Instrumental Variable Analysis with a Nonlinear Exposure–Outcome Relationship

PubMed Central

Davies, Neil M.; Thompson, Simon G.

2014-01-01

Background: Instrumental variable methods can estimate the causal effect of an exposure on an outcome using observational data. Many instrumental variable methods assume that the exposure–outcome relation is linear, but in practice this assumption is often in doubt, or perhaps the shape of the relation is a target for investigation. We investigate this issue in the context of Mendelian randomization, the use of genetic variants as instrumental variables. Methods: Using simulations, we demonstrate the performance of a simple linear instrumental variable method when the true shape of the exposure–outcome relation is not linear. We also present a novel method for estimating the effect of the exposure on the outcome within strata of the exposure distribution. This enables the estimation of localized average causal effects within quantile groups of the exposure or as a continuous function of the exposure using a sliding window approach. Results: Our simulations suggest that linear instrumental variable estimates approximate a population-averaged causal effect. This is the average difference in the outcome if the exposure for every individual in the population is increased by a fixed amount. Estimates of localized average causal effects reveal the shape of the exposure–outcome relation for a variety of models. These methods are used to investigate the relations between body mass index and a range of cardiovascular risk factors. Conclusions: Nonlinear exposure–outcome relations should not be a barrier to instrumental variable analyses. When the exposure–outcome relation is not linear, either a population-averaged causal effect or the shape of the exposure–outcome relation can be estimated. PMID:25166881

Estimation and Identification of the Complier Average Causal Effect Parameter in Education RCTs

ERIC Educational Resources Information Center

Schochet, Peter Z.; Chiang, Hanley S.

2011-01-01

In randomized control trials (RCTs) in the education field, the complier average causal effect (CACE) parameter is often of policy interest, because it pertains to intervention effects for students who receive a meaningful dose of treatment services. This article uses a causal inference and instrumental variables framework to examine the…

Causal transfer function analysis to describe closed loop interactions between cardiovascular and cardiorespiratory variability signals.

PubMed

Faes, L; Porta, A; Cucino, R; Cerutti, S; Antolini, R; Nollo, G

2004-06-01

Although the concept of transfer function is intrinsically related to an input-output relationship, the traditional and widely used estimation method merges both feedback and feedforward interactions between the two analyzed signals. This limitation may endanger the reliability of transfer function analysis in biological systems characterized by closed loop interactions. In this study, a method for estimating the transfer function between closed loop interacting signals was proposed and validated in the field of cardiovascular and cardiorespiratory variability. The two analyzed signals x and y were described by a bivariate autoregressive model, and the causal transfer function from x to y was estimated after imposing causality by setting to zero the model coefficients representative of the reverse effects from y to x. The method was tested in simulations reproducing linear open and closed loop interactions, showing a better adherence of the causal transfer function to the theoretical curves with respect to the traditional approach in presence of non-negligible reverse effects. It was then applied in ten healthy young subjects to characterize the transfer functions from respiration to heart period (RR interval) and to systolic arterial pressure (SAP), and from SAP to RR interval. In the first two cases, the causal and non-causal transfer function estimates were comparable, indicating that respiration, acting as exogenous signal, sets an open loop relationship upon SAP and RR interval. On the contrary, causal and traditional transfer functions from SAP to RR were significantly different, suggesting the presence of a considerable influence on the opposite causal direction. Thus, the proposed causal approach seems to be appropriate for the estimation of parameters, like the gain and the phase lag from SAP to RR interval, which have a large clinical and physiological relevance.

Evaluating Candidate Principal Surrogate Endpoints

PubMed Central

Gilbert, Peter B.; Hudgens, Michael G.

2009-01-01

Summary Frangakis and Rubin (2002, Biometrics 58, 21–29) proposed a new definition of a surrogate endpoint (a “principal” surrogate) based on causal effects. We introduce an estimand for evaluating a principal surrogate, the causal effect predictiveness (CEP) surface, which quantifies how well causal treatment effects on the biomarker predict causal treatment effects on the clinical endpoint. Although the CEP surface is not identifiable due to missing potential outcomes, it can be identified by incorporating a baseline covariate(s) that predicts the biomarker. Given case–cohort sampling of such a baseline predictor and the biomarker in a large blinded randomized clinical trial, we develop an estimated likelihood method for estimating the CEP surface. This estimation assesses the “surrogate value” of the biomarker for reliably predicting clinical treatment effects for the same or similar setting as the trial. A CEP surface plot provides a way to compare the surrogate value of multiple biomarkers. The approach is illustrated by the problem of assessing an immune response to a vaccine as a surrogate endpoint for infection. PMID:18363776

Causal Mediation Analysis: Warning! Assumptions Ahead

ERIC Educational Resources Information Center

Keele, Luke

2015-01-01

In policy evaluations, interest may focus on why a particular treatment works. One tool for understanding why treatments work is causal mediation analysis. In this essay, I focus on the assumptions needed to estimate mediation effects. I show that there is no "gold standard" method for the identification of causal mediation effects. In…

Estimation of effective connectivity using multi-layer perceptron artificial neural network.

PubMed

Talebi, Nasibeh; Nasrabadi, Ali Motie; Mohammad-Rezazadeh, Iman

2018-02-01

Studies on interactions between brain regions estimate effective connectivity, (usually) based on the causality inferences made on the basis of temporal precedence. In this study, the causal relationship is modeled by a multi-layer perceptron feed-forward artificial neural network, because of the ANN's ability to generate appropriate input-output mapping and to learn from training examples without the need of detailed knowledge of the underlying system. At any time instant, the past samples of data are placed in the network input, and the subsequent values are predicted at its output. To estimate the strength of interactions, the measure of " Causality coefficient " is defined based on the network structure, the connecting weights and the parameters of hidden layer activation function. Simulation analysis demonstrates that the method, called "CREANN" (Causal Relationship Estimation by Artificial Neural Network), can estimate time-invariant and time-varying effective connectivity in terms of MVAR coefficients. The method shows robustness with respect to noise level of data. Furthermore, the estimations are not significantly influenced by the model order (considered time-lag), and the different initial conditions (initial random weights and parameters of the network). CREANN is also applied to EEG data collected during a memory recognition task. The results implicate that it can show changes in the information flow between brain regions, involving in the episodic memory retrieval process. These convincing results emphasize that CREANN can be used as an appropriate method to estimate the causal relationship among brain signals.

Translating context to causality in cardiovascular disparities research.

PubMed

Benn, Emma K T; Goldfeld, Keith S

2016-04-01

Moving from a descriptive focus to a comprehensive analysis grounded in causal inference can be particularly daunting for disparities researchers. However, even a simple model supported by the theoretical underpinnings of causality gives researchers a better chance to make correct inferences about possible interventions that can benefit our most vulnerable populations. This commentary provides a brief description of how race/ethnicity and context relate to questions of causality, and uses a hypothetical scenario to explore how different researchers might analyze the data to estimate causal effects of interest. Perhaps although not entirely removed of bias, these causal estimates will move us a step closer to understanding how to intervene. (PsycINFO Database Record (c) 2016 APA, all rights reserved).

Interpreting findings from Mendelian randomization using the MR-Egger method.

PubMed

Burgess, Stephen; Thompson, Simon G

2017-05-01

Mendelian randomization-Egger (MR-Egger) is an analysis method for Mendelian randomization using summarized genetic data. MR-Egger consists of three parts: (1) a test for directional pleiotropy, (2) a test for a causal effect, and (3) an estimate of the causal effect. While conventional analysis methods for Mendelian randomization assume that all genetic variants satisfy the instrumental variable assumptions, the MR-Egger method is able to assess whether genetic variants have pleiotropic effects on the outcome that differ on average from zero (directional pleiotropy), as well as to provide a consistent estimate of the causal effect, under a weaker assumption-the InSIDE (INstrument Strength Independent of Direct Effect) assumption. In this paper, we provide a critical assessment of the MR-Egger method with regard to its implementation and interpretation. While the MR-Egger method is a worthwhile sensitivity analysis for detecting violations of the instrumental variable assumptions, there are several reasons why causal estimates from the MR-Egger method may be biased and have inflated Type 1 error rates in practice, including violations of the InSIDE assumption and the influence of outlying variants. The issues raised in this paper have potentially serious consequences for causal inferences from the MR-Egger approach. We give examples of scenarios in which the estimates from conventional Mendelian randomization methods and MR-Egger differ, and discuss how to interpret findings in such cases.

Adolescent Drug Use and the Deterrent Effect of School-Imposed Penalties

ERIC Educational Resources Information Center

Waddell, G. R.

2012-01-01

Estimates of the effect of school-imposed penalties for drug use on a student's consumption of marijuana are biased if both are determined by unobservable school or individual attributes. Reverse causality is also a potential challenge to retrieving estimates of the causal relationship, as the severity of school sanctions may simply reflect the…

Assessing the suitability of summary data for two-sample Mendelian randomization analyses using MR-Egger regression: the role of the I2 statistic.

