of background radiances in rain-free regions. Bias and random error estimates are derived from applications of the algorithm to synthetic radiance data, based upon a subset of cloud resolving model simulations, and from the Bayesian formulation itself. Synthetic rain rate and latent heating estimates exhibit a trend of high (low) bias for low (high) retrieved values. The Bayesian estimates of random error are propagated to represent errors at coarser time and space resolutions, based upon applications of the algorithm to TRMM Microwave Imager (TMI) data. Errors in instantaneous rain rate estimates at 0.5 deg resolution range from approximately 50% at 1 mm/h to 20% at 14 mm/h. These errors represent about 70-90% of the mean random deviation between collocated passive microwave and spaceborne radar rain rate estimates. The cumulative algorithm error in TMI estimates at monthly, 2.5 deg resolution is relatively small (less than 6% at 5 mm/day) compared to the random error due to infrequent satellite temporal sampling (8-35% at the same rain rate).

Job Satisfaction among Health-Care Staff in Township Health Centers in Rural China: Results from a Latent Class Analysis.

PubMed

Wang, Haipeng; Tang, Chengxiang; Zhao, Shichao; Meng, Qingyue; Liu, Xiaoyun

2017-09-22

Background : The lower job satisfaction of health-care staff will lead to more brain drain, worse work performance, and poorer health-care outcomes. The aim of this study was to identify patterns of job satisfaction among health-care staff in rural China, and to investigate the association between the latent clusters and health-care staff's personal and professional features; Methods : We selected 12 items of five-point Likert scale questions to measure job satisfaction. A latent-class analysis was performed to identify subgroups based on the items of job satisfaction; Results : Four latent classes of job satisfaction were identified: 8.9% had high job satisfaction, belonging to "satisfied class"; 38.2% had low job satisfaction, named as "unsatisfied class"; 30.5% were categorized into "unsatisfied class with the exception of interpersonal relationships"; 22.4% were identified as "pseudo-satisfied class", only satisfied with management-oriented items. Low job satisfaction was associated with specialty, training opportunity, and income inequality. Conclusions : The minority of health-care staff belong to the "satisfied class". Three among four subgroups are not satisfied with income, benefit, training, and career development. Targeting policy interventions should be implemented to improve the items of job satisfaction based on the patterns and health-care staff's features.

Examination of the Predictors of Latent Class Typologies of Bullying Involvement among Middle School Students

ERIC Educational Resources Information Center

Lovegrove, Peter J.; Henry, Kimberly L.; Slater, Michael D.

2012-01-01

This study employs latent class analysis to construct bullying involvement typologies among 3,114 students (48% male, 58% White) in 40 middle schools across the United States. Four classes were constructed: victims (15%); bullies (13%); bully/victims (13%); and noninvolved (59%). Respondents who were male and participated in fewer conventional…

Using Latent Class Analysis to Identify Academic and Behavioral Risk Status in Elementary Students

ERIC Educational Resources Information Center

King, Kathleen R.; Lembke, Erica S.; Reinke, Wendy M.

2016-01-01

Identifying classes of children on the basis of academic and behavior risk may have important implications for the allocation of intervention resources within Response to Intervention (RTI) and Multi-Tiered System of Support (MTSS) models. Latent class analysis (LCA) was conducted with a sample of 517 third grade students. Fall screening scores in…

Life-Course Body Mass Index Trajectories Are Predicted by Childhood Socioeconomic Status but Not Exposure to Improved Nutrition during the First 1000 Days after Conception in Guatemalan Adults.

PubMed

Ford, Nicole D; Martorell, Reynaldo; Mehta, Neil K; Ramirez-Zea, Manuel; Stein, Aryeh D

2016-11-01

Latin America has experienced increases in obesity. Little is known about the role of early life factors on body mass index (BMI) gain over the life course. The objective of this research was to examine the role of early life factors [specifically, nutrition supplementation during the first 1000 d (from conception to 2 y of age) and childhood household socioeconomic status (SES)] on the pattern of BMI gain from birth or early childhood through midadulthood by using latent class growth analysis. Study participants (711 women, 742 men) who were born in 4 villages in Guatemala (1962-1977) were followed prospectively since participating in a randomized nutrition supplementation trial as children. Sex-specific BMI latent class trajectories were derived from 22 possible measures of height and weight from 1969 to 2004. To characterize early life determinants of BMI latent class membership, we used logistic regression modeling and estimated the difference-in-difference (DD) effect of nutrition supplementation during the first 1000 d. We identified 2 BMI latent classes in women [low (57%) and high (43%)] and 3 classes in men [low (38%), medium (47%), and high (15%)]. Nutrition supplementation during the first 1000 d after conception was not associated with BMI latent class membership (DD test: P > 0.15 for men and women), whereas higher SES was associated with increased odds of high BMI latent class membership in both men (OR: 1.98; 95% CI: 1.09, 3.61) and women (OR: 1.62; 95% CI: 1.07, 2.45) for the highest relative to the lowest tertile. In a cohort of Guatemalan men and women, nutrition supplementation provided during the first 1000 d was not significantly associated with higher BMI trajectory. Higher childhood household SES was associated with increased odds of high BMI latent class membership relative to the poorest households. The pathways through which this operates still need to be explored. © 2016 American Society for Nutrition.

Interactions between lower urinary tract symptoms and cardiovascular risk factors determine distinct patterns of erectile dysfunction: a latent class analysis.

PubMed

Barbosa, João A B A; Muracca, Eduardo; Nakano, Élcio; Assalin, Adriana R; Cordeiro, Paulo; Paranhos, Mario; Cury, José; Srougi, Miguel; Antunes, Alberto A

2013-12-01

An epidemiological association between lower urinary tract symptoms and erectile dysfunction is well established. However, interactions among multiple risk factors and the role of each in pathological mechanisms are not fully elucidated We enrolled 898 men undergoing prostate cancer screening for evaluation with the International Prostate Symptom Score (I-PSS) and simplified International Index of Erectile Function-5 (IIEF-5) questionnaires. Age, race, hypertension, diabetes, dyslipidemia, metabolic syndrome, cardiovascular disease, serum hormones and anthropometric parameters were also evaluated. Risk factors for erectile dysfunction were identified by logistic regression. The 333 men with at least mild to moderate erectile dysfunction (IIEF 16 or less) were included in a latent class model to identify relationships across erectile dysfunction risk factors. Age, hypertension, diabetes, lower urinary tract symptoms and cardiovascular event were independent predictors of erectile dysfunction (p<0.05). We identified 3 latent classes of patients with erectile dysfunction (R2 entropy=0.82). Latent class 1 had younger men at low cardiovascular risk and a moderate/high prevalence of lower urinary tract symptoms. Latent class 2 had the oldest patients at moderate cardiovascular risk with an increased prevalence of lower urinary tract symptoms. Latent class 3 had men of intermediate age with the highest prevalence of cardiovascular risk factors and lower urinary tract symptoms. Erectile dysfunction severity and lower urinary tract symptoms increased from latent class 1 to 3. Risk factor interactions determined different severities of lower urinary tract symptoms and erectile dysfunction. The effect of lower urinary tract symptoms and cardiovascular risk outweighed that of age. While in the youngest patients lower urinary tract symptoms acted as a single risk factor for erectile dysfunction, the contribution of vascular disease resulted in significantly more severe dysfunction. Applying a risk factor interaction model to prospective trials could reveal distinct classes of drug responses and help define optimal treatment strategies for specific groups. Copyright © 2013 American Urological Association Education and Research, Inc. Published by Elsevier Inc. All rights reserved.

Uncovering a latent multinomial: Analysis of mark-recapture data with misidentification

USGS Publications Warehouse

Link, W.A.; Yoshizaki, J.; Bailey, L.L.; Pollock, K.H.

2010-01-01

Natural tags based on DNA fingerprints or natural features of animals are now becoming very widely used in wildlife population biology. However, classic capture-recapture models do not allow for misidentification of animals which is a potentially very serious problem with natural tags. Statistical analysis of misidentification processes is extremely difficult using traditional likelihood methods but is easily handled using Bayesian methods. We present a general framework for Bayesian analysis of categorical data arising from a latent multinomial distribution. Although our work is motivated by a specific model for misidentification in closed population capture-recapture analyses, with crucial assumptions which may not always be appropriate, the methods we develop extend naturally to a variety of other models with similar structure. Suppose that observed frequencies f are a known linear transformation f = A???x of a latent multinomial variable x with cell probability vector ?? = ??(??). Given that full conditional distributions [?? | x] can be sampled, implementation of Gibbs sampling requires only that we can sample from the full conditional distribution [x | f, ??], which is made possible by knowledge of the null space of A???. We illustrate the approach using two data sets with individual misidentification, one simulated, the other summarizing recapture data for salamanders based on natural marks. ?? 2009, The International Biometric Society.

Uncovering a Latent Multinomial: Analysis of Mark-Recapture Data with Misidentification

USGS Publications Warehouse

Link, W.A.; Yoshizaki, J.; Bailey, L.L.; Pollock, K.H.

2009-01-01

Natural tags based on DNA fingerprints or natural features of animals are now becoming very widely used in wildlife population biology. However, classic capture-recapture models do not allow for misidentification of animals which is a potentially very serious problem with natural tags. Statistical analysis of misidentification processes is extremely difficult using traditional likelihood methods but is easily handled using Bayesian methods. We present a general framework for Bayesian analysis of categorical data arising from a latent multinomial distribution. Although our work is motivated by a specific model for misidentification in closed population capture-recapture analyses, with crucial assumptions which may not always be appropriate, the methods we develop extend naturally to a variety of other models with similar structure. Suppose that observed frequencies f are a known linear transformation f=A'x of a latent multinomial variable x with cell probability vector pi= pi(theta). Given that full conditional distributions [theta | x] can be sampled, implementation of Gibbs sampling requires only that we can sample from the full conditional distribution [x | f, theta], which is made possible by knowledge of the null space of A'. We illustrate the approach using two data sets with individual misidentification, one simulated, the other summarizing recapture data for salamanders based on natural marks.

Personality and changes in comorbidity patterns among anxiety and depressive disorders.

PubMed

Spinhoven, Philip; de Rooij, Mark; Heiser, Willem; Smit, Jan H; Penninx, Brenda W J H

2012-11-01

This prospective study examined the prognostic value of the Big Five personality model for changes in comorbidity patterns of emotional disorders both from a person- and trait-centered perspective. Moreover, it is investigated whether the predictive effect of personality can be attributed to symptom severity at baseline. We followed a cohort of 2566 persons (18-65 years) recruited in primary and specialized mental health care during two years. Personality dimensions at baseline were assessed with the NEO-FFI. The Diagnostic and Statistical Manual of Mental Disorders (4th ed.)-based diagnostic interviews with the CIDI allowed assessment of changes in comorbidity patterns of anxiety and depressive disorders over two years. Data were analyzed with latent class analysis (LCA) and latent transition analysis (LTA). LCA identified a four-class latent comorbidity class solution (Few Disorders, Fear Disorders, Distress Disorders, and Comorbid Fear and Distress Disorders) and a five-class latent personality class solution (High Resilients, Medium Resilients, Low Overcontrollers, Medium Overcontrollers, and High Overcontrollers). LTA showed that the likelihood of remaining in the same latent class was larger than that of transitioning to a less severe comorbidity class. Also, after correcting for symptom severity, medium and high Overcontrollers as well as participants with lower levels of conscientiousness were less likely to transition to a less severe comorbidity class. In particular, the individual trait of conscientiousness may be less dependent on current levels of anxiety and depressive symptoms and be a key pathoplastic or even predisposing variable in anxiety and depression and needs more theoretical and empirical study. (PsycINFO Database Record (c) 2012 APA, all rights reserved).

Bayesian networks for maritime traffic accident prevention: benefits and challenges.

PubMed

Hänninen, Maria

2014-12-01

Bayesian networks are quantitative modeling tools whose applications to the maritime traffic safety context are becoming more popular. This paper discusses the utilization of Bayesian networks in maritime safety modeling. Based on literature and the author's own experiences, the paper studies what Bayesian networks can offer to maritime accident prevention and safety modeling and discusses a few challenges in their application to this context. It is argued that the capability of representing rather complex, not necessarily causal but uncertain relationships makes Bayesian networks an attractive modeling tool for the maritime safety and accidents. Furthermore, as the maritime accident and safety data is still rather scarce and has some quality problems, the possibility to combine data with expert knowledge and the easy way of updating the model after acquiring more evidence further enhance their feasibility. However, eliciting the probabilities from the maritime experts might be challenging and the model validation can be tricky. It is concluded that with the utilization of several data sources, Bayesian updating, dynamic modeling, and hidden nodes for latent variables, Bayesian networks are rather well-suited tools for the maritime safety management and decision-making. Copyright © 2014 Elsevier Ltd. All rights reserved.

Friendship networks of inner-city adults: a latent class analysis and multi-level regression of supporter types and the association of supporter latent class membership with supporter and recipient drug use.

PubMed

Bohnert, Amy S B; German, Danielle; Knowlton, Amy R; Latkin, Carl A

2010-03-01

Social support is a multi-dimensional construct that is important to drug use cessation. The present study identified types of supportive friends among the social network members in a community-based sample and examined the relationship of supporter-type classes with supporter, recipient, and supporter-recipient relationship characteristics. We hypothesized that the most supportive network members and their support recipients would be less likely to be current heroin/cocaine users. Participants (n=1453) were recruited from low-income neighborhoods with a high prevalence of drug use. Participants identified their friends via a network inventory, and all nominated friends were included in a latent class analysis and grouped based on their probability of providing seven types of support. These latent classes were included as the dependent variable in a multi-level regression of supporter drug use, recipient drug use, and other characteristics. The best-fitting latent class model identified five support patterns: friends who provided Little/No Support, Low/Moderate Support, High Support, Socialization Support, and Financial Support. In bivariate models, friends in the High, Low/Moderate, and Financial Support were less likely to use heroin or cocaine and had less conflict with and were more trusted by the support recipient than friends in the Low/No Support class. Individuals with supporters in those same support classes compared to the Low/No Support class were less likely to use heroin or cocaine, or to be homeless or female. Multivariable models suggested similar trends. Those with current heroin/cocaine use were less likely to provide or receive comprehensive support from friends. Published by Elsevier Ireland Ltd.

Investigating preferences for mosquito-control technologies in Mozambique with latent class analysis.

PubMed

Smith, Rachel A; Barclay, Victoria C; Findeis, Jill L

2011-07-21

It is common practice to seek the opinions of future end-users during the development of innovations. Thus, the aim of this study is to investigate latent classes of users in Mozambique based on their preferences for mosquito-control technology attributes and covariates of these classes, as well as to explore which current technologies meet these preferences. Surveys were administered in five rural villages in Mozambique. The data were analysed with latent class analysis. This study showed that users' preferences for malaria technologies varied, and people could be categorized into four latent classes based on shared preferences. The largest class, constituting almost half of the respondents, would not avoid a mosquito-control technology because of its cost, heat, odour, potential to make other health issues worse, ease of keeping clean, or inadequate mosquito control. The other three groups are characterized by the attributes which would make them avoid a technology; these groups are labelled as the bites class, by-products class, and multiple-concerns class. Statistically significant covariates included literacy, self-efficacy, willingness to try new technologies, and perceived seriousness of malaria for the household. To become widely diffused, best practices suggest that end-users should be included in product development to ensure that preferred attributes or traits are considered. This study demonstrates that end-user preferences can be very different and that one malaria control technology will not satisfy everyone.

Interlocking directorates in Irish companies using a latent space model for bipartite networks

PubMed Central

Friel, Nial; Rastelli, Riccardo; Wyse, Jason; Raftery, Adrian E.

2016-01-01

We analyze the temporal bipartite network of the leading Irish companies and their directors from 2003 to 2013, encompassing the end of the Celtic Tiger boom and the ensuing financial crisis in 2008. We focus on the evolution of company interlocks, whereby a company director simultaneously sits on two or more boards. We develop a statistical model for this dataset by embedding the positions of companies and directors in a latent space. The temporal evolution of the network is modeled through three levels of Markovian dependence: one on the model parameters, one on the companies’ latent positions, and one on the edges themselves. The model is estimated using Bayesian inference. Our analysis reveals that the level of interlocking, as measured by a contraction of the latent space, increased before and during the crisis, reaching a peak in 2009, and has generally stabilized since then. PMID:27247395

Interlocking directorates in Irish companies using a latent space model for bipartite networks.

PubMed

Friel, Nial; Rastelli, Riccardo; Wyse, Jason; Raftery, Adrian E

2016-06-14

We analyze the temporal bipartite network of the leading Irish companies and their directors from 2003 to 2013, encompassing the end of the Celtic Tiger boom and the ensuing financial crisis in 2008. We focus on the evolution of company interlocks, whereby a company director simultaneously sits on two or more boards. We develop a statistical model for this dataset by embedding the positions of companies and directors in a latent space. The temporal evolution of the network is modeled through three levels of Markovian dependence: one on the model parameters, one on the companies' latent positions, and one on the edges themselves. The model is estimated using Bayesian inference. Our analysis reveals that the level of interlocking, as measured by a contraction of the latent space, increased before and during the crisis, reaching a peak in 2009, and has generally stabilized since then.

Latent lifestyle preferences and household location decisions

NASA Astrophysics Data System (ADS)

Walker, Joan L.; Li, Jieping

2007-04-01

Lifestyle, indicating preferences towards a particular way of living, is a key driver of the decision of where to live. We employ latent class choice models to represent this behavior, where the latent classes are the lifestyles and the choice model is the choice of residential location. Thus, we simultaneously estimate lifestyle groups and how lifestyle impacts location decisions. Empirical results indicate three latent lifestyle segments: suburban dwellers, urban dwellers, and transit-riders. The suggested lifestyle segments have intriguing policy implications. Lifecycle characteristics are used to predict lifestyle preferences, although there remain significant aspects that cannot be explained by observable variables.

A Latent Class Multidimensional Scaling Model for Two-Way One-Mode Continuous Rating Dissimilarity Data

ERIC Educational Resources Information Center

Vera, J. Fernando; Macias, Rodrigo; Heiser, Willem J.

2009-01-01

In this paper, we propose a cluster-MDS model for two-way one-mode continuous rating dissimilarity data. The model aims at partitioning the objects into classes and simultaneously representing the cluster centers in a low-dimensional space. Under the normal distribution assumption, a latent class model is developed in terms of the set of…

Latent Class Detection and Class Assignment: A Comparison of the MAXEIG Taxometric Procedure and Factor Mixture Modeling Approaches

ERIC Educational Resources Information Center

Lubke, Gitta; Tueller, Stephen

2010-01-01

Taxometric procedures such as MAXEIG and factor mixture modeling (FMM) are used in latent class clustering, but they have very different sets of strengths and weaknesses. Taxometric procedures, popular in psychiatric and psychopathology applications, do not rely on distributional assumptions. Their sole purpose is to detect the presence of latent…

Latent trajectory studies: the basics, how to interpret the results, and what to report.

PubMed

van de Schoot, Rens

2015-01-01

In statistics, tools have been developed to estimate individual change over time. Also, the existence of latent trajectories, where individuals are captured by trajectories that are unobserved (latent), can be evaluated (Muthén & Muthén, 2000). The method used to evaluate such trajectories is called Latent Growth Mixture Modeling (LGMM) or Latent Class Growth Modeling (LCGA). The difference between the two models is whether variance within latent classes is allowed for (Jung & Wickrama, 2008). The default approach most often used when estimating such models begins with estimating a single cluster model, where only a single underlying group is presumed. Next, several additional models are estimated with an increasing number of clusters (latent groups or classes). For each of these models, the software is allowed to estimate all parameters without any restrictions. A final model is chosen based on model comparison tools, for example, using the BIC, the bootstrapped chi-square test, or the Lo-Mendell-Rubin test. To ease the use of LGMM/LCGA step by step in this symposium (Van de Schoot, 2015) guidelines are presented which can be used for researchers applying the methods to longitudinal data, for example, the development of posttraumatic stress disorder (PTSD) after trauma (Depaoli, van de Schoot, van Loey, & Sijbrandij, 2015; Galatzer-Levy, 2015). The guidelines include how to use the software Mplus (Muthén & Muthén, 1998-2012) to run the set of models needed to answer the research question: how many latent classes exist in the data? The next step described in the guidelines is how to add covariates/predictors to predict class membership using the three-step approach (Vermunt, 2010). Lastly, it described what essentials to report in the paper. When applying LGMM/LCGA models for the first time, the guidelines presented can be used to guide what models to run and what to report.

Associations Between Neurotransmitter Genes and Fatigue and Energy Levels in Women Following Breast Cancer Surgery

PubMed Central

Eshragh, Jasmine; Dhruva, Anand; Paul, Steven M.; Cooper, Bruce A.; Mastick, Judy; Hamolsky, Deborah; Levine, Jon D.; Miaskowski, Christine; Kober, Kord M.

2016-01-01

Context Fatigue is a common problem in oncology patients. Less is known about decrements in energy levels and the mechanisms that underlie both fatigue and energy. Objectives In patients with breast cancer, variations in neurotransmitter genes between Lower and Higher Fatigue latent classes and between the Higher and Lower Energy latent classes were evaluated. Methods Patients completed assessments prior to and monthly for 6 months following surgery. Growth mixture modeling was used to identify distinct latent classes for fatigue severity and energy levels. Thirty candidate genes involved in various aspects of neurotransmission were evaluated. Results Eleven single nucleotide polymorphisms (SNPs) or haplotypes (i.e., ADRB2 rs1042718, BDNF rs6265, COMT rs9332377, CYP3A4 rs4646437, GALR1 rs949060, GCH1 rs3783642, NOS1 rs9658498, NOS1 rs2293052, NPY1R Haplotype A04, SLC6A2 rs17841327 and 5HTTLPR + rs25531 in SLC6A4) were associated with latent class membership for fatigue. Seven SNPs or haplotypes (i.e., NOS1 rs471871, SLC6A1 rs2675163, SLC6A1 Haplotype D01, SLC6A2 rs36027, SLC6A3 rs37022, SLC6A4 rs2020942, and TAC1 rs2072100) were associated with latent class membership for energy. Three of thirteen genes (i.e., NOS1, SLC6A2, SLC6A4) were associated with latent class membership for both fatigue and energy. Conclusions Molecular findings support the hypothesis that fatigue and energy are distinct, yet related symptoms. Results suggest that a large number of neurotransmitters play a role in the development and maintenance of fatigue and energy levels in breast cancer patients. PMID:27720787

Classifying Patients with Chronic Pelvic Pain into Levels of Biopsychosocial Dysfunction Using Latent Class Modeling of Patient Reported Outcome Measures

PubMed Central

Fenton, Bradford W.; Grey, Scott F.; Tossone, Krystel; McCarroll, Michele; Von Gruenigen, Vivian E.

2015-01-01

Chronic pelvic pain affects multiple aspects of a patient's physical, social, and emotional functioning. Latent class analysis (LCA) of Patient Reported Outcome Measures Information System (PROMIS) domains has the potential to improve clinical insight into these patients' pain. Based on the 11 PROMIS domains applied to n=613 patients referred for evaluation in a chronic pelvic pain specialty center, exploratory factor analysis (EFA) was used to identify unidimensional superdomains. Latent profile analysis (LPA) was performed to identify the number of homogeneous classes present and to further define the pain classification system. The EFA combined the 11 PROMIS domains into four unidimensional superdomains of biopsychosocial dysfunction: Pain, Negative Affect, Fatigue, and Social Function. Based on multiple fit criteria, a latent class model revealed four distinct classes of CPP: No dysfunction (3.2%); Low Dysfunction (17.8%); Moderate Dysfunction (53.2%); and High Dysfunction (25.8%). This study is the first description of a novel approach to the complex disease process such as chronic pelvic pain and was validated by demographic, medical, and psychosocial variables. In addition to an essentially normal class, three classes of increasing biopsychosocial dysfunction were identified. The LCA approach has the potential for application to other complex multifactorial disease processes. PMID:26355825

Evaluation of a Computer-aided Lung Auscultation System for Diagnosis of Bovine Respiratory Disease in Feedlot Cattle.

PubMed

Mang, A V; Buczinski, S; Booker, C W; Timsit, E

2015-01-01

A computer-aided lung auscultation (CALA) system was recently developed to diagnose bovine respiratory disease (BRD) in feedlot cattle. To determine, in a case-control study, the level of agreement between CALA and veterinary lung auscultation and to evaluate the sensitivity (Se) and specificity (Sp) of CALA to diagnose BRD in feedlot cattle. A total of 561 Angus cross-steers (initial body weight = 246 ± 45 kg) were observed during the first 50 day after entry to a feedlot. Case-control study. Steers with visual signs of BRD identified by pen checkers were examined by a veterinarian, including lung auscultation using a conventional stethoscope and CALA that produced a lung score from 1 (normal) to 5 (chronic). For each steer examined for BRD, 1 apparently healthy steer was selected as control and similarly examined. Agreement between CALA and veterinary auscultation was assessed by kappa statistic. CALA's Se and Sp were estimated using Bayesian latent class analysis. Of the 561 steers, 35 were identified with visual signs of BRD and 35 were selected as controls. Comparison of veterinary auscultation and CALA (using a CALA score ≥2 as a cut off) revealed a substantial agreement (kappa = 0.77). Using latent class analysis, CALA had a relatively high Se (92.9%; 95% credible interval [CI] = 0.71-0.99) and Sp (89.6%; 95% CI = 0.64-0.99) for diagnosing BRD compared with pen checking. CALA had good diagnostic accuracy (albeit with a relatively wide CI). Its use in feedlots could increase the proportion of cattle accurately diagnosed with BRD. Copyright © 2015 The Authors. Journal of Veterinary Internal Medicine published by Wiley Periodicals, Inc. on behalf of the American College of Veterinary Internal Medicine.

Alexithymia and psychosocial problems among Italian preadolescents. A latent class analysis approach.

PubMed

Mannarini, Stefania; Balottin, Laura; Toldo, Irene; Gatta, Michela

2016-10-01

The study, conducted on Italian preadolscents aged 11 to 13 belonging to the general population, aims to investigate the relationship between the emotional functioning, namely, alexithymia, and the risk of developing behavioral and emotional problems measured using the Strength and Difficulty Questionnaire. The latent class analysis approach allowed to identify two latent variables, accounting for the internalizing (emotional symptoms and difficulties in emotional awareness) and for the externalizing problems (conduct problems and hyperactivity, problematic relationships with peers, poor prosocial behaviors and externally oriented thinking). The two latent variables featured two latent classes: the difficulty in dealing with problems and the strength to face problems that was representative of most of the healthy participants with specific gender differences. Along with the analysis of psychopathological behaviors, the study of resilience and strengths can prove to be a key step in order to develop valuable preventive approaches to tackle psychiatric disorders. © 2016 Scandinavian Psychological Associations and John Wiley & Sons Ltd.

Bayesian classification theory

NASA Technical Reports Server (NTRS)

Hanson, Robin; Stutz, John; Cheeseman, Peter

1991-01-01

The task of inferring a set of classes and class descriptions most likely to explain a given data set can be placed on a firm theoretical foundation using Bayesian statistics. Within this framework and using various mathematical and algorithmic approximations, the AutoClass system searches for the most probable classifications, automatically choosing the number of classes and complexity of class descriptions. A simpler version of AutoClass has been applied to many large real data sets, has discovered new independently-verified phenomena, and has been released as a robust software package. Recent extensions allow attributes to be selectively correlated within particular classes, and allow classes to inherit or share model parameters though a class hierarchy. We summarize the mathematical foundations of AutoClass.

Latent Classes of PTSD Symptoms in Vietnam Veterans

ERIC Educational Resources Information Center

Steenkamp, Maria M.; Nickerson, Angela; Maguen, Shira; Dickstein, Benjamin D.; Nash, William P.; Litz, Brett T.

2012-01-01

The authors examined heterogeneity in posttraumatic stress disorder (PTSD) symptom presentation among veterans (n = 335) participating in the clinical interview subsample of the National Vietnam Veterans Readjustment Study. Latent class analysis was used to identify clinically homogeneous subgroups of Vietnam War combat veterans. Consistent with…

Race Differences in Patterns of Risky Behavior and Associated Risk Factors in Adolescence.

PubMed

Childs, Kristina K; Ray, James V

2017-05-01

Using data from the National Longitudinal Study of Adolescent Health (Add Health), this study expands on previous research by (a) examining differences across race in patterns or "subgroups" of adolescents based on nine self-reported behaviors (e.g., delinquency, substance use, risky sexual practices) and (b) comparing the risk factors (e.g., peer association, parenting, neighborhood cohesion), both within and across the race-specific subgroups, related to membership into the identified latent classes. The data used in this study include respondents aged 13 to 17 who participated in Waves 1 and 2 of the Add Health in-home interview. Latent class analysis (LCA) identified key differences in the number and characteristics of the latent classes across the racial subgroups. In addition, both similarities and differences in the risk factors for membership into the latent classes were identified across and within the race-specific subgroups. Implications for understanding risky behavior in adolescence, as well as directions for future research, are discussed.

Making sense of sparse rating data in collaborative filtering via topographic organization of user preference patterns.

PubMed

Polcicová, Gabriela; Tino, Peter

2004-01-01

We introduce topographic versions of two latent class models (LCM) for collaborative filtering. Latent classes are topologically organized on a square grid. Topographic organization of latent classes makes orientation in rating/preference patterns captured by the latent classes easier and more systematic. The variation in film rating patterns is modelled by multinomial and binomial distributions with varying independence assumptions. In the first stage of topographic LCM construction, self-organizing maps with neural field organized according to the LCM topology are employed. We apply our system to a large collection of user ratings for films. The system can provide useful visualization plots unveiling user preference patterns buried in the data, without loosing potential to be a good recommender model. It appears that multinomial distribution is most adequate if the model is regularized by tight grid topologies. Since we deal with probabilistic models of the data, we can readily use tools from probability and information theories to interpret and visualize information extracted by our system.

Prediction of incidence and stability of alcohol use disorders by latent internalizing psychopathology risk profiles in adolescence and young adulthood.

PubMed

Behrendt, Silke; Bühringer, Gerhard; Höfler, Michael; Lieb, Roselind; Beesdo-Baum, Katja

2017-10-01

Comorbid internalizing mental disorders in alcohol use disorders (AUD) can be understood as putative independent risk factors for AUD or as expressions of underlying shared psychopathology vulnerabilities. However, it remains unclear whether: 1) specific latent internalizing psychopathology risk-profiles predict AUD-incidence and 2) specific latent internalizing comorbidity-profiles in AUD predict AUD-stability. To investigate baseline latent internalizing psychopathology risk profiles as predictors of subsequent AUD-incidence and -stability in adolescents and young adults. Data from the prospective-longitudinal EDSP study (baseline age 14-24 years) were used. The study-design included up to three follow-up assessments in up to ten years. DSM-IV mental disorders were assessed with the DIA-X/M-CIDI. To investigate risk-profiles and their associations with AUD-outcomes, latent class analysis with auxiliary outcome variables was applied. AUD-incidence: a 4-class model (N=1683) was identified (classes: normative-male [45.9%], normative-female [44.2%], internalizing [5.3%], nicotine dependence [4.5%]). Compared to the normative-female class, all other classes were associated with a higher risk of subsequent incident alcohol dependence (p<0.05). AUD-stability: a 3-class model (N=1940) was identified with only one class (11.6%) with high probabilities for baseline AUD. This class was further characterized by elevated substance use disorder (SUD) probabilities and predicted any subsequent AUD (OR 8.5, 95% CI 5.4-13.3). An internalizing vulnerability may constitute a pathway to AUD incidence in adolescence and young adulthood. In contrast, no indication for a role of internalizing comorbidity profiles in AUD-stability was found, which may indicate a limited importance of such profiles - in contrast to SUD-related profiles - in AUD stability. Copyright © 2017 Elsevier B.V. All rights reserved.

Using latent class analysis to model prescription medications in the measurement of falling among a community elderly population

PubMed Central

2013-01-01

Background Falls among the elderly are a major public health concern. Therefore, the possibility of a modeling technique which could better estimate fall probability is both timely and needed. Using biomedical, pharmacological and demographic variables as predictors, latent class analysis (LCA) is demonstrated as a tool for the prediction of falls among community dwelling elderly. Methods Using a retrospective data-set a two-step LCA modeling approach was employed. First, we looked for the optimal number of latent classes for the seven medical indicators, along with the patients’ prescription medication and three covariates (age, gender, and number of medications). Second, the appropriate latent class structure, with the covariates, were modeled on the distal outcome (fall/no fall). The default estimator was maximum likelihood with robust standard errors. The Pearson chi-square, likelihood ratio chi-square, BIC, Lo-Mendell-Rubin Adjusted Likelihood Ratio test and the bootstrap likelihood ratio test were used for model comparisons. Results A review of the model fit indices with covariates shows that a six-class solution was preferred. The predictive probability for latent classes ranged from 84% to 97%. Entropy, a measure of classification accuracy, was good at 90%. Specific prescription medications were found to strongly influence group membership. Conclusions In conclusion the LCA method was effective at finding relevant subgroups within a heterogenous at-risk population for falling. This study demonstrated that LCA offers researchers a valuable tool to model medical data. PMID:23705639

Associations Between Latent Classes of Interpersonal Polyvictimization and Polyperpetration and Sexual Risk Behaviors Among Young Pregnant Couples: A Dyadic Analysis.

PubMed

Willie, Tiara; Kershaw, Trace S

2018-05-24

Interpersonal violence victimization and perpetration have been associated with sexual risk behaviors among adolescents and young adults, but research is lacking on: (1) how patterns of interpersonal polyvictimization and polyperpetration are associated with sexual risk among young pregnant couples, and (2) how individual and partner experiences of violence differentially impact sexual risk. The current analyses used baseline data from a longitudinal study that followed 296 pregnant young couples from pregnancy to 12 months postpartum. Couples were recruited at obstetrics and gynecology clinics, and an ultrasound clinic in the U.S. Latent class analysis identified subgroups based on polyvictimization and polyperpetration. Using the Actor-Partner Interdependence Model, path analyses assessed actor-partner effects of class membership on sexual risk. Three latent classes were used for women: Class 1: Polyvictim-Polyperpetrator; Class 2: Nonvictim-Nonperpetrator; and Class 3: Community and Prior IPV Victim. Four latent classes were used for men: Class 1: Community and Prior IPV Victim; Class 2: Polyvictim-Nonpartner Perpetrator; Class 3: Prior IPV and Peer Victim; and Class 4: Nonvictim-Nonperpetrator. Path analyses revealed that females in Class 2 and their male partners had higher condom use than females in Class 3. Males in Class 2 had more sexual partners than males in Class 1. Among nonmonogamous couples, males in Class 2 were less likely to be involved with a female partner reporting unprotected sex than males in Class 1. Among nonmonogamous couples, females in Class 2 had more acts of unprotected sex than females in Class 1. Males in Class 4 were less likely to have concurrent sexual partners compared to males in Class 1. Risk reduction interventions should address both victimization and perpetration. Additional research is needed to understand how mechanisms driving differential sexual risk by patterns of interpersonal polyvictimization and polyperpetration.

Post-traumatic stress symptoms and structure among orphan and vulnerable children and adolescents in Zambia.

PubMed

Familiar, Itziar; Murray, Laura; Gross, Alden; Skavenski, Stephanie; Jere, Elizabeth; Bass, Judith

2014-11-01

Scant information exists on PTSD symptoms and structure in youth from developing countries. We describe the symptom profile and exposure to trauma experiences among 343 orphan and vulnerable children and adolescents from Zambia. We distinguished profiles of post-traumatic stress symptoms using latent class analysis. Average number of trauma-related symptoms (21.6; range 0-38) was similar across sex and age. Latent class model suggested 3 classes varying by level of severity: low (31% of the sample), medium (45% of the sample), and high (24% of the sample) symptomatology. Results suggest that PTSD is a continuously distributed latent trait.

Modeling Individual Differences in Unfolding Preference Data: A Restricted Latent Class Approach.

ERIC Educational Resources Information Center

Bockenholt, Ulf; Bockenholt, Ingo

1990-01-01

A latent-class scaling approach is presented for modeling paired comparison and "pick any/t" data obtained in preference studies. The utility of this approach is demonstrated through analysis of data from studies involving consumer preference and preference for political candidates. (SLD)

Assets as a Socioeconomic Status Index: Categorical Principal Components Analysis vs. Latent Class Analysis.

PubMed

Sartipi, Majid; Nedjat, Saharnaz; Mansournia, Mohammad Ali; Baigi, Vali; Fotouhi, Akbar

2016-11-01

Some variables like Socioeconomic Status (SES) cannot be directly measured, instead, so-called 'latent variables' are measured indirectly through calculating tangible items. There are different methods for measuring latent variables such as data reduction methods e.g. Principal Components Analysis (PCA) and Latent Class Analysis (LCA). The purpose of our study was to measure assets index- as a representative of SES- through two methods of Non-Linear PCA (NLPCA) and LCA, and to compare them for choosing the most appropriate model. This was a cross sectional study in which 1995 respondents filled the questionnaires about their assets in Tehran. The data were analyzed by SPSS 19 (CATPCA command) and SAS 9.2 (PROC LCA command) to estimate their socioeconomic status. The results were compared based on the Intra-class Correlation Coefficient (ICC). The 6 derived classes from LCA based on BIC, were highly consistent with the 6 classes from CATPCA (Categorical PCA) (ICC = 0.87, 95%CI: 0.86 - 0.88). There is no gold standard to measure SES. Therefore, it is not possible to definitely say that a specific method is better than another one. LCA is a complicated method that presents detailed information about latent variables and required one assumption (local independency), while NLPCA is a simple method, which requires more assumptions. Generally, NLPCA seems to be an acceptable method of analysis because of its simplicity and high agreement with LCA.

A Rational Analysis of the Acquisition of Multisensory Representations

ERIC Educational Resources Information Center

Yildirim, Ilker; Jacobs, Robert A.

2012-01-01

How do people learn multisensory, or amodal, representations, and what consequences do these representations have for perceptual performance? We address this question by performing a rational analysis of the problem of learning multisensory representations. This analysis makes use of a Bayesian nonparametric model that acquires latent multisensory…

Bayesian Estimation of Multi-Unidimensional Graded Response IRT Models

ERIC Educational Resources Information Center

Kuo, Tzu-Chun

2015-01-01

Item response theory (IRT) has gained an increasing popularity in large-scale educational and psychological testing situations because of its theoretical advantages over classical test theory. Unidimensional graded response models (GRMs) are useful when polytomous response items are designed to measure a unified latent trait. They are limited in…

Specifying and Refining a Complex Measurement Model.

ERIC Educational Resources Information Center

Levy, Roy; Mislevy, Robert J.

This paper aims to describe a Bayesian approach to modeling and estimating cognitive models both in terms of statistical machinery and actual instrument development. Such a method taps the knowledge of experts to provide initial estimates for the probabilistic relationships among the variables in a multivariate latent variable model and refines…

Latent Class Subtyping of Attention-Deficit/Hyperactivity Disorder and Comorbid Conditions

ERIC Educational Resources Information Center

Acosta, Maria T.; Castellanos, F. Xavier; Bolton, Kelly L.; Balog, Joan Z.; Eagen, Patricia; Nee, Linda; Jones, Janet; Palacio, Luis; Sarampote, Christopher; Russell, Heather F.; Berg, Kate; Arcos-Burgos, Mauricio; Muenke, Maximilian

2008-01-01

The study attempts to carry out latent class analysis (LCA) in a sample of 1010 individuals, some with Attention-Deficit/Hyperactivity disorder (ADHD) and others normal. Results indicate that LCA can feasibly allow the combination of externalizing and internalizing symptoms for future tests regarding specific genetic risk factors.

Latent Profiles of Temperament and Their Relations of Psychopathology and Wellness

ERIC Educational Resources Information Center

Rettew, David C.; Althoff, Robert R.; Dumenci, Levent; Ayer, Lynsay; Hudziak, James J.

2008-01-01

The study applies latent profiles analysis to a group of children and adolescents to test temperament phenotypes in order to examine their association to wellness and psychopathology. One of the results concluded that lifetime disorder was lower in the steady class as compared to moderate class.

Bayesian evaluation of clinical diagnostic test characteristics of visual observations and remote monitoring to diagnose bovine respiratory disease in beef calves.

PubMed

White, Brad J; Goehl, Dan R; Amrine, David E; Booker, Calvin; Wildman, Brian; Perrett, Tye

2016-04-01

Accurate diagnosis of bovine respiratory disease (BRD) in beef cattle is a critical facet of therapeutic programs through promotion of prompt treatment of diseased calves in concert with judicious use of antimicrobials. Despite the known inaccuracies, visual observation (VO) of clinical signs is the conventional diagnostic modality for BRD diagnosis. Objective methods of remotely monitoring cattle wellness could improve diagnostic accuracy; however, little information exists describing the accuracy of this method compared to traditional techniques. The objective of this research is to employ Bayesian methodology to elicit diagnostic characteristics of conventional VO compared to remote early disease identification (REDI) to diagnose BRD. Data from previous literature on the accuracy of VO were combined with trial data consisting of direct comparison between VO and REDI for BRD in two populations. No true gold standard diagnostic test exists for BRD; therefore, estimates of diagnostic characteristics of each test were generated using Bayesian latent class analysis. Results indicate a 90.0% probability that the sensitivity of REDI (median 81.3%; 95% probability interval [PI]: 55.5, 95.8) was higher than VO sensitivity (64.5%; PI: 57.9, 70.8). The specificity of REDI (median 92.9%; PI: 88.2, 96.9) was also higher compared to VO (median 69.1%; PI: 66.3, 71.8). The differences in sensitivity and specificity resulted in REDI exhibiting higher positive and negative predictive values in both high (41.3%) and low (2.6%) prevalence situations. This research illustrates the potential of remote cattle monitoring to augment conventional methods of BRD diagnosis resulting in more accurate identification of diseased cattle. Copyright © 2016 Elsevier B.V. All rights reserved.

Sensitivity and specificity of a hand-held milk electrical conductivity meter compared to the California mastitis test for mastitis in dairy cattle.

PubMed

Fosgate, G T; Petzer, I M; Karzis, J

2013-04-01

Screening tests for mastitis can play an important role in proactive mastitis control programs. The primary objective of this study was to compare the sensitivity and specificity of milk electrical conductivity (EC) to the California mastitis test (CMT) in commercial dairy cattle in South Africa using Bayesian methods without a perfect reference test. A total of 1848 quarter milk specimens were collected from 173 cows sampled during six sequential farm visits. Of these samples, 25.8% yielded pathogenic bacterial isolates. The most frequently isolated species were coagulase negative Staphylococci (n=346), Streptococcus agalactiae (n=54), and Staphylococcus aureus (n=42). The overall cow-level prevalence of mastitis was 54% based on the Bayesian latent class (BLC) analysis. The CMT was more accurate than EC for classification of cows having somatic cell counts >200,000/mL and for isolation of a bacterial pathogen. BLC analysis also suggested an overall benefit of CMT over EC but the statistical evidence was not strong (P=0.257). The Bayesian model estimated the sensitivity and specificity of EC (measured via resistance) at a cut-point of >25 mΩ/cm to be 89.9% and 86.8%, respectively. The CMT had a sensitivity and specificity of 94.5% and 77.7%, respectively, when evaluated at the weak positive cut-point. EC was useful for identifying milk specimens harbouring pathogens but was not able to differentiate among evaluated bacterial isolates. Screening tests can be used to improve udder health as part of a proactive management plan. Copyright © 2012 Elsevier Ltd. All rights reserved.

Psychological Distress among Victimized Women on Probation and Parole: A Latent Class Analysis

PubMed Central

Golder, Seana; Engstrom, Malitta; Hall, Martin T.; Higgins, George; Logan, TK

2015-01-01

Latent class analysis was used to identify subgroups of victimized women (N=406) on probation and parole differentiated by levels of general psychological distress. The nine primary symptom dimensions from the Brief Symptom Inventory (BSI) were used individually as latent class indicators (Derogatis, 1993). Results identified three classes of women characterized by increasing levels of psychological distress; classes were further differentiated by posttraumatic stress disorder symptoms, cumulative victimization, substance use and other domains of psychosocial functioning (i.e., sociodemographic characteristics; informal social support and formal service utilization; perceived life stress; and resource loss). The present research was effective in uncovering important heterogeneity in psychological distress using a highly reliable and easily accessible measure of general psychological distress. Differentiating levels of psychological distress and associated patterns of psychosocial risk can be used to develop intervention strategies targeting the needs of different subgroups of women. Implications for treatment and future research are presented. PMID:25915692

Beyond the average marital communication: Latent profiles of the observed interactions among Chinese newlywed couples.

PubMed

Cao, Hongjian; Fang, Xiaoyi; Fine, Mark A; Ju, Xiaoyan; Lan, Jing; Liu, Xuanwen

2015-12-01

Employing a multicontext observational design, using a person-centered approach, and treating the marital dyad as the unit of analysis, this study examined the within-couple communication patterning of 144 Chinese newlywed couples and its association with relationship satisfaction. Latent profile analysis consistently revealed 3 profiles of spouses' interactive behaviors across contexts differing in both topic nature (i.e., problem-solving vs. social support) and initiator (i.e., husbands vs. wives): (a) traditionally undemonstrative profile, (b) emotionally quarrelling profile, and (c) warmly supportive profile. The prevalence of communication profiles changed markedly with the nature of the discussion topic and the topic initiator. Further, using latent class analysis, we classified couples into subgroups based on their identified profile memberships across contexts (i.e., consistency of interaction mode across contexts). Three classes were identified: (a) consistently quarrelling class, (b) consistently supportive class, and (c) modestly traditional class. Both the consistently supportive class and the modestly traditional class reported significantly higher levels of marital satisfaction than did the consistently quarrelling class. (c) 2015 APA, all rights reserved).

A Bayesian hierarchical model with novel prior specifications for estimating HIV testing rates

PubMed Central

An, Qian; Kang, Jian; Song, Ruiguang; Hall, H. Irene

2016-01-01

Human immunodeficiency virus (HIV) infection is a severe infectious disease actively spreading globally, and acquired immunodeficiency syndrome (AIDS) is an advanced stage of HIV infection. The HIV testing rate, that is, the probability that an AIDS-free HIV infected person seeks a test for HIV during a particular time interval, given no previous positive test has been obtained prior to the start of the time, is an important parameter for public health. In this paper, we propose a Bayesian hierarchical model with two levels of hierarchy to estimate the HIV testing rate using annual AIDS and AIDS-free HIV diagnoses data. At level one, we model the latent number of HIV infections for each year using a Poisson distribution with the intensity parameter representing the HIV incidence rate. At level two, the annual numbers of AIDS and AIDS-free HIV diagnosed cases and all undiagnosed cases stratified by the HIV infections at different years are modeled using a multinomial distribution with parameters including the HIV testing rate. We propose a new class of priors for the HIV incidence rate and HIV testing rate taking into account the temporal dependence of these parameters to improve the estimation accuracy. We develop an efficient posterior computation algorithm based on the adaptive rejection metropolis sampling technique. We demonstrate our model using simulation studies and the analysis of the national HIV surveillance data in the USA. PMID:26567891

The Integration of Continuous and Discrete Latent Variable Models: Potential Problems and Promising Opportunities

ERIC Educational Resources Information Center

Bauer, Daniel J.; Curran, Patrick J.