PubMed

Bowden, Jack; Del Greco M, Fabiola; Minelli, Cosetta; Davey Smith, George; Sheehan, Nuala A; Thompson, John R

2016-12-01

: MR-Egger regression has recently been proposed as a method for Mendelian randomization (MR) analyses incorporating summary data estimates of causal effect from multiple individual variants, which is robust to invalid instruments. It can be used to test for directional pleiotropy and provides an estimate of the causal effect adjusted for its presence. MR-Egger regression provides a useful additional sensitivity analysis to the standard inverse variance weighted (IVW) approach that assumes all variants are valid instruments. Both methods use weights that consider the single nucleotide polymorphism (SNP)-exposure associations to be known, rather than estimated. We call this the `NO Measurement Error' (NOME) assumption. Causal effect estimates from the IVW approach exhibit weak instrument bias whenever the genetic variants utilized violate the NOME assumption, which can be reliably measured using the F-statistic. The effect of NOME violation on MR-Egger regression has yet to be studied. An adaptation of the I2 statistic from the field of meta-analysis is proposed to quantify the strength of NOME violation for MR-Egger. It lies between 0 and 1, and indicates the expected relative bias (or dilution) of the MR-Egger causal estimate in the two-sample MR context. We call it IGX2 . The method of simulation extrapolation is also explored to counteract the dilution. Their joint utility is evaluated using simulated data and applied to a real MR example. In simulated two-sample MR analyses we show that, when a causal effect exists, the MR-Egger estimate of causal effect is biased towards the null when NOME is violated, and the stronger the violation (as indicated by lower values of IGX2 ), the stronger the dilution. When additionally all genetic variants are valid instruments, the type I error rate of the MR-Egger test for pleiotropy is inflated and the causal effect underestimated. Simulation extrapolation is shown to substantially mitigate these adverse effects. We demonstrate our proposed approach for a two-sample summary data MR analysis to estimate the causal effect of low-density lipoprotein on heart disease risk. A high value of IGX2 close to 1 indicates that dilution does not materially affect the standard MR-Egger analyses for these data. : Care must be taken to assess the NOME assumption via the IGX2 statistic before implementing standard MR-Egger regression in the two-sample summary data context. If IGX2 is sufficiently low (less than 90%), inferences from the method should be interpreted with caution and adjustment methods considered. © The Author 2016. Published by Oxford University Press on behalf of the International Epidemiological Association.

Instrumental variables I: instrumental variables exploit natural variation in nonexperimental data to estimate causal relationships.

PubMed

Rassen, Jeremy A; Brookhart, M Alan; Glynn, Robert J; Mittleman, Murray A; Schneeweiss, Sebastian

2009-12-01

The gold standard of study design for treatment evaluation is widely acknowledged to be the randomized controlled trial (RCT). Trials allow for the estimation of causal effect by randomly assigning participants either to an intervention or comparison group; through the assumption of "exchangeability" between groups, comparing the outcomes will yield an estimate of causal effect. In the many cases where RCTs are impractical or unethical, instrumental variable (IV) analysis offers a nonexperimental alternative based on many of the same principles. IV analysis relies on finding a naturally varying phenomenon, related to treatment but not to outcome except through the effect of treatment itself, and then using this phenomenon as a proxy for the confounded treatment variable. This article demonstrates how IV analysis arises from an analogous but potentially impossible RCT design, and outlines the assumptions necessary for valid estimation. It gives examples of instruments used in clinical epidemiology and concludes with an outline on estimation of effects.

Instrumental variables I: instrumental variables exploit natural variation in nonexperimental data to estimate causal relationships

PubMed Central

Rassen, Jeremy A.; Brookhart, M. Alan; Glynn, Robert J.; Mittleman, Murray A.; Schneeweiss, Sebastian

2010-01-01

The gold standard of study design for treatment evaluation is widely acknowledged to be the randomized controlled trial (RCT). Trials allow for the estimation of causal effect by randomly assigning participants either to an intervention or comparison group; through the assumption of “exchangeability” between groups, comparing the outcomes will yield an estimate of causal effect. In the many cases where RCTs are impractical or unethical, instrumental variable (IV) analysis offers a nonexperimental alternative based on many of the same principles. IV analysis relies on finding a naturally varying phenomenon, related to treatment but not to outcome except through the effect of treatment itself, and then using this phenomenon as a proxy for the confounded treatment variable. This article demonstrates how IV analysis arises from an analogous but potentially impossible RCT design, and outlines the assumptions necessary for valid estimation. It gives examples of instruments used in clinical epidemiology and concludes with an outline on estimation of effects. PMID:19356901

Estimating causal effects with a non-paranormal method for the design of efficient intervention experiments

PubMed Central

2014-01-01

Background Knockdown or overexpression of genes is widely used to identify genes that play important roles in many aspects of cellular functions and phenotypes. Because next-generation sequencing generates high-throughput data that allow us to detect genes, it is important to identify genes that drive functional and phenotypic changes of cells. However, conventional methods rely heavily on the assumption of normality and they often give incorrect results when the assumption is not true. To relax the Gaussian assumption in causal inference, we introduce the non-paranormal method to test conditional independence in the PC-algorithm. Then, we present the non-paranormal intervention-calculus when the directed acyclic graph (DAG) is absent (NPN-IDA), which incorporates the cumulative nature of effects through a cascaded pathway via causal inference for ranking causal genes against a phenotype with the non-paranormal method for estimating DAGs. Results We demonstrate that causal inference with the non-paranormal method significantly improves the performance in estimating DAGs on synthetic data in comparison with the original PC-algorithm. Moreover, we show that NPN-IDA outperforms the conventional methods in exploring regulators of the flowering time in Arabidopsis thaliana and regulators that control the browning of white adipocytes in mice. Our results show that performance improvement in estimating DAGs contributes to an accurate estimation of causal effects. Conclusions Although the simplest alternative procedure was used, our proposed method enables us to design efficient intervention experiments and can be applied to a wide range of research purposes, including drug discovery, because of its generality. PMID:24980787

Estimating causal effects with a non-paranormal method for the design of efficient intervention experiments.

PubMed

Teramoto, Reiji; Saito, Chiaki; Funahashi, Shin-ichi

2014-06-30

Knockdown or overexpression of genes is widely used to identify genes that play important roles in many aspects of cellular functions and phenotypes. Because next-generation sequencing generates high-throughput data that allow us to detect genes, it is important to identify genes that drive functional and phenotypic changes of cells. However, conventional methods rely heavily on the assumption of normality and they often give incorrect results when the assumption is not true. To relax the Gaussian assumption in causal inference, we introduce the non-paranormal method to test conditional independence in the PC-algorithm. Then, we present the non-paranormal intervention-calculus when the directed acyclic graph (DAG) is absent (NPN-IDA), which incorporates the cumulative nature of effects through a cascaded pathway via causal inference for ranking causal genes against a phenotype with the non-paranormal method for estimating DAGs. We demonstrate that causal inference with the non-paranormal method significantly improves the performance in estimating DAGs on synthetic data in comparison with the original PC-algorithm. Moreover, we show that NPN-IDA outperforms the conventional methods in exploring regulators of the flowering time in Arabidopsis thaliana and regulators that control the browning of white adipocytes in mice. Our results show that performance improvement in estimating DAGs contributes to an accurate estimation of causal effects. Although the simplest alternative procedure was used, our proposed method enables us to design efficient intervention experiments and can be applied to a wide range of research purposes, including drug discovery, because of its generality.

Causal Methods for Observational Research: A Primer.

PubMed

Almasi-Hashiani, Amir; Nedjat, Saharnaz; Mansournia, Mohammad Ali

2018-04-01

The goal of many observational studies is to estimate the causal effect of an exposure on an outcome after adjustment for confounders, but there are still some serious errors in adjusting confounders in clinical journals. Standard regression modeling (e.g., ordinary logistic regression) fails to estimate the average effect of exposure in total population in the presence of interaction between exposure and covariates, and also cannot adjust for time-varying confounding appropriately. Moreover, stepwise algorithms of the selection of confounders based on P values may miss important confounders and lead to bias in effect estimates. Causal methods overcome these limitations. We illustrate three causal methods including inverse-probability-of-treatment-weighting (IPTW) and parametric g-formula, with an emphasis on a clever combination of these 2 methods: targeted maximum likelihood estimation (TMLE) which enjoys a double-robust property against bias. © 2018 The Author(s). This is an open-access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

A Causal Inference Analysis of the Effect of Wildland Fire ...

EPA Pesticide Factsheets

Wildfire smoke is a major contributor to ambient air pollution levels. In this talk, we develop a spatio-temporal model to estimate the contribution of fire smoke to overall air pollution in different regions of the country. We combine numerical model output with observational data within a causal inference framework. Our methods account for aggregation and potential bias of the numerical model simulation, and address uncertainty in the causal estimates. We apply the proposed method to estimation of ozone and fine particulate matter from wildland fires and the impact on health burden assessment. We develop a causal inference framework to assess contributions of fire to ambient PM in the presence of spatial interference.

Can chance cause cancer? A causal consideration.

PubMed

Stensrud, Mats Julius; Strohmaier, Susanne; Valberg, Morten; Aalen, Odd Olai

2017-04-01

The role of randomness, environment and genetics in cancer development is debated. We approach the discussion by using the potential outcomes framework for causal inference. By briefly considering the underlying assumptions, we suggest that the antagonising views arise due to estimation of substantially different causal effects. These effects may be hard to interpret, and the results cannot be immediately compared. Indeed, it is not clear whether it is possible to define a causal effect of chance at all. Copyright © 2017 Elsevier Ltd. All rights reserved.

Three Cs in measurement models: causal indicators, composite indicators, and covariates.

PubMed

Bollen, Kenneth A; Bauldry, Shawn

2011-09-01

In the last 2 decades attention to causal (and formative) indicators has grown. Accompanying this growth has been the belief that one can classify indicators into 2 categories: effect (reflective) indicators and causal (formative) indicators. We argue that the dichotomous view is too simple. Instead, there are effect indicators and 3 types of variables on which a latent variable depends: causal indicators, composite (formative) indicators, and covariates (the "Three Cs"). Causal indicators have conceptual unity, and their effects on latent variables are structural. Covariates are not concept measures, but are variables to control to avoid bias in estimating the relations between measures and latent variables. Composite (formative) indicators form exact linear combinations of variables that need not share a concept. Their coefficients are weights rather than structural effects, and composites are a matter of convenience. The failure to distinguish the Three Cs has led to confusion and questions, such as, Are causal and formative indicators different names for the same indicator type? Should an equation with causal or formative indicators have an error term? Are the coefficients of causal indicators less stable than effect indicators? Distinguishing between causal and composite indicators and covariates goes a long way toward eliminating this confusion. We emphasize the key role that subject matter expertise plays in making these distinctions. We provide new guidelines for working with these variable types, including identification of models, scaling latent variables, parameter estimation, and validity assessment. A running empirical example on self-perceived health illustrates our major points.

Estimating the causal effects of smoking.

PubMed

Rubin, D B

An important application of statistics in recent years has been to address the causal effects of smoking. There is little doubt that there are health risks associated with smoking. However, more general issues concern the causal effects due to the alleged misconduct of the tobacco industry or due to programmes designed to curtail tobacco use. To address any such causal question, assumptions must be made. Although some of the issues are well known in the statistical and epidemiological literature, there does not appear to be a unified treatment that provides prescriptive guidance on the estimation of these causal effects with explication of the needed assumptions. A 'conduct attributable fraction' is derived, which allows for arbitrary changes in smoking and non-smoking health care expenditure related factors in a counterfactual world without the alleged misconduct, and therefore generalizes the traditional 'smoking attributable fraction'. The formulation presented here, although described for the problem of estimating excess health care expenditures due to the alleged misconduct of the tobacco industry, is more general. It can be applied to any outcome, such as mortality, morbidity, or income from excise taxes, as well as to any situation in which consequences due to alleged misconduct (for example, of two entities, such as the tobacco and the asbestos industries) or due to hypothetical programmes (for example, extra smoking reduction initiatives) are to be estimated. Copyright 2001 John Wiley & Sons, Ltd.

Do causal concentration-response functions exist? A critical review of associational and causal relations between fine particulate matter and mortality.

PubMed

Cox, Louis Anthony Tony

2017-08-01

Concentration-response (C-R) functions relating concentrations of pollutants in ambient air to mortality risks or other adverse health effects provide the basis for many public health risk assessments, benefits estimates for clean air regulations, and recommendations for revisions to existing air quality standards. The assumption that C-R functions relating levels of exposure and levels of response estimated from historical data usefully predict how future changes in concentrations would change risks has seldom been carefully tested. This paper critically reviews literature on C-R functions for fine particulate matter (PM2.5) and mortality risks. We find that most of them describe historical associations rather than valid causal models for predicting effects of interventions that change concentrations. The few papers that explicitly attempt to model causality rely on unverified modeling assumptions, casting doubt on their predictions about effects of interventions. A large literature on modern causal inference algorithms for observational data has been little used in C-R modeling. Applying these methods to publicly available data from Boston and the South Coast Air Quality Management District around Los Angeles shows that C-R functions estimated for one do not hold for the other. Changes in month-specific PM2.5 concentrations from one year to the next do not help to predict corresponding changes in average elderly mortality rates in either location. Thus, the assumption that estimated C-R relations predict effects of pollution-reducing interventions may not be true. Better causal modeling methods are needed to better predict how reducing air pollution would affect public health.

Links between causal effects and causal association for surrogacy evaluation in a gaussian setting.

PubMed

Conlon, Anna; Taylor, Jeremy; Li, Yun; Diaz-Ordaz, Karla; Elliott, Michael

2017-11-30

Two paradigms for the evaluation of surrogate markers in randomized clinical trials have been proposed: the causal effects paradigm and the causal association paradigm. Each of these paradigms rely on assumptions that must be made to proceed with estimation and to validate a candidate surrogate marker (S) for the true outcome of interest (T). We consider the setting in which S and T are Gaussian and are generated from structural models that include an unobserved confounder. Under the assumed structural models, we relate the quantities used to evaluate surrogacy within both the causal effects and causal association frameworks. We review some of the common assumptions made to aid in estimating these quantities and show that assumptions made within one framework can imply strong assumptions within the alternative framework. We demonstrate that there is a similarity, but not exact correspondence between the quantities used to evaluate surrogacy within each framework, and show that the conditions for identifiability of the surrogacy parameters are different from the conditions, which lead to a correspondence of these quantities. Copyright © 2017 John Wiley & Sons, Ltd.

Kernel canonical-correlation Granger causality for multiple time series

NASA Astrophysics Data System (ADS)

Wu, Guorong; Duan, Xujun; Liao, Wei; Gao, Qing; Chen, Huafu

2011-04-01

Canonical-correlation analysis as a multivariate statistical technique has been applied to multivariate Granger causality analysis to infer information flow in complex systems. It shows unique appeal and great superiority over the traditional vector autoregressive method, due to the simplified procedure that detects causal interaction between multiple time series, and the avoidance of potential model estimation problems. However, it is limited to the linear case. Here, we extend the framework of canonical correlation to include the estimation of multivariate nonlinear Granger causality for drawing inference about directed interaction. Its feasibility and effectiveness are verified on simulated data.

Network Mendelian randomization: using genetic variants as instrumental variables to investigate mediation in causal pathways

PubMed Central

Burgess, Stephen; Daniel, Rhian M; Butterworth, Adam S; Thompson, Simon G

2015-01-01

Background: Mendelian randomization uses genetic variants, assumed to be instrumental variables for a particular exposure, to estimate the causal effect of that exposure on an outcome. If the instrumental variable criteria are satisfied, the resulting estimator is consistent even in the presence of unmeasured confounding and reverse causation. Methods: We extend the Mendelian randomization paradigm to investigate more complex networks of relationships between variables, in particular where some of the effect of an exposure on the outcome may operate through an intermediate variable (a mediator). If instrumental variables for the exposure and mediator are available, direct and indirect effects of the exposure on the outcome can be estimated, for example using either a regression-based method or structural equation models. The direction of effect between the exposure and a possible mediator can also be assessed. Methods are illustrated in an applied example considering causal relationships between body mass index, C-reactive protein and uric acid. Results: These estimators are consistent in the presence of unmeasured confounding if, in addition to the instrumental variable assumptions, the effects of both the exposure on the mediator and the mediator on the outcome are homogeneous across individuals and linear without interactions. Nevertheless, a simulation study demonstrates that even considerable heterogeneity in these effects does not lead to bias in the estimates. Conclusions: These methods can be used to estimate direct and indirect causal effects in a mediation setting, and have potential for the investigation of more complex networks between multiple interrelated exposures and disease outcomes. PMID:25150977

The alarming problems of confounding equivalence using logistic regression models in the perspective of causal diagrams.

PubMed

Yu, Yuanyuan; Li, Hongkai; Sun, Xiaoru; Su, Ping; Wang, Tingting; Liu, Yi; Yuan, Zhongshang; Liu, Yanxun; Xue, Fuzhong

2017-12-28

Confounders can produce spurious associations between exposure and outcome in observational studies. For majority of epidemiologists, adjusting for confounders using logistic regression model is their habitual method, though it has some problems in accuracy and precision. It is, therefore, important to highlight the problems of logistic regression and search the alternative method. Four causal diagram models were defined to summarize confounding equivalence. Both theoretical proofs and simulation studies were performed to verify whether conditioning on different confounding equivalence sets had the same bias-reducing potential and then to select the optimum adjusting strategy, in which logistic regression model and inverse probability weighting based marginal structural model (IPW-based-MSM) were compared. The "do-calculus" was used to calculate the true causal effect of exposure on outcome, then the bias and standard error were used to evaluate the performances of different strategies. Adjusting for different sets of confounding equivalence, as judged by identical Markov boundaries, produced different bias-reducing potential in the logistic regression model. For the sets satisfied G-admissibility, adjusting for the set including all the confounders reduced the equivalent bias to the one containing the parent nodes of the outcome, while the bias after adjusting for the parent nodes of exposure was not equivalent to them. In addition, all causal effect estimations through logistic regression were biased, although the estimation after adjusting for the parent nodes of exposure was nearest to the true causal effect. However, conditioning on different confounding equivalence sets had the same bias-reducing potential under IPW-based-MSM. Compared with logistic regression, the IPW-based-MSM could obtain unbiased causal effect estimation when the adjusted confounders satisfied G-admissibility and the optimal strategy was to adjust for the parent nodes of outcome, which obtained the highest precision. All adjustment strategies through logistic regression were biased for causal effect estimation, while IPW-based-MSM could always obtain unbiased estimation when the adjusted set satisfied G-admissibility. Thus, IPW-based-MSM was recommended to adjust for confounders set.