2004-01-01

Structural equation mixture modeling (SEMM) integrates continuous and discrete latent variable models. Drawing on prior research on the relationships between continuous and discrete latent variable models, the authors identify 3 conditions that may lead to the estimation of spurious latent classes in SEMM: misspecification of the structural model,…

Polydrug Use and HIV Risk Among People Who Inject Heroin in Tijuana, Mexico: A Latent Class Analysis.

PubMed

Meacham, Meredith C; Rudolph, Abby E; Strathdee, Steffanie A; Rusch, Melanie L; Brouwer, Kimberly C; Patterson, Thomas L; Vera, Alicia; Rangel, Gudelia; Roesch, Scott C

2015-01-01

Although most people who inject drugs (PWID) in Tijuana, Mexico, primarily inject heroin, injection and non-injection use of methamphetamine and cocaine is common. We examined patterns of polydrug use among heroin injectors to inform prevention and treatment of drug use and its health and social consequences. Participants were PWID residing in Tijuana, aged ≥18 years who reported heroin injection in the past six months and were recruited through respondent-driven sampling (n = 1,025). Latent class analysis was conducted to assign individuals to classes on a probabilistic basis, using four indicators of past six-month polydrug and polyroute use: cocaine injecting, cocaine smoking or snorting, methamphetamine injecting, and methamphetamine smoking or snorting. Latent class membership was regressed onto covariates in a multinomial logistic regression. Latent class analyses testing 1, 2, 3, and 4 classes were fit, with the 3-class solution fitting best. Class 1 was defined by predominantly heroin use (50.2%, n = 515); class 2 by methamphetamine and heroin use (43.7%, n = 448), and class 3 by methamphetamine, cocaine, and heroin use (6.0%, n = 62). Bivariate and multivariate analyses indicated a group of methamphetamine and cocaine users that exhibited higher-risk sexual practices and lower heroin injecting frequency, and a group of methamphetamine users who were younger and more likely to be female. Discrete subtypes of heroin PWID were identified based on methamphetamine and cocaine use patterns. These findings have identified subtypes of heroin injectors who require more tailored interventions to reduce the health and social harms of injecting drug use.

Polydrug use and HIV risk among people who inject heroin in Tijuana, Mexico: A Latent class analysis

PubMed Central

Meacham, M.C.; Rudolph, A.E.; Strathdee, S.A.; Rusch, M.L.; Brouwer, K.C.; Patterson, T.L.; Vera, A.; Rangel, G.; Roesch, S.C.

2016-01-01

Background Although most people who inject drugs (PWID) in Tijuana, Mexico, primarily inject heroin, injection and non-injection use of methamphetamine and cocaine is common. We examined patterns of polydrug use among heroin injectors to inform prevention and treatment of drug use and its health and social consequences. Methods Participants were PWID residing in Tijuana aged ≥ 18 years who reported heroin injection in the past 6 months and were recruited through respondent driven sampling (n=1025). Latent class analysis was conducted to assign individuals to classes on a probabilistic basis, using four indicators of past 6 month polydrug and polyroute use: cocaine injecting, cocaine smoking or snorting, methamphetamine injecting, methamphetamine smoking or snorting. Latent class membership was regressed onto covariates in a multinomial logistic regression. Results Latent class analyses testing 1, 2, 3, and 4 classes were fit, with the 3-class solution fitting best. Class 1 was defined by predominantly heroin use (50.2%, n=515); class 2 by methamphetamine and heroin use (43.7%, n=448), and class 3 by methamphetamine, cocaine, and heroin use (6.0%, n=62). Bivariate and multivariate analyses indicated a group of methamphetamine and cocaine users that exhibited higher risk sexual practices and lower heroin injecting frequency, and a group of methamphetamine users who were younger and more likely to be female. Conclusions Discrete subtypes of heroin PWID were identified based on methamphetamine and cocaine use patterns. These findings have identified subtypes of heroin injectors who require more tailored interventions to reduce the health and social harms of injecting drug use. PMID:26444185

Examining the Latent Class Structure of CO2 Hypersensitivity using Time Course Trajectories of Panic Response Systems

PubMed Central

Roberson-Nay, Roxann; Beadel, Jessica R.; Gorlin, Eugenia I.; Latendresse, Shawn J.; Teachman, Bethany A.

2014-01-01

Background and Objectives Carbon dioxide (CO2) hypersensitivity is hypothesized to be a robust endophenotypic marker of panic spectrum vulnerability. The goal of the current study was to explore the latent class trajectories of three primary response systems theoretically associated with CO2 hypersensitivity: subjective anxiety, panic symptoms, and respiratory rate (fR). Methods Participants (n=376; 56% female) underwent a maintained 7.5% CO2 breathing task that included three phases: baseline, CO2 air breathing, and recovery. Growth mixture modeling was used to compare response classes (1..n) to identify the best-fit model for each marker. Panic correlates also were examined to determine class differences in panic vulnerability. Results For subjective anxiety ratings, a three-class model was selected, with individuals in one class reporting an acute increase in anxiety during 7.5% CO2 breathing and a return to pre-CO2 levels during recovery. A second, smaller latent class was distinguished by elevated anxiety across all three phases. The third class reported low anxiety reported during room air, a mild increase in anxiety during 7.5% CO2 breathing, and a return to baseline during recovery. Latent class trajectories for fR yielded one class whereas panic symptom response yielded two classes. Limitations This study examined CO2 hypersensitivity in one of the largest samples to date, but did not ascertain a general population sample thereby limiting generalizability. Moreover, a true resting baseline measure of fR was not measured. Conclusions Two classes potentially representing different risk pathways were observed. Implications of results will be discussed in the context of panic risk research. PMID:25496936

Latent Class Analysis of Early Developmental Trajectory in Baby Siblings of Children with Autism

PubMed Central

Landa, Rebecca J.; Gross, Alden L.; Stuart, Elizabeth A.; Bauman, Margaret

2012-01-01

Background Siblings of children with autism (sibs-A) are at increased genetic risk for autism spectrum disorders (ASD) and milder impairments. To elucidate diversity and contour of early developmental trajectories exhibited by sibs-A, regardless of diagnostic classification, latent class modeling was used. Methods Sibs-A (n=204) were assessed with the Mullen Scales of Early Learning from age 6–36 months. Mullen T scores served as dependent variables. Outcome classifications at age 36 months included: ASD (n=52); non-ASD social/communication delay (broader autism phenotype; BAP) (n=31); and unaffected (n=121). Child-specific patterns of performance were studied using latent class growth analysis. Latent class membership was then related to diagnostic outcome through estimation of within-class proportions of children assigned to each diagnostic classification. Results A 4-class model was favored. Class 1 represented accelerated development and consisted of 25.7% of the sample, primarily unaffected children. Class 2 (40.0% of the sample), was characterized by normative development with above-average nonverbal cognitive outcome. Class 3 (22.3% of the sample) was characterized by receptive language, and gross and fine motor delay. Class 4 (12.0% of the sample), was characterized by widespread delayed skill acquisition, reflected by declining trajectories. Children with an outcome diagnosis of ASD were spread across Classes 2, 3, and 4. Conclusions Results support a category of ASD that involves slowing in early non-social development. Receptive language and motor development is vulnerable to early delay in sibs-A with and without ASD outcomes. Non-ASD sibs-A are largely distributed across classes depicting average or accelerated development. Developmental trajectories of motor, language, and cognition appear independent of communication and social delays in non-ASD sibs-A. PMID:22574686

Class Enumeration and Parameter Recovery of Growth Mixture Modeling and Second-Order Growth Mixture Modeling in the Presence of Measurement Noninvariance between Latent Classes

PubMed Central

Kim, Eun Sook; Wang, Yan

2017-01-01

Population heterogeneity in growth trajectories can be detected with growth mixture modeling (GMM). It is common that researchers compute composite scores of repeated measures and use them as multiple indicators of growth factors (baseline performance and growth) assuming measurement invariance between latent classes. Considering that the assumption of measurement invariance does not always hold, we investigate the impact of measurement noninvariance on class enumeration and parameter recovery in GMM through a Monte Carlo simulation study (Study 1). In Study 2, we examine the class enumeration and parameter recovery of the second-order growth mixture modeling (SOGMM) that incorporates measurement models at the first order level. Thus, SOGMM estimates growth trajectory parameters with reliable sources of variance, that is, common factor variance of repeated measures and allows heterogeneity in measurement parameters between latent classes. The class enumeration rates are examined with information criteria such as AIC, BIC, sample-size adjusted BIC, and hierarchical BIC under various simulation conditions. The results of Study 1 showed that the parameter estimates of baseline performance and growth factor means were biased to the degree of measurement noninvariance even when the correct number of latent classes was extracted. In Study 2, the class enumeration accuracy of SOGMM depended on information criteria, class separation, and sample size. The estimates of baseline performance and growth factor mean differences between classes were generally unbiased but the size of measurement noninvariance was underestimated. Overall, SOGMM is advantageous in that it yields unbiased estimates of growth trajectory parameters and more accurate class enumeration compared to GMM by incorporating measurement models. PMID:28928691

Bayesian analysis and classification of two Enzyme-Linked Immunosorbent Assay (ELISA) tests without a gold standard

PubMed Central

Zhang, Jingyang; Chaloner, Kathryn; McLinden, James H.; Stapleton, Jack T.

2013-01-01

Reconciling two quantitative ELISA tests for an antibody to an RNA virus, in a situation without a gold standard and where false negatives may occur, is the motivation for this work. False negatives occur when access of the antibody to the binding site is blocked. Based on the mechanism of the assay, a mixture of four bivariate normal distributions is proposed with the mixture probabilities depending on a two-stage latent variable model including the prevalence of the antibody in the population and the probabilities of blocking on each test. There is prior information on the prevalence of the antibody, and also on the probability of false negatives, and so a Bayesian analysis is used. The dependence between the two tests is modeled to be consistent with the biological mechanism. Bayesian decision theory is utilized for classification. The proposed method is applied to the motivating data set to classify the data into two groups: those with and those without the antibody. Simulation studies describe the properties of the estimation and the classification. Sensitivity to the choice of the prior distribution is also addressed by simulation. The same model with two levels of latent variables is applicable in other testing procedures such as quantitative polymerase chain reaction tests where false negatives occur when there is a mutation in the primer sequence. PMID:23592433

Progressive Elaboration and Cross-Validation of a Latent Class Typology of Adolescent Alcohol Involvement in a National Sample

PubMed Central

Donovan, John E.; Chung, Tammy

2015-01-01

Objective: Most studies of adolescent drinking focus on single alcohol use behaviors (e.g., high-volume drinking, drunkenness) and ignore the patterning of adolescents’ involvement across multiple alcohol behaviors. The present latent class analyses (LCAs) examined a procedure for empirically determining multiple cut points on the alcohol use behaviors in order to establish a typology of adolescent alcohol involvement. Method: LCA was carried out on six alcohol use behavior indicators collected from 6,504 7th through 12th graders who participated in Wave I of the National Longitudinal Study of Adolescent Health (AddHealth). To move beyond dichotomous indicators, a “progressive elaboration” strategy was used, starting with six dichotomous indicators and then evaluating a series of models testing additional cut points on the ordinal indicators at progressively higher points for one indicator at a time. Analyses were performed on one random half-sample, and confirmatory LCAs were performed on the second random half-sample and in the Wave II data. Results: The final model consisted of four latent classes (never or non–current drinkers, low-intake drinkers, non–problem drinkers, and problem drinkers). Confirmatory LCAs in the second random half-sample from Wave I and in Wave II support this four-class solution. The means on the four latent classes were also generally ordered on an array of measures reflecting psychosocial risk for problem behavior. Conclusions: These analyses suggest that there may be four different classes or types of alcohol involvement among adolescents, and, more importantly, they illustrate the utility of the progressive elaboration strategy for moving beyond dichotomous indicators in latent class models. PMID:25978828

Clustering of modifiable biobehavioral risk factors for chronic disease in US adults: a latent class analysis.

PubMed

Leventhal, Adam M; Huh, Jimi; Dunton, Genevieve F

2014-11-01

Examining the co-occurrence patterns of modifiable biobehavioral risk factors for deadly chronic diseases (e.g. cancer, cardiovascular disease, diabetes) can elucidate the etiology of risk factors and guide disease-prevention programming. The aims of this study were to (1) identify latent classes based on the clustering of five key biobehavioral risk factors among US adults who reported at least one risk factor and (2) explore the demographic correlates of the identified latent classes. Participants were respondents of the National Epidemiologic Survey of Alcohol and Related Conditions (2004-2005) with at least one of the following disease risk factors in the past year (N = 22,789), which were also the latent class indicators: (1) alcohol abuse/dependence, (2) drug abuse/dependence, (3) nicotine dependence, (4) obesity, and (5) physical inactivity. Housing sample units were selected to match the US National Census in location and demographic characteristics, with young adults oversampled. Participants were administered surveys by trained interviewers. Five latent classes were yielded: 'obese, active non-substance abusers' (23%); 'nicotine-dependent, active, and non-obese' (19%); 'active, non-obese alcohol abusers' (6%); 'inactive, non-substance abusers' (50%); and 'active, polysubstance abusers' (3.7%). Four classes were characterized by a 100% likelihood of having one risk factor coupled with a low or moderate likelihood of having the other four risk factors. The five classes exhibited unique demographic profiles. Risk factors may cluster together in a non-monotonic fashion, with the majority of the at-risk population of US adults expected to have a high likelihood of endorsing only one of these five risk factors. © Royal Society for Public Health 2013.

Predictors of Patterns of Alcohol-Related Blackouts Over Time in Youth From the Collaborative Study of the Genetics of Alcoholism: The Roles of Genetics and Cannabis

PubMed Central

Schuckit, Marc A.; Smith, Tom L.; Shafir, Alexandra; Clausen, Peyton; Danko, George; Gonçalves, Priscila Dib; Anthenelli, Robert M.; Chan, Grace; Kuperman, Samuel; Hesselbrock, Michie; Hesselbrock, Victor; Kramer, John; Bucholz, Kathleen K.

2017-01-01

Objective: Alcohol-related blackouts (ARBs) are anterograde amnesias related to heavy alcohol intake seen in about 50% of drinkers. Although a major determinant of ARBs relates to blood alcohol concentrations, additional contributions come from genetic vulnerabilities and possible impacts of cannabis use disorders (CUDs). We evaluated relationships of genetics and cannabis use to latent class trajectories of ARBs in 829 subjects from the Collaborative Study of the Genetics of Alcoholism (COGA). Method: The number of ARBs experienced every 2 years from subjects with average ages of 18 to 25 were entered into a latent class growth analysis in Mplus, and resulting class membership was evaluated in light of baseline characteristics, including CUDs. Correlations of number of ARBs across assessments were also compared for sibling pairs versus unrelated subjects. Results: Latent class growth analysis identified ARB-based Classes 1 (consistent low = 42.5%), 2 (moderate low = 28.3%), 3 (moderate high = 22.9%), and 4 (consistent high = 6.3%). A multinomial logistic regression analysis within latent class growth analysis revealed that baseline CUDs related most closely to Classes 3 and 4. The number of ARBs across time correlated .23 for sibling pairs and -.10 for unrelated subjects. Conclusions: Baseline CUDs related to the most severe latent ARB course over time, even when considered along with other trajectory predictors, including baseline alcohol use disorders and maximum number of drinks. Data indicated significant roles for genetic factors for alcohol use disorder patterns over time. Future research is needed to improve understanding of how cannabis adds to the ARB risk and to find genes that contribute to risks for ARBs among drinkers. PMID:27936363

The heterogeneous health latent classes of elderly people and their socio-demographic characteristics in Taiwan.

PubMed

Liu, Li-Fan; Tian, Wei-Hua; Yao, Hui-Ping

2014-01-01

The health care needs of elderly people were influenced by their heterogeneity. This study aimed to identify the health latent classes of elderly people by using latent class analysis to deal with heterogeneity and examine their socio-demographic characteristics. Data came from the 2005 National Health Interview Survey (NHIS) in Taiwan. In total, 2449 elderly individuals with available health indicators were examined in latent class analysis (LCA), and 2217 elderly community-dwellings with complete socio-demographic data were analyzed by multinomial logistic regression. Four health latent classes were identified which included 1066 (43.5%) people in the High Comorbidity (HC), 152 (6.2%) in the Functional Impairment (FI), 252 (10.3%) in the Frail (FR), and 979 (40.0%) in the Relatively Healthy (RH) group. Multinomial logistic regressions revealed socio-demographic characteristics among health classes. The variables associated with an increased likelihood of being in the FR group were age, female, and living with families. They were also correlated to ethnicity and educations. Apart from age and gender, the Functional Impairment group was less likely to be ethnicity of Hakka, more likely to live with others than were the RH group. The HC group tended to be younger, with higher educations, and more likely to live in urban area than the Functional Impairment group. The correlations between health classes and socio-demographic factors were discussed. The health status of elderly people includes a variety of health indicators. A person-centered approach is critical to identify the health heterogeneity of elderly people and manage their care needs by targeting differential aging. Copyright © 2013 Elsevier Ireland Ltd. All rights reserved.

Co-occurrence of Victimization from Five Subtypes of Bullying: Physical, Verbal, Social Exclusion, Spreading Rumors, and Cyber

PubMed Central

Iannotti, Ronald J.; Luk, Jeremy W.; Nansel, Tonja R.

2010-01-01

Objective To examine co-occurrence of five subtypes of peer victimization. Methods Data were obtained from a national sample of 7,475 US adolescents in grades 6 through 10 in the 2005/2006 Health Behavior in School-Aged Children (HBSC) study. Latent class analyses (LCA) were conducted on victimization by physical, verbal, social exclusion, spreading rumors, and cyber bullying. Results Three latent classes were identified, including an all-types victims class (9.7% of males and 6.2% of females), a verbal/relational victims class (28.1% of males and 35.1% of females), and a nonvictim class (62.2% of males and 58.7% of females). Males were more likely to be all-type victims. There was a graded relationship between the three latent classes and level of depression, frequency of medically attended injuries, and medicine use, especially among females. Conclusions  Increased co-occurrence of victimization types put adolescents at greater risks for poorer physical and psychological outcomes. PMID:20488883

Longitudinal analysis of latent classes of psychopathology and patterns of class migration in survivors of severe injury.

PubMed

Forbes, David; Nickerson, Angela; Alkemade, Nathan; Bryant, Richard A; Creamer, Mark; Silove, Derrick; McFarlane, Alexander C; Van Hooff, Miranda; Fletcher, Susan L; O'Donnell, Meaghan

2015-09-01

Little research to date has explored the typologies of psychopathology following trauma, beyond development of particular diagnoses such as posttraumatic stress disorder (PTSD). The objective of this study was to determine the longitudinal patterns of these typologies, especially the movement of persons across clusters of psychopathology. In this 6-year longitudinal study, 1,167 hospitalized severe injury patients who were recruited between April 2004-February 2006 were analyzed, with repeated measures at baseline, 3 months, 12 months, and 72 months after injury. All patients met the DSM-IV criterion A1 for PTSD. Structured clinical interviews were used to assess psychiatric disorders at each follow-up point. Latent class analysis and latent transition analysis were applied to assess clusters of individuals determined by psychopathology. The Mini International Neuropsychiatric Interview (MINI) and Clinician-Administered PTSD Scale (CAPS) were employed to complete diagnoses. Four latent classes were identified at each time point: (1) Alcohol/Depression class (3 months, 2.1%; 12 months, 1.3%; and 72 months, 1.1%), (2) Alcohol class (3 months, 3.3%; 12 months, 3.7%; and 72 months, 5.4%), (3) PTSD/Depression class (3 months, 10.3%; 12 months, 11.5%; and 72 months, 6.4%), and (4) No Disorder class (3 months, 84.2%; 12 months, 83.5%; and 72 months, 87.1%). Latent transition analyses conducted across the 2 transition points (12 months and 72 months) found consistently high levels of stability in the No Disorder class (90.9%, 93.0%, respectively) but lower and reducing levels of consistency in the PTSD/Depression class (81.3%, 46.6%), the Alcohol/Depression class (59.7%, 21.5%), and the Alcohol class (61.0%, 36.5%), demonstrating high levels of between-class migration. Despite the array of psychiatric disorders that may develop following severe injury, a 4-class model best described the data with excellent classification certainty. The high levels of migration across classes indicate a complex pattern of psychopathology expression over time. The findings have considerable implications for tailoring multifocused interventions to class type, as well as flexible stepped care models, and for the potential development and delivery of transdiagnostic interventions targeting underlying mechanisms. © Copyright 2015 Physicians Postgraduate Press, Inc.

Post-traumatic stress symptoms and structure among orphan and vulnerable children and adolescents in Zambia

PubMed Central

Familiar, Itziar; Murray, Laura; Gross, Alden; Skavenski, Stephanie; Jere, Elizabeth; Bass, Judith

2014-01-01

Background Scant information exists on PTSD symptoms and structure in youth from developing countries. Methods We describe the symptom profile and exposure to trauma experiences among 343 orphan and vulnerable children and adolescents from Zambia. We distinguished profiles of post-traumatic stress symptoms using latent class analysis. Results Average number of trauma-related symptoms (21.6; range 0-38) was similar across sex and age. Latent class model suggested 3 classes varying by level of severity: low (31% of the sample), medium (45% of the sample), and high (24% of the sample) symptomatology. Conclusions Results suggest that PTSD is a continuously distributed latent trait. PMID:25382359

Bayesian switching factor analysis for estimating time-varying functional connectivity in fMRI.

PubMed

Taghia, Jalil; Ryali, Srikanth; Chen, Tianwen; Supekar, Kaustubh; Cai, Weidong; Menon, Vinod

2017-07-15

There is growing interest in understanding the dynamical properties of functional interactions between distributed brain regions. However, robust estimation of temporal dynamics from functional magnetic resonance imaging (fMRI) data remains challenging due to limitations in extant multivariate methods for modeling time-varying functional interactions between multiple brain areas. Here, we develop a Bayesian generative model for fMRI time-series within the framework of hidden Markov models (HMMs). The model is a dynamic variant of the static factor analysis model (Ghahramani and Beal, 2000). We refer to this model as Bayesian switching factor analysis (BSFA) as it integrates factor analysis into a generative HMM in a unified Bayesian framework. In BSFA, brain dynamic functional networks are represented by latent states which are learnt from the data. Crucially, BSFA is a generative model which estimates the temporal evolution of brain states and transition probabilities between states as a function of time. An attractive feature of BSFA is the automatic determination of the number of latent states via Bayesian model selection arising from penalization of excessively complex models. Key features of BSFA are validated using extensive simulations on carefully designed synthetic data. We further validate BSFA using fingerprint analysis of multisession resting-state fMRI data from the Human Connectome Project (HCP). Our results show that modeling temporal dependencies in the generative model of BSFA results in improved fingerprinting of individual participants. Finally, we apply BSFA to elucidate the dynamic functional organization of the salience, central-executive, and default mode networks-three core neurocognitive systems with central role in cognitive and affective information processing (Menon, 2011). Across two HCP sessions, we demonstrate a high level of dynamic interactions between these networks and determine that the salience network has the highest temporal flexibility among the three networks. Our proposed methods provide a novel and powerful generative model for investigating dynamic brain connectivity. Copyright © 2017 Elsevier Inc. All rights reserved.

Heterogeneity of sleep quality in relation to circadian preferences and depressive symptomatology among major depressive patients.

PubMed

Selvi, Yavuz; Boysan, Murat; Kandeger, Ali; Uygur, Omer F; Sayin, Ayca A; Akbaba, Nursel; Koc, Basak

2018-08-01

The current study aimed at investigating the latent dimensional structure of sleep quality as indexed by the seven components of the Pittsburgh Sleep Quality Index (PSQI), as well as latent covariance structure between sleep quality, circadian preferences and depressive symptoms. Two hundred twenty-five patients with major depressive disorder (MDD), with an average age of 29.92 ± 10.49 years (aged between 17 and 63), participated in the study. The PSQI, Morningness-Eveningness Questionnaire (MEQ) and Beck Depression Inventory (BDI) were administered to participants. Four sets of latent class analyses were subsequently run to obtain optimal number of latent classes best fit to the data. Mixture models revealed that sleep quality is multifaceted in MDD. The data best fit to four-latent-class model: Poor Habitual Sleep Quality (PHSQ), Poor Subjective Sleep Quality (PSSQ), Intermediate Sleep Quality (ISQ), and Good Sleep Quality (GSQ). MDD patients classified into GSQ latent class (23.6%) reported the lowest depressive symptoms and were more prone to morningness diurnal preferences compared to other three homogenous sub-groups. Finally, the significant association between eveningness diurnal preferences and depressive symptomatology was significantly mediated by poor sleep quality. The cross-sectional nature of the study and the lack of an objective measurement of sleep such as polysomnography recordings was the most striking limitation of the study. We concluded sleep quality in relation to circadian preferences and depressive symptoms has a heterogeneous nature in MDD. Copyright © 2018. Published by Elsevier B.V.

A personality-based latent class typology of outpatients with major depressive disorder: association with symptomatology, prescription pattern and social function.

PubMed

Hori, Hiroaki; Teraishi, Toshiya; Nagashima, Anna; Koga, Norie; Ota, Miho; Hattori, Kotaro; Kim, Yoshiharu; Higuchi, Teruhiko; Kunugi, Hiroshi

2017-08-01

While major depressive disorder (MDD) is considered to be a heterogeneous disorder, the nature of the heterogeneity remains unclear. Studies have attempted to classify patients with MDD using latent variable techniques, yet the empirical approaches to symptom-based subtyping of MDD have not provided conclusive evidence. Here we aimed to identify homogeneous classes of MDD based on personality traits, using a latent profile analysis. We studied 238 outpatients with DSM-IV MDD recruited from our specialized depression outpatient clinic and assessed their dimensional personality traits with the Temperament and Character Inventory. Latent profile analysis was conducted with 7 dimensions of the Temperament and Character Inventory as indicators. Relationships of the identified classes with symptomatology, prescription pattern, and social function were then examined. The latent profile analysis indicated that a 3-class solution best fit the data. Of the sample, 46.2% was classified into a "neurotic" group characterized by high harm avoidance and low self-directedness; 30.3% into an "adaptive" group characterized by high self-directedness and cooperativeness; and 23.5% into a "socially-detached" group characterized by low reward dependence and cooperativeness and high self-transcendence. The 2 maladaptive groups, namely neurotic and socially-detached groups, demonstrated unique patterns of symptom expression, different classes of psychotropic medication use, and lower social functioning. Generalizability of the findings was limited since our patients were recruited from the specialized depression outpatient clinic. Our personality-based latent profile analysis identified clinically meaningful 3 MDD groups that were markedly different in their personality profiles associated with distinct symptomatology and functioning. Copyright © 2017 Elsevier B.V. All rights reserved.

The Latent Classes of Subclinical ADHD Symptoms: Convergences of Multiple Informant Reports

ERIC Educational Resources Information Center

Kobor, Andrea; Takacs, Adam; Urban, Robert; Csepe, Valeria

2012-01-01

The purpose of the present study was to conduct latent class analysis on the Hyperactivity scale of the Strengths and Difficulties Questionnaire in order to identify distinct subgroups of subclinical ADHD in a multi-informant framework. We hypothesized a similar structure between teachers and parents, and differences in symptom severity across…

Some Factor Analytic Approximations to Latent Class Structure.

ERIC Educational Resources Information Center

Dziuban, Charles D.; Denton, William T.

Three procedures, alpha, image, and uniqueness rescaling, were applied to a joint occurrence probability matrix. That matrix was the basis of a well-known latent class structure. The values of the recurring subscript elements were varied as follows: Case 1 - The known elements were input; Case 2 - The upper bounds to the recurring subscript…

A Mixture Rasch Model-Based Computerized Adaptive Test for Latent Class Identification

ERIC Educational Resources Information Center

Jiao, Hong; Macready, George; Liu, Junhui; Cho, Youngmi

2012-01-01

This study explored a computerized adaptive test delivery algorithm for latent class identification based on the mixture Rasch model. Four item selection methods based on the Kullback-Leibler (KL) information were proposed and compared with the reversed and the adaptive KL information under simulated testing conditions. When item separation was…

Latent Class Analysis of Conduct Problems of Elementary Students Receiving Special Education Services

ERIC Educational Resources Information Center

Toupin, Jean; Déry, Michèle; Verlaan, Pierrette; Lemelin, Jean-Pascal; Lecocq, Aurélie; Jagiellowicz, Jadwiga

2016-01-01

Students with conduct problems (CPs) may present heterogeneity in terms of behavioral manifestations and service needs. Previous studies using Latent Class Analysis (LCA) to capture this heterogeneity have been conducted mostly with community samples and have often applied a narrow definition of CP. Considering this context, this study…

Multilevel Latent Class Analysis: Parametric and Nonparametric Models

ERIC Educational Resources Information Center

Finch, W. Holmes; French, Brian F.

2014-01-01

Latent class analysis is an analytic technique often used in educational and psychological research to identify meaningful groups of individuals within a larger heterogeneous population based on a set of variables. This technique is flexible, encompassing not only a static set of variables but also longitudinal data in the form of growth mixture…

A comparison of latent class, K-means, and K-median methods for clustering dichotomous data.

PubMed

Brusco, Michael J; Shireman, Emilie; Steinley, Douglas

2017-09-01

The problem of partitioning a collection of objects based on their measurements on a set of dichotomous variables is a well-established problem in psychological research, with applications including clinical diagnosis, educational testing, cognitive categorization, and choice analysis. Latent class analysis and K-means clustering are popular methods for partitioning objects based on dichotomous measures in the psychological literature. The K-median clustering method has recently been touted as a potentially useful tool for psychological data and might be preferable to its close neighbor, K-means, when the variable measures are dichotomous. We conducted simulation-based comparisons of the latent class, K-means, and K-median approaches for partitioning dichotomous data. Although all 3 methods proved capable of recovering cluster structure, K-median clustering yielded the best average performance, followed closely by latent class analysis. We also report results for the 3 methods within the context of an application to transitive reasoning data, in which it was found that the 3 approaches can exhibit profound differences when applied to real data. (PsycINFO Database Record (c) 2017 APA, all rights reserved).

Relationships among Behavioral Profiles, Reading Performance, and Disability Labels: A Latent Class Analysis of Students in the ED, LD, and OHI Special Education Categories

ERIC Educational Resources Information Center

Weingarten, Zachary

2018-01-01

The aim of this study was to explore the variation in student behavior across the ED, LD, and OHI disability categories and to examine demographic, behavioral, and academic factors that may place students at risk for negative outcomes. This study used teachers' rating of students' in-class behavior to identify latent classes of students in the ED,…

Latent classes of sexual behaviors: Prevalence, predictors, and consequences

PubMed Central

Wesche, Rose; Lefkowitz, Eva S.; Vasilenko, Sara A.

2016-01-01

Scholars of adolescent and emerging adult sexuality have recently begun to study how diverse patterns of sexual behaviors contribute to development and well-being. A person-oriented approach to studying sexual behaviors provides a nuanced understanding of sexual repertoires. The goals of this paper were to document patterns of sexual behaviors ranging from kissing to penetrative sex, and to examine how latent classes of behaviors, gender, and partner type (romantic vs. nonromantic) predict intra- and interpersonal consequences of sexual behaviors. Latent class analysis of a stratified random sample of U.S. college students revealed four classes of sexual behaviors: Kissing Only, Kissing and Touching, All Behaviors, and Oral and Penetrative Only. Compared to individuals in the All Behaviors class, individuals in the Kissing Only class were less likely to experience a positive or a negative intrapersonal consequence of sexual behaviors. Men were less likely to report a negative intrapersonal consequence than women were. Partner type predicted negative interpersonal consequences for the All Behaviors class. Implications are discussed in terms of normative sexual development, prevention, and sexual and relationship education. PMID:28163800

Parent–Child Relationships in Stepfather Families and Adolescent Adjustment: A Latent Class Analysis

PubMed Central

Amato, Paul R.; King, Valarie; Thorsen, Maggie L.

2015-01-01

In the current study the authors drew on Waves I and III from Add Health to examine the closeness of parent–adolescent relationships in married mother–stepfather families (N = 1,934). They used latent class analysis to identify family constellations defined by adolescents’ relationships with all of their parents: mothers, stepfathers, and biological nonresident fathers. In particular, the authors (a) identified the most common underlying patterns of adolescent–parent relationships in stepfamilies; (b) determined the background characteristics that predict membership in these groups; and (c) examined how adolescents in these groups fare with respect to depressive symptoms, delinquency, and substance use. The results indicate that adolescents’ relationships can be represented with 4 latent classes. Adolescents in these classes differ on measures of adjustment, and many of these differences persist into the early adult years. PMID:27022199

The association between school exclusion, delinquency and subtypes of cyber- and F2F-victimizations: identifying and predicting risk profiles and subtypes using latent class analysis.

PubMed

Barboza, Gia Elise

2015-01-01

This purpose of this paper is to identify risk profiles of youth who are victimized by on- and offline harassment and to explore the consequences of victimization on school outcomes. Latent class analysis is used to explore the overlap and co-occurrence of different clusters of victims and to examine the relationship between class membership and school exclusion and delinquency. Participants were a random sample of youth between the ages of 12 and 18 selected for inclusion to participate in the 2011 National Crime Victimization Survey: School Supplement. The latent class analysis resulted in four categories of victims: approximately 3.1% of students were highly victimized by both bullying and cyberbullying behaviors; 11.6% of youth were classified as being victims of relational bullying, verbal bullying and cyberbullying; a third class of students were victims of relational bullying, verbal bullying and physical bullying but were not cyberbullied (8%); the fourth and final class, characteristic of the majority of students (77.3%), was comprised of non-victims. The inclusion of covariates to the latent class model indicated that gender, grade and race were significant predictors of at least one of the four victim classes. School delinquency measures were included as distal outcomes to test for both overall and pairwise associations between classes. With one exception, the results were indicative of a significant relationship between school delinquency and the victim subtypes. Implications for these findings are discussed. Copyright © 2014 Elsevier Ltd. All rights reserved.

Heterogeneity in patterns of DSM-5 posttraumatic stress disorder and depression symptoms: Latent profile analyses.

PubMed

Contractor, Ateka A; Roley-Roberts, Michelle E; Lagdon, Susan; Armour, Cherie

2017-04-01

Posttraumatic stress disorder (PTSD) and depression co-occur frequently following the experience of potentially traumatizing events (PTE; Morina et al., 2013). A person-centered approach to discern heterogeneous patterns of such co-occurring symptoms is recommended (Galatzer-Levy and Bryant, 2013). We assessed heterogeneity in PTSD and depression symptomatology; and subsequently assessed relations between class membership with psychopathology constructs (alcohol use, distress tolerance, dissociative experiences). The sample consisted of 268 university students who had experienced a PTE and susequently endorsed clinical levels of PTSD or depression severity. Latent profile analyses (LPA) was used to identify the best-fitting class solution accouring to recommended fit indices (Nylund et al., 2007a); and the effects of covariates was analyzed using a 3-step approach (Vermunt, 2010). Results of the LPA indicated an optimal 3-class solutions: high severity (Class 2), lower PTSD-higher depression (Class 1), and higher PTSD-lower depression (Class 3). Covariates of distress tolerance, and different kinds of dissociative experiences differentiated the latent classes. Use of self-report measure could lead to response biases; and the specific nature of the sample limits generalizability of results. We found evidence for a depressive subtype of PTSD differentiated from other classes in terms of lower distress tolerance and greater dissociative experiences. Thus, transdiagnostic treatment protocols may be most beneficial for these latent class members. Further, the distinctiveness of PTSD and depression at comparatively lower levels of PTSD severity was supported (mainly in terms of distress tolerance abilities); hence supporting the current classification system placement of these disorders. Copyright © 2017 Elsevier B.V. All rights reserved.

Latent Cognitive Phenotypes in De Novo Parkinson's Disease: A Person-Centered Approach.

PubMed

LaBelle, Denise R; Walsh, Ryan R; Banks, Sarah J

2017-08-01

Cognitive impairment is an important aspect of Parkinson's disease (PD), but there is considerable heterogeneity in its presentation. This investigation aims to identify and characterize latent cognitive phenotypes in early PD. Latent class analysis, a data-driven, person-centered, cluster analysis was performed on cognitive data from the Parkinson's Progressive Markers Initiative baseline visit. This analytic method facilitates identification of naturally occurring endophenotypes. Resulting classes were compared across biomarker, symptom, and demographic data. Six cognitive phenotypes were identified. Three demonstrated consistent performance across indicators, representing poor ("Weak-Overall"), average ("Typical-Overall"), and strong ("Strong-Overall") cognition. The remaining classes demonstrated unique patterns of cognition, characterized by "Strong-Memory," "Weak-Visuospatial," and "Amnestic" profiles. The Amnestic class evidenced greater tremor severity and anosmia, but was unassociated with biomarkers linked with Alzheimer's disease. The Weak-Overall class was older and reported more non-motor features associated with cognitive decline, including anxiety, depression, autonomic dysfunction, anosmia, and REM sleep behaviors. The Strong-Overall class was younger, more female, and reported less dysautonomia and anosmia. Classes were unrelated to disease duration, functional independence, or available biomarkers. Latent cognitive phenotypes with focal patterns of impairment were observed in recently diagnosed individuals with PD. Cognitive profiles were found to be independent of traditional biomarkers and motoric indices of disease progression. Only globally impaired class was associated with previously reported indicators of cognitive decline, suggesting this group may drive the effects reported in studies using variable-based analysis. Longitudinal and neuroanatomical characterization of classes will yield further insight into the evolution of cognitive change in the disease. (JINS, 2017, 23, 551-563).

A Bayesian method for assessing multiscalespecies-habitat relationships

USGS Publications Warehouse

Stuber, Erica F.; Gruber, Lutz F.; Fontaine, Joseph J.

2017-01-01

ContextScientists face several theoretical and methodological challenges in appropriately describing fundamental wildlife-habitat relationships in models. The spatial scales of habitat relationships are often unknown, and are expected to follow a multi-scale hierarchy. Typical frequentist or information theoretic approaches often suffer under collinearity in multi-scale studies, fail to converge when models are complex or represent an intractable computational burden when candidate model sets are large.ObjectivesOur objective was to implement an automated, Bayesian method for inference on the spatial scales of habitat variables that best predict animal abundance.MethodsWe introduce Bayesian latent indicator scale selection (BLISS), a Bayesian method to select spatial scales of predictors using latent scale indicator variables that are estimated with reversible-jump Markov chain Monte Carlo sampling. BLISS does not suffer from collinearity, and substantially reduces computation time of studies. We present a simulation study to validate our method and apply our method to a case-study of land cover predictors for ring-necked pheasant (Phasianus colchicus) abundance in Nebraska, USA.ResultsOur method returns accurate descriptions of the explanatory power of multiple spatial scales, and unbiased and precise parameter estimates under commonly encountered data limitations including spatial scale autocorrelation, effect size, and sample size. BLISS outperforms commonly used model selection methods including stepwise and AIC, and reduces runtime by 90%.ConclusionsGiven the pervasiveness of scale-dependency in ecology, and the implications of mismatches between the scales of analyses and ecological processes, identifying the spatial scales over which species are integrating habitat information is an important step in understanding species-habitat relationships. BLISS is a widely applicable method for identifying important spatial scales, propagating scale uncertainty, and testing hypotheses of scaling relationships.

A Hierarchical Bayesian Multidimensional Scaling Methodology for Accommodating Both Structural and Preference Heterogeneity

ERIC Educational Resources Information Center

Park, Joonwook; Desarbo, Wayne S.; Liechty, John

2008-01-01

Multidimensional scaling (MDS) models for the analysis of dominance data have been developed in the psychometric and classification literature to simultaneously capture subjects' "preference heterogeneity" and the underlying dimentional structure for a set of designated stimuli in a parsimonious manner. There are two major types of latent utility…

Eye Tracking and Pupillometry are Indicators of Dissociable Latent Decision Processes

PubMed Central

Cavanagh, James F.; Wiecki, Thomas V.; Kochar, Angad; Frank, Michael J.

2014-01-01

Can you predict what someone is going to do just by watching them? This is certainly difficult: it would require a clear mapping between observable indicators and unobservable cognitive states. In this report we demonstrate how this is possible by monitoring eye gaze and pupil dilation, which predict dissociable biases during decision making. We quantified decision making using the Drift Diffusion Model (DDM), which provides an algorithmic account of how evidence accumulation and response caution contribute to decisions through separate latent parameters of drift rate and decision threshold, respectively. We used a hierarchical Bayesian estimation approach to assess the single trial influence of observable physiological signals on these latent DDM parameters. Increased eye gaze dwell time specifically predicted an increased drift rate toward the fixated option, irrespective of the value of the option. In contrast, greater pupil dilation specifically predicted an increase in decision threshold during difficult decisions. These findings suggest that eye tracking and pupillometry reflect the operations of dissociated latent decision processes. PMID:24548281

Spatial Bayesian Latent Factor Regression Modeling of Coordinate-based Meta-analysis Data

PubMed Central

Montagna, Silvia; Wager, Tor; Barrett, Lisa Feldman; Johnson, Timothy D.; Nichols, Thomas E.

2017-01-01

Summary Now over 20 years old, functional MRI (fMRI) has a large and growing literature that is best synthesised with meta-analytic tools. As most authors do not share image data, only the peak activation coordinates (foci) reported in the paper are available for Coordinate-Based Meta-Analysis (CBMA). Neuroimaging meta-analysis is used to 1) identify areas of consistent activation; and 2) build a predictive model of task type or cognitive process for new studies (reverse inference). To simultaneously address these aims, we propose a Bayesian point process hierarchical model for CBMA. We model the foci from each study as a doubly stochastic Poisson process, where the study-specific log intensity function is characterised as a linear combination of a high-dimensional basis set. A sparse representation of the intensities is guaranteed through latent factor modeling of the basis coefficients. Within our framework, it is also possible to account for the effect of study-level covariates (meta-regression), significantly expanding the capabilities of the current neuroimaging meta-analysis methods available. We apply our methodology to synthetic data and neuroimaging meta-analysis datasets. PMID:28498564

A Taxonomy of Latent Structure Assumptions for Probability Matrix Decomposition Models.

ERIC Educational Resources Information Center

Meulders, Michel; De Boeck, Paul; Van Mechelen, Iven

2003-01-01

Proposed a taxonomy of latent structure assumptions for probability matrix decomposition (PMD) that includes the original PMD model and a three-way extension of the multiple classification latent class model. Simulation study results show the usefulness of the taxonomy. (SLD)

Effects of additional data on Bayesian clustering.

PubMed

Yamazaki, Keisuke

2017-10-01

Hierarchical probabilistic models, such as mixture models, are used for cluster analysis. These models have two types of variables: observable and latent. In cluster analysis, the latent variable is estimated, and it is expected that additional information will improve the accuracy of the estimation of the latent variable. Many proposed learning methods are able to use additional data; these include semi-supervised learning and transfer learning. However, from a statistical point of view, a complex probabilistic model that encompasses both the initial and additional data might be less accurate due to having a higher-dimensional parameter. The present paper presents a theoretical analysis of the accuracy of such a model and clarifies which factor has the greatest effect on its accuracy, the advantages of obtaining additional data, and the disadvantages of increasing the complexity. Copyright © 2017 Elsevier Ltd. All rights reserved.

Characteristics of cyclist crashes in Italy using latent class analysis and association rule mining

PubMed Central

De Angelis, Marco; Marín Puchades, Víctor; Fraboni, Federico; Pietrantoni, Luca

2017-01-01

The factors associated with severity of the bicycle crashes may differ across different bicycle crash patterns. Therefore, it is important to identify distinct bicycle crash patterns with homogeneous attributes. The current study aimed at identifying subgroups of bicycle crashes in Italy and analyzing separately the different bicycle crash types. The present study focused on bicycle crashes that occurred in Italy during the period between 2011 and 2013. We analyzed categorical indicators corresponding to the characteristics of infrastructure (road type, road signage, and location type), road user (i.e., opponent vehicle and cyclist’s maneuver, type of collision, age and gender of the cyclist), vehicle (type of opponent vehicle), and the environmental and time period variables (time of the day, day of the week, season, pavement condition, and weather). To identify homogenous subgroups of bicycle crashes, we used latent class analysis. Using latent class analysis, the bicycle crash data set was segmented into 19 classes, which represents 19 different bicycle crash types. Logistic regression analysis was used to identify the association between class membership and severity of the bicycle crashes. Finally, association rules were conducted for each of the latent classes to uncover the factors associated with an increased likelihood of severity. Association rules highlighted different crash characteristics associated with an increased likelihood of severity for each of the 19 bicycle crash types. PMID:28158296

Classification and Short-Term Course of DSM-IV Cannabis, Hallucinogen, Cocaine, and Opioid Disorders in Treated Adolescents

ERIC Educational Resources Information Center

Chung, Tammy; Martin, Christoper S.

2005-01-01

This study examined the latent class structure of Diagnostic and Statistical Manual of Mental Disorders (text rev.; DSM-IV; American Psychiatric Association, 2000) symptoms used to diagnose cannabis, hallucinogen, cocaine, and opiate disorders among 501 adolescents recruited from addictions treatment. Latent class results were compared with the…

Evidence for Latent Classes of IQ in Young Children with Autism Spectrum Disorder

ERIC Educational Resources Information Center

Munson, Jeffrey; Dawson, Geraldine; Sterling, Lindsey; Beauchaine, Theodore; Zhou, Andrew; Koehler, Elizabeth; Lord, Catherine; Rogers, Sally; Sigman, Marian; Estes, Annette; Abbott, Robert

2008-01-01

Autism is currently viewed as a spectrum condition that includes strikingly different severity levels; IQ is consistently described as one of the primary aspects of the heterogeneity in autism. To investigate the possibility of more than one distinct subtype of autism based on IQ, both latent class analysis and taxometrics methods were used to…

Multilevel Latent Class Analysis: An Application of Adolescent Smoking Typologies with Individual and Contextual Predictors

ERIC Educational Resources Information Center

Henry, Kimberly L.; Muthen, Bengt

2010-01-01

Latent class analysis (LCA) is a statistical method used to identify subtypes of related cases using a set of categorical or continuous observed variables. Traditional LCA assumes that observations are independent. However, multilevel data structures are common in social and behavioral research and alternative strategies are needed. In this…

Latent Class Analysis of Early Developmental Trajectory in Baby Siblings of Children with Autism

ERIC Educational Resources Information Center

Landa, Rebecca J.; Gross, Alden L.; Stuart, Elizabeth A.; Bauman, Margaret

2012-01-01

Background: Siblings of children with autism (sibs-A) are at increased genetic risk for autism spectrum disorders (ASD) and milder impairments. To elucidate diversity and contour of early developmental trajectories exhibited by sibs-A, regardless of diagnostic classification, latent class modeling was used. Methods: Sibs-A (N = 204) were assessed…

A Latent Class Analysis of Adolescent Gambling: Application of Resilience Theory

ERIC Educational Resources Information Center

Goldstein, Abby L.; Faulkner, Breanne; Cunningham, Rebecca M.; Zimmerman, Marc A.; Chermack, Stephen; Walton, Maureen A.

2013-01-01

The current study examined the application of resilience theory to adolescent gambling using Latent Class Analysis (LCA) to establish subtypes of adolescent gamblers and to explore risk and promotive factors associated with gambling group membership. Participants were a diverse sample of 249 adolescents ages 14 to 18 (30.1 % female, 59.4 % African…

Hybrid Model of IRT and Latent Class Models.