Long-Term Consequences of Early Sexual Initiation on Young Adult Health: A Causal Inference Approach

ERIC Educational Resources Information Center

Kugler, Kari C.; Vasilenko, Sara A.; Butera, Nicole M.; Coffman, Donna L.

2017-01-01

Although early sexual initiation has been linked to negative outcomes, it is unknown whether these effects are causal. In this study, we use propensity score methods to estimate the causal effect of early sexual initiation on young adult sexual risk behaviors and health outcomes using data from the National Longitudinal Study of Adolescent to…

Investigating causal associations between use of nicotine, alcohol, caffeine and cannabis: a two-sample bidirectional Mendelian randomization study.

PubMed

Verweij, Karin J H; Treur, Jorien L; Vink, Jacqueline M

2018-07-01

Epidemiological studies consistently show co-occurrence of use of different addictive substances. Whether these associations are causal or due to overlapping underlying influences remains an important question in addiction research. Methodological advances have made it possible to use published genetic associations to infer causal relationships between phenotypes. In this exploratory study, we used Mendelian randomization (MR) to examine the causality of well-established associations between nicotine, alcohol, caffeine and cannabis use. Two-sample MR was employed to estimate bidirectional causal effects between four addictive substances: nicotine (smoking initiation and cigarettes smoked per day), caffeine (cups of coffee per day), alcohol (units per week) and cannabis (initiation). Based on existing genome-wide association results we selected genetic variants associated with the exposure measure as an instrument to estimate causal effects. Where possible we applied sensitivity analyses (MR-Egger and weighted median) more robust to horizontal pleiotropy. Most MR tests did not reveal causal associations. There was some weak evidence for a causal positive effect of genetically instrumented alcohol use on smoking initiation and of cigarettes per day on caffeine use, but these were not supported by the sensitivity analyses. There was also some suggestive evidence for a positive effect of alcohol use on caffeine use (only with MR-Egger) and smoking initiation on cannabis initiation (only with weighted median). None of the suggestive causal associations survived corrections for multiple testing. Two-sample Mendelian randomization analyses found little evidence for causal relationships between nicotine, alcohol, caffeine and cannabis use. © 2018 Society for the Study of Addiction.

Effects of BMI, Fat Mass, and Lean Mass on Asthma in Childhood: A Mendelian Randomization Study

PubMed Central

Granell, Raquel; Henderson, A. John; Evans, David M.; Smith, George Davey; Ness, Andrew R.; Lewis, Sarah; Palmer, Tom M.; Sterne, Jonathan A. C.

2014-01-01

Background Observational studies have reported associations between body mass index (BMI) and asthma, but confounding and reverse causality remain plausible explanations. We aim to investigate evidence for a causal effect of BMI on asthma using a Mendelian randomization approach. Methods and Findings We used Mendelian randomization to investigate causal effects of BMI, fat mass, and lean mass on current asthma at age 7½ y in the Avon Longitudinal Study of Parents and Children (ALSPAC). A weighted allele score based on 32 independent BMI-related single nucleotide polymorphisms (SNPs) was derived from external data, and associations with BMI, fat mass, lean mass, and asthma were estimated. We derived instrumental variable (IV) estimates of causal risk ratios (RRs). 4,835 children had available data on BMI-associated SNPs, asthma, and BMI. The weighted allele score was strongly associated with BMI, fat mass, and lean mass (all p-values<0.001) and with childhood asthma (RR 2.56, 95% CI 1.38–4.76 per unit score, p = 0.003). The estimated causal RR for the effect of BMI on asthma was 1.55 (95% CI 1.16–2.07) per kg/m2, p = 0.003. This effect appeared stronger for non-atopic (1.90, 95% CI 1.19–3.03) than for atopic asthma (1.37, 95% CI 0.89–2.11) though there was little evidence of heterogeneity (p = 0.31). The estimated causal RRs for the effects of fat mass and lean mass on asthma were 1.41 (95% CI 1.11–1.79) per 0.5 kg and 2.25 (95% CI 1.23–4.11) per kg, respectively. The possibility of genetic pleiotropy could not be discounted completely; however, additional IV analyses using FTO variant rs1558902 and the other BMI-related SNPs separately provided similar causal effects with wider confidence intervals. Loss of follow-up was unlikely to bias the estimated effects. Conclusions Higher BMI increases the risk of asthma in mid-childhood. Higher BMI may have contributed to the increase in asthma risk toward the end of the 20th century. Please see later in the article for the Editors' Summary PMID:24983943

Three Cs in Measurement Models: Causal Indicators, Composite Indicators, and Covariates

PubMed Central

Bollen, Kenneth A.; Bauldry, Shawn

2013-01-01

In the last two decades attention to causal (and formative) indicators has grown. Accompanying this growth has been the belief that we can classify indicators into two categories, effect (reflective) indicators and causal (formative) indicators. This paper argues that the dichotomous view is too simple. Instead, there are effect indicators and three types of variables on which a latent variable depends: causal indicators, composite (formative) indicators, and covariates (the “three Cs”). Causal indicators have conceptual unity and their effects on latent variables are structural. Covariates are not concept measures, but are variables to control to avoid bias in estimating the relations between measures and latent variable(s). Composite (formative) indicators form exact linear combinations of variables that need not share a concept. Their coefficients are weights rather than structural effects and composites are a matter of convenience. The failure to distinguish the “three Cs” has led to confusion and questions such as: are causal and formative indicators different names for the same indicator type? Should an equation with causal or formative indicators have an error term? Are the coefficients of causal indicators less stable than effect indicators? Distinguishing between causal and composite indicators and covariates goes a long way toward eliminating this confusion. We emphasize the key role that subject matter expertise plays in making these distinctions. We provide new guidelines for working with these variable types, including identification of models, scaling latent variables, parameter estimation, and validity assessment. A running empirical example on self-perceived health illustrates our major points. PMID:21767021

Causality Analysis of fMRI Data Based on the Directed Information Theory Framework.

PubMed

Wang, Zhe; Alahmadi, Ahmed; Zhu, David C; Li, Tongtong

2016-05-01

This paper aims to conduct fMRI-based causality analysis in brain connectivity by exploiting the directed information (DI) theory framework. Unlike the well-known Granger causality (GC) analysis, which relies on the linear prediction technique, the DI theory framework does not have any modeling constraints on the sequences to be evaluated and ensures estimation convergence. Moreover, it can be used to generate the GC graphs. In this paper, first, we introduce the core concepts in the DI framework. Second, we present how to conduct causality analysis using DI measures between two time series. We provide the detailed procedure on how to calculate the DI for two finite-time series. The two major steps involved here are optimal bin size selection for data digitization and probability estimation. Finally, we demonstrate the applicability of DI-based causality analysis using both the simulated data and experimental fMRI data, and compare the results with that of the GC analysis. Our analysis indicates that GC analysis is effective in detecting linear or nearly linear causal relationship, but may have difficulty in capturing nonlinear causal relationships. On the other hand, DI-based causality analysis is more effective in capturing both linear and nonlinear causal relationships. Moreover, it is observed that brain connectivity among different regions generally involves dynamic two-way information transmissions between them. Our results show that when bidirectional information flow is present, DI is more effective than GC to quantify the overall causal relationship.

Quantifying the causal effects of 20mph zones on road casualties in London via doubly robust estimation.

PubMed

Li, Haojie; Graham, Daniel J

2016-08-01

This paper estimates the causal effect of 20mph zones on road casualties in London. Potential confounders in the key relationship of interest are included within outcome regression and propensity score models, and the models are then combined to form a doubly robust estimator. A total of 234 treated zones and 2844 potential control zones are included in the data sample. The propensity score model is used to select a viable control group which has common support in the covariate distributions. We compare the doubly robust estimates with those obtained using three other methods: inverse probability weighting, regression adjustment, and propensity score matching. The results indicate that 20mph zones have had a significant causal impact on road casualty reduction in both absolute and proportional terms. Copyright © 2016 Elsevier Ltd. All rights reserved.

A new approach to hierarchical data analysis: Targeted maximum likelihood estimation for the causal effect of a cluster-level exposure.