ERIC Educational Resources Information Center

Yamamoto, Kentaro

This study developed a hybrid of item response theory (IRT) models and latent class models, which combined the strengths of each type of model. The primary motivation for developing the new model is to describe characteristics of examinees' knowledge at the time of the examination. Hence, the application of the model lies mainly in so-called…

Identifying Students' Expectancy-Value Beliefs: A Latent Class Analysis Approach to Analyzing Middle School Students' Science Self-Perceptions

ERIC Educational Resources Information Center

Phelan, Julia; Ing, Marsha; Nylund-Gibson, Karen; Brown, Richard S.

2017-01-01

This study extends current research by organizing information about students' expectancy-value achievement motivation, in a way that helps parents and teachers identify specific entry points to encourage and support students' science aspirations. This study uses latent class analysis to describe underlying differences in ability beliefs, task…

Exploring the Relationship between Autism Spectrum Disorder and Epilepsy Using Latent Class Cluster Analysis

ERIC Educational Resources Information Center

Cuccaro, Michael L.; Tuchman, Roberto F.; Hamilton, Kara L.; Wright, Harry H.; Abramson, Ruth K.; Haines, Jonathan L.; Gilbert, John R.; Pericak-Vance, Margaret

2012-01-01

Epilepsy co-occurs frequently in autism spectrum disorders (ASD). Understanding this co-occurrence requires a better understanding of the ASD-epilepsy phenotype (or phenotypes). To address this, we conducted latent class cluster analysis (LCCA) on an ASD dataset (N = 577) which included 64 individuals with epilepsy. We identified a 5-cluster…

The Structure of Student Satisfaction with College Services: A Latent Class Model

ERIC Educational Resources Information Center

Adwere-Boamah, Joseph

2011-01-01

Latent Class Analysis (LCA) was used to identify distinct groups of Community college students based on their self-ratings of satisfaction with student service programs. The programs were counseling, financial aid, health center, student programs and student government. The best fitting model to describe the data was a two Discrete-Factor model…

Patterns of Adolescent Bullying Behaviors: Physical, Verbal, Exclusion, Rumor, and Cyber

ERIC Educational Resources Information Center

Wang, Jing; Iannotti, Ronald J.; Luk, Jeremy W.

2012-01-01

Patterns of engagement in cyber bullying and four types of traditional bullying were examined using latent class analysis (LCA). Demographic differences and externalizing problems were evaluated across latent class membership. Data were obtained from the 2005-2006 Health Behavior in School-aged Survey and the analytic sample included 7,508 U.S.…

Solidarity and Conflict between Adult Children and Parents: A Latent Class Analysis

ERIC Educational Resources Information Center

van Gaalen, Ruben I.; Dykstra, Pearl A.

2006-01-01

Using multiple dimensions of solidarity and conflict in a latent class analysis, we develop a typology of adult child-parent relationships. The data (N = 4,990) are from the first wave of the Netherlands Kinship Panel Study. In descending order of relationship quality, the 5 types are harmonious (akin to relationships with friends), ambivalent…

Tailoring community-based wellness initiatives with latent class analysis--Massachusetts Community Transformation Grant projects.

PubMed

Arcaya, Mariana; Reardon, Timothy; Vogel, Joshua; Andrews, Bonnie K; Li, Wenjun; Land, Thomas

2014-02-13

Community-based approaches to preventing chronic diseases are attractive because of their broad reach and low costs, and as such, are integral components of health care reform efforts. Implementing community-based initiatives across Massachusetts' municipalities presents both programmatic and evaluation challenges. For effective delivery and evaluation of the interventions, establishing a community typology that groups similar municipalities provides a balanced and cost-effective approach. Through a series of key informant interviews and exploratory data analysis, we identified 55 municipal-level indicators of 6 domains for the typology analysis. The domains were health behaviors and health outcomes, housing and land use, transportation, retail environment, socioeconomics, and demographic composition. A latent class analysis was used to identify 10 groups of municipalities based on similar patterns of municipal-level indicators across the domains. Our model with 10 latent classes yielded excellent classification certainty (relative entropy = .995, minimum class probability for any class = .871), and differentiated distinct groups of municipalities based on health-relevant needs and resources. The classes differentiated healthy and racially and ethnically diverse urban areas from cities with similar population densities and diversity but worse health outcomes, affluent communities from lower-income rural communities, and mature suburban areas from rapidly suburbanizing communities with different healthy-living challenges. Latent class analysis is a tool that may aid in the planning, communication, and evaluation of community-based wellness initiatives such as Community Transformation Grants projects administrated by the Centers for Disease Control and Prevention.

Characterizing Longitudinal Patterns of Physical Activity in Mid-Adulthood Using Latent Class Analysis: Results From a Prospective Cohort Study

PubMed Central

Silverwood, Richard J.; Nitsch, Dorothea; Pierce, Mary; Kuh, Diana; Mishra, Gita D.

2011-01-01

The authors aimed to describe how longitudinal patterns of physical activity during mid-adulthood (ages 31–53 years) can be characterized using latent class analysis in a population-based birth cohort study, the Medical Research Council’s 1946 National Survey of Health and Development. Three different types of physical activity—walking, cycling, and leisure-time physical activity—were analyzed separately using self-reported data collected from questionnaires between 1977 and 1999; 3,847 study members were included in the analysis for one or more types of activity. Patterns of activity differed by sex, so stratified analyses were conducted. Two walking latent classes were identified representing low (52.8% of males in the cohort, 33.5% of females) and high (47.2%, 66.5%) levels of activity. Similar low (91.4%, 82.1%) and high (8.6%, 17.9%) classes were found for cycling, while 3 classes were identified for leisure-time physical activity: “low activity” (46.2%, 48.2%), “sports and leisure activity” (31.0%, 35.3%), and “gardening and do-it-yourself activities” (22.8%, 16.5%). The classes were reasonably or very well separated, with the exception of walking in females. Latent class analysis was found to be a useful tool for characterizing longitudinal patterns of physical activity, even when the measurement instrument differs slightly across ages, which added value in comparison with observed activity at a single age. PMID:22074812

Latent Classes of Adolescent Sexual and Romantic Relationship Experiences: Implications for Adult Sexual Health and Relationship Outcomes

PubMed Central

Vasilenko, Sara A.; Kugler, Kari C.; Lanza, Stephanie T.

2015-01-01

Adolescents’ sexual and romantic relationship experiences are multidimensional, but often studied as single constructs. Thus, it is not clear how different patterns of sexual and relationship experience may interact to differentially predict later outcomes. In this study we used latent class analysis to model patterns (latent classes) of adolescent sexual and romantic experiences, and then examined how these classes are associated with young adult sexual health and relationship outcomes in data from the National Longitudinal Study of Adolescent to Adult Health. We identified six adolescent relationship classes: No Relationship (33%), Waiting (22%), Intimate (38%), Private (3%), Low Involvement (3%), and Physical (2%). Adolescents in the Waiting and Intimate classes were more likely to have married by young adulthood than those in other classes, and those in the Physical class had a greater number of sexual partners and higher rates of STIs. Some gender differences were found; for example, women in the Low-involvement and Physical classes in adolescence had average or high odds of marriage, whereas men in these classes had relatively low odds of marriage. Our findings identify more and less normative patterns of romantic and sexual experiences in late adolescence, and elucidate associations between adolescent experiences and adult outcomes. PMID:26445133

Latent class analysis of gambling subtypes and impulsive/compulsive associations: Time to rethink diagnostic boundaries for gambling disorder?

PubMed

Chamberlain, Samuel R; Stochl, Jan; Redden, Sarah A; Odlaug, Brian L; Grant, Jon E

2017-09-01

Gambling disorder has been associated with cognitive dysfunction and impaired quality of life. The current definition of non-pathological, problem, and pathological types of gambling is based on total symptom scores, which may overlook nuanced underlying presentations of gambling symptoms. The aims of the current study were (i) to identify subtypes of gambling in young adults, using latent class analysis, based on individual responses from the Structured Clinical Interview for Gambling Disorder (SCI-GD); and (ii) to explore relationships between these gambling subtypes, and clinical/cognitive measures. Total 582 non-treatment seeking young adults were recruited from two US cities, on the basis of gambling five or more times per year. Participants undertook clinical and neurocognitive assessment, including stop-signal, decision-making, and set-shifting tasks. Data from individual items of the Structured Clinical Interview for Gambling Disorder (SCI-GD) were entered into latent class analysis. Optimal number of classes representing gambling subtypes was identified using Bayesian Information Criterion and differences between them were explored using multivariate analysis of variance. Three subtypes of gambling were identified, termed recreational gamblers (60.2% of the sample; reference group), problem gamblers (29.2%), and pathological gamblers (10.5%). Common quality of life impairment, elevated Barratt Impulsivity scores, occurrence of mainstream mental disorders, having a first degree relative with an addiction, and impaired decision-making were evident in both problem and pathological gambling groups. The diagnostic item 'chasing losses' most discriminated recreational from problem gamblers, while endorsement of 'social, financial, or occupational losses due to gambling' most discriminated pathological gambling from both other groups. Significantly higher rates of impulse control disorders occurred in the pathological group, versus the problem group, who in turn showed significantly higher rates than the reference group. The pathological group also had higher set-shifting errors and nicotine consumption. Even problem gamblers who had a relatively low total SCI-PG scores (mean endorsement of two items) exhibited impaired quality of life, objective cognitive impairment on decision-making, and occurrence of other mental disorders that did not differ significantly from those seen in the pathological gamblers. Furthermore, problem/pathological gambling was associated with other impulse control disorders, but not increased alcohol use. Groups differed on quality of life when classified using the data-driven approach, but not when classified using DSM cut-offs. Thus, the current DSM-5 approach will fail to discriminate a significant fraction of patients with biologically plausible, functionally impairing illness, and may not be ideal in terms of diagnostic classification. Cognitive distortions related to 'chasing losses' represent a particularly important candidate treatment target for early intervention. Copyright © 2017 The Authors. Published by Elsevier Ltd.. All rights reserved.

Severity of mental illness as a result of multiple childhood adversities: US National Epidemiologic Survey.

PubMed

Curran, Emma; Adamson, Gary; Stringer, Maurice; Rosato, Michael; Leavey, Gerard

2016-05-01

To examine patterns of childhood adversity, their long-term consequences and the combined effect of different childhood adversity patterns as predictors of subsequent psychopathology. Secondary analysis of data from the US National Epidemiologic Survey on alcohol and related conditions. Using latent class analysis to identify childhood adversity profiles; and using multinomial logistic regression to validate and further explore these profiles with a range of associated demographic and household characteristics. Finally, confirmatory factor analysis substantiated initial latent class analysis findings by investigating a range of mental health diagnoses. Latent class analysis generated a three-class model of childhood adversity in which 60 % of participants were allocated to a low adversity class; 14 % to a global adversities class (reporting exposures for all the derived latent classes); and 26 % to a domestic emotional and physical abuse class (exposed to a range of childhood adversities). Confirmatory Factor analysis defined an internalising-externalising spectrum to represent lifetime reporting patterns of mental health disorders. Using logistic regression, both adversity groups showed specific gender and race/ethnicity differences, related family discord and increased psychopathology. We identified underlying patterns in the exposure to childhood adversity and associated mental health. These findings are informative in their description of the configuration of adversities, rather than focusing solely on the cumulative aspect of experience. Amelioration of longer-term negative consequences requires early identification of psychopathology risk factors that can inform protective and preventive interventions. This study highlights the utility of screening for childhood adversities when individuals present with symptoms of psychiatric disorders.

Bayesian 2-Stage Space-Time Mixture Modeling With Spatial Misalignment of the Exposure in Small Area Health Data.

PubMed

Lawson, Andrew B; Choi, Jungsoon; Cai, Bo; Hossain, Monir; Kirby, Russell S; Liu, Jihong

2012-09-01

We develop a new Bayesian two-stage space-time mixture model to investigate the effects of air pollution on asthma. The two-stage mixture model proposed allows for the identification of temporal latent structure as well as the estimation of the effects of covariates on health outcomes. In the paper, we also consider spatial misalignment of exposure and health data. A simulation study is conducted to assess the performance of the 2-stage mixture model. We apply our statistical framework to a county-level ambulatory care asthma data set in the US state of Georgia for the years 1999-2008.

Lay Americans' views of why scientists disagree with each other.

PubMed

Johnson, Branden B; Dieckmann, Nathan F

2017-10-01

A survey experiment assessed response to five explanations of scientific disputes: problem complexity, self-interest, values, competence, and process choices (e.g. theories and methods). A US lay sample ( n = 453) did not distinguish interests from values, nor competence from process, as explanations of disputes. Process/competence was rated most likely and interests/values least; all, on average, were deemed likely to explain scientific disputes. Latent class analysis revealed distinct subgroups varying in their explanation preferences, with a more complex latent class structure for participants who had heard of scientific disputes in the past. Scientific positivism and judgments of science's credibility were the strongest predictors of latent class membership, controlling for scientific reasoning, political ideology, confidence in choice, scenario, education, gender, age, and ethnicity. The lack of distinction observed overall between different explanations, as well as within classes, raises challenges for further research on explanations of scientific disputes people find credible and why.

Latent transition models with latent class predictors: attention deficit hyperactivity disorder subtypes and high school marijuana use

PubMed Central

Reboussin, Beth A.; Ialongo, Nicholas S.

2011-01-01

Summary Attention deficit hyperactivity disorder (ADHD) is a neurodevelopmental disorder which is most often diagnosed in childhood with symptoms often persisting into adulthood. Elevated rates of substance use disorders have been evidenced among those with ADHD, but recent research focusing on the relationship between subtypes of ADHD and specific drugs is inconsistent. We propose a latent transition model (LTM) to guide our understanding of how drug use progresses, in particular marijuana use, while accounting for the measurement error that is often found in self-reported substance use data. We extend the LTM to include a latent class predictor to represent empirically derived ADHD subtypes that do not rely on meeting specific diagnostic criteria. We begin by fitting two separate latent class analysis (LCA) models by using second-order estimating equations: a longitudinal LCA model to define stages of marijuana use, and a cross-sectional LCA model to define ADHD subtypes. The LTM model parameters describing the probability of transitioning between the LCA-defined stages of marijuana use and the influence of the LCA-defined ADHD subtypes on these transition rates are then estimated by using a set of first-order estimating equations given the LCA parameter estimates. A robust estimate of the LTM parameter variance that accounts for the variation due to the estimation of the two sets of LCA parameters is proposed. Solving three sets of estimating equations enables us to determine the underlying latent class structures independently of the model for the transition rates and simplifying assumptions about the correlation structure at each stage reduces the computational complexity. PMID:21461139

Microaggressions, Discrimination, and Phenotype among African Americans: A Latent Class Analysis of the Impact of Skin Tone and BMI.

PubMed

Keith, Verna M; Nguyen, Ann W; Taylor, Robert Joseph; Mouzon, Dawne M; Chatters, Linda M

2017-05-01

Data from the 2001-2003National Survey of American Life are used to investigate the effects of phenotype on everyday experiences with discrimination among African Americans (N=3343). Latent class analysis is used to identify four classes of discriminatory treatment: 1) low levels of discrimination, 2) disrespect and condescension, 3) character-based discrimination, and 4) high levels of discrimination. We then employ latent class multinomial logistic regression to evaluate the association between skin tone and body weight and these four classes of discrimination. Designating the low level discrimination class as the reference group, findings revealed that respondents with darker skin were more likely to be classified into the disrespect/condescension and the high level microaggression types. BMI was unrelated to the discrimination type, although there was a significant interaction effect between gender and BMI. BMI was strongly and positively associated with membership in the disrespect and condescension type among men but not among women. These findings indicate that skin tone and body weight are two phenotypic characteristics that influence the type and frequency of discrimination experienced by African Americans.

Latent class analysis to evaluate performance of point-of-care CCA for low-intensity Schistosoma mansoni infections in Burundi.

PubMed

Clements, Michelle N; Corstjens, Paul L A M; Binder, Sue; Campbell, Carl H; de Dood, Claudia J; Fenwick, Alan; Harrison, Wendy; Kayugi, Donatien; King, Charles H; Kornelis, Dieuwke; Ndayishimiye, Onesime; Ortu, Giuseppina; Lamine, Mariama Sani; Zivieri, Antonio; Colley, Daniel G; van Dam, Govert J

2018-02-23

Kato-Katz examination of stool smears is the field-standard method for detecting Schistosoma mansoni infection. However, Kato-Katz misses many active infections, especially of light intensity. Point-of-care circulating cathodic antigen (CCA) is an alternative field diagnostic that is more sensitive than Kato-Katz when intensity is low, but interpretation of CCA-trace results is unclear. To evaluate trace results, we tested urine and stool specimens from 398 pupils from eight schools in Burundi using four approaches: two in Burundi and two in a laboratory in Leiden, the Netherlands. In Burundi, we used Kato-Katz and point-of-care CCA (CCAB). In Leiden, we repeated the CCA (CCAL) and also used Up-Converting Phosphor Circulating Anodic Antigen (CAA). We applied Bayesian latent class analyses (LCA), first considering CCA traces as negative and then as positive. We used the LCA output to estimate validity of the prevalence estimates of each test in comparison to the population-level infection prevalence and estimated the proportion of trace results that were likely true positives. Kato-Katz yielded the lowest prevalence (6.8%), and CCAB with trace considered positive yielded the highest (53.5%). There were many more trace results recorded by CCA in Burundi (32.4%) than in Leiden (2.3%). Estimated prevalence with CAA was 46.5%. LCA indicated that Kato-Katz had the lowest sensitivity: 15.9% [Bayesian Credible Interval (BCI): 9.2-23.5%] with CCA-trace considered negative and 15.0% with trace as positive (BCI: 9.6-21.4%), implying that Kato-Katz missed approximately 85% of infections. CCAB underestimated disease prevalence when trace was considered negative and overestimated disease prevalence when trace was considered positive, by approximately 12 percentage points each way, and CAA overestimated prevalence in both models. Our results suggest that approximately 52.2% (BCI: 37.8-5.8%) of the CCAB trace readings were true infections. Whether measured in the laboratory or the field, CCA outperformed Kato-Katz at the low infection intensities in Burundi. CCA with trace as negative likely missed many infections, whereas CCA with trace as positive overestimated prevalence. In the absence of a field-friendly gold standard diagnostic, the use of a variety of diagnostics with differing properties will become increasingly important as programs move towards elimination of schistosomiasis. It is clear that CCA is a valuable tool for the detection and mapping of S. mansoni infection in the field and CAA may be a valuable field tool in the future.

Symptom Trajectories in Children Receiving Treatment for Leukemia: A Latent Class Growth Analysis With Multitrajectory Modeling.

PubMed

Hockenberry, Marilyn J; Hooke, Mary C; Rodgers, Cheryl; Taylor, Olga; Koerner, Kari M; Mitby, Pauline; Moore, Ida; Scheurer, Michael E; Pan, Wei

2017-07-01

Cancer treatment symptoms play a major role in determining the health of children with cancer. Symptom toxicity often results in complications, treatment delays, and therapy dose reductions that can compromise leukemia therapy and jeopardize chances for long-term survival. Critical to understanding symptom experiences during treatment is the need for exploration of "why" inter-individual symptom differences occur; this will determine who may be most susceptible to treatment toxicities. This study examined specific symptom trajectories during the first 18 months of childhood leukemia treatment. Symptom measures included fatigue, sleep disturbances, pain, nausea, and depression. Symptom trajectories of 236 children with leukemia three to 18 years old were explored prospectively over four periods: initiation of post-induction therapy, four and eight post-induction therapy, and the last time point was at the beginning of maintenance/continuation therapy. Latent class growth analysis was used to classify patients into distinctive groups with similar symptom trajectories based on patients' response patterns on the symptom measures over time. Three latent classes of symptom trajectories were identified and classified into mild, moderate, and severe symptom trajectories. The only demographic characteristic with a significant relationship to membership in the latent class symptom trajectories was race/ethnicity. All other demographic characteristics including leukemia risk levels showed no significant relationships. This study is unique in that groups of patients with similar symptoms were identified rather than groups of symptoms. Further research using latent class growth analysis is needed. Copyright © 2017 American Academy of Hospice and Palliative Medicine. Published by Elsevier Inc. All rights reserved.

An examination of generalized anxiety disorder and dysthymic disorder by latent class analysis.

PubMed

Rhebergen, D; van der Steenstraten, I M; Sunderland, M; de Graaf, R; Ten Have, M; Lamers, F; Penninx, B W J H; Andrews, G

2014-06-01

The nosological status of generalized anxiety disorder (GAD) versus dysthymic disorder (DD) has been questioned. The aim of this study was to examine qualitative differences within (co-morbid) GAD and DD symptomatology. Latent class analysis was applied to anxious and depressive symptomatology of respondents from three population-based studies (2007 Australian National Survey of Mental Health and Wellbeing; National Comorbidity Survey Replication; and Netherlands Mental Health Survey and Incidence Study-2; together known as the Triple study) and respondents from a multi-site naturalistic cohort [Netherlands Study of Depression and Anxiety (NESDA)]. Sociodemographics and clinical characteristics of each class were examined. A three-class (Triple study) and two-class (NESDA) model best fitted the data, reflecting mainly different levels of severity of symptoms. In the Triple study, no division into a predominantly GAD or DD co-morbidity subtype emerged. Likewise, in spite of the presence of pure GAD and DD cases in the NESDA sample, latent class analysis did not identify specific anxiety or depressive profiles in the NESDA study. Next, sociodemographics and clinical characteristics of each class were examined. Classes only differed in levels of severity. The absence of qualitative differences in anxious or depressive symptomatology in empirically derived classes questions the differentiation between GAD and DD.

A Latent Class Analysis of Online Sexual Experiences and Offline Sexual Behaviors Among Female Adolescents.

PubMed

Maas, Megan K; Bray, Bethany C; Noll, Jennie G

2017-11-20

This study used latent class analysis to identify patterns (i.e., classes) across a broad range of online sexual experiences among female adolescents (n = 312) and to explore offline sexual behavior and substance use correlates of as well as maltreatment differences in class membership. The following four classes were identified: Online Abstinent, Online Inclusive, Attractors, and Seekers. Maltreated female adolescents were more likely to be members of the Online Inclusive class and less likely to be members of the Online Abstinent class than nonmaltreated female adolescents. Offline sexual behaviors and substance use differentially predicted class membership. These results suggest online sexual experiences vary greatly and should not be aggregated together as a global risk factor for all female adolescents. © 2017 Society for Research on Adolescence.

Latent Class Analysis of Antisocial Behavior: Interaction of Serotonin Transporter Genotype and Maltreatment

ERIC Educational Resources Information Center

Li, James J.; Lee, Steve S.

2010-01-01

To improve understanding about genetic and environmental influences on antisocial behavior (ASB), we tested the association of the 44-base pair polymorphism of the serotonin transporter gene (5-HTTLPR) and maltreatment using latent class analysis in 2,488 boys and girls from Wave 1 of the National Longitudinal Study of Adolescent Health. In boys,…

Investigating Subtypes of Child Development: A Comparison of Cluster Analysis and Latent Class Cluster Analysis in Typology Creation

ERIC Educational Resources Information Center

DiStefano, Christine; Kamphaus, R. W.

2006-01-01

Two classification methods, latent class cluster analysis and cluster analysis, are used to identify groups of child behavioral adjustment underlying a sample of elementary school children aged 6 to 11 years. Behavioral rating information across 14 subscales was obtained from classroom teachers and used as input for analyses. Both the procedures…

Multilevel Latent Class Analysis for Large-Scale Educational Assessment Data: Exploring the Relation between the Curriculum and Students' Mathematical Strategies

ERIC Educational Resources Information Center

Fagginger Auer, Marije F.; Hickendorff, Marian; Van Putten, Cornelis M.; Béguin, Anton A.; Heiser, Willem J.

2016-01-01

A first application of multilevel latent class analysis (MLCA) to educational large-scale assessment data is demonstrated. This statistical technique addresses several of the challenges that assessment data offers. Importantly, MLCA allows modeling of the often ignored teacher effects and of the joint influence of teacher and student variables.…

Clustering Educational Digital Library Usage Data: A Comparison of Latent Class Analysis and K-Means Algorithms

ERIC Educational Resources Information Center

Xu, Beijie; Recker, Mimi; Qi, Xiaojun; Flann, Nicholas; Ye, Lei

2013-01-01

This article examines clustering as an educational data mining method. In particular, two clustering algorithms, the widely used K-means and the model-based Latent Class Analysis, are compared, using usage data from an educational digital library service, the Instructional Architect (IA.usu.edu). Using a multi-faceted approach and multiple data…

Identifying Students at Risk: An Examination of Computer-Adaptive Measures and Latent Class Growth Analysis

ERIC Educational Resources Information Center

Keller-Margulis, Milena; McQuillin, Samuel D.; Castañeda, Juan Javier; Ochs, Sarah; Jones, John H.

2018-01-01

Multitiered systems of support depend on screening technology to identify students at risk. The purpose of this study was to examine the use of a computer-adaptive test and latent class growth analysis (LCGA) to identify students at risk in reading with focus on the use of this methodology to characterize student performance in screening.…

Examining the Stability of "DSM-IV" and Empirically Derived Eating Disorder Classification: Implications for "DSM-5"

ERIC Educational Resources Information Center

Peterson, Carol B.; Crow, Scott J.; Swanson, Sonja A.; Crosby, Ross D.; Wonderlich, Stephen A.; Mitchell, James E.; Agras, W. Stewart; Halmi, Katherine A.

2011-01-01

Objective: The purpose of this investigation was to derive an empirical classification of eating disorder symptoms in a heterogeneous eating disorder sample using latent class analysis (LCA) and to examine the longitudinal stability of these latent classes (LCs) and the stability of DSM-IV eating disorder (ED) diagnoses. Method: A total of 429…

Developmental Typologies of Identity Formation and Adjustment in Female Emerging Adults: A Latent Class Growth Analysis Approach

ERIC Educational Resources Information Center

Luyckx, Koen; Schwartz, Seth J.; Goossens, Luc; Soenens, Bart; Beyers, Wim

2008-01-01

The developmental interplay between identity and adjustment was examined in a seven-wave longitudinal study of 428 European female college students (M[subscript age] = 18.8 years) over a period of 3 years, with semi-annual measurement waves each year. Latent Class Growth Analysis (LCGA) was used to identify developmental typologies of both…

Latent Classes of Adolescent Posttraumatic Stress Disorder Predict Functioning and Disorder after 1 Year

ERIC Educational Resources Information Center

Ayer, Lynsay; Danielson, Carla Kmett; Amstadter, Ananda B.; Ruggiero, Ken; Saunders, Ben; Kilpatrick, Dean

2011-01-01

Objective: To identify latent classes of posttraumatic stress disorder (PTSD) symptoms in a national sample of adolescents, and to test their associations with PTSD and functional impairment 1 year later. Method: A total of 1,119 trauma-exposed youth aged 12 through 17 years (mean = 14.99 years, 51% female and 49% male) participating in the…

The Co-Occurrence of Substance Use and Bullying Behaviors among U.S. Adolescents: Understanding Demographic Characteristics and Social Influences

ERIC Educational Resources Information Center

Luk, Jeremy W.; Wang, Jing; Simons-Morton, Bruce G.

2012-01-01

This study examined the co-occurrence of subtypes of substance use and bullying behaviors using latent class analysis and evaluated latent class differences in demographic characteristics, peer and parental influences. Self-reported questionnaire data were collected from a nationally representative sample (N = 7508) of 6-10th grade adolescents in…

Longitudinal burnout-collaboration patterns in Japanese medical care workers at special needs schools: a latent class growth analysis

PubMed Central

Kanayama, Mieko; Suzuki, Machiko; Yuma, Yoshikazu

2016-01-01

The present study aimed to identify and characterize potential burnout types and the relationship between burnout and collaboration over time. Latent class growth analysis and the growth mixture model were used to identify and characterize heterogeneous patterns of longitudinal stability and change in burnout, and the relationship between burnout and collaboration. We collected longitudinal data at three time points based on Japanese academic terms. The 396 study participants included academic teachers, yogo teachers, and registered nurses in Japanese special needs schools. The best model included four types of both burnout and collaboration in latent class growth analysis with intercept, slope, and quadratic terms. The four types of burnout were as follows: low stable, moderate unstable, high unstable, and high decreasing. They were identified as involving inverse collaboration function. The results indicated that there could be dynamic burnout types, namely moderate unstable, high unstable, and high decreasing, when focusing on growth trajectories in latent class analyses. The finding that collaboration was dynamic for dynamic burnout types and stable for stable burnout types is of great interest. This was probably related to the inverse relationship between the two constructs. PMID:27366107

[Latent Class Analysis of Gambling Activities among Korean Adolescents].

PubMed

Kang, Kyonghwa; Kim, Hyeongsu; Park, Ae Ran; Kim, Hee Young; Lee, Kunsei

2018-04-01

The aim of this study is to identify the types of gambling among adolescents and provide basic prevention information regarding adolescents' gambling problems. Secondary data from representative national survey on 2015 Youth Gambling Problems of Korea Center on Gambling Problems were used. Using latent class analysis (LCA), 13 gambling types such as offline and online games of 14,011 adolescents were classified, and gambling experiences and characteristics were analyzed. The subgroups of adolescent gambling were identified as four latent classes: a rare group (84.5% of the sample), a risk group (1.0%), an offline group (11.9%), and an expanded group (2.6%). The types and characteristics of gambling among the latent classes differed. In the risk group, adolescents participated in online illegal sports betting and internet casino, and gambling time, gambling expenses, and the number of gambling types were higher than other groups. Gambling frequently occur among adolescent, and the subtypes of gambling did not reveal homogeneous characteristics. In order to prevent adolescent gambling problems, it is a necessary to develop tailored prevention intervention in the nursing field, which is appropriate to the characteristics of adolescent gambling group and can help with early identification. © 2018 Korean Society of Nursing Science.

Collinear Latent Variables in Multilevel Confirmatory Factor Analysis: A Comparison of Maximum Likelihood and Bayesian Estimations

ERIC Educational Resources Information Center

Can, Seda; van de Schoot, Rens; Hox, Joop

2015-01-01

Because variables may be correlated in the social and behavioral sciences, multicollinearity might be problematic. This study investigates the effect of collinearity manipulated in within and between levels of a two-level confirmatory factor analysis by Monte Carlo simulation. Furthermore, the influence of the size of the intraclass correlation…

A Bayesian Hierarchical Selection Model for Academic Growth with Missing Data

ERIC Educational Resources Information Center

Allen, Jeff

2017-01-01

Using a sample of schools testing annually in grades 9-11 with a vertically linked series of assessments, a latent growth curve model is used to model test scores with student intercepts and slopes nested within school. Missed assessments can occur because of student mobility, student dropout, absenteeism, and other reasons. Missing data…

Recovery of Graded Response Model Parameters: A Comparison of Marginal Maximum Likelihood and Markov Chain Monte Carlo Estimation

ERIC Educational Resources Information Center

Kieftenbeld, Vincent; Natesan, Prathiba

2012-01-01

Markov chain Monte Carlo (MCMC) methods enable a fully Bayesian approach to parameter estimation of item response models. In this simulation study, the authors compared the recovery of graded response model parameters using marginal maximum likelihood (MML) and Gibbs sampling (MCMC) under various latent trait distributions, test lengths, and…

Latent class analysis of early developmental trajectory in baby siblings of children with autism.

PubMed

Landa, Rebecca J; Gross, Alden L; Stuart, Elizabeth A; Bauman, Margaret

2012-09-01

Siblings of children with autism (sibs-A) are at increased genetic risk for autism spectrum disorders (ASD) and milder impairments. To elucidate diversity and contour of early developmental trajectories exhibited by sibs-A, regardless of diagnostic classification, latent class modeling was used. Sibs-A (N = 204) were assessed with the Mullen Scales of Early Learning from age 6 to 36 months. Mullen T scores served as dependent variables. Outcome classifications at age 36 months included: ASD (N = 52); non-ASD social/communication delay (broader autism phenotype; BAP; N = 31); and unaffected (N = 121). Child-specific patterns of performance were studied using latent class growth analysis. Latent class membership was then related to diagnostic outcome through estimation of within-class proportions of children assigned to each diagnostic classification. A 4-class model was favored. Class 1 represented accelerated development and consisted of 25.7% of the sample, primarily unaffected children. Class 2 (40.0% of the sample), was characterized by normative development with above-average nonverbal cognitive outcome. Class 3 (22.3% of the sample) was characterized by receptive language, and gross and fine motor delay. Class 4 (12.0% of the sample), was characterized by widespread delayed skill acquisition, reflected by declining trajectories. Children with an outcome diagnosis of ASD were spread across Classes 2, 3, and 4. Results support a category of ASD that involves slowing in early non-social development. Receptive language and motor development is vulnerable to early delay in sibs-A with and without ASD outcomes. Non-ASD sibs-A are largely distributed across classes depicting average or accelerated development. Developmental trajectories of motor, language, and cognition appear independent of communication and social delays in non-ASD sibs-A. © 2012 The Authors. Journal of Child Psychology and Psychiatry © 2012 Association for Child and Adolescent Mental Health.

Optimization of Modeled Land-Atmosphere Exchanges of Water and Energy in an Isotopically-Enabled Land Surface Model by Bayesian Parameter Calibration

NASA Astrophysics Data System (ADS)

Wong, T. E.; Noone, D. C.; Kleiber, W.

2014-12-01

The single largest uncertainty in climate model energy balance is the surface latent heating over tropical land. Furthermore, the partitioning of the total latent heat flux into contributions from surface evaporation and plant transpiration is of great importance, but notoriously poorly constrained. Resolving these issues will require better exploiting information which lies at the interface between observations and advanced modeling tools, both of which are imperfect. There are remarkably few observations which can constrain these fluxes, placing strict requirements on developing statistical methods to maximize the use of limited information to best improve models. Previous work has demonstrated the power of incorporating stable water isotopes into land surface models for further constraining ecosystem processes. We present results from a stable water isotopically-enabled land surface model (iCLM4), including model experiments partitioning the latent heat flux into contributions from plant transpiration and surface evaporation. It is shown that the partitioning results are sensitive to the parameterization of kinetic fractionation used. We discuss and demonstrate an approach to calibrating select model parameters to observational data in a Bayesian estimation framework, requiring Markov Chain Monte Carlo sampling of the posterior distribution, which is shown to constrain uncertain parameters as well as inform relevant values for operational use. Finally, we discuss the application of the estimation scheme to iCLM4, including entropy as a measure of information content and specific challenges which arise in calibration models with a large number of parameters.

Multilevel Higher-Order Item Response Theory Models

ERIC Educational Resources Information Center

Huang, Hung-Yu; Wang, Wen-Chung

2014-01-01

In the social sciences, latent traits often have a hierarchical structure, and data can be sampled from multiple levels. Both hierarchical latent traits and multilevel data can occur simultaneously. In this study, we developed a general class of item response theory models to accommodate both hierarchical latent traits and multilevel data. The…

Common Mental Disorders among Occupational Groups: Contributions of the Latent Class Model

PubMed Central

Martins Carvalho, Fernando; de Araújo, Tânia Maria

2016-01-01

Background. The Self-Reporting Questionnaire (SRQ-20) is widely used for evaluating common mental disorders. However, few studies have evaluated the SRQ-20 measurements performance in occupational groups. This study aimed to describe manifestation patterns of common mental disorders symptoms among workers populations, by using latent class analysis. Methods. Data derived from 9,959 Brazilian workers, obtained from four cross-sectional studies that used similar methodology, among groups of informal workers, teachers, healthcare workers, and urban workers. Common mental disorders were measured by using SRQ-20. Latent class analysis was performed on each database separately. Results. Three classes of symptoms were confirmed in the occupational categories investigated. In all studies, class I met better criteria for suspicion of common mental disorders. Class II discriminated workers with intermediate probability of answers to the items belonging to anxiety, sadness, and energy decrease that configure common mental disorders. Class III was composed of subgroups of workers with low probability to respond positively to questions for screening common mental disorders. Conclusions. Three patterns of symptoms of common mental disorders were identified in the occupational groups investigated, ranging from distinctive features to low probabilities of occurrence. The SRQ-20 measurements showed stability in capturing nonpsychotic symptoms. PMID:27630999

Bayesian estimation of differential transcript usage from RNA-seq data.

PubMed

Papastamoulis, Panagiotis; Rattray, Magnus

2017-11-27

Next generation sequencing allows the identification of genes consisting of differentially expressed transcripts, a term which usually refers to changes in the overall expression level. A specific type of differential expression is differential transcript usage (DTU) and targets changes in the relative within gene expression of a transcript. The contribution of this paper is to: (a) extend the use of cjBitSeq to the DTU context, a previously introduced Bayesian model which is originally designed for identifying changes in overall expression levels and (b) propose a Bayesian version of DRIMSeq, a frequentist model for inferring DTU. cjBitSeq is a read based model and performs fully Bayesian inference by MCMC sampling on the space of latent state of each transcript per gene. BayesDRIMSeq is a count based model and estimates the Bayes Factor of a DTU model against a null model using Laplace's approximation. The proposed models are benchmarked against the existing ones using a recent independent simulation study as well as a real RNA-seq dataset. Our results suggest that the Bayesian methods exhibit similar performance with DRIMSeq in terms of precision/recall but offer better calibration of False Discovery Rate.

Estimation of diagnostic test accuracy without full verification: a review of latent class methods

PubMed Central

Collins, John; Huynh, Minh

2014-01-01

The performance of a diagnostic test is best evaluated against a reference test that is without error. For many diseases, this is not possible, and an imperfect reference test must be used. However, diagnostic accuracy estimates may be biased if inaccurately verified status is used as the truth. Statistical models have been developed to handle this situation by treating disease as a latent variable. In this paper, we conduct a systematized review of statistical methods using latent class models for estimating test accuracy and disease prevalence in the absence of complete verification. PMID:24910172

Large-scale weakly supervised object localization via latent category learning.

PubMed

Chong Wang; Kaiqi Huang; Weiqiang Ren; Junge Zhang; Maybank, Steve

2015-04-01

Localizing objects in cluttered backgrounds is challenging under large-scale weakly supervised conditions. Due to the cluttered image condition, objects usually have large ambiguity with backgrounds. Besides, there is also a lack of effective algorithm for large-scale weakly supervised localization in cluttered backgrounds. However, backgrounds contain useful latent information, e.g., the sky in the aeroplane class. If this latent information can be learned, object-background ambiguity can be largely reduced and background can be suppressed effectively. In this paper, we propose the latent category learning (LCL) in large-scale cluttered conditions. LCL is an unsupervised learning method which requires only image-level class labels. First, we use the latent semantic analysis with semantic object representation to learn the latent categories, which represent objects, object parts or backgrounds. Second, to determine which category contains the target object, we propose a category selection strategy by evaluating each category's discrimination. Finally, we propose the online LCL for use in large-scale conditions. Evaluation on the challenging PASCAL Visual Object Class (VOC) 2007 and the large-scale imagenet large-scale visual recognition challenge 2013 detection data sets shows that the method can improve the annotation precision by 10% over previous methods. More importantly, we achieve the detection precision which outperforms previous results by a large margin and can be competitive to the supervised deformable part model 5.0 baseline on both data sets.

Sequential Monte Carlo for inference of latent ARMA time-series with innovations correlated in time

NASA Astrophysics Data System (ADS)

Urteaga, Iñigo; Bugallo, Mónica F.; Djurić, Petar M.

2017-12-01

We consider the problem of sequential inference of latent time-series with innovations correlated in time and observed via nonlinear functions. We accommodate time-varying phenomena with diverse properties by means of a flexible mathematical representation of the data. We characterize statistically such time-series by a Bayesian analysis of their densities. The density that describes the transition of the state from time t to the next time instant t+1 is used for implementation of novel sequential Monte Carlo (SMC) methods. We present a set of SMC methods for inference of latent ARMA time-series with innovations correlated in time for different assumptions in knowledge of parameters. The methods operate in a unified and consistent manner for data with diverse memory properties. We show the validity of the proposed approach by comprehensive simulations of the challenging stochastic volatility model.

On the explaining-away phenomenon in multivariate latent variable models.

PubMed

van Rijn, Peter; Rijmen, Frank

2015-02-01

Many probabilistic models for psychological and educational measurements contain latent variables. Well-known examples are factor analysis, item response theory, and latent class model families. We discuss what is referred to as the 'explaining-away' phenomenon in the context of such latent variable models. This phenomenon can occur when multiple latent variables are related to the same observed variable, and can elicit seemingly counterintuitive conditional dependencies between latent variables given observed variables. We illustrate the implications of explaining away for a number of well-known latent variable models by using both theoretical and real data examples. © 2014 The British Psychological Society.

A Bayesian hierarchical model with novel prior specifications for estimating HIV testing rates.

PubMed

An, Qian; Kang, Jian; Song, Ruiguang; Hall, H Irene

2016-04-30

Human immunodeficiency virus (HIV) infection is a severe infectious disease actively spreading globally, and acquired immunodeficiency syndrome (AIDS) is an advanced stage of HIV infection. The HIV testing rate, that is, the probability that an AIDS-free HIV infected person seeks a test for HIV during a particular time interval, given no previous positive test has been obtained prior to the start of the time, is an important parameter for public health. In this paper, we propose a Bayesian hierarchical model with two levels of hierarchy to estimate the HIV testing rate using annual AIDS and AIDS-free HIV diagnoses data. At level one, we model the latent number of HIV infections for each year using a Poisson distribution with the intensity parameter representing the HIV incidence rate. At level two, the annual numbers of AIDS and AIDS-free HIV diagnosed cases and all undiagnosed cases stratified by the HIV infections at different years are modeled using a multinomial distribution with parameters including the HIV testing rate. We propose a new class of priors for the HIV incidence rate and HIV testing rate taking into account the temporal dependence of these parameters to improve the estimation accuracy. We develop an efficient posterior computation algorithm based on the adaptive rejection metropolis sampling technique. We demonstrate our model using simulation studies and the analysis of the national HIV surveillance data in the USA. Copyright © 2015 John Wiley & Sons, Ltd.

Classes of conduct disorder symptoms and their life course correlates in a US national sample.

PubMed

Breslau, J; Saito, N; Tancredi, D J; Nock, M; Gilman, S E

2012-05-01

Population data on conduct disorder (CD) symptoms can help determine whether hypothesized subtypes of CD are sufficiently disparate in their familial, psychiatric and life course correlates to distinguish separate diagnostic entities. Latent class analysis (LCA) of CD symptoms occurring before age 15 was conducted in a national sample of adults aged 18-44 years from the National Epidemiological Study of Alcohol and Related Conditions. Associations of latent class membership with parental behavior problems, onset of psychiatric disorders and anti-social behaviors after age 15, adolescent life events (e.g. high school drop-out), and past-year life events (e.g. divorce/separation, bankruptcy) were estimated. LCA identified a no-CD class with low prevalence of all symptoms, three intermediate classes - deceit/theft, rule violations, aggression - and a severe class. The prevalence of CD, according to DSM-IV criteria, was 0% in the no-CD class, between 13.33% and 33.69% in the intermediate classes and 62.20% in the severe class. Latent class membership is associated with all the familial, psychiatric and life course outcomes examined. Among the intermediate classes, risk for subsequent mood/anxiety disorders and anti-social behavior was higher in the deceit/theft and aggressive classes than in the rule violations class. However, risk for adolescent life events is highest in the rule violations class. CD symptoms tend to occur in a partially ordered set of classes in the general population. Prognostically meaningful distinctions can be drawn between classes, but only at low levels of symptoms.

Stress, Mental Health, and Substance Abuse Problems in a Sample of Diversion Program Youths: An Exploratory Latent Class Analysis

ERIC Educational Resources Information Center

Dembo, Richard; Briones, Rhissa; Gulledge, Laura; Karas, Lora; Winters, Ken C.; Belenko, Steven; Greenbaum, Paul E.

2012-01-01

Reflective of interest in mental health and substance abuse issues among youths involved with the justice system, we performed a latent class analysis on baseline information collected on 100 youths involved in two diversion programs. Results identified two groups of youths: Group 1: a majority of the youths, who had high levels of delinquency,…

Prevention of Substance Use among Adolescents through Social and Emotional Training in School: A Latent-Class Analysis of a Five-Year Intervention in Sweden

ERIC Educational Resources Information Center

Kimber, Birgitta; Sandell, Rolf

2009-01-01

The study considers the impact of a program for social and emotional learning in Swedish schools on use of drugs, volatile substances, alcohol and tobacco. The program was evaluated in an effectiveness study. Intervention students were compared longitudinally with non-intervention students using nonparametric latent class analysis to identify…

A Latent Class Growth Analysis of School Bullying and Its Social Context: The Self-Determination Theory Perspective

ERIC Educational Resources Information Center

Lam, Shui-fong; Law, Wilbert; Chan, Chi-Keung; Wong, Bernard P. H.; Zhang, Xiao

2015-01-01

The contribution of social context to school bullying was examined from the self-determination theory perspective in this longitudinal study of 536 adolescents from 3 secondary schools in Hong Kong. Latent class growth analysis of the student-reported data at 5 time points from grade 7 to grade 9 identified 4 groups of students: bullies (9.8%),…

Tri-city study of Ecstasy use problems: a latent class analysis.

PubMed

Scheier, Lawrence M; Ben Abdallah, Arbi; Inciardi, James A; Copeland, Jan; Cottler, Linda B

2008-12-01

This study used latent class analysis to examine distinctive subtypes of Ecstasy users based on 24 abuse and dependence symptoms underlying standard DSM-IV criteria. Data came from a three site, population-based, epidemiological study to examine diagnostic nosology for Ecstasy use. Subject inclusion criteria included lifetime Ecstasy use exceeding five times and once in the past year, with participants ranging in age between 16 and 47 years of age from St. Louis, Miami, U.S. and Sydney, Australia. A satisfactory model typified four latent classes representing clearly differentiated diagnostic clusters including: (1) a group of sub-threshold users endorsing few abuse and dependence symptoms (negatives), (2) a group of 'diagnostic orphans' who had characteristic features of dependence for a select group of symptoms (mild dependent), (3) a 'transitional group' mimicking the orphans with regard to their profile of dependence also but reporting some abuse symptoms (moderate dependent), and (4) a 'severe dependent' group with a distinct profile of abuse and dependence symptoms. A multinomial logistic regression model indicated that certain latent classes showed unique associations with external non-diagnostic markers. Controlling for demographic characteristics and lifetime quantity of Ecstasy pill use, criminal behavior and motivational cues for Ecstasy use were the most efficient predictors of cluster membership. This study reinforces the heuristic utility of DSM-IV criteria applied to Ecstasy but with a different collage of symptoms that produced four distinct classes of Ecstasy users.

Nonlinear Structured Growth Mixture Models in M"plus" and OpenMx

ERIC Educational Resources Information Center

Grimm, Kevin J.; Ram, Nilam; Estabrook, Ryne

2010-01-01

Growth mixture models (GMMs; B. O. Muthen & Muthen, 2000; B. O. Muthen & Shedden, 1999) are a combination of latent curve models (LCMs) and finite mixture models to examine the existence of latent classes that follow distinct developmental patterns. GMMs are often fit with linear, latent basis, multiphase, or polynomial change models…

Higher-Order Item Response Models for Hierarchical Latent Traits

ERIC Educational Resources Information Center

Huang, Hung-Yu; Wang, Wen-Chung; Chen, Po-Hsi; Su, Chi-Ming

2013-01-01

Many latent traits in the human sciences have a hierarchical structure. This study aimed to develop a new class of higher order item response theory models for hierarchical latent traits that are flexible in accommodating both dichotomous and polytomous items, to estimate both item and person parameters jointly, to allow users to specify…

Locally Dependent Latent Trait Model and the Dutch Identity Revisited.