PubMed

Balzer, Laura B; Zheng, Wenjing; van der Laan, Mark J; Petersen, Maya L

2018-01-01

We often seek to estimate the impact of an exposure naturally occurring or randomly assigned at the cluster-level. For example, the literature on neighborhood determinants of health continues to grow. Likewise, community randomized trials are applied to learn about real-world implementation, sustainability, and population effects of interventions with proven individual-level efficacy. In these settings, individual-level outcomes are correlated due to shared cluster-level factors, including the exposure, as well as social or biological interactions between individuals. To flexibly and efficiently estimate the effect of a cluster-level exposure, we present two targeted maximum likelihood estimators (TMLEs). The first TMLE is developed under a non-parametric causal model, which allows for arbitrary interactions between individuals within a cluster. These interactions include direct transmission of the outcome (i.e. contagion) and influence of one individual's covariates on another's outcome (i.e. covariate interference). The second TMLE is developed under a causal sub-model assuming the cluster-level and individual-specific covariates are sufficient to control for confounding. Simulations compare the alternative estimators and illustrate the potential gains from pairing individual-level risk factors and outcomes during estimation, while avoiding unwarranted assumptions. Our results suggest that estimation under the sub-model can result in bias and misleading inference in an observational setting. Incorporating working assumptions during estimation is more robust than assuming they hold in the underlying causal model. We illustrate our approach with an application to HIV prevention and treatment.

Regression Discontinuity Design: A Guide for Strengthening Causal Inference in HRD

ERIC Educational Resources Information Center

Chambers, Silvana

2016-01-01

Purpose: Regression discontinuity (RD) design is a sophisticated quasi-experimental approach used for inferring causal relationships and estimating treatment effects. This paper aims to educate human resource development (HRD) researchers and practitioners on the implementation of RD design as an ethical alternative for making causal claims about…

Tracking time-varying causality and directionality of information flow using an error reduction ratio test with applications to electroencephalography data.

PubMed

Zhao, Yifan; Billings, Steve A; Wei, Hualiang; Sarrigiannis, Ptolemaios G

2012-11-01

This paper introduces an error reduction ratio-causality (ERR-causality) test that can be used to detect and track causal relationships between two signals. In comparison to the traditional Granger method, one significant advantage of the new ERR-causality test is that it can effectively detect the time-varying direction of linear or nonlinear causality between two signals without fitting a complete model. Another important advantage is that the ERR-causality test can detect both the direction of interactions and estimate the relative time shift between the two signals. Numerical examples are provided to illustrate the effectiveness of the new method together with the determination of the causality between electroencephalograph signals from different cortical sites for patients during an epileptic seizure.

Time-varying effect moderation using the structural nested mean model: estimation using inverse-weighted regression with residuals

PubMed Central

Almirall, Daniel; Griffin, Beth Ann; McCaffrey, Daniel F.; Ramchand, Rajeev; Yuen, Robert A.; Murphy, Susan A.

2014-01-01

This article considers the problem of examining time-varying causal effect moderation using observational, longitudinal data in which treatment, candidate moderators, and possible confounders are time varying. The structural nested mean model (SNMM) is used to specify the moderated time-varying causal effects of interest in a conditional mean model for a continuous response given time-varying treatments and moderators. We present an easy-to-use estimator of the SNMM that combines an existing regression-with-residuals (RR) approach with an inverse-probability-of-treatment weighting (IPTW) strategy. The RR approach has been shown to identify the moderated time-varying causal effects if the time-varying moderators are also the sole time-varying confounders. The proposed IPTW+RR approach provides estimators of the moderated time-varying causal effects in the SNMM in the presence of an additional, auxiliary set of known and measured time-varying confounders. We use a small simulation experiment to compare IPTW+RR versus the traditional regression approach and to compare small and large sample properties of asymptotic versus bootstrap estimators of the standard errors for the IPTW+RR approach. This article clarifies the distinction between time-varying moderators and time-varying confounders. We illustrate the methodology in a case study to assess if time-varying substance use moderates treatment effects on future substance use. PMID:23873437

Comparing methods for estimation of heterogeneous treatment effects using observational data from health care databases.

PubMed

Wendling, T; Jung, K; Callahan, A; Schuler, A; Shah, N H; Gallego, B

2018-06-03

There is growing interest in using routinely collected data from health care databases to study the safety and effectiveness of therapies in "real-world" conditions, as it can provide complementary evidence to that of randomized controlled trials. Causal inference from health care databases is challenging because the data are typically noisy, high dimensional, and most importantly, observational. It requires methods that can estimate heterogeneous treatment effects while controlling for confounding in high dimensions. Bayesian additive regression trees, causal forests, causal boosting, and causal multivariate adaptive regression splines are off-the-shelf methods that have shown good performance for estimation of heterogeneous treatment effects in observational studies of continuous outcomes. However, it is not clear how these methods would perform in health care database studies where outcomes are often binary and rare and data structures are complex. In this study, we evaluate these methods in simulation studies that recapitulate key characteristics of comparative effectiveness studies. We focus on the conditional average effect of a binary treatment on a binary outcome using the conditional risk difference as an estimand. To emulate health care database studies, we propose a simulation design where real covariate and treatment assignment data are used and only outcomes are simulated based on nonparametric models of the real outcomes. We apply this design to 4 published observational studies that used records from 2 major health care databases in the United States. Our results suggest that Bayesian additive regression trees and causal boosting consistently provide low bias in conditional risk difference estimates in the context of health care database studies. Copyright © 2018 John Wiley & Sons, Ltd.

The Causal Effect of Off-Campus Work on Time to Degree

ERIC Educational Resources Information Center

Behr, Andreas; Theune, Katja

2016-01-01

In this paper we analyze the effect of outside university work on time to first degree at German universities. The database is the "Absolventenpanel" 2001, a panel study conducted by the "Hochschul-Informations-System". Aiming to estimate the causal effect correctly, we apply a matching strategy based on the approach put…

Comments: Causal Interpretations of Mediation Effects

ERIC Educational Resources Information Center

Jo, Booil; Stuart, Elizabeth A.

2012-01-01

The authors thank Dr. Lindsay Page for providing a nice illustration of the use of the principal stratification framework to define causal effects, and a Bayesian model for effect estimation. They hope that her well-written article will help expose education researchers to these concepts and methods, and move the field of mediation analysis in…

Propensity Score Estimation With Boosted Regression for Evaluating Causal Effects in Observational Studies

ERIC Educational Resources Information Center

McCaffrey, Daniel F.; Ridgeway, Greg; Morral, Andrew R.

2004-01-01

Causal effect modeling with naturalistic rather than experimental data is challenging. In observational studies participants in different treatment conditions may also differ on pretreatment characteristics that influence outcomes. Propensity score methods can theoretically eliminate these confounds for all observed covariates, but accurate…

Whose statistical reasoning is facilitated by a causal structure intervention?

PubMed

McNair, Simon; Feeney, Aidan

2015-02-01

People often struggle when making Bayesian probabilistic estimates on the basis of competing sources of statistical evidence. Recently, Krynski and Tenenbaum (Journal of Experimental Psychology: General, 136, 430-450, 2007) proposed that a causal Bayesian framework accounts for peoples' errors in Bayesian reasoning and showed that, by clarifying the causal relations among the pieces of evidence, judgments on a classic statistical reasoning problem could be significantly improved. We aimed to understand whose statistical reasoning is facilitated by the causal structure intervention. In Experiment 1, although we observed causal facilitation effects overall, the effect was confined to participants high in numeracy. We did not find an overall facilitation effect in Experiment 2 but did replicate the earlier interaction between numerical ability and the presence or absence of causal content. This effect held when we controlled for general cognitive ability and thinking disposition. Our results suggest that clarifying causal structure facilitates Bayesian judgments, but only for participants with sufficient understanding of basic concepts in probability and statistics.

Stratified exact tests for the weak causal null hypothesis in randomized trials with a binary outcome.

PubMed

Chiba, Yasutaka

2017-09-01

Fisher's exact test is commonly used to compare two groups when the outcome is binary in randomized trials. In the context of causal inference, this test explores the sharp causal null hypothesis (i.e. the causal effect of treatment is the same for all subjects), but not the weak causal null hypothesis (i.e. the causal risks are the same in the two groups). Therefore, in general, rejection of the null hypothesis by Fisher's exact test does not mean that the causal risk difference is not zero. Recently, Chiba (Journal of Biometrics and Biostatistics 2015; 6: 244) developed a new exact test for the weak causal null hypothesis when the outcome is binary in randomized trials; the new test is not based on any large sample theory and does not require any assumption. In this paper, we extend the new test; we create a version of the test applicable to a stratified analysis. The stratified exact test that we propose is general in nature and can be used in several approaches toward the estimation of treatment effects after adjusting for stratification factors. The stratified Fisher's exact test of Jung (Biometrical Journal 2014; 56: 129-140) tests the sharp causal null hypothesis. This test applies a crude estimator of the treatment effect and can be regarded as a special case of our proposed exact test. Our proposed stratified exact test can be straightforwardly extended to analysis of noninferiority trials and to construct the associated confidence interval. © 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

Marginal Structural Models with Counterfactual Effect Modifiers.

PubMed

Zheng, Wenjing; Luo, Zhehui; van der Laan, Mark J

2018-06-08

In health and social sciences, research questions often involve systematic assessment of the modification of treatment causal effect by patient characteristics. In longitudinal settings, time-varying or post-intervention effect modifiers are also of interest. In this work, we investigate the robust and efficient estimation of the Counterfactual-History-Adjusted Marginal Structural Model (van der Laan MJ, Petersen M. Statistical learning of origin-specific statically optimal individualized treatment rules. Int J Biostat. 2007;3), which models the conditional intervention-specific mean outcome given a counterfactual modifier history in an ideal experiment. We establish the semiparametric efficiency theory for these models, and present a substitution-based, semiparametric efficient and doubly robust estimator using the targeted maximum likelihood estimation methodology (TMLE, e.g. van der Laan MJ, Rubin DB. Targeted maximum likelihood learning. Int J Biostat. 2006;2, van der Laan MJ, Rose S. Targeted learning: causal inference for observational and experimental data, 1st ed. Springer Series in Statistics. Springer, 2011). To facilitate implementation in applications where the effect modifier is high dimensional, our third contribution is a projected influence function (and the corresponding projected TMLE estimator), which retains most of the robustness of its efficient peer and can be easily implemented in applications where the use of the efficient influence function becomes taxing. We compare the projected TMLE estimator with an Inverse Probability of Treatment Weighted estimator (e.g. Robins JM. Marginal structural models. In: Proceedings of the American Statistical Association. Section on Bayesian Statistical Science, 1-10. 1997a, Hernan MA, Brumback B, Robins JM. Marginal structural models to estimate the causal effect of zidovudine on the survival of HIV-positive men. 2000;11:561-570), and a non-targeted G-computation estimator (Robins JM. A new approach to causal inference in mortality studies with sustained exposure periods - application to control of the healthy worker survivor effect. Math Modell. 1986;7:1393-1512.). The comparative performance of these estimators is assessed in a simulation study. The use of the projected TMLE estimator is illustrated in a secondary data analysis for the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) trial where effect modifiers are subject to missing at random.

Estimating Temporal Causal Interaction between Spike Trains with Permutation and Transfer Entropy

PubMed Central

Li, Zhaohui; Li, Xiaoli

2013-01-01

Estimating the causal interaction between neurons is very important for better understanding the functional connectivity in neuronal networks. We propose a method called normalized permutation transfer entropy (NPTE) to evaluate the temporal causal interaction between spike trains, which quantifies the fraction of ordinal information in a neuron that has presented in another one. The performance of this method is evaluated with the spike trains generated by an Izhikevich’s neuronal model. Results show that the NPTE method can effectively estimate the causal interaction between two neurons without influence of data length. Considering both the precision of time delay estimated and the robustness of information flow estimated against neuronal firing rate, the NPTE method is superior to other information theoretic method including normalized transfer entropy, symbolic transfer entropy and permutation conditional mutual information. To test the performance of NPTE on analyzing simulated biophysically realistic synapses, an Izhikevich’s cortical network that based on the neuronal model is employed. It is found that the NPTE method is able to characterize mutual interactions and identify spurious causality in a network of three neurons exactly. We conclude that the proposed method can obtain more reliable comparison of interactions between different pairs of neurons and is a promising tool to uncover more details on the neural coding. PMID:23940662

Using register data to estimate causal effects of interventions: An ex post synthetic control-group approach.

PubMed

Bygren, Magnus; Szulkin, Ryszard

2017-07-01

It is common in the context of evaluations that participants have not been selected on the basis of transparent participation criteria, and researchers and evaluators many times have to make do with observational data to estimate effects of job training programs and similar interventions. The techniques developed by researchers in such endeavours are useful not only to researchers narrowly focused on evaluations, but also to social and population science more generally, as observational data overwhelmingly are the norm, and the endogeneity challenges encountered in the estimation of causal effects with such data are not trivial. The aim of this article is to illustrate how register data can be used strategically to evaluate programs and interventions and to estimate causal effects of participation in these. We use propensity score matching on pretreatment-period variables to derive a synthetic control group, and we use this group as a comparison to estimate the employment-treatment effect of participation in a large job-training program. We find the effect of treatment to be small and positive but transient. Our method reveals a strong regression to the mean effect, extremely easy to interpret as a treatment effect had a less advanced design been used (e.g. a within-subjects panel data analysis), and illustrates one of the unique advantages of using population register data for research purposes.

Causal inference in survival analysis using pseudo-observations.

PubMed

Andersen, Per K; Syriopoulou, Elisavet; Parner, Erik T

2017-07-30

Causal inference for non-censored response variables, such as binary or quantitative outcomes, is often based on either (1) direct standardization ('G-formula') or (2) inverse probability of treatment assignment weights ('propensity score'). To do causal inference in survival analysis, one needs to address right-censoring, and often, special techniques are required for that purpose. We will show how censoring can be dealt with 'once and for all' by means of so-called pseudo-observations when doing causal inference in survival analysis. The pseudo-observations can be used as a replacement of the outcomes without censoring when applying 'standard' causal inference methods, such as (1) or (2) earlier. We study this idea for estimating the average causal effect of a binary treatment on the survival probability, the restricted mean lifetime, and the cumulative incidence in a competing risks situation. The methods will be illustrated in a small simulation study and via a study of patients with acute myeloid leukemia who received either myeloablative or non-myeloablative conditioning before allogeneic hematopoetic cell transplantation. We will estimate the average causal effect of the conditioning regime on outcomes such as the 3-year overall survival probability and the 3-year risk of chronic graft-versus-host disease. Copyright © 2017 John Wiley & Sons, Ltd. Copyright © 2017 John Wiley & Sons, Ltd.

Driven by Power? Probe Question and Presentation Format Effects on Causal Judgment

ERIC Educational Resources Information Center

Perales, Jose C.; Shanks, David R.

2008-01-01

It has been proposed that causal power (defined as the probability with which a candidate cause would produce an effect in the absence of any other background causes) can be intuitively computed from cause-effect covariation information. Estimation of power is assumed to require a special type of counterfactual probe question, worded to remove…

Differences in Causal Estimates from Longitudinal Analyses of Residualized versus Simple Gain Scores: Contrasting Controls for Selection and Regression Artifacts

ERIC Educational Resources Information Center

Larzelere, Robert E.; Ferrer, Emilio; Kuhn, Brett R.; Danelia, Ketevan

2010-01-01

This study estimates the causal effects of six corrective actions for children's problem behaviors, comparing four types of longitudinal analyses that correct for pre-existing differences in a cohort of 1,464 4- and 5-year-olds from Canadian National Longitudinal Survey of Children and Youth (NLSCY) data. Analyses of residualized gain scores found…

Estimating Causal Effects of Local Air Pollution on Daily Deaths: Effect of Low Levels.

PubMed

Schwartz, Joel; Bind, Marie-Abele; Koutrakis, Petros

2017-01-01

Although many time-series studies have established associations of daily pollution variations with daily deaths, there are fewer at low concentrations, or focused on locally generated pollution, which is becoming more important as regulations reduce regional transport. Causal modeling approaches are also lacking. We used causal modeling to estimate the impact of local air pollution on mortality at low concentrations. Using an instrumental variable approach, we developed an instrument for variations in local pollution concentrations that is unlikely to be correlated with other causes of death, and examined its association with daily deaths in the Boston, Massachusetts, area. We combined height of the planetary boundary layer and wind speed, which affect concentrations of local emissions, to develop the instrument for particulate matter ≤ 2.5 μm (PM2.5), black carbon (BC), or nitrogen dioxide (NO2) variations that were independent of year, month, and temperature. We also used Granger causality to assess whether omitted variable confounding existed. We estimated that an interquartile range increase in the instrument for local PM2.5 was associated with a 0.90% increase in daily deaths (95% CI: 0.25, 1.56). A similar result was found for BC, and a weaker association with NO2. The Granger test found no evidence of omitted variable confounding for the instrument. A separate test confirmed the instrument was not associated with mortality independent of pollution. Furthermore, the association remained when all days with PM2.5 concentrations > 30 μg/m3 were excluded from the analysis (0.84% increase in daily deaths; 95% CI: 0.19, 1.50). We conclude that there is a causal association of local air pollution with daily deaths at concentrations below U.S. EPA standards. The estimated attributable risk in Boston exceeded 1,800 deaths during the study period, indicating that important public health benefits can follow from further control efforts. Citation: Schwartz J, Bind MA, Koutrakis P. 2017. Estimating causal effects of local air pollution on daily deaths: effect of low levels. Environ Health Perspect 125:23-29; http://dx.doi.org/10.1289/EHP232.