ERIC Educational Resources Information Center

Ip, Edward H.

2002-01-01

Proposes a class of locally dependent latent trait models for responses to psychological and educational tests. Focuses on models based on a family of conditional distributions, or kernel, that describes joint multiple item responses as a function of student latent trait, not assuming conditional independence. Also proposes an EM algorithm for…

Multi-angle backscatter classification and sub-bottom profiling for improved seafloor characterization

NASA Astrophysics Data System (ADS)

Alevizos, Evangelos; Snellen, Mirjam; Simons, Dick; Siemes, Kerstin; Greinert, Jens

2018-06-01

This study applies three classification methods exploiting the angular dependence of acoustic seafloor backscatter along with high resolution sub-bottom profiling for seafloor sediment characterization in the Eckernförde Bay, Baltic Sea Germany. This area is well suited for acoustic backscatter studies due to its shallowness, its smooth bathymetry and the presence of a wide range of sediment types. Backscatter data were acquired using a Seabeam1180 (180 kHz) multibeam echosounder and sub-bottom profiler data were recorded using a SES-2000 parametric sonar transmitting 6 and 12 kHz. The high density of seafloor soundings allowed extracting backscatter layers for five beam angles over a large part of the surveyed area. A Bayesian probability method was employed for sediment classification based on the backscatter variability at a single incidence angle, whereas Maximum Likelihood Classification (MLC) and Principal Components Analysis (PCA) were applied to the multi-angle layers. The Bayesian approach was used for identifying the optimum number of acoustic classes because cluster validation is carried out prior to class assignment and class outputs are ordinal categorical values. The method is based on the principle that backscatter values from a single incidence angle express a normal distribution for a particular sediment type. The resulting Bayesian classes were well correlated to median grain sizes and the percentage of coarse material. The MLC method uses angular response information from five layers of training areas extracted from the Bayesian classification map. The subsequent PCA analysis is based on the transformation of these five layers into two principal components that comprise most of the data variability. These principal components were clustered in five classes after running an external cluster validation test. In general both methods MLC and PCA, separated the various sediment types effectively, showing good agreement (kappa >0.7) with the Bayesian approach which also correlates well with ground truth data (r2 > 0.7). In addition, sub-bottom data were used in conjunction with the Bayesian classification results to characterize acoustic classes with respect to their geological and stratigraphic interpretation. The joined interpretation of seafloor and sub-seafloor data sets proved to be an efficient approach for a better understanding of seafloor backscatter patchiness and to discriminate acoustically similar classes in different geological/bathymetric settings.

Associations among personal care product use patterns and exogenous hormone use in the NIEHS Sister Study.

PubMed

Taylor, Kyla W; Baird, Donna D; Herring, Amy H; Engel, Lawrence S; Nichols, Hazel B; Sandler, Dale P; Troester, Melissa A

2017-09-01

It is hypothesized that certain chemicals in personal care products may alter the risk of adverse health outcomes. The primary aim of this study was to use a data-centered approach to classify complex patterns of exposure to personal care products and to understand how these patterns vary according to use of exogenous hormone exposures, oral contraceptives (OCs) and post-menopausal hormone therapy (HT). The NIEHS Sister Study is a prospective cohort study of 50,884 US women. Limiting the sample to non-Hispanic blacks and whites (N=47,019), latent class analysis (LCA) was used to identify groups of individuals with similar patterns of personal care product use based on responses to 48 survey questions. Personal care products were categorized into three product types (beauty, hair, and skincare products) and separate latent classes were constructed for each type. Adjusted prevalence differences (PD) were calculated to estimate the association between exogenous hormone use, as measured by ever/never OC or HT use, and patterns of personal care product use. LCA reduced data dimensionality by grouping of individuals with similar patterns of personal care product use into mutually exclusive latent classes (three latent classes for beauty product use, three for hair, and four for skin care. There were strong differences in personal care usage by race, particularly for haircare products. For both blacks and whites, exogenous hormone exposures were associated with higher levels of product use, especially beauty and skincare products. Relative to individual product use questions, latent class variables capture complex patterns of personal care product usage. These patterns differed by race and were associated with ever OC and HT use. Future studies should consider personal care product exposures with other exogenous exposures when modeling health risks.

Associations among personal care product use patterns and exogenous hormone use in the NIEHS Sister Study

PubMed Central

Taylor, Kyla W.; Baird, Donna D.; Herring, Amy H.; Engel, Lawrence S.; Nichols, Hazel B.; Sandler, Dale P.; Troester, Melissa A.

2017-01-01

It is hypothesized that certain chemicals in personal care products may alter the risk of adverse health outcomes. The primary aim of this study was to use a data-centered approach to classify complex patterns of exposure to personal care products and to understand how these patterns vary according to use of exogenous hormone exposures, oral contraceptives (OCs) and post-menopausal hormone therapy (HT). The NIEHS Sister Study is a prospective cohort study of 50,884 US women. Limiting the sample to non-Hispanic blacks and whites (N = 47,019), latent class analysis (LCA) was used to identify groups of individuals with similar patterns of personal care product use based on responses to 48 survey questions. Personal care products were categorized into three product types (beauty, hair, and skincare products) and separate latent classes were constructed for each type. Adjusted prevalence differences (PD) were calculated to estimate the association between exogenous hormone use, as measured by ever/never OC or HT use, and patterns of personal care product use. LCA reduced data dimensionality by grouping of individuals with similar patterns of personal care product use into mutually exclusive latent classes (three latent classes for beauty product use, three for hair, and four for skin care. There were strong differences in personal care usage by race, particularly for haircare products. For both blacks and whites, exogenous hormone exposures were associated with higher levels of product use, especially beauty and skincare products. Relative to individual product use questions, latent class variables capture complex patterns of personal care product usage. These patterns differed by race and were associated with ever OC and HT use. Future studies should consider personal care product exposures with other exogenous exposures when modeling health risks. PMID:28120835

Substance Use, Violence, and Antiretroviral Adherence: A Latent Class Analysis of Women Living with HIV in Canada.

PubMed

Carter, Allison; Roth, Eric Abella; Ding, Erin; Milloy, M-J; Kestler, Mary; Jabbari, Shahab; Webster, Kath; de Pokomandy, Alexandra; Loutfy, Mona; Kaida, Angela

2018-03-01

We used latent class analysis to identify substance use patterns for 1363 women living with HIV in Canada and assessed associations with socio-economic marginalization, violence, and sub-optimal adherence to combination antiretroviral therapy (cART). A six-class model was identified consisting of: abstainers (26.3%), Tobacco Users (8.81%), Alcohol Users (31.9%), 'Socially Acceptable' Poly-substance Users (13.9%), Illicit Poly-substance Users (9.81%) and Illicit Poly-substance Users of All Types (9.27%). Multinomial logistic regression showed that women experiencing recent violence had significantly higher odds of membership in all substance use latent classes, relative to Abstainers, while those reporting sub-optimal cART adherence had higher odds of being members of the poly-substance use classes only. Factors significantly associated with Illicit Poly-substance Users of All Types were sexual minority status, lower income, and lower resiliency. Findings underline a need for increased social and structural supports for women who use substances to support them in leading safe and healthy lives with HIV.

Eye tracking and pupillometry are indicators of dissociable latent decision processes.

PubMed

Cavanagh, James F; Wiecki, Thomas V; Kochar, Angad; Frank, Michael J

2014-08-01

Can you predict what people are going to do just by watching them? This is certainly difficult: it would require a clear mapping between observable indicators and unobservable cognitive states. In this report, we demonstrate how this is possible by monitoring eye gaze and pupil dilation, which predict dissociable biases during decision making. We quantified decision making using the drift diffusion model (DDM), which provides an algorithmic account of how evidence accumulation and response caution contribute to decisions through separate latent parameters of drift rate and decision threshold, respectively. We used a hierarchical Bayesian estimation approach to assess the single trial influence of observable physiological signals on these latent DDM parameters. Increased eye gaze dwell time specifically predicted an increased drift rate toward the fixated option, irrespective of the value of the option. In contrast, greater pupil dilation specifically predicted an increase in decision threshold during difficult decisions. These findings suggest that eye tracking and pupillometry reflect the operations of dissociated latent decision processes. PsycINFO Database Record (c) 2014 APA, all rights reserved.

Insights into the latent multinomial model through mark-resight data on female grizzly bears with cubs-of-the-year

USGS Publications Warehouse

Higgs, Megan D.; Link, William; White, Gary C.; Haroldson, Mark A.; Bjornlie, Daniel D.

2013-01-01

Mark-resight designs for estimation of population abundance are common and attractive to researchers. However, inference from such designs is very limited when faced with sparse data, either from a low number of marked animals, a low probability of detection, or both. In the Greater Yellowstone Ecosystem, yearly mark-resight data are collected for female grizzly bears with cubs-of-the-year (FCOY), and inference suffers from both limitations. To overcome difficulties due to sparseness, we assume homogeneity in sighting probabilities over 16 years of bi-annual aerial surveys. We model counts of marked and unmarked animals as multinomial random variables, using the capture frequencies of marked animals for inference about the latent multinomial frequencies for unmarked animals. We discuss undesirable behavior of the commonly used discrete uniform prior distribution on the population size parameter and provide OpenBUGS code for fitting such models. The application provides valuable insights into subtleties of implementing Bayesian inference for latent multinomial models. We tie the discussion to our application, though the insights are broadly useful for applications of the latent multinomial model.

Predicting Viral Infection From High-Dimensional Biomarker Trajectories

PubMed Central

Chen, Minhua; Zaas, Aimee; Woods, Christopher; Ginsburg, Geoffrey S.; Lucas, Joseph; Dunson, David; Carin, Lawrence

2013-01-01

There is often interest in predicting an individual’s latent health status based on high-dimensional biomarkers that vary over time. Motivated by time-course gene expression array data that we have collected in two influenza challenge studies performed with healthy human volunteers, we develop a novel time-aligned Bayesian dynamic factor analysis methodology. The time course trajectories in the gene expressions are related to a relatively low-dimensional vector of latent factors, which vary dynamically starting at the latent initiation time of infection. Using a nonparametric cure rate model for the latent initiation times, we allow selection of the genes in the viral response pathway, variability among individuals in infection times, and a subset of individuals who are not infected. As we demonstrate using held-out data, this statistical framework allows accurate predictions of infected individuals in advance of the development of clinical symptoms, without labeled data and even when the number of biomarkers vastly exceeds the number of individuals under study. Biological interpretation of several of the inferred pathways (factors) is provided. PMID:23704802

Latent Class Analysis of Incomplete Data via an Entropy-Based Criterion

PubMed Central

Larose, Chantal; Harel, Ofer; Kordas, Katarzyna; Dey, Dipak K.

2016-01-01

Latent class analysis is used to group categorical data into classes via a probability model. Model selection criteria then judge how well the model fits the data. When addressing incomplete data, the current methodology restricts the imputation to a single, pre-specified number of classes. We seek to develop an entropy-based model selection criterion that does not restrict the imputation to one number of clusters. Simulations show the new criterion performing well against the current standards of AIC and BIC, while a family studies application demonstrates how the criterion provides more detailed and useful results than AIC and BIC. PMID:27695391

Variation in Latent Classes of Adult Attention-Deficit Hyperactivity Disorder by Sex and Environmental Adversity.

PubMed

Ebejer, Jane L; Medland, Sarah E; van der Werf, Julius; Lynskey, Michael; Martin, Nicholas G; Duffy, David L

2016-11-01

The findings of genetic, imaging and neuropsychological studies of attention-deficit hyperactivity disorder (ADHD) are mixed. To understand why this might be the case we use both dimensional and categorical symptom measurement to provide alternate and detailed perspectives of symptom expression. Interviewers collected ADHD, conduct problems (CP) and sociodemographic data from 3793 twins and their siblings aged 22 to 49 (M = 32.6). We estimate linear weighting of symptoms across ADHD and CP items. Latent class analyses and regression describe associations between measured variables, environmental risk factors and subsequent disadvantage. Additionally, the clinical relevance of each class was estimated. Five classes were found for women and men; few symptoms, hyperactive-impulsive, CP, inattentive, combined symptoms with CP. Women within the inattentive class reported more symptoms and reduced emotional health when compared to men and to women within other latent classes. Women and men with combined ADHD symptoms reported comorbid conduct problems but those with either inattention or hyperactivity-impulsivity only did not. The dual perspective of dimensional and categorical measurement of ADHD provides important detail about symptom variation across sex and with environmental covariates. © The Author(s) 2013.

Bayesian Community Detection in the Space of Group-Level Functional Differences

PubMed Central

Venkataraman, Archana; Yang, Daniel Y.-J.; Pelphrey, Kevin A.; Duncan, James S.

2017-01-01

We propose a unified Bayesian framework to detect both hyper- and hypo-active communities within whole-brain fMRI data. Specifically, our model identifies dense subgraphs that exhibit population-level differences in functional synchrony between a control and clinical group. We derive a variational EM algorithm to solve for the latent posterior distributions and parameter estimates, which subsequently inform us about the afflicted network topology. We demonstrate that our method provides valuable insights into the neural mechanisms underlying social dysfunction in autism, as verified by the Neurosynth meta-analytic database. In contrast, both univariate testing and community detection via recursive edge elimination fail to identify stable functional communities associated with the disorder. PMID:26955022

Bayesian Community Detection in the Space of Group-Level Functional Differences.

PubMed

Venkataraman, Archana; Yang, Daniel Y-J; Pelphrey, Kevin A; Duncan, James S

2016-08-01

We propose a unified Bayesian framework to detect both hyper- and hypo-active communities within whole-brain fMRI data. Specifically, our model identifies dense subgraphs that exhibit population-level differences in functional synchrony between a control and clinical group. We derive a variational EM algorithm to solve for the latent posterior distributions and parameter estimates, which subsequently inform us about the afflicted network topology. We demonstrate that our method provides valuable insights into the neural mechanisms underlying social dysfunction in autism, as verified by the Neurosynth meta-analytic database. In contrast, both univariate testing and community detection via recursive edge elimination fail to identify stable functional communities associated with the disorder.

Maternal eating disorder and infant diet. A latent class analysis based on the Norwegian Mother and Child Cohort Study (MoBa).

PubMed

Torgersen, Leila; Ystrom, Eivind; Siega-Riz, Anna Maria; Berg, Cecilie Knoph; Zerwas, Stephanie C; Reichborn-Kjennerud, Ted; Bulik, Cynthia M

2015-01-01

Knowledge of infant diet and feeding practices among children of mothers with eating disorders is essential to promote healthy eating in these children. This study compared the dietary patterns of 6-month-old children of mothers with anorexia nervosa, bulimia nervosa, binge eating disorder, and eating disorder not otherwise specified-purging subtype, to the diet of children of mothers with no eating disorders (reference group). The study was based on 53,879 mothers in the Norwegian Mother and Child Cohort Study (MoBa). Latent class analysis (LCA) was used to identify discrete latent classes of infant diet based on the mothers' responses to questions about 16 food items. LCA identified five classes, characterized by primarily homemade vegetarian food (4% of infants), homemade traditional food (8%), commercial cereals (35%), commercial jarred baby food (39%), and a mix of all food groups (11%). The association between latent dietary classes and maternal eating disorders were estimated by multinomial logistic regression. Infants of mothers with bulimia nervosa had a lower probability of being in the homemade traditional food class compared to the commercial jarred baby food class, than the referent (O.R. 0.59; 95% CI 0.36-0.99). Infants of mothers with binge eating disorder had a lower probability of being in the homemade vegetarian class compared to the commercial jarred baby food class (O.R. 0.77; 95% CI 0.60-0.99), but only before adjusting for relevant confounders. Anorexia nervosa and eating disorder not otherwise specified-purging subtype were not statistically significantly associated with any of the dietary classes. These results suggest that maternal eating disorders may to some extent influence the child's diet at 6 months; however, the extent to which these differences influence child health and development remains an area for further inquiry. Copyright © 2014 Elsevier Ltd. All rights reserved.

Spatial Bayesian latent factor regression modeling of coordinate-based meta-analysis data.

PubMed

Montagna, Silvia; Wager, Tor; Barrett, Lisa Feldman; Johnson, Timothy D; Nichols, Thomas E

2018-03-01

Now over 20 years old, functional MRI (fMRI) has a large and growing literature that is best synthesised with meta-analytic tools. As most authors do not share image data, only the peak activation coordinates (foci) reported in the article are available for Coordinate-Based Meta-Analysis (CBMA). Neuroimaging meta-analysis is used to (i) identify areas of consistent activation; and (ii) build a predictive model of task type or cognitive process for new studies (reverse inference). To simultaneously address these aims, we propose a Bayesian point process hierarchical model for CBMA. We model the foci from each study as a doubly stochastic Poisson process, where the study-specific log intensity function is characterized as a linear combination of a high-dimensional basis set. A sparse representation of the intensities is guaranteed through latent factor modeling of the basis coefficients. Within our framework, it is also possible to account for the effect of study-level covariates (meta-regression), significantly expanding the capabilities of the current neuroimaging meta-analysis methods available. We apply our methodology to synthetic data and neuroimaging meta-analysis datasets. © 2017, The International Biometric Society.

Gender roles and binge drinking among Latino emerging adults: a latent class regression analysis.

PubMed

Vaughan, Ellen L; Wong, Y Joel; Middendorf, Katharine G

2014-09-01

Gender roles are often cited as a culturally specific predictor of drinking among Latino populations. This study used latent class regression to test the relationships between gender roles and binge drinking in a sample of Latino emerging adults. Participants were Latino emerging adults who participated in Wave III of the National Longitudinal Study of Adolescent Health (N = 2,442). A subsample of these participants (n = 660) completed the Bem Sex Role Inventory--Short. We conducted latent class regression using 3 dimensions of gender roles (femininity, social masculinity, and personal masculinity) to predict binge drinking. Results indicated a 3-class solution. In Class 1, the protective personal masculinity class, personal masculinity (e.g., being a leader, defending one's own beliefs) was associated with a reduction in the odds of binge drinking. In Class 2, the nonsignificant class, gender roles were not related to binge drinking. In Class 3, the mixed masculinity class, personal masculinity was associated with a reduction in the odds of binge drinking, whereas social masculinity (e.g., forceful, dominant) was associated with an increase in the odds of binge drinking. Post hoc analyses found that females, those born outside the United States, and those with greater English language usage were at greater odds of being in Class 1 (vs. Class 2). Males, those born outside the United States, and those with greater Spanish language usage were at greater odds of being in Class 3 (vs. Class 2). Directions for future research and implications for practice with Latino emerging adults are discussed.

Conflicting views on elder care responsibility in Japan.

PubMed

Lee, Kristen Schultz

2016-05-01

I examine the attitudinal ambivalence created by conflicting social expectations regarding parent-child devotion, filial obligation and family membership, and gender norms in a national population of Japanese adults. I ask: in a context of rapidly changing family and elder care norms, how do different beliefs and attitudes overlap and conflict and how are they related to elder care preferences? I analyze data from the 2006 Japanese General Social Survey and use Latent Class Analysis to identify latent groups in the population defined by their beliefs and examine the relationship between class membership and elder care preferences. I found variation in the population with respect to the measured beliefs as well as a relationship between patterns of beliefs and choice of elder caregiver. I found conflicting expectations regarding elder care responsibility in one latent class and this class also expressed elder care preferences that conflict with at least some of their strongly held beliefs. Copyright © 2016 Elsevier Inc. All rights reserved.

The intersectionality of discrimination attributes and bullying among youth: an applied latent class analysis.

PubMed

Garnett, Bernice Raveche; Masyn, Katherine E; Austin, S Bryn; Miller, Matthew; Williams, David R; Viswanath, Kasisomayajula

2014-08-01

Discrimination is commonly experienced among adolescents. However, little is known about the intersection of multiple attributes of discrimination and bullying. We used a latent class analysis (LCA) to illustrate the intersections of discrimination attributes and bullying, and to assess the associations of LCA membership to depressive symptoms, deliberate self harm and suicidal ideation among a sample of ethnically diverse adolescents. The data come from the 2006 Boston Youth Survey where students were asked whether they had experienced discrimination based on four attributes: race/ethnicity, immigration status, perceived sexual orientation and weight. They were also asked whether they had been bullied or assaulted for these attributes. A total of 965 (78%) students contributed to the LCA analytic sample (45% Non-Hispanic Black, 29% Hispanic, 58% Female). The LCA revealed that a 4-class solution had adequate relative and absolute fit. The 4-classes were characterized as: low discrimination (51%); racial discrimination (33%); sexual orientation discrimination (7%); racial and weight discrimination with high bullying (intersectional class) (7%). In multivariate models, compared to the low discrimination class, individuals in the sexual orientation discrimination class and the intersectional class had higher odds of engaging in deliberate self-harm. Students in the intersectional class also had higher odds of suicidal ideation. All three discrimination latent classes had significantly higher depressive symptoms compared to the low discrimination class. Multiple attributes of discrimination and bullying co-occur among adolescents. Research should consider the co-occurrence of bullying and discrimination.

Patterns of HIV Risks and Related Factors among People Who Inject Drugs in Kermanshah, Iran: A Latent Class Analysis.

PubMed

Sharifi, Hamid; Mirzazadeh, Ali; Noroozi, Alireza; Marshall, Brandon D L; Farhoudian, Ali; Higgs, Peter; Vameghi, Meroe; Mohhamadi Shahboulaghi, Farahnaz; Qorbani, Mostafa; Massah, Omid; Armoon, Bahram; Noroozi, Mehdi

2017-01-01

The objective of this study was to explore patterns of drug use and sexual risk behaviors among people who inject drugs (PWID) in Iran. We surveyed 500 PWID in Kermanshah concerning demographic characteristics, sexual risk behaviors, and drug-related risk behaviors in the month prior to study. We used latent class analysis (LCA) to establish a baseline model of risk profiles and to identify the optimal number of latent classes, and we used ordinal regression to identify factors associated with class membership. Three classes of multiple HIV risk were identified. The probability of membership in the high-risk class was 0.33, compared to 0.26 and 0.40 for the low- and moderate-risk classes, respectively. Compared to members in the lowest-risk class (reference group), the highest-risk class members had higher odds of being homeless (OR = 4.5, CI: 1.44-8.22; p = 0.001) in the past 12 months. Members of the high-risk class had lower odds of regularly visiting a needle and syringe exchange program as compared to the lowest-risk class members (AOR = 0.42, CI: 0.2-0.81; p = 0.01). Findings show the sexual and drug-related HIV risk clusters among PWID in Iran, and emphasize the importance of developing targeted prevention and harm reduction programs for all domains of risk behaviors, both sexual and drug use related.

Separating "Rotators" from "Nonrotators" in the Mental Rotations Test: A Multigroup Latent Class Analysis

ERIC Educational Resources Information Center

Geiser, Christian; Lehmann, Wolfgang; Eid, Michael

2006-01-01

Items of mental rotation tests can not only be solved by mental rotation but also by other solution strategies. A multigroup latent class analysis of 24 items of the Mental Rotations Test (MRT) was conducted in a sample of 1,695 German pupils and students to find out how many solution strategies can be identified for the items of this test. The…

What Do Test Score Really Mean? A Latent Class Analysis of Danish Test Score Performance

ERIC Educational Resources Information Center

McIntosh, James; Munk, Martin D.

2014-01-01

Latent class Poisson count models are used to analyse a sample of Danish test score results from a cohort of individuals born in 1954-1955, tested in 1968, and followed until 2011. The procedure takes account of unobservable effects as well as excessive zeros in the data. We show that the test scores measure manifest or measured ability as it has…

Latent transition analysis of pre-service teachers' efficacy in mathematics and science

NASA Astrophysics Data System (ADS)

Ward, Elizabeth Kennedy

This study modeled changes in pre-service teacher efficacy in mathematics and science over the course of the final year of teacher preparation using latent transition analysis (LTA), a longitudinal form of analysis that builds on two modeling traditions (latent class analysis (LCA) and auto-regressive modeling). Data were collected using the STEBI-B, MTEBI-r, and the ABNTMS instruments. The findings suggest that LTA is a viable technique for use in teacher efficacy research. Teacher efficacy is modeled as a construct with two dimensions: personal teaching efficacy (PTE) and outcome expectancy (OE). Findings suggest that the mathematics and science teaching efficacy (PTE) of pre-service teachers is a multi-class phenomena. The analyses revealed a four-class model of PTE at the beginning and end of the final year of teacher training. Results indicate that when pre-service teachers transition between classes, they tend to move from a lower efficacy class into a higher efficacy class. In addition, the findings suggest that time-varying variables (attitudes and beliefs) and time-invariant variables (previous coursework, previous experiences, and teacher perceptions) are statistically significant predictors of efficacy class membership. Further, analyses suggest that the measures used to assess outcome expectancy are not suitable for LCA and LTA procedures.

Evaluation of a multiplex immunoassay for bovine respiratory syncytial virus and bovine coronavirus antibodies in bulk tank milk against two indirect ELISAs using latent class analysis.

PubMed

Toftaker, Ingrid; Toft, Nils; Stokstad, Maria; Sølverød, Liv; Harkiss, Gordon; Watt, Neil; O' Brien, Amanda; Nødtvedt, Ane

2018-06-01

Bovine respiratory syncytial virus (BRSV) and bovine coronavirus (BCV) are responsible for respiratory disease and diarrhea in cattle worldwide. The Norwegian control program against these infections is based on herd-level diagnosis using a new multiplex immunoassay. The objective of this study was to estimate sensitivity and specificity across different cut-off values for the MVD-Enferplex BCV/BRSV multiplex, by comparing them to a commercially available ELISA, the SVANOVIR ® BCV-Ab and SVANOVIR ® BRSV-Ab, respectively. We analyzed bulk tank milk samples from 360 herds in a low- and 360 herds in a high-prevalence area. As none of the tests were considered perfect, estimation of test characteristics was performed using Bayesian latent class models. At the manufacturers' recommended cut-off values, the median sensitivity for the BRSV multiplex and the BRSV ELISA was 94.4 [89.8-98.7 95% Posterior Credibility Interval (PCI)] and 99.8 [98.7-100 95% PCI], respectively. The median specificity for the BRSV multiplex was 90.6 [85.5-94.4 95% PCI], but only 57.4 [50.5-64.4 95% PCI] for the BRSV ELISA. However, increasing the cut-off of the BRSV ELISA increased specificity without compromising sensitivity. For the BCV multiplex we found that by using only one of the three antigens included in the test, the specificity increased, without concurrent loss in sensitivity. At the recommended cut-off this resulted in a sensitivity of 99.9 [99.3-100 95% PCI] and specificity of 93.7 [88.8-97.8 95% PCI] for the multiplex and a sensitivity of 99.5 [98.1-100 95% PCI] and a specificity of 99.6 [97.6-100 95% PCI] for the BCV ELISA. Copyright © 2018 Elsevier B.V. All rights reserved.

Elucidating the association between the self-harm inventory and several borderline personality measures in an inpatient psychiatric sample.

PubMed

Sellbom, Martin; Sansone, Randy A; Songer, Douglas A

2017-09-01

The current study evaluated the utility of the self-harm inventory (SHI) as a proxy for and screening measure of borderline personality disorder (BPD) using several diagnostic and statistical manual of mental disorders (DSM)-based BPD measures as criteria. We used a sample of 145 psychiatric inpatients, who completed the SHI and a series of well-validated, DSM-based self-report measures of BPD. Using a series of latent trait and latent class analyses, we found that the SHI was substantially associated with a latent construct representing BPD, as well as differentiated latent classes of 'high' vs. 'low' BPD, with good accuracy. The SHI can serve as proxy for and a good screening measure for BPD, but future research needs to replicate these findings using structured interview-based measurement of BPD.

Dimensionality of the Latent Structure and Item Selection via Latent Class Multidimensional IRT Models

ERIC Educational Resources Information Center

Bartolucci, F.; Montanari, G. E.; Pandolfi, S.

2012-01-01

With reference to a questionnaire aimed at assessing the performance of Italian nursing homes on the basis of the health conditions of their patients, we investigate two relevant issues: dimensionality of the latent structure and discriminating power of the items composing the questionnaire. The approach is based on a multidimensional item…

Higher Order Testlet Response Models for Hierarchical Latent Traits and Testlet-Based Items

ERIC Educational Resources Information Center

Huang, Hung-Yu; Wang, Wen-Chung

2013-01-01

Both testlet design and hierarchical latent traits are fairly common in educational and psychological measurements. This study aimed to develop a new class of higher order testlet response models that consider both local item dependence within testlets and a hierarchy of latent traits. Due to high dimensionality, the authors adopted the Bayesian…

Latent Transition Analysis of Pre-Service Teachers' Efficacy in Mathematics and Science

ERIC Educational Resources Information Center

Ward, Elizabeth Kennedy

2009-01-01

This study modeled changes in pre-service teacher efficacy in mathematics and science over the course of the final year of teacher preparation using latent transition analysis (LTA), a longitudinal form of analysis that builds on two modeling traditions (latent class analysis (LCA) and auto-regressive modeling). Data were collected using the…

School Victimization and Substance Use among Adolescents in California

PubMed Central

Astor, Ron A.; Estrada, Joey N.; Benbenishty, Rami; Unger, Jennifer B.

2016-01-01

Substance use and violence co-occur among adolescents. However, the extant literature focuses on the substance use behaviors of perpetrators of violence and not on victims. This study identifies patterns of school victimization and substance use and how they co-occur. The California Healthy Kids Survey was used to identify latent classes/clusters of school victimization patterns and lifetime and frequency of recent (past month) alcohol, tobacco, and marijuana use (N =419,698). Demographic characteristics (age, gender, and race/ethnicity) were included as predictors of latent class membership. Analyses revealed four latent classes of school victimization: low victimization (44.4 %), moderate victimization (22.3 %), verbal/relational victimization (20.8 %), and high victimization (with physical threats; 12.5 %). There were also four classes of substance use: non-users (58.5 %), alcohol experimenters (some recent alcohol use; 25.8 %), mild poly-substance users (lifetime use of all substances with few days of recent use; 9.1 %), and frequent poly-substance users (used all substances several times in the past month; 6.5 %). Those in the high victimization class were twice as likely to be frequent poly-substance users, and mild poly-substance use was most salient for those in the verbal victimization class. Few studies have explored latent patterns of substance use and violence victimization concurrently. The findings indicate substantial heterogeneity in victimization and substance use among youth in California schools with implications for targeted and tailored interventions. Understanding how certain types of victimization are associated with particular patterns of substance use will provide schools with opportunities to screen for concurrent behavioral health problems among youth. PMID:24482139

School victimization and substance use among adolescents in California.

PubMed

Gilreath, Tamika D; Astor, Ron A; Estrada, Joey N; Benbenishty, Rami; Unger, Jennifer B

2014-12-01

Substance use and violence co-occur among adolescents. However, the extant literature focuses on the substance use behaviors of perpetrators of violence and not on victims. This study identifies patterns of school victimization and substance use and how they co-occur. The California Healthy Kids Survey was used to identify latent classes/clusters of school victimization patterns and lifetime and frequency of recent (past month) alcohol, tobacco, and marijuana use (N = 419,698). Demographic characteristics (age, gender, and race/ethnicity) were included as predictors of latent class membership. Analyses revealed four latent classes of school victimization: low victimization (44.4 %), moderate victimization (22.3 %), verbal/relational victimization (20.8 %), and high victimization (with physical threats; 12.5 %). There were also four classes of substance use: non-users (58.5 %), alcohol experimenters (some recent alcohol use; 25.8 %), mild poly-substance users (lifetime use of all substances with few days of recent use; 9.1 %), and frequent poly-substance users (used all substances several times in the past month; 6.5 %). Those in the high victimization class were twice as likely to be frequent poly-substance users, and mild poly-substance use was most salient for those in the verbal victimization class. Few studies have explored latent patterns of substance use and violence victimization concurrently. The findings indicate substantial heterogeneity in victimization and substance use among youth in California schools with implications for targeted and tailored interventions. Understanding how certain types of victimization are associated with particular patterns of substance use will provide schools with opportunities to screen for concurrent behavioral health problems among youth.

Multi-class segmentation of neuronal electron microscopy images using deep learning

NASA Astrophysics Data System (ADS)

Khobragade, Nivedita; Agarwal, Chirag

2018-03-01

Study of connectivity of neural circuits is an essential step towards a better understanding of functioning of the nervous system. With the recent improvement in imaging techniques, high-resolution and high-volume images are being generated requiring automated segmentation techniques. We present a pixel-wise classification method based on Bayesian SegNet architecture. We carried out multi-class segmentation on serial section Transmission Electron Microscopy (ssTEM) images of Drosophila third instar larva ventral nerve cord, labeling the four classes of neuron membranes, neuron intracellular space, mitochondria and glia / extracellular space. Bayesian SegNet was trained using 256 ssTEM images of 256 x 256 pixels and tested on 64 different ssTEM images of the same size, from the same serial stack. Due to high class imbalance, we used a class-balanced version of Bayesian SegNet by re-weighting each class based on their relative frequency. We achieved an overall accuracy of 93% and a mean class accuracy of 88% for pixel-wise segmentation using this encoder-decoder approach. On evaluating the segmentation results using similarity metrics like SSIM and Dice Coefficient, we obtained scores of 0.994 and 0.886 respectively. Additionally, we used the network trained using the 256 ssTEM images of Drosophila third instar larva for multi-class labeling of ISBI 2012 challenge ssTEM dataset.

Bayesian inference of uncertainties in precipitation-streamflow modeling in a snow affected catchment

NASA Astrophysics Data System (ADS)

Koskela, J. J.; Croke, B. W. F.; Koivusalo, H.; Jakeman, A. J.; Kokkonen, T.

2012-11-01

Bayesian inference is used to study the effect of precipitation and model structural uncertainty on estimates of model parameters and confidence limits of predictive variables in a conceptual rainfall-runoff model in the snow-fed Rudbäck catchment (142 ha) in southern Finland. The IHACRES model is coupled with a simple degree day model to account for snow accumulation and melt. The posterior probability distribution of the model parameters is sampled by using the Differential Evolution Adaptive Metropolis (DREAM(ZS)) algorithm and the generalized likelihood function. Precipitation uncertainty is taken into account by introducing additional latent variables that were used as multipliers for individual storm events. Results suggest that occasional snow water equivalent (SWE) observations together with daily streamflow observations do not contain enough information to simultaneously identify model parameters, precipitation uncertainty and model structural uncertainty in the Rudbäck catchment. The addition of an autoregressive component to account for model structure error and latent variables having uniform priors to account for input uncertainty lead to dubious posterior distributions of model parameters. Thus our hypothesis that informative priors for latent variables could be replaced by additional SWE data could not be confirmed. The model was found to work adequately in 1-day-ahead simulation mode, but the results were poor in the simulation batch mode. This was caused by the interaction of parameters that were used to describe different sources of uncertainty. The findings may have lessons for other cases where parameterizations are similarly high in relation to available prior information.

The Importance of Isomorphism for Conclusions about Homology: A Bayesian Multilevel Structural Equation Modeling Approach with Ordinal Indicators.

PubMed

Guenole, Nigel

2016-01-01

We describe a Monte Carlo study examining the impact of assuming item isomorphism (i.e., equivalent construct meaning across levels of analysis) on conclusions about homology (i.e., equivalent structural relations across levels of analysis) under varying degrees of non-isomorphism in the context of ordinal indicator multilevel structural equation models (MSEMs). We focus on the condition where one or more loadings are higher on the between level than on the within level to show that while much past research on homology has ignored the issue of psychometric isomorphism, psychometric isomorphism is in fact critical to valid conclusions about homology. More specifically, when a measurement model with non-isomorphic items occupies an exogenous position in a multilevel structural model and the non-isomorphism of these items is not modeled, the within level exogenous latent variance is under-estimated leading to over-estimation of the within level structural coefficient, while the between level exogenous latent variance is overestimated leading to underestimation of the between structural coefficient. When a measurement model with non-isomorphic items occupies an endogenous position in a multilevel structural model and the non-isomorphism of these items is not modeled, the endogenous within level latent variance is under-estimated leading to under-estimation of the within level structural coefficient while the endogenous between level latent variance is over-estimated leading to over-estimation of the between level structural coefficient. The innovative aspect of this article is demonstrating that even minor violations of psychometric isomorphism render claims of homology untenable. We also show that posterior predictive p-values for ordinal indicator Bayesian MSEMs are insensitive to violations of isomorphism even when they lead to severely biased within and between level structural parameters. We highlight conditions where poor estimation of even correctly specified models rules out empirical examination of isomorphism and homology without taking precautions, for instance, larger Level-2 sample sizes, or using informative priors.

The Importance of Isomorphism for Conclusions about Homology: A Bayesian Multilevel Structural Equation Modeling Approach with Ordinal Indicators

PubMed Central

Guenole, Nigel

2016-01-01

We describe a Monte Carlo study examining the impact of assuming item isomorphism (i.e., equivalent construct meaning across levels of analysis) on conclusions about homology (i.e., equivalent structural relations across levels of analysis) under varying degrees of non-isomorphism in the context of ordinal indicator multilevel structural equation models (MSEMs). We focus on the condition where one or more loadings are higher on the between level than on the within level to show that while much past research on homology has ignored the issue of psychometric isomorphism, psychometric isomorphism is in fact critical to valid conclusions about homology. More specifically, when a measurement model with non-isomorphic items occupies an exogenous position in a multilevel structural model and the non-isomorphism of these items is not modeled, the within level exogenous latent variance is under-estimated leading to over-estimation of the within level structural coefficient, while the between level exogenous latent variance is overestimated leading to underestimation of the between structural coefficient. When a measurement model with non-isomorphic items occupies an endogenous position in a multilevel structural model and the non-isomorphism of these items is not modeled, the endogenous within level latent variance is under-estimated leading to under-estimation of the within level structural coefficient while the endogenous between level latent variance is over-estimated leading to over-estimation of the between level structural coefficient. The innovative aspect of this article is demonstrating that even minor violations of psychometric isomorphism render claims of homology untenable. We also show that posterior predictive p-values for ordinal indicator Bayesian MSEMs are insensitive to violations of isomorphism even when they lead to severely biased within and between level structural parameters. We highlight conditions where poor estimation of even correctly specified models rules out empirical examination of isomorphism and homology without taking precautions, for instance, larger Level-2 sample sizes, or using informative priors. PMID:26973580

Exploring evidence of a dissociative subtype in PTSD: Baseline symptom structure, etiology, and treatment efficacy for those who dissociate.

PubMed

Burton, Mark S; Feeny, Norah C; Connell, Arin M; Zoellner, Lori A

2018-05-01

With the inclusion of a dissociative subtype, recent changes to the DSM-5 diagnosis of posttraumatic stress disorder (PTSD) have emphasized the role of dissociation in the experience and treatment of the disorder. However, there is a lack of research exploring the clinical impact for highly dissociative groups receiving treatment for PTSD. The current study examined the presence and clinical impact of a dissociative subtype in a sample of individuals receiving treatment for chronic PTSD. This study used latent transition analyses (LTA), an expanded form of latent profile analyses (LPA), to examine latent profiles of PTSD and dissociation symptoms before and after treatment for individuals (N = 200) receiving prolonged exposure (PE) or sertraline treatment for chronic PTSD. The best fitting LTA model was one with a 4-class solution at both pretreatment and posttreatment. There was a latent class at pretreatment with higher levels of dissociative symptoms. However, this class was also marked by higher reexperiencing symptoms, and membership was not predicted by chronic child abuse. Further, although those in the class were less likely to transition to the responder class overall, this was not the case for exposure-based treatment specifically. These findings are not in line with the dissociative-subtype theoretical literature that proposes those who dissociate represent a clinically distinct group that may respond worse to exposure-based treatments for PTSD. (PsycINFO Database Record (c) 2018 APA, all rights reserved).

Optimal Cutoff and Accuracy of an IgM Enzyme-Linked Immunosorbent Assay for Diagnosis of Acute Scrub Typhus in Northern Thailand: an Alternative Reference Method to the IgM Immunofluorescence Assay

PubMed Central

Blacksell, Stuart D.; Tanganuchitcharnchai, Ampai; Jintaworn, Suthatip; Kantipong, Pacharee; Richards, Allen L.; Day, Nicholas P. J.

2016-01-01

The enzyme-linked immunosorbent assay (ELISA) has been proposed as an alternative serologic diagnostic test to the indirect immunofluorescence assay (IFA) for scrub typhus. Here, we systematically determine the optimal sample dilution and cutoff optical density (OD) and estimate the accuracy of IgM ELISA using Bayesian latent class models (LCMs). Data from 135 patients with undifferentiated fever were reevaluated using Bayesian LCMs. Every patient was evaluated for the presence of an eschar and tested with a blood culture for Orientia tsutsugamushi, three different PCR assays, and an IgM IFA. The IgM ELISA was performed for every sample at sample dilutions from 1:100 to 1:102,400 using crude whole-cell antigens of the Karp, Kato, and Gilliam strains of O. tsutsugamushi developed by the Naval Medical Research Center. We used Bayesian LCMs to generate unbiased receiver operating characteristic curves and found that the sample dilution of 1:400 was optimal for the IgM ELISA. With the optimal cutoff OD of 1.474 at a sample dilution of 1:400, the IgM ELISA had a sensitivity of 85.7% (95% credible interval [CrI], 77.4% to 86.7%) and a specificity of 98.1% (95% CrI, 97.2% to 100%) using paired samples. For the ELISA, the OD could be determined objectively and quickly, in contrast to the reading of IFA slides, which was both subjective and labor-intensive. The IgM ELISA for scrub typhus has high diagnostic accuracy and is less subjective than the IgM IFA. We suggest that the IgM ELISA may be used as an alternative reference test to the IgM IFA for the serological diagnosis of scrub typhus. PMID:27008880

Incremental Value (Bayesian Framework) of Thoracic Ultrasonography over Thoracic Auscultation for Diagnosis of Bronchopneumonia in Preweaned Dairy Calves.

PubMed

Buczinski, S; Ménard, J; Timsit, E

2016-07-01

Thoracic ultrasonography (TUS) is a specific and relatively sensitive method to diagnose bronchopneumonia (BP) in dairy calves. Unfortunately, as it requires specific training and equipment, veterinarians typically base their diagnosis on thoracic auscultation (AUSC), which is rapid and easy to perform. We hypothesized that the use of TUS, in addition to AUSC, can significantly increase accuracy of BP diagnosis. Therefore, the objectives were to (i) determine the incremental value of TUS over AUSC for diagnosis of BP in preweaned dairy calves and (ii) assess diagnostic accuracy of AUSC. Two hundred and nine dairy calves (<1 month of age) were enrolled in this cross-sectional study. Prospective cross-sectional study. All calves from a veal calves unit were examined (independent operators) using the Wisconsin Calf Respiratory Scoring Criteria (CRSC), AUSC, and TUS. A Bayesian latent class approach was used to estimate the incremental value of AUSC over TUS (integrated discrimination improvement [IDI]) and the diagnostic accuracy of AUSC. Abnormal CRSC, AUSC, and TUS were recorded in 3.3, 53.1, and 23.9% of calves, respectively. AUSC was sensitive (72.9%; 95% Bayesian credible interval [BCI]: 50.1-96.4%), but not specific (53.3%; 95% BCI: 43.3-64.0%) to diagnose BP. Compared to AUSC, TUS was more specific (92.9%; 95% BCI: 86.5-97.1%), but had similar sensitivity (76.5%; 95% BCI: 60.2-88.8%). The incremental value of TUS over AUSC was high (IDI = 43.7%; 5% BCI: 22.0-63.0%) significantly improving proportions of sick and healthy calves appropriately classified. The use of TUS over AUSC significantly improved accuracy of BP diagnosis in dairy calves. Copyright © 2016 The Authors. Journal of Veterinary Internal Medicine published by Wiley Periodicals, Inc. on behalf of the American College of Veterinary Internal Medicine.

Substance use, mental illness, and familial conflict non-negotiation among HIV-positive African-Americans: latent class regression and a new syndemic framework.

PubMed

Robinson, Allysha C; Knowlton, Amy R; Gielen, Andrea C; Gallo, Joseph J

2016-02-01

We evaluated a synergistic epidemic (syndemic) of substance use, mental illness, and familial conflict non-negotiation among HIV-positive injection drug users (IDU). Baseline BEACON study data was utilized. Latent class analyses identified syndemic classes. These classes were regressed on sex, viral suppression, and acute care non-utilization. Females were hypothesized to have higher syndemic burden, and worse health outcomes than males. Nine percent of participants had high substance use/mental illness prevalence (Class 4); 23 % had moderate levels of all factors (Class 3); 25 % had high mental illness (Class 2); 43 % had moderate substance use/mental illness (Class 1; N = 331). Compared to Classes 1-3, Class 4 was mostly female (p < .05), less likely to achieve viral suppression, and more likely to utilize acute care (p < .05). Interventions should target African-American IDU females to improve their risk of negative medical outcomes. Findings support comprehensive syndemic approaches to HIV interventions, rather than singular treatment methods.

Causal Latent Markov Model for the Comparison of Multiple Treatments in Observational Longitudinal Studies

ERIC Educational Resources Information Center

Bartolucci, Francesco; Pennoni, Fulvia; Vittadini, Giorgio

2016-01-01

We extend to the longitudinal setting a latent class approach that was recently introduced by Lanza, Coffman, and Xu to estimate the causal effect of a treatment. The proposed approach enables an evaluation of multiple treatment effects on subpopulations of individuals from a dynamic perspective, as it relies on a latent Markov (LM) model that is…

Examination of the Predictors of Latent Class Typologies of Bullying Involvement among Middle School Students

PubMed Central

LOVEGROVE, PETER J.; HENRY, KIMBERLY L.; SLATER, MICHAEL D.

2012-01-01

This study employs latent class analysis to construct bullying involvement typologies among 3114 students (48% male, 58% White) in 40 middle schools across the U.S. Four classes were constructed: victims (15%); bullies (13%); bully-victims (13%); and noninvolved (59%). Respondents who were male and participated in fewer conventional activities were more likely to be members of the victims class. Students who were African-American and reported being less successful at school had a higher likelihood of membership in the bullies class. Bully-victims shared characteristics with bullies and victims: Students with more feelings of anger toward others and a higher tendency toward sensation-seeking had a higher likelihood of membership in the bullies and bully-victims classes, whereas lower levels of social inclusion was associated with membership in the victims and bully-victims classes. PMID:22606069

Nonlinear Structured Growth Mixture Models in Mplus and OpenMx

PubMed Central

Grimm, Kevin J.; Ram, Nilam; Estabrook, Ryne

2014-01-01

Growth mixture models (GMMs; Muthén & Muthén, 2000; Muthén & Shedden, 1999) are a combination of latent curve models (LCMs) and finite mixture models to examine the existence of latent classes that follow distinct developmental patterns. GMMs are often fit with linear, latent basis, multiphase, or polynomial change models because of their common use, flexibility in modeling many types of change patterns, the availability of statistical programs to fit such models, and the ease of programming. In this paper, we present additional ways of modeling nonlinear change patterns with GMMs. Specifically, we show how LCMs that follow specific nonlinear functions can be extended to examine the presence of multiple latent classes using the Mplus and OpenMx computer programs. These models are fit to longitudinal reading data from the Early Childhood Longitudinal Study-Kindergarten Cohort to illustrate their use. PMID:25419006

Negative Treatment by Family as a Predictor of Depressive Symptoms, Life Satisfaction, Suicidality, and Tobacco/Alcohol Use in Vietnamese Sexual Minority Women.

PubMed

Nguyen, Trang Quynh; Bandeen-Roche, Karen; German, Danielle; Nguyen, Nam T T; Bass, Judith K; Knowlton, Amy R

2016-10-01

Research linking family rejection and health outcomes in sexual minority people is mostly limited to North America. We assessed the associations between negative treatment by family members and depressive symptoms, life satisfaction, suicidality, and tobacco/alcohol use in sexual minority women (SMW) in Viet Nam. Data were from an anonymous internet survey (n = 1936). Latent class analysis characterized patterns of negative treatment by family members experienced by respondents. Latent class with distal outcome modeling was used to regress depressive symptoms, life satisfaction, suicidality, and tobacco/alcohol use on family treatment class, controlling for predictors of family treatment and for two other types of sexual prejudice. Five latent family treatment classes were extracted, including four negative classes representing varying patterns of negative family treatment. Overall, more than one negative class predicted lower life satisfaction, more depressive symptoms, and higher odds of attempted suicide (relative to the non-negative class), supporting the minority stress hypothesis that negative family treatment is predictive of poorer outcomes. Only the most negative class had elevated alcohol use. The association between family treatment and smoking status was not statistically significant. The most negative class, unexpectedly, did not have the highest odds of having attempted suicide, raising a question about survivor bias. This population requires public health attention, with emphasis placed on interventions targeting the family to promote acceptance and to prevent negative treatment, and interventions supporting those SMW who encounter the worst types of negative family treatment.

Use of Systematic Methods to Improve Disease Identification in Administrative Data: The Case of Severe Sepsis.

PubMed

Shahraz, Saeid; Lagu, Tara; Ritter, Grant A; Liu, Xiadong; Tompkins, Christopher

2017-03-01

Selection of International Classification of Diseases (ICD)-based coded information for complex conditions such as severe sepsis is a subjective process and the results are sensitive to the codes selected. We use an innovative data exploration method to guide ICD-based case selection for severe sepsis. Using the Nationwide Inpatient Sample, we applied Latent Class Analysis (LCA) to determine if medical coders follow any uniform and sensible coding for observations with severe sepsis. We examined whether ICD-9 codes specific to sepsis (038.xx for septicemia, a subset of 995.9 codes representing Systemic Inflammatory Response syndrome, and 785.52 for septic shock) could all be members of the same latent class. Hospitalizations coded with sepsis-specific codes could be assigned to a latent class of their own. This class constituted 22.8% of all potential sepsis observations. The probability of an observation with any sepsis-specific codes being assigned to the residual class was near 0. The chance of an observation in the residual class having a sepsis-specific code as the principal diagnosis was close to 0. Validity of sepsis class assignment is supported by empirical results, which indicated that in-hospital deaths in the sepsis-specific class were around 4 times as likely as that in the residual class. The conventional methods of defining severe sepsis cases in observational data substantially misclassify sepsis cases. We suggest a methodology that helps reliable selection of ICD codes for conditions that require complex coding.

Consumption Patterns of Nightlife Attendees in Munich: A Latent-Class Analysis.

PubMed

Hannemann, Tessa-Virginia; Kraus, Ludwig; Piontek, Daniela

2017-09-19

The affinity for substance use among patrons of nightclubs has been well established. With novel psychoactive substances (NPS) quickly emerging on the European drug market, trends, and patterns of use are potentially changing. (1) The detection of subgroups of consumers in the electronic dance music scene of a major German metropolitan city, (2) describing the consumption patterns of these subgroups, (3) exploring the prevalence and type of NPS consumption in this population at nightlife events in Munich. A total of 1571 patrons answered questions regarding their own substance use and the emergence of NPS as well as their experience with these substances. A latent class analysis was employed to detect consumption patterns within the sample. A four class model was determined reflecting different consumption patterns: the conservative class (34.9%) whose substance was limited to cannabis; the traditional class (36.6%) which especially consumed traditional club drugs; the psychedelic class (17.5%) which, in addition to traditional club drugs also consumed psychedelic drugs; and an unselective class (10.9%) which displayed the greatest likelihood of consumption of all assessed drugs. "Smoking mixtures" and methylone were the new substances mentioned most often, the number of substances mentioned differed between latent classes. Specific strategies are needed to reduce harm in those displaying the riskiest substance use. Although NPS use is still a fringe phenomenon its prevalence is greater in this subpopulation than in the general population, especially among users in the high-risk unselective class.

Prevalence of paratuberculosis in the dairy goat and dairy sheep industries in Ontario, Canada.

PubMed

Bauman, Cathy A; Jones-Bitton, Andria; Menzies, Paula; Toft, Nils; Jansen, Jocelyn; Kelton, David

2016-02-01

A cross-sectional study was undertaken (October 2010 to August 2011) to estimate the prevalence of paratuberculosis in the small ruminant dairy industries in Ontario, Canada. Blood and feces were sampled from 580 goats and 397 sheep (lactating and 2 y of age or older) that were randomly selected from 29 randomly selected dairy goat herds and 21 convenience-selected dairy sheep flocks. Fecal samples were analyzed using bacterial culture (BD BACTEC MGIT 960) and polymerase chain reaction (Tetracore); serum samples were tested with the Prionics Parachek enzyme-linked immunosorbent assay (ELISA). Using 3-test latent class Bayesian models, true farm-level prevalence was estimated to be 83.0% [95% probability interval (PI): 62.6% to 98.1%] for dairy goats and 66.8% (95% PI: 41.6% to 91.4%) for dairy sheep. The within-farm true prevalence for dairy goats was 35.2% (95% PI: 23.0% to 49.8%) and for dairy sheep was 48.3% (95% PI: 27.6% to 74.3%). These data indicate that a paratuberculosis control program for small ruminants is needed in Ontario.

Evaluation of Two rK39 Dipstick Tests, Direct Agglutination Test, and Indirect Fluorescent Antibody Test for Diagnosis of Visceral Leishmaniasis in a New Epidemic Site in Highland Ethiopia

PubMed Central

Cañavate, Carmen; Herrero, Merce; Nieto, Javier; Cruz, Israel; Chicharro, Carmen; Aparicio, Pilar; Mulugeta, Abate; Argaw, Daniel; Blackstock, Anna J.; Alvar, Jorge; Bern, Caryn

2011-01-01

We assessed the performance characteristics of two rK39 immunochromatographic tests, a direct agglutination test (DAT), and an indirect immunofluorescent antibody test (IFAT) in the site of a new epidemic of visceral leishmaniasis (VL) in northwestern Ethiopia. The study population was composed of 179 patients with suspected VL and 67 controls. The sensitivities of Kalazar Detect®, DiaMed-IT Leish®, DAT, and IFAT in 35 polymerase chain reaction–confirmed VL cases were 94.3%, 91.4%, 91.4%, and 100%, respectively, and the specificities were 98.5%, 94%, 98.5%, and 98.5%, respectively. In a Bayesian latent class analysis of all 246 specimens, the estimated sensitivities were 90.5%, 89%, 88.8%, and 96% for Kalazar Detect®, DiaMed-IT Leish®, DAT, and IFAT, respectively; DAT showed the highest estimated specificity (97.4%). Both rK39 immunochromatographic tests perform as well as DAT, and are suitable for VL diagnosis in first-level health centers in this area of Ethiopia. PMID:21212210

Chronic Disease Risk Typologies among Young Adults in Community College.

PubMed

Jeffries, Jayne K; Lytle, Leslie; Sotres-Alvarez, Daniela; Golden, Shelley; Aiello, Allison E; Linnan, Laura

2018-03-01

To address chronic disease risk holistically from a behavioral perspective, insights are needed to refine understanding of the covariance of key health behaviors. This study aims to identify distinct typologies of young adults based on 4 modifiable risk factors of chronic disease using a latent class analysis approach, and to describe patterns of class membership based on demographic characteristics, living arrangements, and weight. Overall, 441 young adults aged 18-35 attending community colleges in the Minnesota Twin Cities area completed a baseline questionnaire for the Choosing Healthy Options in College Environments and Settings study, a RCT. Behavioral items were used to create indicators for latent classes, and individuals were classified using maximum-probability assignment. Three latent classes were identified: 'active, binge-drinkers with a healthy dietary intake' (13.1%); 'non-active, moderate-smokers and non-drinkers with poor dietary intake' (38.2%); 'moderately active, non-smokers and non-drinkers with moderately healthy dietary intake' (48.7%). Classes exhibited unique demographic and weight-related profiles. This study may contribute to the literature on health behaviors among young adults and provides evidence that there are weight and age differences among subgroups. Understanding how behaviors cluster is important for identifying groups for targeted interventions in community colleges.

Risk-taking behaviors and subgrouping of suicide in Iran: A latent class analysis of national registries data.

PubMed

Hajebi, Ahmad; Abbasi-Ghahramanloo, Abbas; Hashemian, Seyed Sepehr; Khatibi, Seyed Reza; Ghasemzade, Masomeh; Khodadost, Mahmoud

2017-09-01

Suicide is one the most important public health problem which is rapidly growing concerns. The aim of this study was to subgroup suicide using LCA method. This cross-sectional study was conducted in Iran based on 66990 records registered in Ministry of Health in 2014. A case report questionnaire in the form of software was used for case registries. Latent class analysis was used to achieve the research objectives. Four latent classes were identified; (a) Non-lethal attempters without a history of psychiatric disorders, (b) Non-lethal attempters with a history of psychiatric disorders, (c) Lethal attempters without a history of psychiatric disorders, (d) Lethal attempters with a history of psychiatric disorders. The probability of completed/an achieved suicide is high in lethal attempter classes. Being male increases the risk of inclusion in lethal attempters' classes (OR = 4.93). Also, being single (OR = 1.16), having an age lower than 25 years (OR = 1.14) and being a rural citizen (OR = 2.36) associate with lethal attempters classes. The males tend to use more violent methods and have more completed suicide. Majority of the individuals are non-lethal attempters who need to be addressed by implementing preventive interventions and mental support provision. Copyright © 2017. Published by Elsevier B.V.

A Bayesian hierarchical latent trait model for estimating rater bias and reliability in large-scale performance assessment

PubMed Central

2018-01-01

We propose a novel approach to modelling rater effects in scoring-based assessment. The approach is based on a Bayesian hierarchical model and simulations from the posterior distribution. We apply it to large-scale essay assessment data over a period of 5 years. Empirical results suggest that the model provides a good fit for both the total scores and when applied to individual rubrics. We estimate the median impact of rater effects on the final grade to be ± 2 points on a 50 point scale, while 10% of essays would receive a score at least ± 5 different from their actual quality. Most of the impact is due to rater unreliability, not rater bias. PMID:29614129

Job Satisfaction among Health-Care Staff in Township Health Centers in Rural China: Results from a Latent Class Analysis

PubMed Central

Wang, Haipeng; Tang, Chengxiang; Zhao, Shichao; Meng, Qingyue; Liu, Xiaoyun

2017-01-01

Background: The lower job satisfaction of health-care staff will lead to more brain drain, worse work performance, and poorer health-care outcomes. The aim of this study was to identify patterns of job satisfaction among health-care staff in rural China, and to investigate the association between the latent clusters and health-care staff’s personal and professional features; Methods: We selected 12 items of five-point Likert scale questions to measure job satisfaction. A latent-class analysis was performed to identify subgroups based on the items of job satisfaction; Results: Four latent classes of job satisfaction were identified: 8.9% had high job satisfaction, belonging to “satisfied class”; 38.2% had low job satisfaction, named as “unsatisfied class”; 30.5% were categorized into “unsatisfied class with the exception of interpersonal relationships”; 22.4% were identified as “pseudo-satisfied class”, only satisfied with management-oriented items. Low job satisfaction was associated with specialty, training opportunity, and income inequality. Conclusions: The minority of health-care staff belong to the “satisfied class”. Three among four subgroups are not satisfied with income, benefit, training, and career development. Targeting policy interventions should be implemented to improve the items of job satisfaction based on the patterns and health-care staff’s features. PMID:28937609

Classification of multiple sclerosis patients by latent class analysis of magnetic resonance imaging characteristics.

PubMed

Zwemmer, J N P; Berkhof, J; Castelijns, J A; Barkhof, F; Polman, C H; Uitdehaag, B M J

2006-10-01

Disease heterogeneity is a major issue in multiple sclerosis (MS). Classification of MS patients is usually based on clinical characteristics. More recently, a pathological classification has been presented. While clinical subtypes differ by magnetic resonance imaging (MRI) signature on a group level, a classification of individual MS patients based purely on MRI characteristics has not been presented so far. To investigate whether a restricted classification of MS patients can be made based on a combination of quantitative and qualitative MRI characteristics and to test whether the resulting subgroups are associated with clinical and laboratory characteristics. MRI examinations of the brain and spinal cord of 50 patients were scored for 21 quantitative and qualitative characteristics. Using latent class analysis, subgroups were identified, for whom disease characteristics and laboratory measures were compared. Latent class analysis revealed two subgroups that mainly differed in the extent of lesion confluency and MRI correlates of neuronal loss in the brain. Demographics and disease characteristics were comparable except for cognitive deficits. No correlations with laboratory measures were found. Latent class analysis offers a feasible approach for classifying subgroups of MS patients based on the presence of MRI characteristics. The reproducibility, longitudinal evolution and further clinical or prognostic relevance of the observed classification will have to be explored in a larger and independent sample of patients.

Phenotypic and molecular characteristics associated with various domains of quality of life in oncology patients and their family caregivers.

PubMed

Alexander, Kimberly E; Cooper, Bruce A; Paul, Steven M; Yates, Patsy; Aouizerat, Bradley E; Miaskowski, Christine

2016-11-01

Not all oncology patients and their family caregivers (FCs) experience the same quality of life (QOL). The purposes of this study were to identify latent classes of oncology patients (n = 168) and their FCs (n = 85) with distinct physical, psychological, social, and spiritual well-being trajectories from prior to through 4 months after the completion of radiation therapy and to evaluate for demographic, clinical, and genetic characteristics that distinguished between these latent classes. Using growth mixture modeling, two latent classes were found for three (i.e., physical, psychological, and social well-being) of the four QOL domains evaluated. Across these three domains, the largest percentage of participants reported relatively high well-being scores across the 6 months of the study. Across these three QOL domains, patients and FCs who were younger, female, belonged to an ethnic minority group, had children at home, had multiple comorbid conditions, or had a lower functional status, were more likely to be classified in the lower QOL class. The social well-being domain was the only domain that had a polymorphism in nuclear factor kappa beta 2 (NFKB2) associated with latent class membership. Carrying one or two doses of the rare allele for rs7897947 was associated with a 54 % decrease in the odds of belonging to the lower social well-being class [OR (95 % CI) = .46 (.21, .99), p = .049]. These findings suggest that a number of phenotypic and molecular characteristics contribute to differences in QOL in oncology patients and their FCs.

Differentiation of functional constipation and constipation predominant irritable bowel syndrome based on Rome III criteria: a population-based study.

PubMed

Koloski, N A; Jones, M; Young, M; Talley, N J

2015-05-01

While the Rome III classification recognises functional constipation (FC) and constipation predominant IBS (IBS-C) as distinct disorders, recent evidence has suggested that these disorders are difficult to separate in clinical practice. To identify whether clinical and lifestyle factors differentiate Rome III-defined IBS-C from FC based on gastrointestinal symptoms and lifestyle characteristics. 3260 people randomly selected from the Australian population returned a postal survey. FC and IBS-C were defined according to Rome III. The first model used logistic regression to differentiate IBS-C from FC based on lifestyle, quality-of-life and psychological characteristics. The second approach was data-driven employing latent class analysis (LCA) to identify naturally occurring clusters in the data considering all symptoms involved in the Rome III criteria for IBS-C and FC. We found n = 206 (6.5%; 95% CI 5.7-7.4%) people met strict Rome III FC whereas n = 109 (3.5%; 95% CI 2.8-4.1%) met strict Rome III IBS-C. The case-control approach indicated that FC patients reported an older age at onset of constipation, were less likely to exercise, had higher mental QoL and less health care seeking than IBS-C. LCA yielded one latent class that was predominantly (75%) FC, while the other class was approximately half IBS-C and half FC. The FC-dominated latent class had clearly lower levels of symptoms used to classify IBS (pain-related symptoms) and was more likely to be male (P = 0.046) but was otherwise similar in distribution of lifestyle factors to the mixed class. The latent class analysis approach suggests a differentiation based more on symptom severity rather than the Rome III view. © 2015 John Wiley & Sons Ltd.

General practitioners' knowledge and concern about electromagnetic fields.

PubMed

Berg-Beckhoff, Gabriele; Breckenkamp, Jürgen; Larsen, Pia Veldt; Kowall, Bernd

2014-12-01

Our aim is to explore general practitioners' (GPs') knowledge about EMF, and to assess whether different knowledge structures are related to the GPs' concern about EMF. Random samples were drawn from lists of GPs in Germany in 2008. Knowledge about EMF was assessed by seven items. A latent class analysis was conducted to identify latent structures in GPs' knowledge. Further, the GPs' concern about EMF health risk was measured using a score comprising six items. The association between GPs' concern about EMF and their knowledge was analysed using multiple linear regression. In total 435 (response rate 23.3%) GPs participated in the study. Four groups were identified by the latent class analysis: 43.1% of the GPs gave mainly correct answers; 23.7% of the GPs answered low frequency EMF questions correctly; 19.2% answered only the questions relating EMF with health risks, and 14.0% answered mostly "don't know". There was no association between GPs' latent knowledge classes or between the number of correct answers given by the GPs and their EMF concern, whereas the number of incorrect answers was associated with EMF concern. Greater EMF concern in subjects with more incorrect answers suggests paying particular attention to misconceptions regarding EMF in risk communication.

Substance Use Patterns Among Adolescents in Europe: A Latent Class Analysis.

PubMed

Göbel, Kristin; Scheithauer, Herbert; Bräker, Astrid-Britta; Jonkman, Harrie; Soellner, Renate

2016-07-28

Several researchers have investigated substance use patterns using a latent class analysis; however, hardly no studies exist on substance use patterns across countries. Adolescent substance use patterns, demographic factors, and international differences in the prevalence of substance use patterns were explored. Data from 25 European countries were used to identify patterns of adolescent (12-16 years, 50.6% female) substance use (N = 33,566). Latent class analysis revealed four substance use classes: nonusers (68%), low-alcohol users (recent use of beer, wine, and alcopops; 16.1%), alcohol users (recent use of alcohol and lifetime use of marijuana; 11.2%), and polysubstance users (recent use of alcohol, marijuana, and other illicit drugs; 4.7%). Results support a general pattern of adolescent substance use across all countries; however, the prevalence rates of use patterns vary for each country. The present research provides insight into substance use patterns across Europe by using a large international adolescent sample, multidimensional indicators and a variety of substances. Substance use patterns are helpful when targeting policy and prevention strategies.

Classification and Clustering Methods for Multiple Environmental Factors in Gene-Environment Interaction: Application to the Multi-Ethnic Study of Atherosclerosis.

PubMed

Ko, Yi-An; Mukherjee, Bhramar; Smith, Jennifer A; Kardia, Sharon L R; Allison, Matthew; Diez Roux, Ana V

2016-11-01

There has been an increased interest in identifying gene-environment interaction (G × E) in the context of multiple environmental exposures. Most G × E studies analyze one exposure at a time, but we are exposed to multiple exposures in reality. Efficient analysis strategies for complex G × E with multiple environmental factors in a single model are still lacking. Using the data from the Multiethnic Study of Atherosclerosis, we illustrate a two-step approach for modeling G × E with multiple environmental factors. First, we utilize common clustering and classification strategies (e.g., k-means, latent class analysis, classification and regression trees, Bayesian clustering using Dirichlet Process) to define subgroups corresponding to distinct environmental exposure profiles. Second, we illustrate the use of an additive main effects and multiplicative interaction model, instead of the conventional saturated interaction model using product terms of factors, to study G × E with the data-driven exposure subgroups defined in the first step. We demonstrate useful analytical approaches to translate multiple environmental exposures into one summary class. These tools not only allow researchers to consider several environmental exposures in G × E analysis but also provide some insight into how genes modify the effect of a comprehensive exposure profile instead of examining effect modification for each exposure in isolation.

Colorectal Cancer Screening: Preferences, Past Behavior, and Future Intentions.

PubMed

Mansfield, Carol; Ekwueme, Donatus U; Tangka, Florence K L; Brown, Derek S; Smith, Judith Lee; Guy, Gery P; Li, Chunyu; Hauber, Brett

2018-05-09

Screening rates for colorectal cancer are below the Healthy People 2020 goal. There are several colorectal cancer screening tests that differ in terms of accuracy, recommended frequency, and administration. In this article, we compare how a set of personal characteristics correlates with preferences for colorectal cancer screening test attributes, past colorectal cancer screening behavior, and future colorectal cancer screening intentions. We conducted a discrete-choice experiment survey to assess relative preferences for attributes of colorectal cancer screening tests among adults aged 50-75 years in USA. We used a latent class logit model to identify classes of preferences and calculated willingness to pay for changes in test attributes. A set of personal characteristics were included in the latent class analysis and analyses of self-reported past screening behavior and self-assessed likelihood of future colorectal cancer screening. Latent class analysis identified three types of respondents. Class 1 valued test accuracy, class 2 valued removing polyps and avoiding discomfort, and class 3 valued cost. Having had a prior colonoscopy and a higher income were predictors of the likelihood of future screening and membership in classes 1 and 2. Health insurance and a self-reported higher risk of developing colorectal cancer were associated with prior screening and higher future screening intentions, but not class membership. We identified distinct classes of preferences focusing on different test features and personal characteristics associated with reported behavior and intentions. Healthcare providers should engage in a careful assessment of patient preferences when recommending colorectal cancer test options to encourage colorectal cancer screening uptake.

Maternal eating disorder and infant diet. A latent class analysis based on the Norwegian Mother and Child Cohort Study (MoBa)

PubMed Central

Torgersen, Leila; Ystrom, Eivind; Siega-Riz, Anna Maria; Berg, Cecilie Knoph; Zerwas, Stephanie; Reichborn-Kjennerud, Ted; Bulik, Cynthia M.

2015-01-01

Knowledge of infant diet and feeding practices among children of mothers with eating disorders is essential to promote healthy eating in these children. This study compared the dietary patterns of 6-month-old children of mothers with anorexia nervosa, bulimia nervosa, binge eating disorder, and eating disorder not otherwise specified - purging subtype, to the diet of children of mothers with no eating disorders. The study was based on 53,879 mothers in the Norwegian Mother and Child Cohort Study (MoBa). Latent class analysis (LCA) was used to identify discrete latent classes of infant diet based on the mothers’ responses to questions about 16 food items. LCA identified five classes, characterized by primarily homemade vegetarian food (4% of the infants in the sample), homemade traditional food (8%), commercial infant cereals (35%), commercial jarred baby food (39%), and a mix of all food groups (11%). We then estimated the association between the different latent dietary classes and maternal eating disorders using a multinomial logistic regression model. Infants of mothers with bulimia nervosa had a lower probability of being in the homemade traditional food class compared to the commercial jarred baby food class, than the referent without an eating disorder (O.R. 0.59; 95% CI 0.36–0.99). Infants of mothers with binge eating disorder had a lower probability of being in the homemade vegetarian class compared to the commercial jarred baby food class, than the referent (O.R. 0.77; 95% CI 0.60–0.99), but only before controlling for relevant confounders. Anorexia nervosa and eating disorder not otherwise specified-purging subtype were not statistically significant associated with any of the dietary classes. These results suggest that in the general population, maternal eating disorders may to some extent influence the child’s diet as early as 6 months after birth; however, the extent to which these differences influence child health and development remain an area for further inquiry. PMID:25453594

Patterns of Physical Activity Among Older Adults in New York City

PubMed Central

Mooney, Stephen J.; Joshi, Spruha; Cerdá, Magdalena; Quinn, James W.; Beard, John R.; Kennedy, Gary J.; Benjamin, Ebele O.; Ompad, Danielle C.; Rundle, Andrew G.

2015-01-01

Introduction Little research to date has explored typologies of physical activity among older adults. An understanding of physical activity patterns may help to both determine the health benefits of different types of activity and target interventions to increase activity levels in older adults. This analysis, conducted in 2014, used a latent class analysis approach to characterize patterns of physical activity in a cohort of older adults. Methods A total of 3,497 men and women aged 65–75 years living in New York City completed the Physical Activity Scale for the Elderly (PASE) in 2011. PASE scale items were used to classify subjects into latent classes. Multinomial regression was then used to relate individual and neighborhood characteristics to class membership. Results Five latent classes were identified: “least active,” “walkers,” “domestic/gardening,” “athletic,” and “domestic/gardening athletic.” Individual-level predictors, including more education, higher income, and better self-reported health, were associated with membership in the more-active classes, particularly the athletic classes. Residential characteristics, including living in single-family housing and living in the lower-density boroughs of New York City, were predictive of membership in one of the domestic/gardening classes. Class membership was associated with BMI even after controlling for total PASE score. Conclusions This study suggests that individual and neighborhood characteristics are associated with distinct physical activity patterns in a group of older urban adults. These patterns are associated with body habitus independent of overall activity. PMID:26091927

Are poker players all the same? Latent class analysis.

PubMed

Dufour, Magali; Brunelle, Natacha; Roy, Élise

2015-06-01

Poker is the gambling game that is currently gaining the most in popularity. However, there is little information on poker players' characteristics and risk factors. Furthermore, the first studies described poker players, often recruited in universities, as an homogeneous group who played in only one of the modes (land based or on the Internet). This study aims to identify, through latent class analyses, poker player subgroups. A convenience sample of 258 adult poker players was recruited across Quebec during special events or through advertising in various media. Participants filled out a series of questionnaires (Canadian Problem Gambling Index, Beck Depression, Beck Anxiety, erroneous belief and alcohol/drug consumption). The latent class analysis suggests that there are three classes of poker players. Class I (recreational poker players) includes those who have the lowest probability of engaging intensively in different game modes. Participants in class II (Internet poker players) all play poker on the Internet. This class includes the highest proportion of players who consider themselves experts or professionals. They make a living in part or in whole from poker. Class III (multiform players) includes participants with the broadest variety of poker patterns. This group is complex: these players are positioned halfway between professional and recreational players. Results indicate that poker players are not an homogeneous group identified simply on the basis of the form of poker played. The specific characteristics associated with each subgroup points to vulnerabilities that could potentially be targeted for preventive interventions.

Latent profile analysis of teacher perceptions of parent contact and comfort.

PubMed

Stormont, Melissa; Herman, Keith C; Reinke, Wendy M; David, Kimberly B; Goel, Nidhi

2013-09-01

The purpose of the study was to explore patterns of parent involvement as perceived by teachers and identify correlates of these patterns. Parent involvement indicators and correlates were selected from a review of existing research. Participants included 34 teachers and 577 children in kindergarten through third grade. The vast majority of the sample was African American (78%), followed by Caucasian (19%) and other ethnic backgrounds (2%). Two subscales from the Parent Involvement-Teacher scale, contact and comfort, were entered as indicators in a latent profile analysis to determine the number and types of parent involvement classes. Contact included the frequency of interactions between parents and teachers; comfort included the quality of their relationship with the parent and how well their goals were aligned. Subsequent latent class regressions were conducted to identify student, school, and family characteristics associated with class membership. Three classes provided the optimal solution. This included two classes of parents with low contact with teachers but different comfort levels; one with low contact and low comfort (11%), and one with low contact but high comfort (71%). The remaining class, representing 18% of parents, was rated high on both contact and comfort. Low income status, family problems, and social, emotional, academic, and self-regulation problems distinguished the low comfort class from the other two classes. It is imperative to help teachers feel more comfortable working with families who may be experiencing substantial stressors and who also have children who need support across school and home settings.

Bayesian learning

NASA Technical Reports Server (NTRS)

Denning, Peter J.

1989-01-01

In 1983 and 1984, the Infrared Astronomical Satellite (IRAS) detected 5,425 stellar objects and measured their infrared spectra. In 1987 a program called AUTOCLASS used Bayesian inference methods to discover the classes present in these data and determine the most probable class of each object, revealing unknown phenomena in astronomy. AUTOCLASS has rekindled the old debate on the suitability of Bayesian methods, which are computationally intensive, interpret probabilities as plausibility measures rather than frequencies, and appear to depend on a subjective assessment of the probability of a hypothesis before the data were collected. Modern statistical methods have, however, recently been shown to also depend on subjective elements. These debates bring into question the whole tradition of scientific objectivity and offer scientists a new way to take responsibility for their findings and conclusions.

Bayesian validation of a serum and milk ELISA for antibodies against Mycobacterium avium subspecies paratuberculosis in Greek dairy goats across lactation.

PubMed

Angelidou, E; Kostoulas, P; Leontides, L

2014-02-01

We validated a commercial (Idexx Pourquier, Montpellier, France) serum and milk indirect ELISA that detects antibodies against Mycobacterium avium ssp. paratuberculosis (MAP) in Greek dairy goats. Each goat was sampled 4 times, starting from kidding and covering early, mid, and late lactation. A total of 1,268 paired milk (or colostrum) and serum samples were collected during the 7-mo lactation period. Bayesian latent class models, which allow for the continuous interpretation of test results, were used to derive the distribution of the serum and milk ELISA response for healthy and MAP-infected individuals at each lactation stage. Both serum and milk ELISA, in all lactation stages, had average and similar overall discriminatory ability as measured by the area under the curve (AUC). For each test, the smallest overlap between the distribution of the healthy and MAP-infected does was in late lactation. At this stage, the AUC was 0.89 (95% credible interval: 0.70; 0.98) and 0.92 (0.74; 0.99) for the milk and serum ELISA, respectively. Both tests had comparable sensitivities and specificities at the recommended cutoffs across lactation. Lowering the cutoffs led to an increase in sensitivity without serious loss in specificity. In conclusion, the milk ELISA was as accurate as the serum ELISA. Therefore, it could serve as the diagnostic tool of choice, especially during the implementation of MAP control programs that require frequent testing, because milk sampling is a noninvasive, rapid, and easy process. Copyright © 2014 American Dairy Science Association. Published by Elsevier Inc. All rights reserved.

Bayesian estimation of the accuracy of the calf respiratory scoring chart and ultrasonography for the diagnosis of bovine respiratory disease in pre-weaned dairy calves.

PubMed

Buczinski, Sébastien; L Ollivett, Terri; Dendukuri, Nandini

2015-05-01

There is currently no gold standard method for the diagnosis of bovine respiratory disease (BRD) complex in Holstein pre-weaned dairy calves. Systematic thoracic ultrasonography (TUS) has been used as a proxy for BRD, but cannot be directly used by producers. The Wisconsin calf respiratory scoring chart (CRSC) is a simpler alternative, but with unknown accuracy. Our objective was to estimate the accuracy of CRSC, while adjusting for the lack of a gold standard. Two cross sectional study populations with a high BRD prevalence (n=106 pre-weaned Holstein calves) and an average BRD prevalence (n=85 pre-weaned Holstein calves) from North America were studied. All calves were simultaneously assessed using CRSC (cutoff used ≥ 5) and TUS (cutoff used ≥ 1cm of lung consolidation). Bayesian latent class models allowing for conditional dependence were used with informative priors for BRD prevalence and TUS accuracy (sensitivity (Se) and specificity (Sp)) and non-informative priors for CRSC accuracies. Robustness of the model was tested by relaxing priors for prevalence or TUS accuracy. The SeCRSC (95% credible interval (CI)) and SpCRSC were 62.4% (47.9-75.8) and 74.1% (64.9-82.8) respectively. The SeTUS was 79.4% (66.4-90.9) and SpTUS was 93.9% (88.0-97.6). The imperfect accuracy of CRSC and TUS should be taken into account when using those tools to assess BRD status. Copyright © 2015 Elsevier B.V. All rights reserved.

Comparing Factor, Class, and Mixture Models of Cannabis Initiation and DSM Cannabis Use Disorder Criteria, Including Craving, in the Brisbane Longitudinal Twin Study

PubMed Central

Kubarych, Thomas S.; Kendler, Kenneth S.; Aggen, Steven H.; Estabrook, Ryne; Edwards, Alexis C.; Clark, Shaunna L.; Martin, Nicholas G.; Hickie, Ian B.; Neale, Michael C.; Gillespie, Nathan A.

2014-01-01

Accumulating evidence suggests that the Diagnostic and Statistical Manual of Mental Disorders (DSM) diagnostic criteria for cannabis abuse and dependence are best represented by a single underlying factor. However, it remains possible that models with additional factors, or latent class models or hybrid models, may better explain the data. Using structured interviews, 626 adult male and female twins provided complete data on symptoms of cannabis abuse and dependence, plus a craving criterion. We compared latent factor analysis, latent class analysis, and factor mixture modeling using normal theory marginal maximum likelihood for ordinal data. Our aim was to derive a parsimonious, best-fitting cannabis use disorder (CUD) phenotype based on DSM-IV criteria and determine whether DSM-5 craving loads onto a general factor. When compared with latent class and mixture models, factor models provided a better fit to the data. When conditioned on initiation and cannabis use, the association between criteria for abuse, dependence, withdrawal, and craving were best explained by two correlated latent factors for males and females: a general risk factor to CUD and a factor capturing the symptoms of social and occupational impairment as a consequence of frequent use. Secondary analyses revealed a modest increase in the prevalence of DSM-5 CUD compared with DSM-IV cannabis abuse or dependence. It is concluded that, in addition to a general factor with loadings on cannabis use and symptoms of abuse, dependence, withdrawal, and craving, a second clinically relevant factor defined by features of social and occupational impairment was also found for frequent cannabis use. PMID:24588857

Patterns of perceived barriers to medical care in older adults: a latent class analysis.

PubMed

Thorpe, Joshua M; Thorpe, Carolyn T; Kennelty, Korey A; Pandhi, Nancy

2011-08-03

This study examined multiple dimensions of healthcare access in order to develop a typology of perceived barriers to healthcare access in community-dwelling elderly. Secondary aims were to define distinct classes of older adults with similar perceived healthcare access barriers and to examine predictors of class membership to identify risk factors for poor healthcare access. A sample of 5,465 community-dwelling elderly was drawn from the 2004 wave of the Wisconsin Longitudinal Study. Perceived barriers to healthcare access were measured using items from the Group Health Association of America Consumer Satisfaction Survey. We used latent class analysis to assess the constellation of items measuring perceived barriers in access and multinomial logistic regression to estimate how risk factors affected the probability of membership in the latent barrier classes. Latent class analysis identified four classes of older adults. Class 1 (75% of sample) consisted of individuals with an overall low level of risk for perceived access problems (No Barriers). Class 2 (5%) perceived problems with the availability/accessibility of healthcare providers such as specialists or mental health providers (Availability/Accessibility Barriers). Class 3 (18%) perceived problems with how well their providers' operations arise organized to accommodate their needs and preferences (Accommodation Barriers). Class 4 (2%) perceived problems with all dimension of access (Severe Barriers). Results also revealed that healthcare affordability is a problem shared by members of all three barrier groups, suggesting that older adults with perceived barriers tend to face multiple, co-occurring problems. Compared to those classified into the No Barriers group, those in the Severe Barrier class were more likely to live in a rural county, have no health insurance, have depressive symptomatology, and speech limitations. Those classified into the Availability/Accessibility Barriers group were more likely to live in rural and micropolitan counties, have depressive symptomatology, more chronic conditions, and hearing limitations. Those in the Accommodation group were more likely to have depressive symptomatology and cognitive limitations. The current study identified a typology of perceived barriers in healthcare access in older adults. The identified risk factors for membership in perceived barrier classes could potentially assist healthcare organizations and providers with targeting polices and interventions designed to improve access in their most vulnerable older adult populations, particularly those in rural areas, with functional disabilities, or in poor mental health.

Latent Classes of Symptoms related to Clinically Depressed Mood in Adolescents.

PubMed

Blom, Eva Henje; Forsman, Mats; Yang, Tony T; Serlachius, Eva; Larsson, Jan-Olov

2014-01-01

The diagnosis of major depressive disorder (MDD), according to the Diagnostic and Statistical Manual of Mental Disorders , is based only on adult symptomatology of depression and not adapted for age and gender. This may contribute to the low diagnostic specificity and validity of adolescent MDD. In this study, we investigated whether latent classes based on symptoms associated with depressed mood could be identified in a sample of adolescents seeking psychiatric care, regardless of traditionally defined diagnostic categories. Self-reports of the Strengths and Difficulties Questionnaire and the Development and Well-Being Assessment were collected consecutively from all new patients between the ages of 13 and 17 years at two psychiatric outpatient clinics in Stockholm, Sweden. Those who reported depressed mood at intake yielded a sample of 21 boys and 156 girls. Latent class analyses were performed for all screening items and for the depression-specific items of the Development and Well-Being Assessment. The symptoms that were reported in association with depressed mood differentiated the adolescents into two classes. One class had moderate emotional severity scores on the Strengths and Difficulties Questionnaire and mainly symptoms that were congruent with the Diagnostic and Statistical Manual of Mental Disorders criteria for MDD. The other class had higher emotional severity scores and similar symptoms to those reported in the first class. However, in addition, this group demonstrated more diverse symptomatology, including vegetative symptoms, suicidal ideation, anxiety, conduct problems, body dysmorphic symptoms, and deliberate vomiting. The classes predicted functional impairment in that the members of the second class showed more functional impairment. The relatively small sample size limited the generalizability of the results of this study, and the amount of items included in the analysis was restricted by the rules of latent class analysis. No conclusions about gender differences between the classes could be could be drawn as a result of the low number of boys included in the study. Two distinct classes were identified among adolescents with depressed mood. The class with highest emotional symptom severity score and the most functional impairment had a more diverse symptomatology that included symptoms that were not congruent with the traditional diagnostic criteria of MDD. However, this additional symptomatology is clinically important to consider. As a result, the clinical usefulness of the Diagnostic and Statistical Manual of Mental Disorders during the diagnostic process of adolescent depression is questioned.

Gradient of association between parenting styles and patterns of drug use in adolescence: A latent class analysis.

PubMed

Valente, Juliana Y; Cogo-Moreira, Hugo; Sanchez, Zila M

2017-11-01

To identify different patterns of drug use in adolescence and determine if these are associated with parenting styles and other sociodemographic factors. A latent class analysis was conducted using baseline data collected in a sample (n=6381) from a randomized controlled trial conducted to evaluate the effectiveness of the #Tamojunto drug-use prevention program, carried out with 7th- and 8th-grade public school students in six Brazilian cities. Three latent classes were identified among the students: 1) abstainers/low users (81.54%), 2) alcohol users/binge drinkers (16.65%), and 3) polydrug users (1.80%). A gradient of inverse association was found between parenting styles (authoritative, authoritarian, and indulgent, with the neglectful style as a reference point) and the classes "alcohol users/binge drinkers" (aOR=0.36, 95%CI=0.27-0.47; aOR=0.56, 95%CI=0.43-0.72; and aOR=0.64, 95%CI=0.51-0.80, respectively) and "polydrug users" (aOR=0.09, 95%CI=0.03-0.24; aOR=0.23, 95%CI=0.11-0.52; and aOR=0.24, 95%CI=0.08-0.74, respectively). Associations were also revealed between the latent classes and the adolescent's age and socioeconomic status. The results suggest that activities to develop parenting skills should be included in school programs aimed at the prevention of drug use among adolescents in order to reduce neglectful practices and thereby possibly reduce drug use among the children. Copyright © 2017. Published by Elsevier B.V.

Illustration of Step-Wise Latent Class Modeling With Covariates and Taxometric Analysis in Research Probing Children's Mental Models in Learning Sciences

PubMed Central

Stamovlasis, Dimitrios; Papageorgiou, George; Tsitsipis, Georgios; Tsikalas, Themistoklis; Vaiopoulou, Julie

2018-01-01

This paper illustrates two psychometric methods, latent class analysis (LCA) and taxometric analysis (TA) using empirical data from research probing children's mental representation in science learning. LCA is used to obtain a typology based on observed variables and to further investigate how the encountered classes might be related to external variables, where the effectiveness of classification process and the unbiased estimations of parameters become the main concern. In the step-wise LCA, the class membership is assigned and subsequently its relationship with covariates is established. This leading-edge modeling approach suffers from severe downward-biased estimations. The illustration of LCA is focused on alternative bias correction approaches and demonstrates the effect of modal and proportional class-membership assignment along with BCH and ML correction procedures. The illustration of LCA is presented with three covariates, which are psychometric variables operationalizing formal reasoning, divergent thinking and field dependence-independence, respectively. Moreover, taxometric analysis, a method designed to detect the type of the latent structural model, categorical or dimensional, is introduced, along with the relevant basic concepts and tools. TA was applied complementarily in the same data sets to answer the fundamental hypothesis about children's naïve knowledge on the matters under study and it comprises an additional asset in building theory which is fundamental for educational practices. Taxometric analysis provided results that were ambiguous as far as the type of the latent structure. This finding initiates further discussion and sets a problematization within this framework rethinking fundamental assumptions and epistemological issues. PMID:29713300

Patterns of Adolescent Bullying Behaviors: Physical, Verbal, Exclusion, Rumor, and Cyber

PubMed Central

Wang, Jing; Iannotti, Ronald J.; Luk, Jeremy W.

2012-01-01

Patterns of engagement in cyber bullying and four types of traditional bullying were examined using latent class analysis (LCA). Demographic differences and externalizing problems were evaluated across latent class membership. Data were obtained from the 2005–2006 Health Behavior in School-aged Survey and the analytic sample included 7,508 U.S. adolescents in grades 6 through 10. LCA models were tested on physical bullying, verbal bullying, social exclusion, spreading rumors, and cyber bullying behaviors. Three latent classes were identified for each gender: All-Types Bullies (10.5% for boys and 4.0% for girls), Verbal/Social Bullies (29.3% for boys and 29.4% for girls), and a Non-Involved class (60.2% for boys and 66.6% for girls). Boys were more likely to be All-Types Bullies than girls. The prevalence rates of All-Types and Verbal/Social Bullies peaked during grades 6 to 8 and grades 7 & 8, respectively. Pairwise comparisons across the three latent classes on externalizing problems were conducted. Overall, the All-Types Bullies were at highest risk of using substances and carrying weapons, the Non-Involved were at lowest risk, and the Verbal/Social Bullies were in the middle. Results also suggest that most cyber bullies belong to a group of highly aggressive adolescents who conduct all types of bullying. This finding does not only improve our understanding of the relation between cyber bullying and traditional bullying, but it also suggests that prevention and intervention efforts could target cyber bullies as a high-risk group for elevated externalizing problems. PMID:22710019

Gender Differences in Anxiety Trajectories from Middle to Late Adolescence

PubMed Central

Ohannessian, Christine McCauley; Milan, Stephanie; Vannucci, Anna

2016-01-01

Although developmental trajectories of anxiety symptomatology have begun to be explored, most research has focused on total anxiety symptom scores during childhood and early adolescence, using racially/ethnically homogenous samples. Understanding the heterogeneous courses of anxiety disorder symptoms during middle to late adolescence has the potential to clarify developmental risk models of anxiety and to inform prevention programs. Therefore, this study specifically examined gender differences in developmental trajectories of anxiety disorder symptoms (generalized anxiety disorder, panic disorder, and social anxiety disorder) from middle to late adolescence in a diverse community sample (N=1,000; 57% female; 65% White), assessed annually over two years. Latent growth curve modeling revealed that girls exhibited a slight linear decrease in generalized anxiety disorder, panic disorder, and social anxiety disorder symptoms, whereas boys exhibited a stable course. These models suggested that one trajectory was appropriate for panic disorder symptoms in both girls and boys. Growth mixture models indicated the presence of four latent generalized anxiety disorder symptom trajectory classes: low increasing, moderate decreasing slightly, high decreasing, and very high decreasing rapidly. Growth mixture models also suggested the presence of five latent social anxiety disorder symptom trajectory classes: a low stable trajectory class and four classes that were qualitatively similar to the latent generalized anxiety disorder trajectories. For both generalized anxiety disorder and social anxiety disorder symptoms, girls were significantly more likely than boys to be in trajectory classes characterized by moderate or high initial symptoms that subsequently decreased over time. These findings provide novel information regarding the developmental course of anxiety disorder symptoms in adolescents. PMID:27889856

The multicategory case of the sequential Bayesian pixel selection and estimation procedure

NASA Technical Reports Server (NTRS)

Pore, M. D.; Dennis, T. B. (Principal Investigator)

1980-01-01

A Bayesian technique for stratified proportion estimation and a sampling based on minimizing the mean squared error of this estimator were developed and tested on LANDSAT multispectral scanner data using the beta density function to model the prior distribution in the two-class case. An extention of this procedure to the k-class case is considered. A generalization of the beta function is shown to be a density function for the general case which allows the procedure to be extended.

Patterns of Adolescent Sexual Behavior Predicting Young Adult Sexually Transmitted Infections: A Latent Class Analysis Approach

PubMed Central

Vasilenko, Sara A.; Kugler, Kari C.; Butera, Nicole M.; Lanza, Stephanie T.

2014-01-01

Adolescent sexual behavior is multidimensional, yet most studies of the topic use variable-oriented methods that reduce behaviors to a single dimension. In this study, we used a person-oriented approach to model adolescent sexual behavior comprehensively, using data from the National Longitudinal Study of Adolescent Health. We identified five latent classes of adolescent sexual behavior: Abstinent (39%), Oral Sex (10%), Low-Risk (25%), Multi-Partner Normative (12%), and Multi-Partner Early (13%). Membership in riskier classes of sexual behavior was predicted by substance use and depressive symptoms. Class membership was also associated with young adult STI outcomes although these associations differed by gender. Male adolescents' STI rates increased with membership in classes with more risky behaviors whereas females' rates were consistent among all sexually active classes. These findings demonstrate the advantages of examining adolescent sexuality in a way that emphasizes its complexity. PMID:24449152

Qualitative and quantitative aspects of information processing in first psychosis: latent class analyses in patients, at-risk subjects, and controls.

PubMed

van Tricht, Mirjam J; Bour, Lo J; Koelman, Johannes H T M; Derks, Eske M; Braff, David L; de Wilde, Odette M; Boerée, Thijs; Linszen, Don H; de Haan, Lieuwe; Nieman, Dorien H

2015-04-01

We aimed to determine profiles of information processing deficits in the pathway to first psychosis. Sixty-one subjects at ultrahigh risk (UHR) for psychosis were assessed, of whom 18 converted to a first episode of psychosis (FEP) within the follow-up period. Additionally, 47 FEP and 30 control subjects were included. Using 10 neurophysiological parameters associated with information processing, latent class analyses yielded three classes at baseline. Class membership was related to group status. Within the UHR sample, two classes were found. Transition to psychosis was nominally associated with class membership. Neurophysiological profiles were unstable over time, but associations between specific neurophysiological components at baseline and follow-up were found. We conclude that certain constellations of neurophysiological variables aid in the differentiation between controls and patients in the prodrome and after first psychosis. Copyright © 2014 Society for Psychophysiological Research.

Patterns of adolescent sexual behavior predicting young adult sexually transmitted infections: a latent class analysis approach.