Age- and Sex-Specific Causal Effects of Adiposity on Cardiovascular Risk Factors

PubMed Central

Fall, Tove; Hägg, Sara; Ploner, Alexander; Mägi, Reedik; Fischer, Krista; Draisma, Harmen H.M.; Sarin, Antti-Pekka; Benyamin, Beben; Ladenvall, Claes; Åkerlund, Mikael; Kals, Mart; Esko, Tõnu; Nelson, Christopher P.; Kaakinen, Marika; Huikari, Ville; Mangino, Massimo; Meirhaeghe, Aline; Kristiansson, Kati; Nuotio, Marja-Liisa; Kobl, Michael; Grallert, Harald; Dehghan, Abbas; Kuningas, Maris; de Vries, Paul S.; de Bruijn, Renée F.A.G.; Willems, Sara M.; Heikkilä, Kauko; Silventoinen, Karri; Pietiläinen, Kirsi H.; Legry, Vanessa; Giedraitis, Vilmantas; Goumidi, Louisa; Syvänen, Ann-Christine; Strauch, Konstantin; Koenig, Wolfgang; Lichtner, Peter; Herder, Christian; Palotie, Aarno; Menni, Cristina; Uitterlinden, André G.; Kuulasmaa, Kari; Havulinna, Aki S.; Moreno, Luis A.; Gonzalez-Gross, Marcela; Evans, Alun; Tregouet, David-Alexandre; Yarnell, John W.G.; Virtamo, Jarmo; Ferrières, Jean; Veronesi, Giovanni; Perola, Markus; Arveiler, Dominique; Brambilla, Paolo; Lind, Lars; Kaprio, Jaakko; Hofman, Albert; Stricker, Bruno H.; van Duijn, Cornelia M.; Ikram, M. Arfan; Franco, Oscar H.; Cottel, Dominique; Dallongeville, Jean; Hall, Alistair S.; Jula, Antti; Tobin, Martin D.; Penninx, Brenda W.; Peters, Annette; Gieger, Christian; Samani, Nilesh J.; Montgomery, Grant W.; Whitfield, John B.; Martin, Nicholas G.; Groop, Leif; Spector, Tim D.; Magnusson, Patrik K.; Amouyel, Philippe; Boomsma, Dorret I.; Nilsson, Peter M.; Järvelin, Marjo-Riitta; Lyssenko, Valeriya; Metspalu, Andres; Strachan, David P.; Salomaa, Veikko; Ripatti, Samuli; Pedersen, Nancy L.; Prokopenko, Inga; McCarthy, Mark I.

2015-01-01

Observational studies have reported different effects of adiposity on cardiovascular risk factors across age and sex. Since cardiovascular risk factors are enriched in obese individuals, it has not been easy to dissect the effects of adiposity from those of other risk factors. We used a Mendelian randomization approach, applying a set of 32 genetic markers to estimate the causal effect of adiposity on blood pressure, glycemic indices, circulating lipid levels, and markers of inflammation and liver disease in up to 67,553 individuals. All analyses were stratified by age (cutoff 55 years of age) and sex. The genetic score was associated with BMI in both nonstratified analysis (P = 2.8 × 10−107) and stratified analyses (all P < 3.3 × 10−30). We found evidence of a causal effect of adiposity on blood pressure, fasting levels of insulin, C-reactive protein, interleukin-6, HDL cholesterol, and triglycerides in a nonstratified analysis and in the <55-year stratum. Further, we found evidence of a smaller causal effect on total cholesterol (P for difference = 0.015) in the ≥55-year stratum than in the <55-year stratum, a finding that could be explained by biology, survival bias, or differential medication. In conclusion, this study extends previous knowledge of the effects of adiposity by providing sex- and age-specific causal estimates on cardiovascular risk factors. PMID:25712996

A Kernel Embedding-Based Approach for Nonstationary Causal Model Inference.

PubMed

Hu, Shoubo; Chen, Zhitang; Chan, Laiwan

2018-05-01

Although nonstationary data are more common in the real world, most existing causal discovery methods do not take nonstationarity into consideration. In this letter, we propose a kernel embedding-based approach, ENCI, for nonstationary causal model inference where data are collected from multiple domains with varying distributions. In ENCI, we transform the complicated relation of a cause-effect pair into a linear model of variables of which observations correspond to the kernel embeddings of the cause-and-effect distributions in different domains. In this way, we are able to estimate the causal direction by exploiting the causal asymmetry of the transformed linear model. Furthermore, we extend ENCI to causal graph discovery for multiple variables by transforming the relations among them into a linear nongaussian acyclic model. We show that by exploiting the nonstationarity of distributions, both cause-effect pairs and two kinds of causal graphs are identifiable under mild conditions. Experiments on synthetic and real-world data are conducted to justify the efficacy of ENCI over major existing methods.

Assessing the causal effect of policies: an example using stochastic interventions.

PubMed

Díaz, Iván; van der Laan, Mark J

2013-11-19

Assessing the causal effect of an exposure often involves the definition of counterfactual outcomes in a hypothetical world in which the stochastic nature of the exposure is modified. Although stochastic interventions are a powerful tool to measure the causal effect of a realistic intervention that intends to alter the population distribution of an exposure, their importance to answer questions about plausible policy interventions has been obscured by the generalized use of deterministic interventions. In this article, we follow the approach described in Díaz and van der Laan (2012) to define and estimate the effect of an intervention that is expected to cause a truncation in the population distribution of the exposure. The observed data parameter that identifies the causal parameter of interest is established, as well as its efficient influence function under the non-parametric model. Inverse probability of treatment weighted (IPTW), augmented IPTW and targeted minimum loss-based estimators (TMLE) are proposed, their consistency and efficiency properties are determined. An extension to longitudinal data structures is presented and its use is demonstrated with a real data example.

Leverage effect and its causality in the Korea composite stock price index

NASA Astrophysics Data System (ADS)

Lee, Chang-Yong

2012-02-01

In this paper, we investigate the leverage effect and its causality in the time series of the Korea Composite Stock Price Index from November of 1997 to September of 2010. The leverage effect, which can be quantitatively expressed as a negative correlation between past return and future volatility, is measured by using the cross-correlation coefficient of different time lags between the two time series of the return and the volatility. We find that past return and future volatility are negatively correlated and that the cross correlation is moderate and decays over 60 trading days. We also carry out a partial correlation analysis in order to confirm that the negative correlation between past return and future volatility is neither an artifact nor influenced by the traded volume. To determine the causality of the leverage effect within the decay time, we additionally estimate the cross correlation between past volatility and future return. With the estimate, we perform a statistical hypothesis test to demonstrate that the causal relation is in favor of the return influencing the volatility rather than the other way around.

Covariate selection with group lasso and doubly robust estimation of causal effects

PubMed Central

Koch, Brandon; Vock, David M.; Wolfson, Julian

2017-01-01

Summary The efficiency of doubly robust estimators of the average causal effect (ACE) of a treatment can be improved by including in the treatment and outcome models only those covariates which are related to both treatment and outcome (i.e., confounders) or related only to the outcome. However, it is often challenging to identify such covariates among the large number that may be measured in a given study. In this paper, we propose GLiDeR (Group Lasso and Doubly Robust Estimation), a novel variable selection technique for identifying confounders and predictors of outcome using an adaptive group lasso approach that simultaneously performs coefficient selection, regularization, and estimation across the treatment and outcome models. The selected variables and corresponding coefficient estimates are used in a standard doubly robust ACE estimator. We provide asymptotic results showing that, for a broad class of data generating mechanisms, GLiDeR yields a consistent estimator of the ACE when either the outcome or treatment model is correctly specified. A comprehensive simulation study shows that GLiDeR is more efficient than doubly robust methods using standard variable selection techniques and has substantial computational advantages over a recently proposed doubly robust Bayesian model averaging method. We illustrate our method by estimating the causal treatment effect of bilateral versus single-lung transplant on forced expiratory volume in one year after transplant using an observational registry. PMID:28636276

Covariate selection with group lasso and doubly robust estimation of causal effects.

PubMed

Koch, Brandon; Vock, David M; Wolfson, Julian

2018-03-01

The efficiency of doubly robust estimators of the average causal effect (ACE) of a treatment can be improved by including in the treatment and outcome models only those covariates which are related to both treatment and outcome (i.e., confounders) or related only to the outcome. However, it is often challenging to identify such covariates among the large number that may be measured in a given study. In this article, we propose GLiDeR (Group Lasso and Doubly Robust Estimation), a novel variable selection technique for identifying confounders and predictors of outcome using an adaptive group lasso approach that simultaneously performs coefficient selection, regularization, and estimation across the treatment and outcome models. The selected variables and corresponding coefficient estimates are used in a standard doubly robust ACE estimator. We provide asymptotic results showing that, for a broad class of data generating mechanisms, GLiDeR yields a consistent estimator of the ACE when either the outcome or treatment model is correctly specified. A comprehensive simulation study shows that GLiDeR is more efficient than doubly robust methods using standard variable selection techniques and has substantial computational advantages over a recently proposed doubly robust Bayesian model averaging method. We illustrate our method by estimating the causal treatment effect of bilateral versus single-lung transplant on forced expiratory volume in one year after transplant using an observational registry. © 2017, The International Biometric Society.