PubMed

Vasilenko, Sara A; Kugler, Kari C; Butera, Nicole M; Lanza, Stephanie T

2015-04-01

Adolescent sexual behavior is multidimensional, yet most studies of the topic use variable-oriented methods that reduce behaviors to a single dimension. In this study, we used a person-oriented approach to model adolescent sexual behavior comprehensively, using data from the National Longitudinal Study of Adolescent Health. We identified five latent classes of adolescent sexual behavior: Abstinent (39%), Oral Sex (10%), Low-Risk (25%), Multi-Partner Normative (12%), and Multi-Partner Early (13%). Membership in riskier classes of sexual behavior was predicted by substance use and depressive symptoms. Class membership was also associated with young adult STI outcomes although these associations differed by gender. Male adolescents' STI rates increased with membership in classes with more risky behaviors whereas females' rates were consistent among all sexually active classes. These findings demonstrate the advantages of examining adolescent sexuality in a way that emphasizes its complexity.

Estimation of sensitivity and specificity of pregnancy diagnosis using transrectal ultrasonography and ELISA for pregnancy-associated glycoprotein in dairy cows using a Bayesian latent class model.

PubMed

Shephard, R W; Morton, J M

2018-01-01

To determine the sensitivity (Se) and specificity (Sp) of pregnancy diagnosis using transrectal ultrasonography and an ELISA for pregnancy-associated glycoprotein (PAG) in milk, in lactating dairy cows in seasonally calving herds approximately 85-100 days after the start of the herd's breeding period. Paired results were used from pregnancy diagnosis using transrectal ultrasonography and ELISA for PAG in milk carried out approximately 85 and 100 days after the start of the breeding period, respectively, from 879 cows from four herds in Victoria, Australia. A Bayesian latent class model was used to estimate the proportion of cows pregnant, the Se and Sp of each test, and covariances between test results in pregnant and non-pregnant cows. Prior probability estimates were defined using beta distributions for the expected proportion of cows pregnant, Se and Sp for each test, and covariances between tests. Markov Chain Monte Carlo iterations identified posterior distributions for each of the unknown variables. Posterior distributions for each parameter were described using medians and 95% probability (i.e. credible) intervals (PrI). The posterior median estimates for Se and Sp for each test were used to estimate positive predictive and negative predictive values across a range of pregnancy proportions. The estimate for proportion pregnant was 0.524 (95% PrI = 0.485-0.562). For pregnancy diagnosis using transrectal ultrasonography, Se and Sp were 0.939 (95% PrI = 0.890-0.974) and 0.943 (95% PrI = 0.885-0.984), respectively; for ELISA, Se and Sp were 0.963 (95% PrI = 0.919-0.990) and 0.870 (95% PrI = 0.806-0.931), respectively. The estimated covariance between test results was 0.033 (95% PrI = 0.008-0.046) and 0.035 (95% PrI = 0.018-0.078) for pregnant and non-pregnant cows, respectively. Pregnancy diagnosis results using transrectal ultrasonography had a higher positive predictive value but lower negative predictive value than results from the ELISA across the range of pregnancy proportions assessed. Pregnancy diagnosis using transrectal ultrasonography and ELISA for PAG in milk had similar Se but differed in predictive values. Pregnancy diagnosis in seasonally calving herds around 85-100 days after the start of the breeding period using the ELISA is expected to result in a higher negative predictive value but lower positive predictive value than pregnancy diagnosis using transrectal ultrasonography. Thus, with the ELISA, a higher proportion of the cows with negative results will be non-pregnant, relative to results from transrectal ultrasonography, but a lower proportion of cows with positive results will be pregnant.

PTSD symptom severity and psychiatric comorbidity in recent motor vehicle accident victims: a latent class analysis.

PubMed

Hruska, Bryce; Irish, Leah A; Pacella, Maria L; Sledjeski, Eve M; Delahanty, Douglas L

2014-10-01

We conducted a latent class analysis (LCA) on 249 recent motor vehicle accident (MVA) victims to examine subgroups that differed in posttraumatic stress disorder (PTSD) symptom severity, current major depressive disorder and alcohol/other drug use disorders (MDD/AoDs), gender, and interpersonal trauma history 6-weeks post-MVA. A 4-class model best fit the data with a resilient class displaying asymptomatic PTSD symptom levels/low levels of comorbid disorders; a mild psychopathology class displaying mild PTSD symptom severity and current MDD; a moderate psychopathology class displaying severe PTSD symptom severity and current MDD/AoDs; and a severe psychopathology class displaying extreme PTSD symptom severity and current MDD. Classes also differed with respect to gender composition and history of interpersonal trauma experience. These findings may aid in the development of targeted interventions for recent MVA victims through the identification of subgroups distinguished by different patterns of psychiatric problems experienced 6-weeks post-MVA. Copyright © 2014 Elsevier Ltd. All rights reserved.

PTSD Symptom Severity and Psychiatric Comorbidity in Recent Motor Vehicle Accident Victims: A Latent Class Analysis

PubMed Central

Hruska, Bryce; Irish, Leah A.; Pacella, Maria L.; Sledjeski, Eve M.; Delahanty, Douglas L.

2014-01-01

We conducted a latent class analysis (LCA) on 249 recent motor vehicle accident (MVA) victims to examine subgroups that differed in posttraumatic stress disorder (PTSD) symptom severity, current major depressive disorder and alcohol/other drug use disorders (MDD/AoDs), gender, and interpersonal trauma history 6-weeks post-MVA. A 4-class model best fit the data with a resilient class displaying asymptomatic PTSD symptom levels/low levels of comorbid disorders; a mild psychopathology class displaying mild PTSD symptom severity and current MDD; a moderate psychopathology class displaying severe PTSD symptom severity and current MDD/AoDs; and a severe psychopathology class displaying extreme PTSD symptom severity and current MDD. Classes also differed with respect to gender composition and history of interpersonal trauma experience. These findings may aid in the development of targeted interventions for recent MVA victims through the identification of subgroups distinguished by different patterns of psychiatric problems experienced 6-weeks post-MVA. PMID:25124501

Impact of an equality constraint on the class-specific residual variances in regression mixtures: A Monte Carlo simulation study

PubMed Central

Kim, Minjung; Lamont, Andrea E.; Jaki, Thomas; Feaster, Daniel; Howe, George; Van Horn, M. Lee

2015-01-01

Regression mixture models are a novel approach for modeling heterogeneous effects of predictors on an outcome. In the model building process residual variances are often disregarded and simplifying assumptions made without thorough examination of the consequences. This simulation study investigated the impact of an equality constraint on the residual variances across latent classes. We examine the consequence of constraining the residual variances on class enumeration (finding the true number of latent classes) and parameter estimates under a number of different simulation conditions meant to reflect the type of heterogeneity likely to exist in applied analyses. Results showed that bias in class enumeration increased as the difference in residual variances between the classes increased. Also, an inappropriate equality constraint on the residual variances greatly impacted estimated class sizes and showed the potential to greatly impact parameter estimates in each class. Results suggest that it is important to make assumptions about residual variances with care and to carefully report what assumptions were made. PMID:26139512

Identifying classes of persons with mild intellectual disability or borderline intellectual functioning: a latent class analysis.

PubMed

Nouwens, Peter J G; Lucas, Rosanne; Smulders, Nienke B M; Embregts, Petri J C M; van Nieuwenhuizen, Chijs

2017-07-17

Persons with mild intellectual disability or borderline intellectual functioning are often studied as a single group with similar characteristics. However, there are indications that differences exist within this population. Therefore, the aim of this study was to identify classes of persons with mild intellectual disability or borderline intellectual functioning and to examine whether these classes are related to individual and/or environmental characteristics. Latent class analysis was performed using file data of 250 eligible participants with a mean age of 26.1 (SD 13.8, range 3-70) years. Five distinct classes of persons with mild intellectual disability or borderline intellectual functioning were found. These classes significantly differed in individual and environmental characteristics. For example, persons with a mild intellectual disability experienced fewer problems than those with borderline intellectual disability. The identification of five classes implies that a differentiated approach is required towards persons with mild intellectual disability or borderline intellectual functioning.

Symptom Cluster Research With Biomarkers and Genetics Using Latent Class Analysis.

PubMed

Conley, Samantha

2017-12-01

The purpose of this article is to provide an overview of latent class analysis (LCA) and examples from symptom cluster research that includes biomarkers and genetics. A review of LCA with genetics and biomarkers was conducted using Medline, Embase, PubMed, and Google Scholar. LCA is a robust latent variable model used to cluster categorical data and allows for the determination of empirically determined symptom clusters. Researchers should consider using LCA to link empirically determined symptom clusters to biomarkers and genetics to better understand the underlying etiology of symptom clusters. The full potential of LCA in symptom cluster research has not yet been realized because it has been used in limited populations, and researchers have explored limited biologic pathways.

Instructor and peer bullying in college students: Distinct typologies based on Latent Class Analysis.

PubMed

Marraccini, Marisa E; Brick, Leslie Ann D; Weyandt, Lisa L

2018-03-22

Although bullying is traditionally considered within the context of primary and secondary school, recent evidence suggests that bullying continues into college and workplace settings. Participants/Method: Latent class analysis (LCA) was employed to classify college bullying involvement typologies among 325 college students attending a northeastern university. Four classes concerning bullying involvement were revealed: Non-involved (36%); Instructor victim (30%); Peer bully-victim (22%); and Peer bully-victim/ Instructor victim (12%). Findings from this study, which classified college bullying experiences by incorporating both peer and instructor (teacher and professor) bullying, add substantially to the literature by providing insight into patterns of relatively unexplored bullying behaviors.

Do marginalized neighbourhoods have less healthy retail food environments? An analysis using Bayesian spatial latent factor and hurdle models.

PubMed

Luan, Hui; Minaker, Leia M; Law, Jane

2016-08-22

Findings of whether marginalized neighbourhoods have less healthy retail food environments (RFE) are mixed across countries, in part because inconsistent approaches have been used to characterize RFE 'healthfulness' and marginalization, and researchers have used non-spatial statistical methods to respond to this ultimately spatial issue. This study uses in-store features to categorize healthy and less healthy food outlets. Bayesian spatial hierarchical models are applied to explore the association between marginalization dimensions and RFE healthfulness (i.e., relative healthy food access that modelled via a probability distribution) at various geographical scales. Marginalization dimensions are derived from a spatial latent factor model. Zero-inflation occurring at the walkable-distance scale is accounted for with a spatial hurdle model. Neighbourhoods with higher residential instability, material deprivation, and population density are more likely to have access to healthy food outlets within a walkable distance from a binary 'have' or 'not have' access perspective. At the walkable distance scale however, materially deprived neighbourhoods are found to have less healthy RFE (lower relative healthy food access). Food intervention programs should be developed for striking the balance between healthy and less healthy food access in the study region as well as improving opportunities for residents to buy and consume foods consistent with dietary recommendations.

Discriminative Relational Topic Models.

PubMed

Chen, Ning; Zhu, Jun; Xia, Fei; Zhang, Bo

2015-05-01

Relational topic models (RTMs) provide a probabilistic generative process to describe both the link structure and document contents for document networks, and they have shown promise on predicting network structures and discovering latent topic representations. However, existing RTMs have limitations in both the restricted model expressiveness and incapability of dealing with imbalanced network data. To expand the scope and improve the inference accuracy of RTMs, this paper presents three extensions: 1) unlike the common link likelihood with a diagonal weight matrix that allows the-same-topic interactions only, we generalize it to use a full weight matrix that captures all pairwise topic interactions and is applicable to asymmetric networks; 2) instead of doing standard Bayesian inference, we perform regularized Bayesian inference (RegBayes) with a regularization parameter to deal with the imbalanced link structure issue in real networks and improve the discriminative ability of learned latent representations; and 3) instead of doing variational approximation with strict mean-field assumptions, we present collapsed Gibbs sampling algorithms for the generalized relational topic models by exploring data augmentation without making restricting assumptions. Under the generic RegBayes framework, we carefully investigate two popular discriminative loss functions, namely, the logistic log-loss and the max-margin hinge loss. Experimental results on several real network datasets demonstrate the significance of these extensions on improving prediction performance.

Sex-related and non-sex-related comorbidity subtypes of tic disorders: a latent class approach.

PubMed

Rodgers, S; Müller, M; Kawohl, W; Knöpfli, D; Rössler, W; Castelao, E; Preisig, M; Ajdacic-Gross, V

2014-05-01

Recent evidence suggests that there may be more than one Gilles de la Tourette syndrome (GTS)/tic disorder phenotype. However, little is known about the common patterns of these GTS/tic disorder-related comorbidities. In addition, sex-specific phenomenological data of GTS/tic disorder-affected adults are rare. Therefore, this community-based study used latent class analyses (LCA) to investigate sex-related and non-sex-related subtypes of GTS/tic disorders and their most common comorbidities. The data were drawn from the PsyCoLaus study (n = 3691), a population-based survey conducted in Lausanne, Switzerland. LCA were performed on the data of 80 subjects manifesting motor/vocal tics during their childhood/adolescence. Comorbid attention-deficit hyperactivity disorder (ADHD), obsessive-compulsive disorder, depressive, phobia and panic symptoms/syndromes comprised the selected indicators. The resultant classes were characterized by psychosocial correlates. In LCA, four latent classes provided the best fit to the data. We identified two male-related classes. The first class exhibited both ADHD and depression. The second class comprised males with only depression. Class three was a female-related class depicting obsessive thoughts/compulsive acts, phobias and panic attacks. This class manifested high psychosocial impairment. Class four had a balanced sex proportion and comorbid symptoms/syndromes such as phobias and panic attacks. The complementary occurrence of comorbid obsessive thoughts/compulsive acts and ADHD impulsivity was remarkable. To the best of our knowledge, this is the first study applying LCA to community data of GTS symptoms/tic disorder-affected persons. Our findings support the utility of differentiating GTS/tic disorder subphenotypes on the basis of comorbid syndromes. © 2013 The Author(s) European Journal of Neurology © 2013 EFNS.

Examining the Heterogeneity and Cost Effectiveness of a Complex Intervention by Segmentation of Patients with Chronic Obstructive Pulmonary Disease.

PubMed

Sørensen, Sabrina Storgaard; Jensen, Morten Berg; Pedersen, Kjeld Møller; Ehlers, Lars

2018-02-01

To examine the heterogeneity in cost-effectiveness analyses of patient-tailored complex interventions. Latent class analysis (LCA) was performed on data from a randomized controlled trial evaluating a patient-tailored case management strategy for patients suffering from chronic obstructive pulmonary disease (COPD). LCA was conducted on detailed process variables representing service variation in the intervention group. Features of the identified latent classes were compared for consistency with baseline demographic, clinical, and economic characteristics for each class. Classes for the control group, corresponding to the identified latent classes for the intervention group, were identified using multinomial logistic regression. Cost-utility analyses were then conducted at the class level, and uncertainty surrounding the point estimates was assessed by probabilistic sensitivity analysis. The LCA identified three distinct classes: the psychologically care class, the extensive COPD care class, and the limited COPD care class. Patient baseline characteristics were in line with the features identified in the LCA. Evaluation of cost-effectiveness revealed highly disparate results, and case management for only the extensive COPD care class appeared cost-effective with an incremental cost-effectiveness ratio of £26,986 per quality-adjusted life-year gained using the threshold value set by the National Institute of Health and Care Excellence. Findings indicate that researchers evaluating patient-tailored complex interventions need to address both supply-side variation and demand-side heterogeneity to link findings with outcome. The article specifically proposes the use of LCA because it is believed to have the potential to enable more appropriate targeting of complex care strategies. Copyright © 2018 International Society for Pharmacoeconomics and Outcomes Research (ISPOR). Published by Elsevier Inc. All rights reserved.

Who Should Be Targeted for the Prevention of Birth Defects? A Latent Class Analysis Based on a Large, Population-Based, Cross-Sectional Study in Shaanxi Province, Western China.

PubMed

Zhu, Zhonghai; Cheng, Yue; Yang, Wenfang; Li, Danyang; Yang, Xue; Liu, Danli; Zhang, Min; Yan, Hong; Zeng, Lingxia

2016-01-01

The wide range and complex combinations of factors that cause birth defects impede the development of primary prevention strategies targeted at high-risk subpopulations. Latent class analysis (LCA) was conducted to identify mutually exclusive profiles of factors associated with birth defects among women between 15 and 49 years of age using data from a large, population-based, cross-sectional study conducted in Shaanxi Province, western China, between August and October, 2013. The odds ratios (ORs) and 95% confidence intervals (CIs) of associated factors and the latent profiles of indicators of birth defects and congenital heart defects were computed using a logistic regression model. Five discrete subpopulations of participants were identified as follows: No folic acid supplementation in the periconceptional period (reference class, 21.37%); low maternal education level + unhealthy lifestyle (class 2, 39.75%); low maternal education level + unhealthy lifestyle + disease (class 3, 23.71%); unhealthy maternal lifestyle + advanced age (class 4, 4.71%); and multi-risk factor exposure (class 5, 10.45%). Compared with the reference subgroup, the other subgroups consistently had a significantly increased risk of birth defects (ORs and 95% CIs: class 2, 1.75 and 1.21-2.54; class 3, 3.13 and 2.17-4.52; class 4, 5.02 and 3.20-7.88; and class 5, 12.25 and 8.61-17.42, respectively). For congenital heart defects, the ORs and 95% CIs were all higher, and the magnitude of OR differences ranged from 1.59 to 16.15. A comprehensive intervention strategy targeting maternal exposure to multiple risk factors is expected to show the strongest results in preventing birth defects.

Latent profile analysis of sixth graders based on teacher ratings: Association with school dropout.

PubMed

Orpinas, Pamela; Raczynski, Katherine; Peters, Jaclyn Wetherington; Colman, Laura; Bandalos, Deborah

2015-12-01

The goal of this study was to identify meaningful groups of sixth graders with common characteristics based on teacher ratings of assets and maladaptive behaviors, describe dropout rates for each group, and examine the validity of these groups using students' self-reports. The sample consisted of racially diverse students (n = 675) attending sixth grade in public schools in Northeast Georgia. The majority of the sample was randomly selected; a smaller group was identified by teachers as high risk for aggression. Based on teacher ratings of externalizing behaviors, internalizing problems, academic skills, leadership, and social assets, latent profile analysis yielded 7 classes that can be displayed along a continuum: Well-Adapted, Average, Average-Social Skills Deficit, Internalizing, Externalizing, Disruptive Behavior with School Problems, and Severe Problems. Dropout rate was lowest for the Well-adapted class (4%) and highest for the Severe Problems class (58%). However, students in the Average-Social Skills Deficit class did not follow the continuum, with a large proportion of students who abandoned high school (29%). The proportion of students identified by teachers as high in aggression consistently increased across the continuum from none in the Well-Adapted class to 84% in the Severe Problems class. Students' self-reports were generally consistent with the latent profile classes. Students in the Well-Adapted class reported low aggression, drug use, and delinquency, and high life satisfaction; self-reports went in the opposite direction for the Disruptive Behaviors with School Problems class. Results highlight the importance of early interventions to improve academic performance, reduce externalizing behaviors, and enhance social assets. (c) 2015 APA, all rights reserved).

Latent Structure Agreement Analysis

DTIC Science & Technology

1989-11-01

correct for bias in estimation of disease prevalence due to misclassification error [39]. Software Varying panel latent class agreement models can be...D., and L. M. Irwig, "Estimation of Test Error Rates, Disease Prevalence and Relative Risk from Misclassified Data: A Review," Journal of Clinical

Further insights on the French WISC-IV factor structure through Bayesian structural equation modeling.

PubMed

Golay, Philippe; Reverte, Isabelle; Rossier, Jérôme; Favez, Nicolas; Lecerf, Thierry

2013-06-01

The interpretation of the Wechsler Intelligence Scale for Children--Fourth Edition (WISC-IV) is based on a 4-factor model, which is only partially compatible with the mainstream Cattell-Horn-Carroll (CHC) model of intelligence measurement. The structure of cognitive batteries is frequently analyzed via exploratory factor analysis and/or confirmatory factor analysis. With classical confirmatory factor analysis, almost all cross-loadings between latent variables and measures are fixed to zero in order to allow the model to be identified. However, inappropriate zero cross-loadings can contribute to poor model fit, distorted factors, and biased factor correlations; most important, they do not necessarily faithfully reflect theory. To deal with these methodological and theoretical limitations, we used a new statistical approach, Bayesian structural equation modeling (BSEM), among a sample of 249 French-speaking Swiss children (8-12 years). With BSEM, zero-fixed cross-loadings between latent variables and measures are replaced by approximate zeros, based on informative, small-variance priors. Results indicated that a direct hierarchical CHC-based model with 5 factors plus a general intelligence factor better represented the structure of the WISC-IV than did the 4-factor structure and the higher order models. Because a direct hierarchical CHC model was more adequate, it was concluded that the general factor should be considered as a breadth rather than a superordinate factor. Because it was possible for us to estimate the influence of each of the latent variables on the 15 subtest scores, BSEM allowed improvement of the understanding of the structure of intelligence tests and the clinical interpretation of the subtest scores. PsycINFO Database Record (c) 2013 APA, all rights reserved.

Substance use predictors of victimization profiles among homeless youth: a latent class analysis.

PubMed

Bender, Kimberly; Thompson, Sanna; Ferguson, Kristin; Langenderfer, Lisa

2014-02-01

Although a substantial body of literature demonstrates high prevalence of street victimization among homeless youth, few studies have investigated the existence of victimization classes that differ on the type and frequency of victimization experienced. Nor do we know how substance use patterns relate to victimization classes. Using latent class analysis (LCA), we examined the existence of victimization classes of homeless youth and investigated substance use predictors of class membership utilizing a large purposive sample (N=601) recruited from homeless youth-serving host agencies in three disparate regions of the U.S. Results of the LCA suggest the presence of three distinct victimization profiles - youth fit into a low-victimization class, a witness class, or a high-victimization class. These three victimization classes demonstrated differences in their substance use, including rates of substance abuse/dependence on alcohol and/or drugs. The presence of distinct victimization profiles suggests the need for screening and referral for differential services. Copyright © 2013 The Foundation for Professionals in Services for Adolescents. Published by Elsevier Ltd. All rights reserved.

Latent profile analyses of posttraumatic stress disorder, depression and generalized anxiety disorder symptoms in trauma-exposed soldiers.

PubMed

Contractor, Ateka A; Elhai, Jon D; Fine, Thomas H; Tamburrino, Marijo B; Cohen, Gregory; Shirley, Edwin; Chan, Philip K; Liberzon, Israel; Galea, Sandro; Calabrese, Joseph R

2015-09-01

Posttraumatic stress disorder (PTSD) is comorbid with major depressive disorder (MDD; Kessler et al., 1995) and generalized anxiety disorder (GAD; Brown et al., 2001). We aimed to (1) assess discrete patterns of post-trauma PTSD-depression-GAD symptoms using latent profile analyses (LPAs), and (2) assess covariates (gender, income, education, age) in defining the best fitting class solution. The PTSD Checklist (assessing PTSD symptoms), GAD-7 scale (assessing GAD symptoms), and Patient Health Questionnaire-9 (assessing depression) were administered to 1266 trauma-exposed Ohio National Guard soldiers. Results indicated three discrete subgroups based on symptom patterns with mild (class 1), moderate (class 2) and severe (class 3) levels of symptomatology. Classes differed in symptom severity rather than symptom type. Income and education significantly predicted class 1 versus class 3 membership, and class 2 versus class 3. In conclusion, there is heterogeneity regarding severity of PTSD-depression-GAD symptomatology among trauma-exposed soldiers, with income and education predictive of class membership. Copyright © 2015 Elsevier Ltd. All rights reserved.

Defining and Predicting Patterns of Early Response in a Web-Based Intervention for Depression

PubMed Central

Arndt, Alice; Rubel, Julian; Berger, Thomas; Schröder, Johanna; Späth, Christina; Meyer, Björn; Greiner, Wolfgang; Gräfe, Viola; Hautzinger, Martin; Fuhr, Kristina; Rose, Matthias; Nolte, Sandra; Löwe, Bernd; Hohagen, Fritz; Klein, Jan Philipp; Moritz, Steffen

2017-01-01

Background Web-based interventions for individuals with depressive disorders have been a recent focus of research and may be an effective adjunct to face-to-face psychotherapy or pharmacological treatment. Objective The aim of our study was to examine the early change patterns in Web-based interventions to identify differential effects. Methods We applied piecewise growth mixture modeling (PGMM) to identify different latent classes of early change in individuals with mild-to-moderate depression (n=409) who underwent a CBT-based web intervention for depression. Results Overall, three latent classes were identified (N=409): Two early response classes (n=158, n=185) and one early deterioration class (n=66). Latent classes differed in terms of outcome (P<.001) and adherence (P=.03) in regard to the number of modules (number of modules with a duration of at least 10 minutes) and the number of assessments (P<.001), but not in regard to the overall amount of time using the system. Class membership significantly improved outcome prediction by 24.8% over patient intake characteristics (P<.001) and significantly added to the prediction of adherence (P=.04). Conclusions These findings suggest that in Web-based interventions outcome and adherence can be predicted by patterns of early change, which can inform treatment decisions and potentially help optimize the allocation of scarce clinical resources. PMID:28600278

Psychometric properties and a latent class analysis of the 12-item World Health Organization Disability Assessment Schedule 2.0 (WHODAS 2.0) in a pooled dataset of community samples.

PubMed

MacLeod, Melissa A; Tremblay, Paul F; Graham, Kathryn; Bernards, Sharon; Rehm, Jürgen; Wells, Samantha

2016-12-01

The 12-item World Health Organization Disability Assessment Schedule 2.0 (WHODAS 2.0) is a brief measurement tool used cross-culturally to capture the multi-dimensional nature of disablement through six domains, including: understanding and interacting with the world; moving and getting around; self-care; getting on with people; life activities; and participation in society. Previous psychometric research supports that the WHODAS 2.0 functions as a general factor of disablement. In a pooled dataset from community samples of adults (N = 447) we used confirmatory factor analysis to confirm a one-factor structure. Latent class analysis was used to identify subgroups of individuals based on their patterns of responses. We identified four distinct classes, or patterns of disablement: (1) pervasive disability; (2) physical disability; (3) emotional, cognitive, or interpersonal disability; (4) no/low disability. Convergent validity of the latent class subgroups was found with respect to socio-demographic characteristics, number of days affected by disabilities, stress, mental health, and substance use. These classes offer a simple and meaningful way to classify people with disabilities based on the 12-item WHODAS 2.0. Focusing on individuals with a high probability of being in the first three classes may help guide interventions. Copyright © 2016 John Wiley & Sons, Ltd.

Revised standards for statistical evidence.

PubMed

Johnson, Valen E

2013-11-26

Recent advances in Bayesian hypothesis testing have led to the development of uniformly most powerful Bayesian tests, which represent an objective, default class of Bayesian hypothesis tests that have the same rejection regions as classical significance tests. Based on the correspondence between these two classes of tests, it is possible to equate the size of classical hypothesis tests with evidence thresholds in Bayesian tests, and to equate P values with Bayes factors. An examination of these connections suggest that recent concerns over the lack of reproducibility of scientific studies can be attributed largely to the conduct of significance tests at unjustifiably high levels of significance. To correct this problem, evidence thresholds required for the declaration of a significant finding should be increased to 25-50:1, and to 100-200:1 for the declaration of a highly significant finding. In terms of classical hypothesis tests, these evidence standards mandate the conduct of tests at the 0.005 or 0.001 level of significance.

Organizational Supports for Research Evidence Use in State Public Health Agencies: A Latent Class Analysis.

PubMed

Hu, Hengrui; Allen, Peg; Yan, Yan; Reis, Rodrigo S; Jacob, Rebekah R; Brownson, Ross C

2018-05-30

Use of research evidence in public health decision making can be affected by organizational supports. Study objectives are to identify patterns of organizational supports and explore associations with research evidence use for job tasks among public health practitioners. In this longitudinal study, we used latent class analysis to identify organizational support patterns, followed by mixed logistic regression analysis to quantify associations with research evidence use. The setting included 12 state public health department chronic disease prevention units and their external partnering organizations involved in chronic disease prevention. Chronic disease prevention staff from 12 US state public health departments and partnering organizations completed self-report surveys at 2 time points, in 2014 and 2016 (N = 872). Latent class analysis was employed to identify subgroups of survey participants with distinct patterns of perceived organizational supports. Two classify-analyze approaches (maximum probability assignment and multiple pseudo-class draws) were used in 2017 to investigate the association between latent class membership and research evidence use. The optimal model identified 4 latent classes, labeled as "unsupportive workplace," "low agency leadership support," "high agency leadership support," and "supportive workplace." With maximum probability assignment, participants in "high agency leadership support" (odds ratio = 2.08; 95% CI, 1.35-3.23) and "supportive workplace" (odds ratio = 1.74; 95% CI, 1.10-2.74) were more likely to use research evidence in job tasks than "unsupportive workplace." The multiple pseudo-class draws produced comparable results with odds ratio = 2.09 (95% CI, 1.31-3.30) for "high agency leadership support" and odds ratio = 1.74 (95% CI, 1.07-2.82) for "supportive workplace." Findings suggest that leadership support may be a crucial element of organizational supports to encourage research evidence use. Organizational supports such as supervisory expectations, access to evidence, and participatory decision-making may need leadership support as well to improve research evidence use in public health job tasks.This is an open-access article distributed under the terms of the Creative Commons Attribution-Non Commercial-No Derivatives License 4.0 (CCBY-NC-ND), where it is permissible to download and share the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal.

Anxiety sensitivity class membership moderates the effects of pre-quit reduction in anxiety sensitivity on quit-day tobacco craving.

PubMed

Bakhshaie, Jafar; Zvolensky, Michael J; Langdon, Kirsten J; Leventhal, Adam M; Smits, Jasper A J; Allan, Nicholas; Schmidt, Norman B

2016-04-01

Although anxiety sensitivity has been primarily conceptualized as a dimensional latent construct, empirical evidence suggests that it also maintains a latent class structure, reflecting low-, moderate-, and high-risk underlying classes. The present study sought to explore whether these anxiety sensitivity classes moderated the relations between the degree of pre-quit reductions in anxiety sensitivity and the severity of nicotine withdrawal symptoms and craving experienced on quit-day. Participants included 195 adult smokers (47% female; Mage=39.4) participating in a larger "anxiety sensitivity reduction-smoking cessation" intervention trial. Anxiety sensitivity class significantly moderated relations between pre-quit reduction in anxiety sensitivity and quit-day craving. Specifically, smokers within the anxiety sensitivity high-risk class, who also demonstrated lesser pre-quit reductions in anxiety sensitivity, experienced the highest levels of craving on quit-day. These findings highlight the importance of 'high-risk' classes of anxiety sensitivity to better understand the experience of craving on quit day. Copyright © 2016. Published by Elsevier Ltd.

The Four U's: Latent Classes of Hookup Motivations Among College Students.

PubMed

Uecker, Jeremy E; Pearce, Lisa D; Andercheck, Brita

2015-06-01

College students' "hookups" have been the subject of a great deal of research in recent years. Motivations for hooking up have been linked to differences in well-being after the hookup, but studies detailing college students' motivations for engaging in hookups focus on single motivations. Using data from the 2010 Duke Hookup Survey, we consider how motivations for hooking up cluster to produce different classes, or profiles, of students who hook up, and how these classes are related to hookup regret. Four distinct classes of motivations emerged from our latent class analysis: Utilitarians (50%), Uninhibiteds (27%), Uninspireds (19%), and Unreflectives (4%). We find a number of differences in hookup motivation classes across social characteristics, including gender, year in school, race-ethnicity, self-esteem, and attitudes about sexual behavior outside committed relationships. Additionally, Uninspireds regret hookups more frequently than members of the other classes, and Uninhibiteds report regret less frequently than Utilitarians and Uninspireds. These findings reveal the complexity of motivations for hooking up and the link between motivations and regret.

Behavioral and Mental Health Correlates of Youth Stalking Victimization: A Latent Class Approach.

PubMed

Reidy, Dennis E; Smith-Darden, Joanne P; Kernsmith, Poco D

2016-12-01

Although recognized as a public health problem, little attention has been paid to the problem of stalking among youth. Latent profile analysis was used to identify latent groups of adolescent stalking victims and their behavioral and mental health correlates. A cross-sectional sample of 1,236 youths were randomly selected from 13 schools stratified by community risk level (i.e., low, moderate, and high risk) and gender. Students completed surveys assessing behavioral indicators of stalking victimization, as well as substance use, sexual behavior, dating violence, and psychiatric symptoms. Data were collected in 2013 and data analyses were performed in 2015. Analysis indicated the presence of a non-victim class, a minimal exposure class, and a victim class for boys and girls alike. Approximately 14% of girls and 13% of boys were in the stalking victim class. Adolescents in the victim class reported more symptoms of post-traumatic stress, mood disorder, and hopelessness, as well as more instances of alcohol use, binge drinking, and physical dating violence victimization. Girls in the victim class also reported engaging in sexting behaviors and oral sex with significantly more partners than their non-victim peers. These findings provide valuable knowledge of the prevalence and pertinent health correlates of stalking victimization in adolescence. The data suggest a substantial proportion of adolescents are victims of stalking and are likewise at risk for a number of deleterious health outcomes. As such, this population merits further attention by prevention researchers and practitioners. Published by Elsevier Inc.

Variations in Care Quality Outcomes of Dying People: Latent Class Analysis of an Adult National Register Population.

PubMed

Öhlén, Joakim; Russell, Lara; Håkanson, Cecilia; Alvariza, Anette; Fürst, Carl Johan; Årestedt, Kristofer; Sawatzky, Richard

2017-01-01

Symptom relief is a key goal of palliative care. There is a need to consider complexities in symptom relief patterns for groups of people to understand and evaluate symptom relief as an indicator of quality of care at end of life. The aims of this study were to distinguish classes of patients who have different symptom relief patterns during the last week of life and to identify predictors of these classes in an adult register population. In a cross-sectional retrospective design, data were used from 87,026 decedents with expected deaths registered in the Swedish Register of Palliative Care in 2011 and 2012. Study variables were structured into patient characteristics, and processes and outcomes of quality of care. A latent class analysis was used to identify symptom relief patterns. Multivariate multinomial regression analyses were used to identify predictors of class membership. Five latent classes were generated: "relieved pain," "relieved pain and rattles," "relieved pain and anxiety," "partly relieved shortness of breath, rattles and anxiety," and "partly relieved pain, anxiety and confusion." Important predictors of class membership were age, sex, cause of death, and having someone present at death, individual prescriptions as needed (PRN) and expert consultations. Interindividual variability and complexity in symptom relief patterns may inform quality of care and its evaluation for dying people across care settings. Copyright © 2016 American Academy of Hospice and Palliative Medicine. Published by Elsevier Inc. All rights reserved.

Stability of ARDS subphenotypes over time in two randomised controlled trials.

PubMed

Delucchi, Kevin; Famous, Katie R; Ware, Lorraine B; Parsons, Polly E; Thompson, B Taylor; Calfee, Carolyn S

2018-05-01

Two distinct acute respiratory distress syndrome (ARDS) subphenotypes have been identified using data obtained at time of enrolment in clinical trials; it remains unknown if these subphenotypes are durable over time. To determine the stability of ARDS subphenotypes over time. Secondary analysis of data from two randomised controlled trials in ARDS, the ARMA trial of lung protective ventilation (n=473; patients randomised to low tidal volumes only) and the ALVEOLI trial of low versus high positive end-expiratory pressure (n=549). Latent class analysis (LCA) and latent transition analysis (LTA) were applied to data from day 0 and day 3, independent of clinical outcomes. In ALVEOLI, LCA indicated strong evidence of two ARDS latent classes at days 0 and 3; in ARMA, evidence of two classes was stronger at day 0 than at day 3. The clinical and biological features of these two classes were similar to those in our prior work and were largely stable over time, though class 2 demonstrated evidence of progressive organ failures by day 3, compared with class 1. In both LCA and LTA models, the majority of patients (>94%) stayed in the same class from day 0 to day 3. Clinical outcomes were statistically significantly worse in class 2 than class 1 and were more strongly associated with day 3 class assignment. ARDS subphenotypes are largely stable over the first 3 days of enrolment in two ARDS Network trials, suggesting that subphenotype identification may be feasible in the context of clinical trials. © Article author(s) (or their employer(s) unless otherwise stated in the text of the article) 2018. All rights reserved. No commercial use is permitted unless otherwise expressly granted.

Inferring Markov chains: Bayesian estimation, model comparison, entropy rate, and out-of-class modeling.

PubMed

Strelioff, Christopher C; Crutchfield, James P; Hübler, Alfred W

2007-07-01

Markov chains are a natural and well understood tool for describing one-dimensional patterns in time or space. We show how to infer kth order Markov chains, for arbitrary k , from finite data by applying Bayesian methods to both parameter estimation and model-order selection. Extending existing results for multinomial models of discrete data, we connect inference to statistical mechanics through information-theoretic (type theory) techniques. We establish a direct relationship between Bayesian evidence and the partition function which allows for straightforward calculation of the expectation and variance of the conditional relative entropy and the source entropy rate. Finally, we introduce a method that uses finite data-size scaling with model-order comparison to infer the structure of out-of-class processes.

Trust from the past: Bayesian Personalized Ranking based Link Prediction in Knowledge Graphs

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

Zhang, Baichuan; Choudhury, Sutanay; Al-Hasan, Mohammad

2016-02-01

Estimating the confidence for a link is a critical task for Knowledge Graph construction. Link prediction, or predicting the likelihood of a link in a knowledge graph based on prior state is a key research direction within this area. We propose a Latent Feature Embedding based link recommendation model for prediction task and utilize Bayesian Personalized Ranking based optimization technique for learning models for each predicate. Experimental results on large-scale knowledge bases such as YAGO2 show that our approach achieves substantially higher performance than several state-of-art approaches. Furthermore, we also study the performance of the link prediction algorithm in termsmore » of topological properties of the Knowledge Graph and present a linear regression model to reason about its expected level of accuracy.« less

Inference on cancer screening exam accuracy using population-level administrative data.

PubMed

Jiang, H; Brown, P E; Walter, S D

2016-01-15

This paper develops a model for cancer screening and cancer incidence data, accommodating the partially unobserved disease status, clustered data structures, general covariate effects, and dependence between exams. The true unobserved cancer and detection status of screening participants are treated as latent variables, and a Markov Chain Monte Carlo algorithm is used to estimate the Bayesian posterior distributions of the diagnostic error rates and disease prevalence. We show how the Bayesian approach can be used to draw inferences about screening exam properties and disease prevalence while allowing for the possibility of conditional dependence between two exams. The techniques are applied to the estimation of the diagnostic accuracy of mammography and clinical breast examination using data from the Ontario Breast Screening Program in Canada. Copyright © 2015 John Wiley & Sons, Ltd.

Simulation-based Bayesian inference for latent traits of item response models: Introduction to the ltbayes package for R.

PubMed

Johnson, Timothy R; Kuhn, Kristine M

2015-12-01

This paper introduces the ltbayes package for R. This package includes a suite of functions for investigating the posterior distribution of latent traits of item response models. These include functions for simulating realizations from the posterior distribution, profiling the posterior density or likelihood function, calculation of posterior modes or means, Fisher information functions and observed information, and profile likelihood confidence intervals. Inferences can be based on individual response patterns or sets of response patterns such as sum scores. Functions are included for several common binary and polytomous item response models, but the package can also be used with user-specified models. This paper introduces some background and motivation for the package, and includes several detailed examples of its use.

Elastic Properties of Novel Co- and CoNi-Based Superalloys Determined through Bayesian Inference and Resonant Ultrasound Spectroscopy

NASA Astrophysics Data System (ADS)

Goodlet, Brent R.; Mills, Leah; Bales, Ben; Charpagne, Marie-Agathe; Murray, Sean P.; Lenthe, William C.; Petzold, Linda; Pollock, Tresa M.

2018-06-01

Bayesian inference is employed to precisely evaluate single crystal elastic properties of novel γ -γ ' Co- and CoNi-based superalloys from simple and non-destructive resonant ultrasound spectroscopy (RUS) measurements. Nine alloys from three Co-, CoNi-, and Ni-based alloy classes were evaluated in the fully aged condition, with one alloy per class also evaluated in the solution heat-treated condition. Comparisons are made between the elastic properties of the three alloy classes and among the alloys of a single class, with the following trends observed. A monotonic rise in the c_{44} (shear) elastic constant by a total of 12 pct is observed between the three alloy classes as Co is substituted for Ni. Elastic anisotropy ( A) is also increased, with a large majority of the nearly 13 pct increase occurring after Co becomes the dominant constituent. Together the five CoNi alloys, with Co:Ni ratios from 1:1 to 1.5:1, exhibited remarkably similar properties with an average A 1.8 pct greater than the Ni-based alloy CMSX-4. Custom code demonstrating a substantial advance over previously reported methods for RUS inversion is also reported here for the first time. CmdStan-RUS is built upon the open-source probabilistic programing language of Stan and formulates the inverse problem using Bayesian methods. Bayesian posterior distributions are efficiently computed with Hamiltonian Monte Carlo (HMC), while initial parameterization is randomly generated from weakly informative prior distributions. Remarkably robust convergence behavior is demonstrated across multiple independent HMC chains in spite of initial parameterization often very far from actual parameter values. Experimental procedures are substantially simplified by allowing any arbitrary misorientation between the specimen and crystal axes, as elastic properties and misorientation are estimated simultaneously.

Augmenting Latent Dirichlet Allocation and Rank Threshold Detection with Ontologies

DTIC Science & Technology

2010-03-01

Probabilistic Latent Semantic Indexing (PLSI) is an automated indexing information retrieval model [20]. It is based on a statistical latent class model which is...uses a statistical foundation that is more accurate in finding hidden semantic relationships [20]. The model uses factor analysis of count data, number...principle of statistical infer- ence which asserts that all of the information in a sample is contained in the likelihood function [20]. The statistical

Latent classes of polydrug and polyroute use and associations with human immunodeficiency virus risk behaviours and overdose among people who inject drugs in Tijuana, Baja California, Mexico.

PubMed

Meacham, Meredith C; Roesch, Scott C; Strathdee, Steffanie A; Lindsay, Suzanne; Gonzalez-Zuniga, Patricia; Gaines, Tommi L

2018-01-01

Patterns of polydrug use among people who inject drugs (PWID) may be differentially associated with overdose and unique human immunodeficiency virus (HIV) risk factors. Subgroups of PWID in Tijuana, Mexico, were identified based on substances used, route of administration, frequency of use and co-injection indicators. Participants were PWID residing in Tijuana age ≥18 years sampled from 2011 to 2012 who reported injecting an illicit substance in the past month (n = 735). Latent class analysis identified discrete classes of polydrug use characterised by 11 indicators of past 6 months substance use. Multinomial logistic regression examined class membership association with HIV risk behaviours, overdose and other covariates using an automated three-step procedure in mplus to account for classification error. Participants were classified into five subgroups. Two polydrug and polyroute classes were defined by use of multiple substances through several routes of administration and were primarily distinguished from each other by cocaine use (class 1: 5%) or no cocaine use (class 2: 29%). The other classes consisted primarily of injectors: cocaine, methamphetamine and heroin injection (class 3: 4%); methamphetamine and heroin injection (class 4: 10%); and heroin injection (class 5: 52%). Compared with the heroin-only injection class, memberships in the two polydrug and polyroute use classes were independently associated with both HIV injection and sexual risk behaviours. Substance use patterns among PWID in Tijuana are highly heterogeneous, and polydrug and polyroute users are a high-risk subgroup who may require more tailored prevention and treatment interventions. [Meacham MC, Roesch SC, Strathdee SA, Lindsay S, Gonzalez-Zuniga P, Gaines TL. Latent classes of polydrug and polyroute use and associations with human immunodeficiency virus risk behaviours and overdose among people who inject drugs in Tijuana, Baja California, Mexico. Drug Alcohol Rev 2018;37:128-136]. © 2017 Australasian Professional Society on Alcohol and other Drugs.

The Influence of Static and Dynamic Intrapersonal Factors on Longitudinal Patterns of Peer Victimization through Mid-adolescence: a Latent Transition Analysis.

PubMed

Haltigan, John D; Vaillancourt, Tracy

2018-01-01

Using 6 cycles (grade 5 through grade 10) of data obtained from a large prospective sample of Canadian school children (N = 700; 52.6% girls), we replicated previous findings concerning the empirical definition of peer victimization (i.e., being bullied) and examined static and dynamic intrapersonal factors associated with its emergence and experiential continuity through mid-adolescence. Latent class analyses consistently revealed a low victimization and an elevated victimization class across time, supporting previous work suggesting peer victimization was defined by degree rather than by type (e.g., physical). Using latent transition analyses (LTA), we found that child sex, parent-perceived pubertal development, and internalizing symptoms influenced the probability of transitioning from the low to the elevated victimization class across time. Higher-order extensions within the LTA modeling framework revealed a lasting effect of grade 5 victimization status on grade 10 victimization status and a large effect of chronic victimization on later parent-reported youth internalizing symptoms (net of prior parent-reported internalizing symptoms) in later adolescence (grade 11). Implications of the current findings for the experience of peer victimization, as well as the application of latent transition analysis as a useful approach for peer victimization research, are discussed.

Quality of life and patient preferences: identification of subgroups of multiple sclerosis patients.

PubMed

Rosato, Rosalba; Testa, Silvia; Oggero, Alessandra; Molinengo, Giorgia; Bertolotto, Antonio

2015-09-01

The aim of this study was to estimate preferences related to quality of life attributes in people with multiple sclerosis, by keeping heterogeneity of patient preference in mind, using the latent class approach. A discrete choice experiment survey was developed using the following attributes: activities of daily living, instrumental activities of daily living, pain/fatigue, anxiety/depression and attention/concentration. Choice sets were presented as pairs of hypothetical health status, based upon a fractional factorial design. The latent class logit model estimated on 152 patients identified three subpopulations, which, respectively, attached more importance to: (1) the physical dimension; (2) pain/fatigue and anxiety/depression; and (3) instrumental activities of daily living impairments, anxiety/depression and attention/concentration. A posterior analysis suggests that the latent class membership may be related to an individual's age to some extent, or to diagnosis and treatment, while apart from energy dimension, no significant difference exists between latent groups, with regard to Multiple Sclerosis Quality of Life-54 scales. A quality of life preference-based utility measure for people with multiple sclerosis was developed. These utility values allow identification of a hierarchic priority among different aspects of quality of life and may allow physicians to develop a care programme tailored to patient needs.

A social network-informed latent class analysis of patterns of substance use, sexual behavior, and mental health: Social Network Study III, Winnipeg, Manitoba, Canada.

PubMed

Hopfer, Suellen; Tan, Xianming; Wylie, John L

2014-05-01

We assessed whether a meaningful set of latent risk profiles could be identified in an inner-city population through individual and network characteristics of substance use, sexual behaviors, and mental health status. Data came from 600 participants in Social Network Study III, conducted in 2009 in Winnipeg, Manitoba, Canada. We used latent class analysis (LCA) to identify risk profiles and, with covariates, to identify predictors of class. A 4-class model of risk profiles fit the data best: (1) solitary users reported polydrug use at the individual level, but low probabilities of substance use or concurrent sexual partners with network members; (2) social-all-substance users reported polydrug use at the individual and network levels; (3) social-noninjection drug users reported less likelihood of injection drug and solvent use; (4) low-risk users reported low probabilities across substances. Unstable housing, preadolescent substance use, age, and hepatitis C status predicted risk profiles. Incorporation of social network variables into LCA can distinguish important subgroups with varying patterns of risk behaviors that can lead to sexually transmitted and bloodborne infections.

Latent class models for joint analysis of disease prevalence and high-dimensional semicontinuous biomarker data.

PubMed

Zhang, Bo; Chen, Zhen; Albert, Paul S

2012-01-01

High-dimensional biomarker data are often collected in epidemiological studies when assessing the association between biomarkers and human disease is of interest. We develop a latent class modeling approach for joint analysis of high-dimensional semicontinuous biomarker data and a binary disease outcome. To model the relationship between complex biomarker expression patterns and disease risk, we use latent risk classes to link the 2 modeling components. We characterize complex biomarker-specific differences through biomarker-specific random effects, so that different biomarkers can have different baseline (low-risk) values as well as different between-class differences. The proposed approach also accommodates data features that are common in environmental toxicology and other biomarker exposure data, including a large number of biomarkers, numerous zero values, and complex mean-variance relationship in the biomarkers levels. A Monte Carlo EM (MCEM) algorithm is proposed for parameter estimation. Both the MCEM algorithm and model selection procedures are shown to work well in simulations and applications. In applying the proposed approach to an epidemiological study that examined the relationship between environmental polychlorinated biphenyl (PCB) exposure and the risk of endometriosis, we identified a highly significant overall effect of PCB concentrations on the risk of endometriosis.

Patterns of adolescent bullying behaviors: physical, verbal, exclusion, rumor, and cyber.

PubMed

Wang, Jing; Iannotti, Ronald J; Luk, Jeremy W

2012-08-01

Patterns of engagement in cyber bullying and four types of traditional bullying were examined using latent class analysis (LCA). Demographic differences and externalizing problems were evaluated across latent class membership. Data were obtained from the 2005-2006 Health Behavior in School-aged Survey and the analytic sample included 7,508 U.S. adolescents in grades 6 through 10. LCA models were tested on physical bullying, verbal bullying, social exclusion, spreading rumors, and cyber bullying behaviors. Three latent classes were identified for each gender: All-Types Bullies (10.5% for boys and 4.0% for girls), Verbal/Social Bullies (29.3% for boys and 29.4% for girls), and a Non-Involved class (60.2% for boys and 66.6% for girls). Boys were more likely to be All-Types Bullies than girls. The prevalence rates of All-Types and Verbal/Social Bullies peaked during grades 6 to 8 and grades 7 and 8, respectively. Pairwise comparisons across the three latent classes on externalizing problems were conducted. Overall, the All-Types Bullies were at highest risk of using substances and carrying weapons, the Non-Involved were at lowest risk, and the Verbal/Social Bullies were in the middle. Results also suggest that most cyber bullies belong to a group of highly aggressive adolescents who conduct all types of bullying. This finding does not only improve our understanding of the relation between cyber bullying and traditional bullying, but it also suggests that prevention and intervention efforts could target cyber bullies as a high-risk group for elevated externalizing problems. Copyright © 2012 Society for the Study of School Psychology. All rights reserved.

Race-Specific Transition Patterns among Alcohol Use Classes in Adolescent Girls

ERIC Educational Resources Information Center

Dauber, Sarah E.; Paulson, James F.; Leiferman, Jenn A.

2011-01-01

We used data from the National Longitudinal Study of Adolescent Health to examine transitions among alcohol use classes in 2225 White and African American adolescent girls, and race differences in predictors of transition into and out of problematic drinking classes. Latent class analysis confirmed four classes for White girls and three for AA…

Modeling Heterogeneity in Relationships between Initial Status and Rates of Change: Treating Latent Variable Regression Coefficients as Random Coefficients in a Three-Level Hierarchical Model

ERIC Educational Resources Information Center

Choi, Kilchan; Seltzer, Michael

2010-01-01

In studies of change in education and numerous other fields, interest often centers on how differences in the status of individuals at the start of a period of substantive interest relate to differences in subsequent change. In this article, the authors present a fully Bayesian approach to estimating three-level Hierarchical Models in which latent…

Adolescent stalking and risk of violence✩

PubMed Central

Smith-Darden, Joanne P.; Reidy, Dennis E.; Kernsmith, Poco D.

2018-01-01

Stalking perpetration and the associated risk for violence among adolescents has generally been neglected. In the present study, 1236 youth completed surveys assessing empirically established stalking indicators, threats and aggression toward stalking victims, dating violence, and violent delinquency. Latent Profile Analysis identified 3 latent classes of boys: non-perpetrators (NP), hyper-intimate pursuit (HIP), and comprehensive stalking perpetrators (CSP) and, and 2 classes for girls: NP and HIP. Boys in the CSP class were the most violent youth on nearly all indices with boys in the HIP class demonstrating an intermediate level of violence compared to NP boys. Girls in the HIP class were more violent than NP girls on all indices. These findings suggest stalking in adolescence merits attention by violence prevention experts. In particular, juvenile stalking may signify youth at risk for multiple forms of violence perpetrated against multiple types of victims, not just the object of their infatuation. PMID:27641644

Latino cigarette smoking patterns by gender in a US national sample

PubMed Central

Kristman-Valente, Allison; Flaherty, Brian P.

2015-01-01

Background Latino smokers are a rising public health concern who experience elevated tobacco related health disparities. Purpose Additional information on Latino smoking is needed to inform screening and treatment. Analysis Latent class analysis using smoking frequency, cigarette preferences, onset, smoking duration, cigarettes per day and minutes to first cigarette were used to create multivariate latent smoking profiles for Latino men and women. Results Final models found seven classes for Latinas and nine classes for Latinos. Despite a common finding in the literature that Latino smokers are more likely to be low-risk, intermittent smokers, the majority of classes, for both males and females, described patterns of high-risk, daily smoking. Gender variations in smoking classes were noted. Conclusions Several markers of smoking risk were identified among both male and female Latino smokers including long durations of smoking, daily smoking and preference for specialty cigarettes, all factors associated with long-term health consequences. PMID:26304857

RACE-SPECIFIC TRANSITION PATTERNS AMONG ALCOHOL USE CLASSES IN ADOLESCENT GIRLS

PubMed Central

Dauber, Sarah E.; Paulson, James F.; Leiferman, Jenn A.

2010-01-01

We used data from the National Longitudinal Study of Adolescent Health to examine transitions among alcohol use classes in 2225 White and African American adolescent girls, and race differences in predictors of transition into and out of problematic drinking classes. Latent class analysis confirmed four classes for White girls and three for AA girls, defined in a previous study. Latent transition analysis revealed more stable abstainers and decreasing alcohol use among AA girls, and more increasing alcohol use among White girls, though stable abstainers were the largest group among both races. Increasing use was predicted by delinquency, academic misbehavior, substance use, and peer support for White girls, and by older age and delinquency for AA girls. Decreasing use was predicted by older age and depressive symptoms for White girls, and by family relationship quality and substance use for AA girls. Study limitations and implications of findings are discussed. PMID:20708254

Gender Differences in Patterns of Substance Use and Delinquency: A Latent Transition Analysis

PubMed Central

Bright, Charlotte Lyn; Sacco, Paul; Kolivoski, Karen M.; Stapleton, Laura M.; Jun, Hyun-Jin; Morris-Compton, Darnell

2017-01-01

This study explores gender-specific patterns and transitions of adolescent substance use and delinquency in a sample of youths at ages 12, 15, and 18 (N = 803). Latent transition analysis identified “Primary Delinquent,” “Delinquency and Substance Use,” and “Low Risk” classes. Females were less likely to be in the “Primary Delinquent” class at age 12 than males. From 15 to 18, females were approximately equally likely to transition from “Primary Delinquent” to both other classes, whereas males were more likely to transition from “Primary Delinquent” to “Delinquency and Substance Use.” These gender differences in behavior can inform services. PMID:28603406

Dietary and exercise change following acute cardiac syndrome onset: A latent class growth modelling analysis.

PubMed

Bennett, Paul; Gruszczynska, Ewa; Marke, Victoria

2016-10-01

The present study aim determine sub-group trajectories of change on measures of diet and exercise following acute coronary syndrome. 150 participants were assessed in hospital, 1 month and 6 months subsequently on measures including physical activity, diet, illness beliefs, coping and mood. Change trajectories were measured using latent class growth modelling. Multinomial logistic regression was used to predict class membership. These analyses revealed changes in exercise were confined to a sub-group of participants already reporting relatively high exercise levels; those eating less healthily evidenced modest dietary improvements. Coping, gender, depression and perceived control predicted group membership to a modest degree. © The Author(s) 2015.

Scale Reliability Evaluation with Heterogeneous Populations

ERIC Educational Resources Information Center

Raykov, Tenko; Marcoulides, George A.

2015-01-01

A latent variable modeling approach for scale reliability evaluation in heterogeneous populations is discussed. The method can be used for point and interval estimation of reliability of multicomponent measuring instruments in populations representing mixtures of an unknown number of latent classes or subpopulations. The procedure is helpful also…

Measurement of Psychological Disorders Using Cognitive Diagnosis Models

ERIC Educational Resources Information Center

Templin, Jonathan L.; Henson, Robert A.

2006-01-01

Cognitive diagnosis models are constrained (multiple classification) latent class models that characterize the relationship of questionnaire responses to a set of dichotomous latent variables. Having emanated from educational measurement, several aspects of such models seem well suited to use in psychological assessment and diagnosis. This article…

Patient factors and quality of life outcomes differ among four subgroups of oncology patients based on symptom occurrence.

PubMed

Astrup, Guro Lindviksmoen; Hofsø, Kristin; Bjordal, Kristin; Guren, Marianne Grønlie; Vistad, Ingvild; Cooper, Bruce; Miaskowski, Christine; Rustøen, Tone

2017-03-01

Reviews of the literature on symptoms in oncology patients undergoing curative treatment, as well as patients receiving palliative care, suggest that they experience multiple, co-occurring symptoms and side effects. The purposes of this study were to determine if subgroups of oncology patients could be identified based on symptom occurrence rates and if these subgroups differed on a number of demographic and clinical characteristics, as well as on quality of life (QoL) outcomes. Latent class analysis (LCA) was used to identify subgroups (i.e. latent classes) of patients with distinct symptom experiences based on the occurrence rates for the 13 most common symptoms from the Memorial Symptom Assessment Scale. In total, 534 patients with breast, head and neck, colorectal, or ovarian cancer participated. Four latent classes of patients were identified based on probability of symptom occurrence: all low class [i.e. low probability for all symptoms (n = 152)], all high class (n = 149), high psychological class (n = 121), and low psychological class (n = 112). Patients in the all high class were significantly younger compared with patients in the all low class. Furthermore, compared to the other three classes, patients in the all high class had lower functional status and higher comorbidity scores, and reported poorer QoL scores. Patients in the high and low psychological classes had a moderate probability of reporting physical symptoms. Patients in the low psychological class reported a higher number of symptoms, a lower functional status, and poorer physical and total QoL scores. Distinct subgroups of oncology patients can be identified based on symptom occurrence rates. Patient characteristics that are associated with these subgroups can be used to identify patients who are at greater risk for multiple co-occurring symptoms and diminished QoL, so that these patients can be offered appropriate symptom management interventions.

Predominant typologies of psychopathology in the United States: a latent class analysis.

PubMed

El-Gabalawy, Renée; Tsai, Jack; Harpaz-Rotem, Ilan; Hoff, Rani; Sareen, Jitender; Pietrzak, Robert H

2013-11-01

Latent class analysis (LCA) offers a parsimonious way of classifying common typologies of psychiatric comorbidity. We used LCA to identify the nature and correlates of predominant typologies of Axis I and II disorders in a large and comprehensive population-based sample of U.S. adults. We analyzed data from Wave 2 of the National Epidemiologic Survey on Alcohol and Related Conditions (2004-2005; n = 34,653), a population-based sample of U.S. adults. We derived latent classes based on all assessed Axis I and II disorders and examined the relationship between the identified Axis I classes and lifetime psychiatric disorders and suicide attempts, and physical and mental health-related quality of life. A four-class solution was optimal in characterizing predominant typologies of both Axis I and II disorders. For Axis I disorders, these included low psychopathology (n = 28,935, 84.0%), internalizing (n = 3693, 9.9%), externalizing (n = 1426, 4.5%), and high psychopathology (n = 599, 1.6%) classes. For Axis II disorders, these included no/low personality disorders (n = 31,265, 90.9%), obsessive/paranoid (n = 1635, 4.6%), borderline/dysregulated (n = 1319, 3.4%), and highly comorbid (n = 434, 1.1%) classes. Compared to the low psychopathology class, all other Axis I classes had significantly increased odds of mental disorders, elevated Axis II classes, suicide attempts and poorer quality of life, with the high psychopathology class having the overall highest rates of these correlates, with the exception of substance use disorders. Compared to the low psychopathology class, the internalizing and externalizing classes had increased rates of mood and anxiety disorders, and substance use disorders, respectively. Axis I and II psychopathology among U.S. adults may be best represented by four predominant typologies. Characterizing co-occurring patterns of psychopathology using person-based typologies represents a higher-order classification system that may be useful in clinical and research settings. Published by Elsevier Ltd.

The three latent classes of adolescent delinquency and the risk factors for membership in each class.

PubMed

Hasking, Penelope Anne; Scheier, Lawrence M; Abdallah, Arbi Ben

2011-01-01

This study used latent class analysis to examine subpopulation membership based on self-reports of delinquent behaviors obtained from Australian youth. Three discrete identifiable classes were derived based on 51 indicators of physical violence, property damage, minor infractions, drug use, and social delinquency. One class of youth engaged in primarily rule breaking and norm violations including underage alcohol use, typical of this age period. A second class was more actively delinquent emphasizing drug use, trespassing, and various forms of disobedience. A third class of highly delinquent youth differed from their counterparts by endorsing drug use, thievery that involved stealing money, goods, and cars, property damage, gambling, precocious sexual experiences, involvement with pornographic materials, and fighting. Multinomial logistic regression predicting class membership indicated highly delinquent youth were more likely to be older males, use venting coping strategies, and be fun or novelty seeking compared with rule breakers. Findings are discussed in terms of refining current taxonomic arguments regarding the structure of delinquency and implications for prevention of early-stage antisocial behavior. © 2010 Wiley-Liss, Inc.

Impact of an equality constraint on the class-specific residual variances in regression mixtures: A Monte Carlo simulation study.

PubMed

Kim, Minjung; Lamont, Andrea E; Jaki, Thomas; Feaster, Daniel; Howe, George; Van Horn, M Lee

2016-06-01

Regression mixture models are a novel approach to modeling the heterogeneous effects of predictors on an outcome. In the model-building process, often residual variances are disregarded and simplifying assumptions are made without thorough examination of the consequences. In this simulation study, we investigated the impact of an equality constraint on the residual variances across latent classes. We examined the consequences of constraining the residual variances on class enumeration (finding the true number of latent classes) and on the parameter estimates, under a number of different simulation conditions meant to reflect the types of heterogeneity likely to exist in applied analyses. The results showed that bias in class enumeration increased as the difference in residual variances between the classes increased. Also, an inappropriate equality constraint on the residual variances greatly impacted on the estimated class sizes and showed the potential to greatly affect the parameter estimates in each class. These results suggest that it is important to make assumptions about residual variances with care and to carefully report what assumptions are made.

Latent Classes of Psychiatric Symptoms among Chinese Children Living in Poverty

ERIC Educational Resources Information Center

Herman, Keith C.; Bi, Yu; Borden, Lindsay A.; Reinke, Wendy M.

2012-01-01

Describing co-occurring symptom patterns among children in nonwestern contexts may have important implications for how emotional and behavior problems are defined, conceptualized, studied, and ultimately prevented. A latent profile analysis (LPA) was conducted on the co-occurring psychiatric symptoms of 196 Chinese children living in poverty.…

Comparisons of Mathematics Intervention Effects in Resource and Inclusive Classrooms

ERIC Educational Resources Information Center

Bottge, Brian A.; Cohen, Allan S.; Choi, Hye-Jeong

2018-01-01

In this article, we describe results of a reanalysis of two randomized studies that tested the effects of enhanced anchored instruction (EAI) on the fractions computation performance of students in special education resource rooms and inclusive mathematics classrooms. Latent class analysis and latent transition analysis classified students…

A person-centered analysis of posttraumatic stress disorder symptoms following a natural disaster: predictors of latent class membership.

PubMed

Rosellini, Anthony J; Coffey, Scott F; Tracy, Melissa; Galea, Sandro

2014-01-01

The present study applied latent class analysis to a sample of 810 participants residing in southern Mississippi at the time of Hurricane Katrina to determine if people would report distinct, meaningful PTSD symptom classes following a natural disaster. We found a four-class solution that distinguished persons on the basis of PTSD symptom severity/pervasiveness (Severe, Moderate, Mild, and Negligible Classes). Multinomial logistic regression models demonstrated that membership in the Severe and Moderate Classes was associated with potentially traumatic hurricane-specific experiences (e.g., being physically injured, seeing dead bodies), pre-hurricane traumatic events, co-occurring depression symptom severity and suicidal ideation, certain religious beliefs, and post-hurricane stressors (e.g., social support). Collectively, the findings suggest that more severe/pervasive typologies of natural disaster PTSD may be predicted by the frequency and severity of exposure to stressful/traumatic experiences (before, during, and after the disaster), co-occurring psychopathology, and specific internal beliefs. Copyright © 2013 Elsevier Ltd. All rights reserved.

A Person-Centered Analysis of Posttraumatic Stress Disorder Symptoms Following a Natural Disaster: Predictors of Latent Class Membership

PubMed Central

Rosellini, Anthony J.; Coffey, Scott F.; Tracy, Melissa; Galea, Sandro

2014-01-01

The present study applied latent class analysis to a sample of 810 participants residing in southern Mississippi at the time of Hurricane Katrina to determine if people would report distinct, meaningful PTSD symptom classes following a natural disaster. We found a four-class solution that distinguished persons on the basis of PTSD symptom severity/pervasiveness (Severe, Moderate, Mild, and Negligible Classes). Multinomial logistic regression models demonstrated that membership in the Severe and Moderate Classes was associated with potentially traumatic hurricane-specific experiences (e.g., being physically injured, seeing dead bodies), pre-hurricane traumatic events, co-occurring depression symptom severity and suicidal ideation, certain religious beliefs, and post-hurricane stressors (e.g., social support). Collectively, the findings suggest that more severe/pervasive typologies of natural disaster PTSD may be predicted by the frequency and severity of exposure to stressful/traumatic experiences (before, during, and after the disaster), co-occurring psychopathology, and specific internal beliefs. PMID:24334161

Latent typologies of posttraumatic stress disorder in World Trade Center responders.

PubMed

Horn, Sarah R; Pietrzak, Robert H; Schechter, Clyde; Bromet, Evelyn J; Katz, Craig L; Reissman, Dori B; Kotov, Roman; Crane, Michael; Harrison, Denise J; Herbert, Robin; Luft, Benjamin J; Moline, Jacqueline M; Stellman, Jeanne M; Udasin, Iris G; Landrigan, Philip J; Zvolensky, Michael J; Southwick, Steven M; Feder, Adriana

2016-12-01

Posttraumatic stress disorder (PTSD) is a debilitating and often chronic psychiatric disorder. Following the 9/11/2001 World Trade Center (WTC) attacks, thousands of individuals were involved in rescue, recovery and clean-up efforts. While a growing body of literature has documented the prevalence and correlates of PTSD in WTC responders, no study has evaluated predominant typologies of PTSD in this population. Participants were 4352 WTC responders with probable WTC-related DSM-IV PTSD. Latent class analyses were conducted to identify predominant typologies of PTSD symptoms and associated correlates. A 3-class solution provided the optimal representation of latent PTSD symptom typologies. The first class, labeled "High-Symptom (n = 1,973, 45.3%)," was characterized by high probabilities of all PTSD symptoms. The second class, "Dysphoric (n = 1,371, 31.5%)," exhibited relatively high probabilities of emotional numbing and dysphoric arousal (e.g., sleep disturbance). The third class, "Threat (n = 1,008, 23.2%)," was characterized by high probabilities of re-experiencing, avoidance and anxious arousal (e.g., hypervigilance). Compared to the Threat class, the Dysphoric class reported a greater number of life stressors after 9/11/2001 (OR = 1.06). The High-Symptom class was more likely than the Threat class to have a positive psychiatric history before 9/11/2001 (OR = 1.7) and reported a greater number of life stressors after 9/11/2001 (OR = 1.1). The High-Symptom class was more likely than the Dysphoric class, which was more likely than the Threat class, to screen positive for depression (83% > 74% > 53%, respectively), and to report greater functional impairment (High-Symptom > Dysphoric [Cohen d = 0.19], Dysphoric > Threat [Cohen d = 0.24]). These results may help inform assessment, risk stratification, and treatment approaches for PTSD in WTC and disaster responders. Copyright © 2016 Elsevier Ltd. All rights reserved.

Emotional Psychological and Related Problems among Truant Youths: An Exploratory Latent Class Analysis

ERIC Educational Resources Information Center

Dembo, Richard; Briones-Robinson, Rhissa; Ungaro, Rocio Aracelis; Gulledge, Laura M.; Karas, Lora M.; Winters, Ken C.; Belenko, Steven; Greenbaum, Paul E.

2012-01-01

Intervention Project. Results identified two classes of youths: Class 1(n=9) - youths with low levels of delinquency, mental health and substance abuse issues; and Class 2(n=37) - youths with high levels of these problems. Comparison of these two classes on their urine analysis test results and parent/guardian reports of traumatic events found…

Exploring the latent trait of Opioid Use Disorder criteria among frequent nonmedical prescription opioid users

PubMed Central

Castaldelli-Maia, João Mauricio; Andrade, Laura H.; Keyes, Katherine M.; Cerdá, Magdalena; Pilowsky, Daniel J.; Martins, Silvia S.

2016-01-01

Background There is a need to explore the dimensional and categorical phenotypes of criteria of opioid use disorder among frequent nonmedical users of prescription opioids (NMUPO) users. Methods We used pooled data of 2011–2012 National Survey on Drug Use and Health to examine reliability and phenotypic variability in the diagnosis of OUD secondary to NMUPO in a nationally-representative sample of 18+ years-old frequent past-year NMUPO users (120+ days, n=806). Through exploratory factor analysis (EFA) and latent class analysis (LCA), we examined 10 past-year OUD criteria. We examined associations between the latent classes and sociodemographic/psychiatric/NMUPO correlates. Results OUD criteria were unidimensional, and a three-class model was the overall best fitting solution for characterizing individuals into phenotypes along this unidimensional continuum: a “non-symptomatic class” (40.7%), “Tolerance-Time spent class” (29.0%) with high probability of endorsing Tolerance/Time Spent criteria, and a “High-moderate symptomatic class” (30.1%). The last class was significantly associated with being male, having insurance and obtaining prescription opioids (PO) nonmedically via “doctor shopping” as compared to the non-symptomatic class. “Tolerance-Time spent class” was significantly associated with being younger (18–25 years) and obtaining PO nonmedically from family/friends as compared to the non-symptomatic class. Conclusion This study revealed the different characteristics and routes of access to PO of different classes of frequent NMUPO users. It is possible that these groups may respond to different interventions, however such conclusions would require a clinical study. PMID:27302873

Collinear Latent Variables in Multilevel Confirmatory Factor Analysis: A Comparison of Maximum Likelihood and Bayesian Estimations.

PubMed

Can, Seda; van de Schoot, Rens; Hox, Joop

2015-06-01

Because variables may be correlated in the social and behavioral sciences, multicollinearity might be problematic. This study investigates the effect of collinearity manipulated in within and between levels of a two-level confirmatory factor analysis by Monte Carlo simulation. Furthermore, the influence of the size of the intraclass correlation coefficient (ICC) and estimation method; maximum likelihood estimation with robust chi-squares and standard errors and Bayesian estimation, on the convergence rate are investigated. The other variables of interest were rate of inadmissible solutions and the relative parameter and standard error bias on the between level. The results showed that inadmissible solutions were obtained when there was between level collinearity and the estimation method was maximum likelihood. In the within level multicollinearity condition, all of the solutions were admissible but the bias values were higher compared with the between level collinearity condition. Bayesian estimation appeared to be robust in obtaining admissible parameters but the relative bias was higher than for maximum likelihood estimation. Finally, as expected, high ICC produced less biased results compared to medium ICC conditions.

Topics in Computational Bayesian Statistics With Applications to Hierarchical Models in Astronomy and Sociology

NASA Astrophysics Data System (ADS)

Sahai, Swupnil

This thesis includes three parts. The overarching theme is how to analyze structured hierarchical data, with applications to astronomy and sociology. The first part discusses how expectation propagation can be used to parallelize the computation when fitting big hierarchical bayesian models. This methodology is then used to fit a novel, nonlinear mixture model to ultraviolet radiation from various regions of the observable universe. The second part discusses how the Stan probabilistic programming language can be used to numerically integrate terms in a hierarchical bayesian model. This technique is demonstrated on supernovae data to significantly speed up convergence to the posterior distribution compared to a previous study that used a Gibbs-type sampler. The third part builds a formal latent kernel representation for aggregate relational data as a way to more robustly estimate the mixing characteristics of agents in a network. In particular, the framework is applied to sociology surveys to estimate, as a function of ego age, the age and sex composition of the personal networks of individuals in the United States.

Profiles of Community Violence Exposure Among African American Youth: An Examination of Desensitization to Violence Using Latent Class Analysis.

PubMed

Gaylord-Harden, Noni K; Dickson, Daniel; Pierre, Cynthia

2016-07-01

The current study employed latent class analysis (LCA) to identify distinct profiles of community violence exposure and their associations to desensitization outcomes in 241 African American early adolescents (M age = 12.86, SD = 1.28) in the sixth through eighth grade from under-resourced urban communities. Participants self-reported on their exposure to community violence, as well as on depressive and anxiety symptoms. The LCA revealed three distinct classes: a class exposed to low levels of violence (low exposure class), a class exposed to moderately high levels of victimization (victimization class), and a class exposed to high levels of all types of violence (high exposure class). Consistent with predictions, the high exposure class showed the lowest levels of depressive symptoms, suggesting a desensitization outcome. Gender and age were also examined in relation to the classes, and age was significantly associated with an increased risk of being a member of the high exposure class relative to the low exposure class. Using person-based analyses to examine desensitization outcomes provides useful information for prevention and intervention efforts, as it helps to identify a specific subgroup of youth that may be more likely to show desensitization outcomes in the context of community violence. © The Author(s) 2015.

A Three-Step Latent Class Analysis to Identify How Different Patterns of Teen Dating Violence and Psychosocial Factors Influence Mental Health.

PubMed

Choi, Hye Jeong; Weston, Rebecca; Temple, Jeff R

2017-04-01

Although multiple forms (i.e., physical, threatening, psychological, sexual, and relational abuse) and patterns (i.e., perpetration and victimization) of violence can co-occur, most existing research examines these experiences individually. Thus, the purpose of this study is to investigate: (1) homogenous subgroups based on victimization and perpetration of multiple forms of teen dating violence; (2) predictors of membership in these subgroups; and (3) mental health consequences associated with membership in each subgroup. Nine hundred eighteen adolescents in the 9 th or 10 th grade at seven public high schools in Texas participated in the survey (56 % female, White: 30 %, Hispanic: 32 %, African American: 29 %, others: 9 %). A three-step latent class analysis was employed. Five latent teen dating violence classes were identified: (1) nonviolence; (2) emotional/verbal abuse; (3) forced sexual contact; (4) psychological + physical violence; and (5) psychological abuse. Females, African Americans, and youth who had higher acceptance of couple violence scores and whose parents had less education were more likely to members of dating violence classes compared with the nonviolence class. Adolescents who experienced multiple types of dating violence reported greater mental health concerns. Prevention programs may benefit by identifying the homogenous subgroups of teen dating violence and targeting adolescent teen dating violence accordingly.

Symptoms of prolonged grief and posttraumatic stress following loss: A latent class analysis.

PubMed

Maccallum, Fiona; Bryant, Richard A

2018-04-01

Individuals vary in how they respond to bereavement. Those who experience poor bereavement outcomes often report symptoms from more than one diagnostic category. This study sought to identify groups of individuals who share similar patterns of prolonged grief disorder and posttraumatic stress disorder symptoms to determine whether these profiles are differentially related to negative appraisals thought to contribute to prolonged grief disorder and posttraumatic stress disorder symptomatology. Participants were 185 bereaved adults. Latent class analysis was used to identify subgroups of individuals who showed similar patterns of co-occurrence of prolonged grief disorder and posttraumatic stress disorder symptoms. Multinomial regression was used to examine the extent to which appraisal domains and sociodemographic and loss factors predicted class membership. Latent class analysis revealed three classes of participants: a low symptom group, a high prolonged grief disorder symptom group, and a high prolonged grief disorder and posttraumatic stress disorder symptom group. Membership of the prolonged grief disorder group and prolonged grief disorder and posttraumatic stress disorder group was predicted by higher mean negative self-related appraisals. Demographic and loss-related factors did not predict group membership. These findings have implications for understanding co-occurrence of prolonged grief disorder and posttraumatic stress disorder symptoms following bereavement. Findings are consistent with theoretical models highlighting the importance of negative self-related beliefs in prolonged grief disorder.

Mental toughness latent profiles in endurance athletes

PubMed Central

Zeiger, Robert S.

2018-01-01

Mental toughness in endurance athletes, while an important factor for success, has been scarcely studied. An online survey was used to examine eight mental toughness factors in endurance athletes. The study aim was to determine mental toughness profiles via latent profile analysis in endurance athletes and whether associations exist between the latent profiles and demographics and sports characteristics. Endurance athletes >18 years of age were recruited via social media outlets (n = 1245, 53% female). Mental toughness was measured using the Sports Mental Toughness Questionnaire (SMTQ), Psychological Performance Inventory-Alternative (PPI-A), and self-esteem was measured using the Rosenberg Self-Esteem Scale (RSE). A three-class solution emerged, designated as high mental toughness (High MT), moderate mental toughness (Moderate MT) and low mental toughness (Low MT). ANOVA tests showed significant differences between all three classes on all 8 factors derived from the SMTQ, PPI-A and the RSE. There was an increased odds of being in the High MT class compared to the Low MT class for males (OR = 1.99; 95% CI, 1.39, 2.83; P<0.001), athletes who were over 55 compared to those who were 18–34 (OR = 2.52; 95% CI, 1.37, 4.62; P<0.01), high sports satisfaction (OR = 8.17; 95% CI, 5.63, 11.87; P<0.001), and high division placement (OR = 2.18; 95% CI, 1.46,3.26; P<0.001). The data showed that mental toughness latent profiles exist in endurance athletes. High MT is associated with demographics and sports characteristics. Mental toughness screening in athletes may help direct practitioners with mental skills training. PMID:29474398

Delirium superimposed on dementia: defining disease states and course from longitudinal measurements of a multivariate index using latent class analysis and hidden Markov chains.

PubMed

Ciampi, Antonio; Dyachenko, Alina; Cole, Martin; McCusker, Jane

2011-12-01

The study of mental disorders in the elderly presents substantial challenges due to population heterogeneity, coexistence of different mental disorders, and diagnostic uncertainty. While reliable tools have been developed to collect relevant data, new approaches to study design and analysis are needed. We focus on a new analytic approach. Our framework is based on latent class analysis and hidden Markov chains. From repeated measurements of a multivariate disease index, we extract the notion of underlying state of a patient at a time point. The course of the disorder is then a sequence of transitions among states. States and transitions are not observable; however, the probability of being in a state at a time point, and the transition probabilities from one state to another over time can be estimated. Data from 444 patients with and without diagnosis of delirium and dementia were available from a previous study. The Delirium Index was measured at diagnosis, and at 2 and 6 months from diagnosis. Four latent classes were identified: fairly healthy, moderately ill, clearly sick, and very sick. Dementia and delirium could not be separated on the basis of these data alone. Indeed, as the probability of delirium increased, so did the probability of decline of mental functions. Eight most probable courses were identified, including good and poor stable courses, and courses exhibiting various patterns of improvement. Latent class analysis and hidden Markov chains offer a promising tool for studying mental disorders in the elderly. Its use may show its full potential as new data become available.

Maternal anaemia at delivery and haemoglobin evolution in children during their first 18 months of life using latent class analysis.

PubMed

Koura, Kobto G; Ouédraogo, Smaïla; Cottrell, Gilles; Le Port, Agnès; Massougbodji, Achille; Garcia, André

2012-01-01

Anaemia during pregnancy and at delivery is an important public health problem in low- and middle-income countries. Its association with the children's haemoglobin level over time remains unclear. Our goals were to identify distinct haemoglobin level trajectories using latent class analysis and to assess the association between these trajectories and maternal anaemia and other risk factors. A prospective study of children from birth to 18 months of life was conducted in a rural setting in Tori-Bossito, Benin. The main outcome measure was the haemoglobin levels repeatedly measured at 3, 6, 9, 12, 15 and 18 months. Variables were collected from the mothers at delivery and from their children at birth and during the follow-up. The analyses were performed by means of Latent Class Analysis which has never been used for this kind of data. All the analyses were performed with Stata software, version 11.0, using the generalized linear latent and mixed model (GLLAMM) framework. We showed that 33.7% of children followed a low haemoglobin trajectory and 66.3% a high trajectory during the first 18 months of life. Newborn anaemia, placental malaria, malaria attack, sickle cell trait and male gender were significantly associated with a lower children's haemoglobin level over time, whereas maternal age, children living in a polygamous family and with good feeding practices had a higher Hb level in the first18 months. We also showed that maternal anaemia was a predictor for 'low haemoglobin level trajectory' group membership but have no significant effect on children haemoglobin level over time. Latent Class Analyses framework seems well suited to analyse longitudinal data under the hypothesis that different subpopulations of subjects are present in the data, each with its own set of parameters, with distinctive evolutions that themselves may reflect distinctive aetiologies.

Maternal Anaemia at Delivery and Haemoglobin Evolution in Children during Their First 18 Months of Life Using Latent Class Analysis

PubMed Central

Koura, Kobto G.; Ouédraogo, Smaïla; Cottrell, Gilles; Le Port, Agnès; Massougbodji, Achille; Garcia, André

2012-01-01

Background Anaemia during pregnancy and at delivery is an important public health problem in low- and middle-income countries. Its association with the children’s haemoglobin level over time remains unclear. Our goals were to identify distinct haemoglobin level trajectories using latent class analysis and to assess the association between these trajectories and maternal anaemia and other risk factors. Method A prospective study of children from birth to 18 months of life was conducted in a rural setting in Tori-Bossito, Benin. The main outcome measure was the haemoglobin levels repeatedly measured at 3, 6, 9, 12, 15 and 18 months. Variables were collected from the mothers at delivery and from their children at birth and during the follow-up. The analyses were performed by means of Latent Class Analysis which has never been used for this kind of data. All the analyses were performed with Stata software, version 11.0, using the generalized linear latent and mixed model (GLLAMM) framework. Results We showed that 33.7% of children followed a low haemoglobin trajectory and 66.3% a high trajectory during the first 18 months of life. Newborn anaemia, placental malaria, malaria attack, sickle cell trait and male gender were significantly associated with a lower children’s haemoglobin level over time, whereas maternal age, children living in a polygamous family and with good feeding practices had a higher Hb level in the first18 months. We also showed that maternal anaemia was a predictor for ‘low haemoglobin level trajectory’ group membership but have no significant effect on children haemoglobin level over time. Conclusion Latent Class Analyses framework seems well suited to analyse longitudinal data under the hypothesis that different subpopulations of subjects are present in the data, each with its own set of parameters, with distinctive evolutions that themselves may reflect distinctive aetiologies. PMID:23185556

Relation Extraction with Weak Supervision and Distributional Semantics

DTIC Science & Technology

2013-05-01

country is no longer a member of the organization), a player and an event, a team and a sport, etc. Multiple meanings of a relation phrase are success ...Zimbabwe, the Commonwealth> <force, country> <American forces, Vietnam>; <Roman Legions, Britain> < player , event> <Brandon Bass, the NBA draft>; <Agassi...training data. We found that dealing with incorrectly labeled examples is critical for its success . We develop a latent Bayesian framework for this

Statistically Derived Subtypes and Associations with Cerebrospinal Fluid and Genetic Biomarkers in Mild Cognitive Impairment: A Latent Profile Analysis.

PubMed

Eppig, Joel S; Edmonds, Emily C; Campbell, Laura; Sanderson-Cimino, Mark; Delano-Wood, Lisa; Bondi, Mark W

2017-08-01

Research demonstrates heterogeneous neuropsychological profiles among individuals with mild cognitive impairment (MCI). However, few studies have included visuoconstructional ability or used latent mixture modeling to statistically identify MCI subtypes. Therefore, we examined whether unique neuropsychological MCI profiles could be ascertained using latent profile analysis (LPA), and subsequently investigated cerebrospinal fluid (CSF) biomarkers, genotype, and longitudinal clinical outcomes between the empirically derived classes. A total of 806 participants diagnosed by means of the Alzheimer's Disease Neuroimaging Initiative (ADNI) MCI criteria received a comprehensive neuropsychological battery assessing visuoconstructional ability, language, attention/executive function, and episodic memory. Test scores were adjusted for demographic characteristics using standardized regression coefficients based on "robust" normal control performance (n=260). Calculated Z-scores were subsequently used in the LPA, and CSF-derived biomarkers, genotype, and longitudinal clinical outcome were evaluated between the LPA-derived MCI classes. Statistical fit indices suggested a 3-class model was the optimal LPA solution. The three-class LPA consisted of a mixed impairment MCI class (n=106), an amnestic MCI class (n=455), and an LPA-derived normal class (n=245). Additionally, the amnestic and mixed classes were more likely to be apolipoprotein e4+ and have worse Alzheimer's disease CSF biomarkers than LPA-derived normal subjects. Our study supports significant heterogeneity in MCI neuropsychological profiles using LPA and extends prior work (Edmonds et al., 2015) by demonstrating a lower rate of progression in the approximately one-third of ADNI MCI individuals who may represent "false-positive" diagnoses. Our results underscore the importance of using sensitive, actuarial methods for diagnosing MCI, as current diagnostic methods may be over-inclusive. (JINS, 2017, 23, 564-576).

What Doesn't Work for Whom? Exploring Heterogeneity in Responsiveness to the Family Check-Up in Early Childhood Using a Mixture Model Approach.

PubMed

Pelham, William E; Dishion, Thomas J; Tein, Jenn-Yun; Shaw, Daniel S; Wilson, Melvin N

2017-11-01

This study applied latent class analysis to a family-centered prevention trial in early childhood to identify subgroups of families with differential responsiveness to the Family Check-Up (FCU) intervention. The sample included 731 families with 2-year-olds randomized to the FCU or control condition and followed through age 5 with yearly follow-up assessments. A two-step mixture model was used to examine whether specific constellations of family characteristics at age 2 (baseline) were related to intervention response across ages 3, 4, and 5. The first step empirically identified latent classes of families based on several family risk and adjustment variables selected on the basis of previous research. The second step modeled the effect of the FCU on longitudinal change in children's problem behavior in each of the empirically derived latent classes. Results suggested a five-class solution, where a significant intervention effect of moderate to large size was observed in one of the five classes-the class characterized by child neglect, legal problems, and parental mental health issues. Pairwise comparisons revealed that the intervention effect was significantly greater in this class of families than in two other classes that were generally less at risk for the development of child disruptive behavior problems, albeit still low-income. Thus, findings suggest that (a) the FCU is most successful in reducing child problem behavior in more highly distressed, low-income families, and (b) the FCU may have little impact for relatively low-risk, low-income families. Future directions include the development of a brief screening process that can triage low-income families into groups that should be targeted for intervention, redirected to other services, monitored prospectively, or left alone.

Neuron’s eye view: Inferring features of complex stimuli from neural responses

PubMed Central

Chen, Xin; Beck, Jeffrey M.

2017-01-01

Experiments that study neural encoding of stimuli at the level of individual neurons typically choose a small set of features present in the world—contrast and luminance for vision, pitch and intensity for sound—and assemble a stimulus set that systematically varies along these dimensions. Subsequent analysis of neural responses to these stimuli typically focuses on regression models, with experimenter-controlled features as predictors and spike counts or firing rates as responses. Unfortunately, this approach requires knowledge in advance about the relevant features coded by a given population of neurons. For domains as complex as social interaction or natural movement, however, the relevant feature space is poorly understood, and an arbitrary a priori choice of features may give rise to confirmation bias. Here, we present a Bayesian model for exploratory data analysis that is capable of automatically identifying the features present in unstructured stimuli based solely on neuronal responses. Our approach is unique within the class of latent state space models of neural activity in that it assumes that firing rates of neurons are sensitive to multiple discrete time-varying features tied to the stimulus, each of which has Markov (or semi-Markov) dynamics. That is, we are modeling neural activity as driven by multiple simultaneous stimulus features rather than intrinsic neural dynamics. We derive a fast variational Bayesian inference algorithm and show that it correctly recovers hidden features in synthetic data, as well as ground-truth stimulus features in a prototypical neural dataset. To demonstrate the utility of the algorithm, we also apply it to cluster neural responses and demonstrate successful recovery of features corresponding to monkeys and faces in the image set. PMID:28827790

Bayesian Inference for the Stereotype Regression Model: Application to a Case-control Study of Prostate Cancer

PubMed Central

Ahn, Jaeil; Mukherjee, Bhramar; Banerjee, Mousumi; Cooney, Kathleen A.

2011-01-01

Summary The stereotype regression model for categorical outcomes, proposed by Anderson (1984) is nested between the baseline category logits and adjacent category logits model with proportional odds structure. The stereotype model is more parsimonious than the ordinary baseline-category (or multinomial logistic) model due to a product representation of the log odds-ratios in terms of a common parameter corresponding to each predictor and category specific scores. The model could be used for both ordered and unordered outcomes. For ordered outcomes, the stereotype model allows more flexibility than the popular proportional odds model in capturing highly subjective ordinal scaling which does not result from categorization of a single latent variable, but are inherently multidimensional in nature. As pointed out by Greenland (1994), an additional advantage of the stereotype model is that it provides unbiased and valid inference under outcome-stratified sampling as in case-control studies. In addition, for matched case-control studies, the stereotype model is amenable to classical conditional likelihood principle, whereas there is no reduction due to sufficiency under the proportional odds model. In spite of these attractive features, the model has been applied less, as there are issues with maximum likelihood estimation and likelihood based testing approaches due to non-linearity and lack of identifiability of the parameters. We present comprehensive Bayesian inference and model comparison procedure for this class of models as an alternative to the classical frequentist approach. We illustrate our methodology by analyzing data from The Flint Men’s Health Study, a case-control study of prostate cancer in African-American men aged 40 to 79 years. We use clinical staging of prostate cancer in terms of Tumors, Nodes and Metastatsis (TNM) as the categorical response of interest. PMID:19731262

Combination of five diagnostic tests to estimate the prevalence of hookworm infection among school-aged children from a rural area of colombia.

PubMed

Barreto, Rafael E; Narváez, Javier; Sepúlveda, Natalia A; Velásquez, Fabián C; Díaz, Sandra C; López, Myriam Consuelo; Reyes, Patricia; Moncada, Ligia I

2017-09-01

Public health programs for the control of soil-transmitted helminthiases require valid diagnostic tests for surveillance and parasitic control evaluation. However, there is currently no agreement about what test should be used as a gold standard for the diagnosis of hookworm infection. Still, in presence of concurrent data for multiple tests it is possible to use statistical models to estimate measures of test performance and prevalence. The aim of this study was to estimate the diagnostic accuracy of five parallel tests (direct microscopic examination, Kato-Katz, Harada-Mori, modified Ritchie-Frick, and culture in agar plate) to detect hookworm infections in a sample of school-aged children from a rural area in Colombia. We used both, a frequentist approach, and Bayesian latent class models to estimate the sensitivity and specificity of five tests for hookworm detection, and to estimate the prevalence of hookworm infection in absence of a Gold Standard. The Kato-Katz and agar plate methods had an overall agreement of 95% and kappa coefficient of 0.76. Different models estimated a sensitivity between 76% and 92% for the agar plate technique, and 52% to 87% for the Kato-Katz technique. The other tests had lower sensitivity. All tests had specificity between 95% and 98%. The prevalence estimated by the Kato-Katz and Agar plate methods for different subpopulations varied between 10% and 14%, and was consistent with the prevalence estimated from the combination of all tests. The Harada-Mori, Ritchie-Frick and direct examination techniques resulted in lower and disparate prevalence estimates. Bayesian approaches assuming imperfect specificity resulted in lower prevalence estimates than the frequentist approach. Copyright © 2017 Elsevier B.V. All rights reserved.

Field evaluation of three blood-based assays for elk (Cervus canadensis) naturally infected with Mycobacterium bovis.

PubMed

Shury, Todd K; Bergeson, Doug; Surujballi, Om; Lyashchenko, Konstantin P; Greenwald, Rena

2014-08-01

Diagnosis of Mycobacterium bovis in wild populations is very challenging due to complications imposed by the use of traditional skin tests, poor sensitivity of gold standard tests which rely on culture of M. bovis from tissues and wide variations in severity of disease. Various combinations of a lymphocyte stimulation test (LST), fluorescence polarization assay (FPA) and the Cervid TB Stat-Pak were evaluated using two different validation approaches: a latent class analysis and classical statistical approach using culture as a gold standard. A validation subsample consisting of animals culled for population control and mortalities from capture provided an unbiased estimate of test performance for comparison. The sensitivity of the LST (0.83, 95% CI: [0.70-0.97] as a single test was similar to existing tuberculin skin tests, but the sensitivity of the FPA (0.40, 95% CI: [0.22-0.58]) and Cervid TB Stat-Pak (0.62, 95% CI: [0.41-0.83]) were lower in this population. Test performance of the LST and Cervid TB Stat-Pak in parallel was similar to the use of all three tests in parallel and inclusion of the FPA did not greatly enhance test performance. Prevalence of M. bovis in elk varied substantially between the high risk area of southern Manitoba (9.1%, 95% CI: [6.09-12.1%]) and lower risk areas outside this zone (0.76%, 95% CI: [0-2.26%]). Bayesian latent class analysis indicated lack of covariance between the two antibody tests (FPA and Cervid TB Stat-Pak) while the classical two-stage analysis indicated there was conditional dependence between the tests. All three tests when used in parallel resulted in 100% NPV using all three validation methods, indicating few elk were misclassified as false negative by post mortem culture. Similar to previous studies, this study found that combinations of blood tests that utilize cell mediated responses along with humoral antibody responses maximize the sensitivity of tests for diagnosis of M. bovis in wild cervid populations. Copyright © 2013 CROWN COPYRIGHT and Elsevier B.V. Published by Elsevier B.V. All rights reserved.

Evaluating diagnostic tests for bovine tuberculosis in the southern part of Germany: A latent class analysis

PubMed Central

Knubben-Schweizer, Gabriela; Döpfer, Dörte; Groll, Andreas; Hafner-Marx, Angela; Hörmansdorfer, Stefan; Sauter-Louis, Carola; Straubinger, Reinhard K.; Zimmermann, Pia; Hartnack, Sonja

2017-01-01

Germany has been officially free of bovine tuberculosis since 1996. However, in the last years there has been an increase of bovine tuberculosis cases, particularly in the southern part of Germany, in the Allgäu region. As a consequence a one-time tuberculosis surveillance program was revisited with different premortal and postmortal tests. The aim of this paper was to estimate diagnostic sensitivities and specificities of the different tests used within this surveillance program. In the absence of a perfect test with 100% sensitivity and 100% specificity, thus in the absence of a gold standard, a Bayesian latent class approach with two different datasets was performed. The first dataset included 389 animals, tested with single intra-dermal comparative cervical tuberculin (SICCT) test, PCR and pathology; the second dataset contained 175 animals, tested with single intra-dermal cervical tuberculin (SICT) test, Bovigam® assay, pathology and culture. Two-way conditional dependencies were considered within the models. Additionally, inter-laboratory agreement (five officially approved laboratories) of the Bovigam® assay was assessed with Cohen's kappa test (21 blood samples). The results are given in posterior means and 95% credibility intervals. The specificities of the SICT test, SICCT test, PCR and pathology ranged between 75.8% [68.8–82.2%] and 99.0% [96.8–100%]. The Bovigam® assay stood out with a very low specificity (6.9% [3.6–11.1%]), though it had the highest sensitivity (95.7% [91.3–99.2%]). The sensitivities of the SICCT test, PCR, SICT test, pathology and culture varied from 57.8% [48.0–67.6%] to 88.9% [65.5–99.7%]. The prevalences were 19.8% [14.6–26.5%] (three-test dataset) and 7.7% [4.2–12.3%] (four-test dataset). Among all pairwise comparisons the highest agreement was 0.62 [0.15–1]). In conclusion, the specificity of the Bovigam® assay and the inter-laboratory agreement were lower than expected. PMID:28640908

Tuberculosis infection among young nursing trainees in South India.

PubMed

Christopher, Devasahayam J; Daley, Peter; Armstrong, Lois; James, Prince; Gupta, Richa; Premkumar, Beulah; Michael, Joy Sarojini; Radha, Vedha; Zwerling, Alice; Schiller, Ian; Dendukuri, Nandini; Pai, Madhukar

2010-04-29

Among healthcare workers in developing countries, nurses spend a large amount of time in direct contact with tuberculosis (TB) patients, and are at high risk for acquisition of TB infection and disease. To better understand the epidemiology of nosocomial TB among nurses, we recruited a cohort of young nursing trainees at Christian Medical College, a large, tertiary medical school hospital in Southern India. Among 535 nursing students enrolled in 2007, 468 gave consent to participate, and 436 underwent two-step tuberculin skin testing (TST). A majority (95%) were females, and almost 80% were under 22 years of age. Detailed TB exposure information was obtained using interviews and clinical log books. Prevalence of latent TB infection (LTBI) was estimated using Bayesian latent class analyses (LCA). Logistic regression analyses were done to determine the association between LTBI prevalence and TB exposure and risk factors. 219 of 436 students (50.2%, 95% CI: 45.4-55.0) were TST positive using the 10 mm or greater cut-off. Based on the LCA, the prevalence of LTBI was 47.8% (95% credible interval 17.8% to 65.6%). In the multivariate analysis, TST positivity was strongly associated with time spent in health care, after adjusting for age at entry into healthcare. Our study showed a high prevalence of LTBI even in young nursing trainees. With the recent TB infection control (TBIC) policy guidance from the World Health Organization as the reference, Indian healthcare providers and the Indian Revised National TB Control Programme will need to implement TBIC interventions, and enhance capacity for TBIC at the country level. Young trainees and nurses, in particular, will need to be targeted for TBIC interventions.

Discriminative Bayesian Dictionary Learning for Classification.

PubMed

Akhtar, Naveed; Shafait, Faisal; Mian, Ajmal

2016-12-01

We propose a Bayesian approach to learn discriminative dictionaries for sparse representation of data. The proposed approach infers probability distributions over the atoms of a discriminative dictionary using a finite approximation of Beta Process. It also computes sets of Bernoulli distributions that associate class labels to the learned dictionary atoms. This association signifies the selection probabilities of the dictionary atoms in the expansion of class-specific data. Furthermore, the non-parametric character of the proposed approach allows it to infer the correct size of the dictionary. We exploit the aforementioned Bernoulli distributions in separately learning a linear classifier. The classifier uses the same hierarchical Bayesian model as the dictionary, which we present along the analytical inference solution for Gibbs sampling. For classification, a test instance is first sparsely encoded over the learned dictionary and the codes are fed to the classifier. We performed experiments for face and action recognition; and object and scene-category classification using five public datasets and compared the results with state-of-the-art discriminative sparse representation approaches. Experiments show that the proposed Bayesian approach consistently outperforms the existing approaches.

The Co-occurrence of Substance Use and Bullying Behaviors among U.S. Adolescents: Understanding Demographic Characteristics and Social Influences

PubMed Central

Luk, Jeremy W.; Wang, Jing; Simons-Morton, Bruce G.

2012-01-01

This study examined the co-occurrence of subtypes of substance use and bullying behaviors using latent class analysis and evaluated latent class differences in demographic characteristics, peer and parental influences. Self-reported questionnaire data were collected from a nationally representative sample (N = 7508) of 6–10th grade adolescents in the United States. Four latent classes were identified: the non-involved (57.7%), substance users (19.4%), bullies (17.5%), and substance-using bullies (5.4%). Older and Hispanic adolescents were more likely to be substance users and substance-using bullies, whereas younger and African American adolescents were more likely to be bullies. Females were more likely to be substance users, whereas males were more likely to be bullies and substance-using bullies. Spending more evenings with peers posed greater risks for substance use, bullying, and the co-occurrence of both problem behaviors. Paternal knowledge exerted protective effects over-and-above the effects of maternal knowledge. Implications for prevention and intervention efforts are discussed. PMID:22698675

Application of Fractal theory for crash rate prediction: Insights from random parameters and latent class tobit models.

PubMed

Chand, Sai; Dixit, Vinayak V

2018-03-01

The repercussions from congestion and accidents on major highways can have significant negative impacts on the economy and environment. It is a primary objective of transport authorities to minimize the likelihood of these phenomena taking place, to improve safety and overall network performance. In this study, we use the Hurst Exponent metric from Fractal Theory, as a congestion indicator for crash-rate modeling. We analyze one month of traffic speed data at several monitor sites along the M4 motorway in Sydney, Australia and assess congestion patterns with the Hurst Exponent of speed (H speed ). Random Parameters and Latent Class Tobit models were estimated, to examine the effect of congestion on historical crash rates, while accounting for unobserved heterogeneity. Using a latent class modeling approach, the motorway sections were probabilistically classified into two segments, based on the presence of entry and exit ramps. This will allow transportation agencies to implement appropriate safety/traffic countermeasures when addressing accident hotspots or inadequately managed sections of motorway. Copyright © 2017 Elsevier Ltd. All rights reserved.

Literacy Profiles of At-Risk Young Adults Enrolled in Career and Technical Education

ERIC Educational Resources Information Center

Mellard, Daryl F.; Woods, Kari L.; Lee, Jae Hoon

2016-01-01

A latent profile analysis of 323 economically and academically at-risk adolescent and young adult learners yielded two classes: an average literacy class (92%) and a low literacy class (8%). The class profiles significantly differed in their word reading and math skills, and in their processing speeds and self-reported learning disabilities. The…

Pubertal Development and Prepubertal Height and Weight Jointly Predict Young Adult Height and Body Mass Index in a Prospective Study in South Africa.

PubMed

Stein, Aryeh D; Lundeen, Elizabeth A; Martorell, Reynaldo; Suchdev, Parminder S; Mehta, Neil K; Richter, Linda M; Norris, Shane A

2016-07-01

Height and adiposity track over childhood, but few studies, to our knowledge, have longitudinally examined the mediating relation of the timing and progression of puberty. We assessed interrelations between prepubertal height and body mass index, the progression through puberty, and young adult height and adiposity. We analyzed data from the Birth to Twenty Plus study (females, n = 823; males, n = 765). Serial measures of anthropometry and pubertal development were obtained between ages 9 and 16 y. We used latent class growth analysis to categorize pubertal development with respect to pubic hair (females and males), breasts (females), and genitalia (males) development. Adult height and weight were obtained at ages 18 to 20 y. Among females, higher latent class (earlier initiation and faster progression through puberty) was associated with an increased risk of obesity [pubic hair class 3 compared with class 1: RR, 3.41 (95% CI: 1.57, 7.44)] and inconsistent associations with height. Among males, higher latent class was associated with increased adult height [pubic hair development class 3 compared with class 1: 2.43 cm (95% CI: 0.88, 4.00)] and increased risk of overweight/obesity [pubic hair development class 3 compared with class 1: OR, 3.44 (95% CI: 1.44, 8.20)]. In females, the association with adult height became inverse after adjusting for prepubertal height [pubic hair development class 3 compared with class 1: females, -1.31 cm (95% CI: -2.32, -0.31)]; in males, the association with height was attenuated with this adjustment [-0.56 cm (95% CI: -1.63, 0.52)]. Associations with adiposity were attenuated after adjusting for prepubertal adiposity. Progression through puberty modifies the relation between prepubertal and adult anthropometry. Screening for early or rapid progression of puberty might identify children at an increased risk of becoming overweight or obese adults.

A Latent Class Analysis of Maternal Responsiveness and Autonomy-Granting in Early Adolescence: Prediction to Later Adolescent Sexual Risk-Taking

PubMed Central

Lanza, H. Isabella; Huang, David Y. C.; Murphy, Debra A.; Hser, Yih-Ing

2013-01-01

The present study sought to extend empirical inquiry related to the role of parenting on adolescent sexual risk-taking by using latent class analysis (LCA) to identify patterns of adolescent-reported mother responsiveness and autonomy-granting in early adolescence and examine associations with sexual risk-taking in mid- and late-adolescence. Utilizing a sample of 12- to 14-year-old adolescents (N = 4,743) from the 1997 National Longitudinal Survey of Youth (NLSY97), results identified a four-class model of maternal responsiveness and autonomy-granting: low responsiveness/high autonomy-granting, moderate responsiveness/moderate autonomy-granting, high responsiveness/low autonomy-granting, high responsiveness/moderate autonomy-granting. Membership in the low responsiveness/high autonomy-granting class predicted greater sexual risk-taking in mid- and late-adolescence compared to all other classes, and membership in the high responsiveness/ moderate autonomy-granting class predicted lower sexual risk-taking. Gender and ethnic differences in responsiveness and autonomy-granting class membership were also found, potentially informing gender and ethnic disparities of adolescent sexual risk-taking. PMID:23828712

The Four U's: Latent Classes of Hookup Motivations Among College Students

PubMed Central

Uecker, Jeremy E.; Pearce, Lisa D.; Andercheck, Brita

2016-01-01

College students’ “hookups” have been the subject of a great deal of research in recent years. Motivations for hooking up have been linked to differences in well-being after the hookup, but studies detailing college students’ motivations for engaging in hookups focus on single motivations. Using data from the 2010 Duke Hookup Survey, we consider how motivations for hooking up cluster to produce different classes, or profiles, of students who hook up, and how these classes are related to hookup regret. Four distinct classes of motivations emerged from our latent class analysis: Utilitarians (50%), Uninhibiteds (27%), Uninspireds (19%), and Unreflectives (4%). We find a number of differences in hookup motivation classes across social characteristics, including gender, year in school, race-ethnicity, self-esteem, and attitudes about sexual behavior outside committed relationships. Additionally, Uninspireds regret hookups more frequently than members of the other classes, and Uninhibiteds report regret less frequently than Utilitarians and Uninspireds. These findings reveal the complexity of motivations for hooking up and the link between motivations and regret. PMID:27066516

Prolonged grief and post-traumatic growth after loss: Latent class analysis.

PubMed

Zhou, Ningning; Yu, Wei; Tang, Suqin; Wang, Jianping; Killikelly, Clare

2018-06-06

Bereavement may trigger different psychological outcomes, such as prolonged grief disorder or post-traumatic growth. The relationship between these two outcomes and potential precipitators remain unknown. The current study aimed to identify classes of Chinese bereaved individuals based on prolonged grief symptoms and post-traumatic growth and to examine predictors for these classes. We used data from 273 Chinese individuals who lost a relative due to disease (92.3%), accident (4.4%) and other reasons (1.8%). Latent class analysis revealed three classes: a resilient class, a growth class, and a combined grief/growth class. A higher level of functional impairment was found for the combined grief/growth class than for the other two classes. Membership in the combined grief/growth class was significantly predicted by the younger age of the deceased and the death of a parent, child or spouse. Subjective closeness with the deceased and gender were marginally significant predictors. When the four variables were included in the multinomial regression analysis, death of a parent, child or spouse significantly predicted the membership to the combined grief/growth class. These findings provide valuable information for the development of tailored interventions that may build on the bereaved individuals' personal strengths. Copyright © 2018. Published by Elsevier B.V.

Paternal Work Stress and Latent Profiles of Father-Infant Parenting Quality

ERIC Educational Resources Information Center

Goodman, W. Benjamin; Crouter, Ann C.; Lanza, Stephanie T.; Cox, Martha J.; Vernon-Feagans, Lynne

2011-01-01

The current study used latent profile analysis (LPA) to examine the implications of fathers' experiences of work stress for paternal behaviors with infants across multiple dimensions of parenting in a sample of fathers living in nonmetropolitan communities (N = 492). LPA revealed five classes of fathers based on levels of social-affective…

Development of Fraction Comparison Strategies: A Latent Transition Analysis

ERIC Educational Resources Information Center

Rinne, Luke F.; Ye, Ai; Jordan, Nancy C.

2017-01-01

The present study investigated the development of fraction comparison strategies through a longitudinal analysis of children's responses to a fraction comparison task in 4th through 6th grades (N = 394). Participants were asked to choose the larger value for 24 fraction pairs blocked by fraction type. Latent class analysis of performance over item…

Investigating Approaches to Estimating Covariate Effects in Growth Mixture Modeling: A Simulation Study

ERIC Educational Resources Information Center

Li, Ming; Harring, Jeffrey R.

2017-01-01

Researchers continue to be interested in efficient, accurate methods of estimating coefficients of covariates in mixture modeling. Including covariates related to the latent class analysis not only may improve the ability of the mixture model to clearly differentiate between subjects but also makes interpretation of latent group membership more…

Comparing Latent Structures of the Grade of Membership, Rasch, and Latent Class Models

ERIC Educational Resources Information Center

Erosheva, Elena A.

2005-01-01

This paper focuses on model interpretation issues and employs a geometric approach to compare the potential value of using the Grade of Membership (GoM) model in representing population heterogeneity. We consider population heterogeneity manifolds generated by letting subject specific parameters vary over their natural range, while keeping other…

Heterogeneity in the Latent Structure of PTSD Symptoms among Canadian Veterans

ERIC Educational Resources Information Center

Naifeh, James A.; Richardson, J. Don; Del Ben, Kevin S.; Elhai, Jon D.

2010-01-01

The current study used factor mixture modeling to identify heterogeneity (i.e., latent classes) in 2 well-supported models of posttraumatic stress disorder's (PTSD) factor structure. Data were analyzed from a clinical sample of 405 Canadian veterans evaluated for PTSD. Results were consistent with our hypotheses. Each PTSD factor model was best…

Optimal Scaling of HIV-Related Sexual Risk Behaviors in Ethnically Diverse Homosexually Active Men.

ERIC Educational Resources Information Center

Cochran, Susan D.; And Others

1995-01-01

Used homogeneity analysis and latent class analysis to analyze sexual behavior patterns in two samples of homosexually active men. Results support the existence of a single, nonlinear, latent dimension underlying male homosexual behaviors consistent with HIV-related risk taking, providing an efficient means to scale sexual behavior patterns. (RJM)

Assigning Cases to Groups Using Taxometric Results: An Empirical Comparison of Classification Techniques

ERIC Educational Resources Information Center

Ruscio, John

2009-01-01

Determining whether individuals belong to different latent classes (taxa) or vary along one or more latent factors (dimensions) has implications for assessment. For example, no instrument can simultaneously maximize the efficiency of categorical and continuous measurement. Methods such as taxometric analysis can test the relative fit of taxonic…

Comparisons of polydrug use at national and inner city levels in England: associations with demographic and socioeconomic factors.

PubMed

Carter, Jennifer L; Strang, John; Frissa, Souci; Hayes, Richard D; Hatch, Stephani L; Hotopf, Matthew

2013-10-01

This study compares polydrug use in national and inner city samples to (1) examine patterns of use underlying different prevalence rates and (2) identify how inner city polydrug use needs targeting in ways not suggested by national research. Latent class analyses on indicators of illicit drug use in the last year, hazardous alcohol use, and cigarette smoking were compared between the inner city 2008-2010 South East London Community Health study (n = 1698) and the nationally representative 2007 Adult Psychiatric Morbidity Survey in England (n = 7403). Multinomial logistic regressions then examined latent class solutions with demographic and socioeconomic factors. Both samples revealed three notably similar classes of polydrug users: a "high-drug" group using multiple substances; a "moderate-drug" group using cannabis, alcohol, and cigarettes; and a "low-drug" group reporting minimal alcohol and cigarette use. However, South East London Community Health reported lower risks of polydrug use for ethnic minorities but not for more educated participants. Despite higher polydrug use prevalence in the inner city, latent classes of polydrug users were similar between samples. Some demographic and socioeconomic factors differed between the samples, suggesting the need for inner city services to use both local and national data for policy planning. Copyright © 2013 Elsevier Inc. All rights reserved.

Attachment in psychosis: A latent profile analysis of attachment styles and association with symptoms in a large psychosis cohort.

PubMed

Bucci, Sandra; Emsley, Richard; Berry, Katherine

2017-01-01

Attachment has been identified as one of various possible mechanisms involved in understanding models of psychosis, but measures that reliably and validly assess attachment styles in psychosis are limited. The aim of this study was to identify attachment patterns in psychosis and examine demographic and clinical correlates across attachment groups. Latent profile analysis on attachment data from 588 participants who met criteria for non-affective psychosis was used to classify people into attachment classes. Four latent classes of attachment were identified: secure, insecure-anxious, insecure-avoidant and disorganised. Secure attachment was the most common attachment style, suggesting that a significant number of clients with psychosis are inherently resilient. Disorganised attachment was associated with a higher proportion of sexual and physical abuse and more severe positive symptoms compared to other attachment classes. This is not only the largest study to examine attachment styles, their demographic and clinical profile, and the clinical profile of disorganised attachment more specifically, in psychosis, but also the first study to use a validated self-report measure of attachment in psychosis to identify four classes of attachment style. Findings advance developmental models of attachment and psychosis; participants with disorganised attachment report more frequent trauma history and more severe psychotic symptoms. Copyright © 2016 Elsevier Ireland Ltd. All rights reserved.

Demographic analysis from summaries of an age-structured population

USGS Publications Warehouse

Link, William A.; Royle, J. Andrew; Hatfield, Jeff S.

2003-01-01

Demographic analyses of age-structured populations typically rely on life history data for individuals, or when individual animals are not identified, on information about the numbers of individuals in each age class through time. While it is usually difficult to determine the age class of a randomly encountered individual, it is often the case that the individual can be readily and reliably assigned to one of a set of age classes. For example, it is often possible to distinguish first-year from older birds. In such cases, the population age structure can be regarded as a latent variable governed by a process prior, and the data as summaries of this latent structure. In this article, we consider the problem of uncovering the latent structure and estimating process parameters from summaries of age class information. We present a demographic analysis for the critically endangered migratory population of whooping cranes (Grus americana), based only on counts of first-year birds and of older birds. We estimate age and year-specific survival rates. We address the controversial issue of whether management action on the breeding grounds has influenced recruitment, relating recruitment rates to the number of seventh-year and older birds, and examining the pattern of variation through time in this rate.

Latent classes of resilience and psychological response among only-child loss parents in China.

PubMed

Wang, An-Ni; Zhang, Wen; Zhang, Jing-Ping; Huang, Fei-Fei; Ye, Man; Yao, Shu-Yu; Luo, Yuan-Hui; Li, Zhi-Hua; Zhang, Jie; Su, Pan

2017-10-01

Only-child loss parents in China recently gained extensive attention as a newly defined social group. Resilience could be a probable solution out of the psychological dilemma. Using a sample of 185 only-child loss people, this study employed latent class analysis (a) to explore whether different classes of resilience could be identified, (b) to determine socio-demographic characteristics of each class, and (c) to compare the depression and the subjective well-being of each class. The results supported a three-class solution, defined as 'high tenacity-strength but moderate optimism class', 'moderate resilience but low self-efficacy class' and 'low tenacity but moderate adaption-dependence class'. Parents with low income and medical insurance of low reimbursement type and without endowment insurance occupied more proportions in the latter two classes. The latter two classes also had a significant higher depression scores and lower subjective well-being scores than high tenacity-strength but moderate optimism class. Future work should care those socio-economically vulnerable bereaved parents, and an elastic economic assistance policy was needed. To develop targeted resilience interventions, the emphasis of high tenacity-strength but moderate optimism class should be the optimism. Moderate resilience but low self-efficacy class should be self-efficacy, and low tenacity but moderate adaption-dependence class should be tenacity. Copyright © 2016 John Wiley & Sons, Ltd.

Topics in Bayesian Hierarchical Modeling and its Monte Carlo Computations

NASA Astrophysics Data System (ADS)

Tak, Hyung Suk

The first chapter addresses a Beta-Binomial-Logit model that is a Beta-Binomial conjugate hierarchical model with covariate information incorporated via a logistic regression. Various researchers in the literature have unknowingly used improper posterior distributions or have given incorrect statements about posterior propriety because checking posterior propriety can be challenging due to the complicated functional form of a Beta-Binomial-Logit model. We derive data-dependent necessary and sufficient conditions for posterior propriety within a class of hyper-prior distributions that encompass those used in previous studies. Frequency coverage properties of several hyper-prior distributions are also investigated to see when and whether Bayesian interval estimates of random effects meet their nominal confidence levels. The second chapter deals with a time delay estimation problem in astrophysics. When the gravitational field of an intervening galaxy between a quasar and the Earth is strong enough to split light into two or more images, the time delay is defined as the difference between their travel times. The time delay can be used to constrain cosmological parameters and can be inferred from the time series of brightness data of each image. To estimate the time delay, we construct a Gaussian hierarchical model based on a state-space representation for irregularly observed time series generated by a latent continuous-time Ornstein-Uhlenbeck process. Our Bayesian approach jointly infers model parameters via a Gibbs sampler. We also introduce a profile likelihood of the time delay as an approximation of its marginal posterior distribution. The last chapter specifies a repelling-attracting Metropolis algorithm, a new Markov chain Monte Carlo method to explore multi-modal distributions in a simple and fast manner. This algorithm is essentially a Metropolis-Hastings algorithm with a proposal that consists of a downhill move in density that aims to make local modes repelling, followed by an uphill move in density that aims to make local modes attracting. The downhill move is achieved via a reciprocal Metropolis ratio so that the algorithm prefers downward movement. The uphill move does the opposite using the standard Metropolis ratio which prefers upward movement. This down-up movement in density increases the probability of a proposed move to a different mode.

Comparison of Suicide Attempters and Decedents in the U.S. Army: A Latent Class Analysis.

PubMed

Skopp, Nancy A; Smolenski, Derek J; Sheppard, Sean C; Bush, Nigel E; Luxton, David D

2016-08-01

A clearer understanding of risk factors for suicidal behavior among soldiers is of principal importance to military suicide prevention. It is unclear whether soldiers who attempt suicide and those who die by suicide have different patterns of risk factors. As such, preventive efforts aimed toward reducing suicide attempts and suicides, respectively, may require different strategies. We conducted a latent class analysis (LCA) to examine classes of risk factors among suicide attempters (n = 1,433) and decedents (n = 424). Both groups were represented by three classes: (1) External/Antisocial Risk Factors, (2) Mental Health Risk Factors, and (3) No Pattern. These findings support the conceptualization that military suicide attempters and decedents represent a single population. Published 2016. This article is a U.S. Government work and is in the public domain in the USA.

Parametric embedding for class visualization.

PubMed

Iwata, Tomoharu; Saito, Kazumi; Ueda, Naonori; Stromsten, Sean; Griffiths, Thomas L; Tenenbaum, Joshua B

2007-09-01

We propose a new method, parametric embedding (PE), that embeds objects with the class structure into a low-dimensional visualization space. PE takes as input a set of class conditional probabilities for given data points and tries to preserve the structure in an embedding space by minimizing a sum of Kullback-Leibler divergences, under the assumption that samples are generated by a gaussian mixture with equal covariances in the embedding space. PE has many potential uses depending on the source of the input data, providing insight into the classifier's behavior in supervised, semisupervised, and unsupervised settings. The PE algorithm has a computational advantage over conventional embedding methods based on pairwise object relations since its complexity scales with the product of the number of objects and the number of classes. We demonstrate PE by visualizing supervised categorization of Web pages, semisupervised categorization of digits, and the relations of words and latent topics found by an unsupervised algorithm, latent Dirichlet allocation.

[Poverty profile regarding households participating in a food assistance program].

PubMed

Álvarez-Uribe, Martha C; Aguirre-Acevedo, Daniel C

2012-06-01

This study was aimed at establishing subgroups having specific socioeconomic characteristics by using latent class analysis as a method for segmenting target population members of the MANA-ICBF supplementary food program in the Antioquia department of Colombia and determine their differences regarding poverty and health conditions in efficiently addressing pertinent resources, programs and policies. The target population consisted of 200,000 children and their households involved in the MANA food assistance program; a representative sample by region was used. Latent class analysis was used, as were the expectation-maximization and Newton Raphson algorithms for identifying the appropriate number of classes. The final model classified the households into four clusters or classes, differing according to well-defined socio-demographic conditions affecting children's health. Some homes had a greater depth of poverty, therefore lowering the families' quality of life and affecting the health of the children in this age group.

Three-year latent class trajectories of attention-deficit/hyperactivity disorder (ADHD) symptoms in a clinical sample not selected for ADHD.

PubMed

Arnold, L Eugene; Ganocy, Stephen J; Mount, Katherine; Youngstrom, Eric A; Frazier, Thomas; Fristad, Mary; Horwitz, Sarah M; Birmaher, Boris; Findling, Robert; Kowatch, Robert A; Demeter, Christine; Axelson, David; Gill, Mary Kay; Marsh, Linda

2014-07-01

This study aims to examine trajectories of attention-deficit/hyperactivity disorder (ADHD) symptoms in the Longitudinal Assessment of Manic Symptoms (LAMS) sample. The LAMS study assessed 684 children aged 6 to 12 years with the Kiddie-Schedule for Affective Disorders and Schizophrenia (K-SADS) and rating scales semi-annually for 3 years. Although they were selected for elevated manic symptoms, 526 children had baseline ADHD diagnoses. With growth mixture modeling (GMM), we separately analyzed inattentive and hyperactive/impulsive symptoms, covarying baseline age. Multiple standard methods determined optimal fit. The χ(2) and Kruskal-Wallis analysis of variance compared resulting latent classes/trajectories on clinical characteristics and medication. Three latent class trajectories best described inattentive symptoms, and 4 classes best described hyperactive/impulsive symptoms. Inattentive trajectories maintained their relative position over time. Hyperactive/impulsive symptoms had 2 consistent trajectories (least and most severe). A third trajectory (4.5%) started mild, then escalated; and a fourth (14%) started severe but improved dramatically. The improving trajectory was associated with the highest rate of ADHD and lowest rate of bipolar diagnoses. Three-fourths of the mildest inattention class were also in the mildest hyperactive/impulsive class; 72% of the severest inattentive class were in the severest hyperactive/impulsive class, but the severest inattention class also included 62% of the improving hyperactive-impulsive class. An ADHD rather than bipolar diagnosis prognosticates a better course of hyperactive/impulsive, but not inattentive, symptoms. High overlap of relative severity between inattention and hyperactivity/impulsivity confirms the link between these symptom clusters. Hyperactive/impulsive symptoms wane more over time. Group means are insufficient to understand individual ADHD prognosis. A small subgroup deteriorates over time in hyperactivity/impulsivity and needs better treatments than currently provided. Copyright © 2014 American Academy of Child and Adolescent Psychiatry. Published by Elsevier Inc. All rights reserved.

Heterogeneity of postpartum depression: a latent class analysis

PubMed Central

2016-01-01

Summary Background Maternal depression in the postpartum period confers substantial morbidity and mortality, but the definition of postpartum depression remains controversial. We investigated the heterogeneity of symptoms with the aim of identifying clinical subtypes of postpartum depression. Methods Data were aggregated from the international perinatal psychiatry consortium Postpartum Depression: Action Towards Causes and Treatment, which represents 19 institutions in seven countries. 17 912 unique subject records with phenotypic data were submitted. We applied latent class analyses in a two-tiered approach to assess the validity of empirically defined subtypes of postpartum depression. Tier one assessed heterogeneity in women with complete data on the Edinburgh postnatal depression scale (EPDS) and tier two in those with postpartum depression case status. Findings 6556 individuals were assessed in tier one and 4245 in tier two. A final model with three latent classes was optimum for both tiers. The most striking characteristics associated with postpartum depression were severity, timing of onset, comorbid anxiety, and suicidal ideation. Women in class 1 had the least severe symptoms (mean EPDS score 10·5), followed by those in class 2 (mean EPDS score 14·8) and those in class 3 (mean EPDS score 20·1). The most severe symptoms of postpartum depression were significantly associated with poor mood (mean EPDS score 20·1), increased anxiety, onset of symptoms during pregnancy, obstetric complications, and suicidal ideation. In class 2, most women (62%) reported symptom onset within 4 weeks postpartum and had more pregnancy complications than in other two classes (69% vs 67% in class 1 and 29% in class 3). Interpretation PPD seems to have several distinct phenotypes. Further assessment of PPD heterogeneity to identify more precise phenotypes will be important for future biological and genetic investigations. Funding Sources of funding are listed at the end of the article. PMID:26359613

Adolescent stalking and risk of violence.

PubMed

Smith-Darden, Joanne P; Reidy, Dennis E; Kernsmith, Poco D

2016-10-01

Stalking perpetration and the associated risk for violence among adolescents has generally been neglected. In the present study, 1236 youth completed surveys assessing empirically established stalking indicators, threats and aggression toward stalking victims, dating violence, and violent delinquency. Latent Profile Analysis identified 3 latent classes of boys: non-perpetrators (NP), hyper-intimate pursuit (HIP), and comprehensive stalking perpetrators (CSP) and, and 2 classes for girls: NP and HIP. Boys in the CSP class were the most violent youth on nearly all indices with boys in the HIP class demonstrating an intermediate level of violence compared to NP boys. Girls in the HIP class were more violent than NP girls on all indices. These findings suggest stalking in adolescence merits attention by violence prevention experts. In particular, juvenile stalking may signify youth at risk for multiple forms of violence perpetrated against multiple types of victims, not just the object of their infatuation. Copyright © 2016 The Foundation for Professionals in Services for Adolescents. Published by Elsevier Ltd. All rights reserved.

Identifying Gender-Specific Developmental Trajectories of Nonviolent and Violent Delinquency from Adolescence to Young Adulthood

PubMed Central

Zheng, Yao; Cleveland, H. Harrington

2013-01-01

Most research examining gender differences in developmental trajectories of antisocial behavior does not consider subtypes of antisocial behavior and is difficult to generalize due to small nonrepresentative samples. The current study investigated gender difference in developmental trajectories from adolescence to young adulthood while addressing those limitations. Analyses were limited to respondents ages 15 and 16 in wave 1 (16–17 in wave 2, and 21–22 in wave 3) of the National Longitudinal Study of Adolescent Health (n = 6244, 49.5% males). Self-report nonviolent and violent delinquencies were simultaneously entered into latent class analysis. Four latent classes were identified: low, desister, decliner, and chronic (male-only). In addition to finding a male-specific chronic class, gender differences included differences in levels of nonviolent and violent delinquency between synonymous classes of males and females, and differences in prevalence of classes across genders. Neighborhood disadvantage and family support predicted trajectories. PMID:23375843

Effect of Alzheimer's disease risk and protective factors on cognitive trajectories in subjective memory complainers.

PubMed

Teipel, Stefan J; Cavedo, Enrica; Lista, Simone; Habert, Marie-Odile; Potier, Marie-Claude; Grothe, Michel J; Epelbaum, Stephane; Sambati, Luisa; Gagliardi, Geoffroy; Toschi, Nicola; Greicius, Michael; Dubois, Bruno; Hampel, Harald

2018-05-21

Cognitive change in people at risk of Alzheimer's disease (AD) such as subjective memory complainers is highly variable across individuals. We used latent class growth modeling to identify distinct classes of nonlinear trajectories of cognitive change over 2 years follow-up from 265 subjective memory complainers individuals (age 70 years and older) of the INSIGHT-preAD cohort. We determined the effect of cortical amyloid load, hippocampus and basal forebrain volumes, and education on the cognitive trajectory classes. Latent class growth modeling identified distinct nonlinear cognitive trajectories. Education was associated with higher performing trajectories, whereas global amyloid load and basal forebrain atrophy were associated with lower performing trajectories. Distinct classes of cognitive trajectories were associated with risk and protective factors of AD. These associations support the notion that the identified cognitive trajectories reflect different risk for AD that may be useful for selecting high-risk individuals for intervention trials. Copyright © 2018. Published by Elsevier Inc.

Examining the Association between Patient-Reported Symptoms of Attention and Memory Dysfunction with Objective Cognitive Performance: A Latent Regression Rasch Model Approach.

PubMed

Li, Yuelin; Root, James C; Atkinson, Thomas M; Ahles, Tim A

2016-06-01

Patient-reported cognition generally exhibits poor concordance with objectively assessed cognitive performance. In this article, we introduce latent regression Rasch modeling and provide a step-by-step tutorial for applying Rasch methods as an alternative to traditional correlation to better clarify the relationship of self-report and objective cognitive performance. An example analysis using these methods is also included. Introduction to latent regression Rasch modeling is provided together with a tutorial on implementing it using the JAGS programming language for the Bayesian posterior parameter estimates. In an example analysis, data from a longitudinal neurocognitive outcomes study of 132 breast cancer patients and 45 non-cancer matched controls that included self-report and objective performance measures pre- and post-treatment were analyzed using both conventional and latent regression Rasch model approaches. Consistent with previous research, conventional analysis and correlations between neurocognitive decline and self-reported problems were generally near zero. In contrast, application of latent regression Rasch modeling found statistically reliable associations between objective attention and processing speed measures with self-reported Attention and Memory scores. Latent regression Rasch modeling, together with correlation of specific self-reported cognitive domains with neurocognitive measures, helps to clarify the relationship of self-report with objective performance. While the majority of patients attribute their cognitive difficulties to memory decline, the Rash modeling suggests the importance of processing speed and initial learning. To encourage the use of this method, a step-by-step guide and programming language for implementation is provided. Implications of this method in cognitive outcomes research are discussed. © The Author 2016. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Replication-Competent Simian Immunodeficiency Virus (SIV) Gag Escape Mutations Archived in Latent Reservoirs during Antiretroviral Treatment of SIV-Infected Macaques▿

PubMed Central

Queen, Suzanne E.; Mears, Brian M.; Kelly, Kathleen M.; Dorsey, Jamie L.; Liao, Zhaohao; Dinoso, Jason B.; Gama, Lucio; Adams, Robert J.; Zink, M. Christine; Clements, Janice E.; Kent, Stephen J.; Mankowski, Joseph L.

2011-01-01

In response to pressure exerted by major histocompatibility complex (MHC) class I-mediated CD8+ T cell control, human immunodeficiency virus (HIV) escape mutations often arise in immunodominant epitopes recognized by MHC class I alleles. While the current standard of care for HIV-infected patients is treatment with highly active antiretroviral therapy (HAART), suppression of viral replication in these patients is not absolute and latently infected cells persist as lifelong reservoirs. To determine whether HIV escape from MHC class I-restricted CD8+ T cell control develops during HAART treatment and then enters latent reservoirs in the periphery and central nervous system (CNS), with the potential to emerge as replication-competent virus, we tracked the longitudinal development of the simian immunodeficiency virus (SIV) Gag escape mutation K165R in HAART-treated SIV-infected pigtailed macaques. Key findings of these studies included: (i) SIV Gag K165R escape mutations emerged in both plasma and cerebrospinal fluid (CSF) during the decaying phase of viremia after HAART initiation before suppression of viral replication, (ii) SIV K165R Gag escape mutations were archived in latent proviral DNA reservoirs, including the brain in animals receiving HAART that suppressed viral replication, and (iii) replication-competent SIV Gag K165R escape mutations were present in the resting CD4+ T cell reservoir in HAART-treated SIV-infected macaques. Despite early administration of aggressive antiretroviral treatment, HIV immune escape from CD8+ T cell control can still develop during the decaying phases of viremia and then persist in latent reservoirs, including the brain, with the potential to emerge if HAART therapy is interrupted. PMID:21715484

Are Student Evaluations of Teaching Effectiveness Valid for Measuring Student Learning Outcomes in Business Related Classes? A Neural Network and Bayesian Analyses

ERIC Educational Resources Information Center

Galbraith, Craig S.; Merrill, Gregory B.; Kline, Doug M.

2012-01-01

In this study we investigate the underlying relational structure between student evaluations of teaching effectiveness (SETEs) and achievement of student learning outcomes in 116 business related courses. Utilizing traditional statistical techniques, a neural network analysis and a Bayesian data reduction and classification algorithm, we find…

Typologies of sexually explicit media use among MSM: An application of latent class analysis

PubMed Central

Erickson, Darin J.; Galos, Dylan L; Smolenski, Derek, J.; Iantaffi, Alex; Rosser, B.R. Simon

2014-01-01

The viewing of sexually explicit media (SEM) is widespread, especially among men, and research linking SEM viewing and sexual behavior has shown a variety of results, some positive (e.g., sexuality education) and some negative (e.g., poorer body image). These results might be due to limitations in measuring SEM consumption, particularly around typology. The goal of the current study was to examine potential patterns of SEM viewing activities. Using data from an online survey of men who have sex with men (MSM), we conducted latent class analyses of 15 SEM activities. Results suggested a three-class solution. The most prevalent class included over 60% of men and was characterized by viewing primarily safer-sex or conventional behaviors. The second class included 32% of men and had a similar albeit amplified pattern of viewing. The final class included just 7% of men and was marked by high levels of viewing of all activities, including fetish and kink. Compared to the conventional or safer-sex class, the other classes had lower internalized homonegativity, lower condom use self-efficacy, and higher SEM consumption or dose. Implications for HIV prevention, sexuality research and the SEM industry are discussed. PMID:25642301

Clustering of adversity in young adults on disability pension due to mental disorders: a latent class analysis.

PubMed

Joensuu, Matti; Mattila-Holappa, Pauliina; Ahola, Kirsi; Ervasti, Jenni; Kivimäki, Mika; Kivekäs, Teija; Koskinen, Aki; Vahtera, Jussi; Virtanen, Marianna

2016-02-01

Mental disorders are the leading cause of work disability among young adults. This study examined whether distinct classes could be identified among young adults on the basis of medical history before receiving a disability pension due to a mental disorder. Medical history was obtained from pension applications and attached medical certificates for 1163 individuals aged 18-34 years who, in 2008, received a disability pension due to a mental disorder. Using latent class analysis, 10 clinical and individual adversities and their associations with sex, age and diagnostic category were examined. Three classes were identified: childhood adversity (prevalence, 33%), comorbidity (23%), and undefined (44%). The childhood adversity class was characterized by adverse events and symptoms reported during childhood and it associated with depressive disorders. The comorbidity class was characterized by comorbid mental disorders, suicide attempts and substance abuse and associated with younger age and bipolar disorder. The undefined class formed no distinct profile; individuals in this class had the lowest number of adversities and it associated with psychotic disorders. The identification of subgroups characterized by childhood circumstances and comorbidity may help planning of prevention and support practices for young adults with mental disorders and risk of work disability.

Bayesian Hierarchical Classes Analysis

ERIC Educational Resources Information Center

Leenen, Iwin; Van Mechelen, Iven; Gelman, Andrew; De Knop, Stijn

2008-01-01

Hierarchical classes models are models for "N"-way "N"-mode data that represent the association among the "N" modes and simultaneously yield, for each mode, a hierarchical classification of its elements. In this paper we present a stochastic extension of the hierarchical classes model for two-way two-mode binary data. In line with the original…

Latent Growth Classes of Alcohol-Related Blackouts over the First Two Years of College

PubMed Central

Merrill, Jennifer E.; Treloar, Hayley; Fernandez, Anne C.; Monnig, Mollie A.; Jackson, Kristina M.; Barnett, Nancy P.

2016-01-01

Alcohol-related blackouts are common among college student drinkers. The present study extends prior work by examining latent growth classes of blackouts and several predictors of class membership. Participants (N=709 college drinkers) completed a baseline survey at college entry and biweekly online assessments throughout freshman and sophomore years. Results revealed five latent growth class trajectories, reflecting varying experiences of blackouts at the beginning of college and differential change in blackouts over time. The largest class represented a relatively low risk group (LOW DECR; 47.3%) characterized by endorsement of no or very low likelihood of blackouts, and decreasing likelihood of blackouts over time. Another decreasing risk group (HIGH DECR; 11.1%) initially reported a high proportion of blackouts and had the steepest decrease in blackout risk over time. A small percentage showed consistently high likelihood of blackouts over time (HIGH STABLE; 4.1%). The remaining two groups were distinguished by relatively moderate (MOD STABLE; 14.9%) and lower (LOW STABLE; 22.6%) likelihood of blackouts, which remained stable over time. Comparisons between classes revealed that students with greater perceived peer drinking, perceived peer approval of drinking, and enhancement motives upon entry to college tended to be in higher-risk groups with consistent experiences of blackouts over time, whereas blackout likelihood decreased over time for students with greater conformity motives. Findings suggest that pre-college preventive interventions may be strengthened by considering not only factors related to current risk for blackouts and other alcohol-related consequences, but also those factors related to persistence of these behaviors over time. PMID:27736145

Identification of Heterogeneous Cognitive Subgroups in Community-Dwelling Older Adults: A Latent Class Analysis of the Einstein Aging Study.

PubMed

Zammit, Andrea R; Hall, Charles B; Lipton, Richard B; Katz, Mindy J; Muniz-Terrera, Graciela

2018-05-01

The aim of this study was to identify natural subgroups of older adults based on cognitive performance, and to establish each subgroup's characteristics based on demographic factors, physical function, psychosocial well-being, and comorbidity. We applied latent class (LC) modeling to identify subgroups in baseline assessments of 1345 Einstein Aging Study (EAS) participants free of dementia. The EAS is a community-dwelling cohort study of 70+ year-old adults living in the Bronx, NY. We used 10 neurocognitive tests and 3 covariates (age, sex, education) to identify latent subgroups. We used goodness-of-fit statistics to identify the optimal class solution and assess model adequacy. We also validated our model using two-fold split-half cross-validation. The sample had a mean age of 78.0 (SD=5.4) and a mean of 13.6 years of education (SD=3.5). A 9-class solution based on cognitive performance at baseline was the best-fitting model. We characterized the 9 identified classes as (i) disadvantaged, (ii) poor language, (iii) poor episodic memory and fluency, (iv) poor processing speed and executive function, (v) low average, (vi) high average, (vii) average, (viii) poor executive and poor working memory, (ix) elite. The cross validation indicated stable class assignment with the exception of the average and high average classes. LC modeling in a community sample of older adults revealed 9 cognitive subgroups. Assignment of subgroups was reliable and associated with external validators. Future work will test the predictive validity of these groups for outcomes such as Alzheimer's disease, vascular dementia and death, as well as markers of biological pathways that contribute to cognitive decline. (JINS, 2018, 24, 511-523).

Medical Complexity among Children with Special Health Care Needs: A Two-Dimensional View.

PubMed

Coller, Ryan J; Lerner, Carlos F; Eickhoff, Jens C; Klitzner, Thomas S; Sklansky, Daniel J; Ehlenbach, Mary; Chung, Paul J

2016-08-01

To identify subgroups of U.S. children with special health care needs (CSHCN) and characterize key outcomes. Secondary analysis of 2009-2010 National Survey of CSHCN. Latent class analysis grouped individuals into substantively meaningful classes empirically derived from measures of pediatric medical complexity. Outcomes were compared among latent classes with weighted logistic or negative binomial regression. LCA identified four unique CSHCN subgroups: broad functional impairment (physical, cognitive, and mental health) with extensive health care (Class 1), broad functional impairment alone (Class 2), predominant physical impairment requiring family-delivered care (Class 3), and physical impairment alone (Class 4). CSHCN from Class 1 had the highest ED visit rates (IRR 3.3, p < .001) and hospitalization odds (AOR: 12.0, p < .001) and lowest odds of a medical home (AOR: 0.17, p < .001). CSHCN in Class 3, despite experiencing more shared decision making and medical home attributes, had more ED visits and missed school than CSHCN in Class 2 (p < .001); the latter, however, experienced more cost-related difficulties, care delays, and parents having to stop work (p < .001). Recognizing distinct impacts of cognitive and mental health impairments and health care delivery needs on CSHCN outcomes may better direct future intervention efforts. © Health Research and Educational Trust.

A Latent Class Analysis of Weight-Related Health Behaviors among 2- and 4-year College Students, and Associated Risk of Obesity

PubMed Central

Mathur, C; Stigler, M; Lust, K; Laska, M

2016-01-01

Little is known about the complex patterning of weight-related health behaviors in 2- and 4-year college students. The objective of this study was to identify and describe unique classes of weight-related health behaviors among college youth. Latent class analysis was used to identify homogenous, mutually exclusive classes of nine health behaviors which represent multiple theoretically/clinically relevant dimensions of obesity risk among 2- versus 4-year college students using cross-sectional statewide surveillance data (n= 17,584). Additionally, differences in class membership on selected sociodemographic characteristics were examined using a model-based approach. Analysis was conducted separately for both college groups, and 5 and 4 classes were identified for 2-and 4-year college students, respectively. Four classes were similar across 2-and 4-year college groups and were characterized as “mostly healthy dietary habits, active”, “moderately high screen time, active”, “moderately healthy dietary habits, inactive”, and “moderately high screen time, inactive”. “Moderately healthy dietary habits, high screen time” was the additional class unique to 2-year college students. These classes differed on a number of sociodemographic characteristics, including the proportion in each class who were classified as obese. Implications for prevention scientists and future intervention programs are considered. PMID:24990599

A latent class analysis of weight-related health behaviors among 2- and 4-year college students and associated risk of obesity.

PubMed

Mathur, Charu; Stigler, Melissa; Lust, Katherine; Laska, Melissa

2014-12-01

Little is known about the complex patterning of weight-related health behaviors in 2- and 4-year college students. The objective of this study was to identify and describe unique classes of weight-related health behaviors among college students. Latent class analysis was used to identify homogenous, mutually exclusive classes of nine health behaviors that represent multiple theoretically/clinically relevant dimensions of obesity risk among 2- versus 4-year college students using cross-sectional statewide surveillance data (N = 17,584). Additionally, differences in class membership on selected sociodemographic characteristics were examined using a model-based approach. Analysis was conducted separately for both college groups, and five and four classes were identified for 2- and 4-year college students, respectively. Four classes were similar across 2- and 4-year college groups and were characterized as "mostly healthy dietary habits, active"; "moderately high screen time, active"; "moderately healthy dietary habits, inactive"; and "moderately high screen time, inactive." "Moderately healthy dietary habits, high screen time" was the additional class unique to 2-year college students. These classes differed on a number of sociodemographic characteristics, including the proportion in each class who were classified as obese. Implications for prevention scientists and future intervention programs are considered. © 2014 Society for Public Health Education.