Multivariate causal attribution and cost-effectiveness of a national mass media campaign in the Philippines.

PubMed

Kincaid, D Lawrence; Do, Mai Phuong

2006-01-01

Cost-effectiveness analysis is based on a simple formula. A dollar estimate of the total cost to conduct a program is divided by the number of people estimated to have been affected by it in terms of some intended outcome. The direct, total costs of most communication campaigns are usually available. Estimating the amount of effect that can be attributed to the communication alone, however is problematical in full-coverage, mass media campaigns where the randomized control group design is not feasible. Single-equation, multiple regression analysis controls for confounding variables but does not adequately address the issue of causal attribution. In this article, multivariate causal attribution (MCA) methods are applied to data from a sample survey of 1,516 married women in the Philippines to obtain a valid measure of the number of new adopters of modern contraceptives that can be causally attributed to a national mass media campaign and to calculate its cost-effectiveness. The MCA analysis uses structural equation modeling to test the causal pathways and to test for endogeneity, biprobit analysis to test for direct effects of the campaign and endogeneity, and propensity score matching to create a statistically equivalent, matched control group that approximates the results that would have been obtained from a randomized control group design. The MCA results support the conclusion that the observed, 6.4 percentage point increase in modern contraceptive use can be attributed to the national mass media campaign and to its indirect effects on attitudes toward contraceptives. This net increase represented 348,695 new adopters in the population of married women at a cost of U.S. $1.57 per new adopter.

Robust inference in summary data Mendelian randomization via the zero modal pleiotropy assumption.

PubMed

Hartwig, Fernando Pires; Davey Smith, George; Bowden, Jack

2017-12-01

Mendelian randomization (MR) is being increasingly used to strengthen causal inference in observational studies. Availability of summary data of genetic associations for a variety of phenotypes from large genome-wide association studies (GWAS) allows straightforward application of MR using summary data methods, typically in a two-sample design. In addition to the conventional inverse variance weighting (IVW) method, recently developed summary data MR methods, such as the MR-Egger and weighted median approaches, allow a relaxation of the instrumental variable assumptions. Here, a new method - the mode-based estimate (MBE) - is proposed to obtain a single causal effect estimate from multiple genetic instruments. The MBE is consistent when the largest number of similar (identical in infinite samples) individual-instrument causal effect estimates comes from valid instruments, even if the majority of instruments are invalid. We evaluate the performance of the method in simulations designed to mimic the two-sample summary data setting, and demonstrate its use by investigating the causal effect of plasma lipid fractions and urate levels on coronary heart disease risk. The MBE presented less bias and lower type-I error rates than other methods under the null in many situations. Its power to detect a causal effect was smaller compared with the IVW and weighted median methods, but was larger than that of MR-Egger regression, with sample size requirements typically smaller than those available from GWAS consortia. The MBE relaxes the instrumental variable assumptions, and should be used in combination with other approaches in sensitivity analyses. © The Author 2017. Published by Oxford University Press on behalf of the International Epidemiological Association

Semiparametric methods for estimation of a nonlinear exposure-outcome relationship using instrumental variables with application to Mendelian randomization.

PubMed

Staley, James R; Burgess, Stephen

2017-05-01

Mendelian randomization, the use of genetic variants as instrumental variables (IV), can test for and estimate the causal effect of an exposure on an outcome. Most IV methods assume that the function relating the exposure to the expected value of the outcome (the exposure-outcome relationship) is linear. However, in practice, this assumption may not hold. Indeed, often the primary question of interest is to assess the shape of this relationship. We present two novel IV methods for investigating the shape of the exposure-outcome relationship: a fractional polynomial method and a piecewise linear method. We divide the population into strata using the exposure distribution, and estimate a causal effect, referred to as a localized average causal effect (LACE), in each stratum of population. The fractional polynomial method performs metaregression on these LACE estimates. The piecewise linear method estimates a continuous piecewise linear function, the gradient of which is the LACE estimate in each stratum. Both methods were demonstrated in a simulation study to estimate the true exposure-outcome relationship well, particularly when the relationship was a fractional polynomial (for the fractional polynomial method) or was piecewise linear (for the piecewise linear method). The methods were used to investigate the shape of relationship of body mass index with systolic blood pressure and diastolic blood pressure. © 2017 The Authors Genetic Epidemiology Published by Wiley Periodicals, Inc.

Semiparametric methods for estimation of a nonlinear exposure‐outcome relationship using instrumental variables with application to Mendelian randomization

PubMed Central

Staley, James R.

2017-01-01

ABSTRACT Mendelian randomization, the use of genetic variants as instrumental variables (IV), can test for and estimate the causal effect of an exposure on an outcome. Most IV methods assume that the function relating the exposure to the expected value of the outcome (the exposure‐outcome relationship) is linear. However, in practice, this assumption may not hold. Indeed, often the primary question of interest is to assess the shape of this relationship. We present two novel IV methods for investigating the shape of the exposure‐outcome relationship: a fractional polynomial method and a piecewise linear method. We divide the population into strata using the exposure distribution, and estimate a causal effect, referred to as a localized average causal effect (LACE), in each stratum of population. The fractional polynomial method performs metaregression on these LACE estimates. The piecewise linear method estimates a continuous piecewise linear function, the gradient of which is the LACE estimate in each stratum. Both methods were demonstrated in a simulation study to estimate the true exposure‐outcome relationship well, particularly when the relationship was a fractional polynomial (for the fractional polynomial method) or was piecewise linear (for the piecewise linear method). The methods were used to investigate the shape of relationship of body mass index with systolic blood pressure and diastolic blood pressure. PMID:28317167

Inverse odds ratio-weighted estimation for causal mediation analysis.

PubMed

Tchetgen Tchetgen, Eric J

2013-11-20

An important scientific goal of studies in the health and social sciences is increasingly to determine to what extent the total effect of a point exposure is mediated by an intermediate variable on the causal pathway between the exposure and the outcome. A causal framework has recently been proposed for mediation analysis, which gives rise to new definitions, formal identification results and novel estimators of direct and indirect effects. In the present paper, the author describes a new inverse odds ratio-weighted approach to estimate so-called natural direct and indirect effects. The approach, which uses as a weight the inverse of an estimate of the odds ratio function relating the exposure and the mediator, is universal in that it can be used to decompose total effects in a number of regression models commonly used in practice. Specifically, the approach may be used for effect decomposition in generalized linear models with a nonlinear link function, and in a number of other commonly used models such as the Cox proportional hazards regression for a survival outcome. The approach is simple and can be implemented in standard software provided a weight can be specified for each observation. An additional advantage of the method is that it easily incorporates multiple mediators of a categorical, discrete or continuous nature. Copyright © 2013 John Wiley & Sons, Ltd.

The causal effect of education on HIV stigma in Uganda: Evidence from a natural experiment.

PubMed

Tsai, Alexander C; Venkataramani, Atheendar S

2015-10-01

HIV is highly stigmatized in sub-Saharan Africa. This is an important public health problem because HIV stigma has many adverse effects that threaten to undermine efforts to control the HIV epidemic. The implementation of a universal primary education policy in Uganda in 1997 provided us with a natural experiment to test the hypothesis that education is causally related to HIV stigma. For this analysis, we pooled publicly available, population-based data from the 2011 Uganda Demographic and Health Survey and the 2011 Uganda AIDS Indicator Survey. The primary outcomes of interest were negative attitudes toward persons with HIV, elicited using four questions about anticipated stigma and social distance. Standard least squares estimates suggested a statistically significant, negative association between years of schooling and HIV stigma (each P < 0.001, with t-statistics ranging from 4.9 to 14.7). We then used a natural experiment design, exploiting differences in birth cohort exposure to universal primary education as an instrumental variable. Participants who were <13 years old at the time of the policy change had 1.36 additional years of schooling compared to those who were ≥13 years old. Adjusting for linear age trends before and after the discontinuity, two-stage least squares estimates suggested no statistically significant causal effect of education on HIV stigma (P-values ranged from 0.21 to 0.69). Three of the four estimated regression coefficients were positive, and in all cases the lower confidence limits convincingly excluded the possibility of large negative effect sizes. These instrumental variables estimates have a causal interpretation and were not overturned by several robustness checks. We conclude that, for young adults in Uganda, additional years of education in the formal schooling system driven by a universal primary school intervention have not had a causal effect on reducing negative attitudes toward persons with HIV. Copyright © 2015 Elsevier Ltd. All rights reserved.

The causal effect of education on HIV stigma in Uganda: evidence from a natural experiment

PubMed Central

Tsai, Alexander C.; Venkataramani, Atheendar S.

2015-01-01

Rationale HIV is highly stigmatized in sub-Saharan Africa. This is an important public health problem because HIV stigma has many adverse effects that threaten to undermine efforts to control the HIV epidemic. Objective The implementation of a universal primary education policy in Uganda in 1997 provided us with a natural experiment to test the hypothesis that education is causally related to HIV stigma. Methods For this analysis, we pooled publicly available, population-based data from the 2011 Uganda Demographic and Health Survey and the 2011 Uganda AIDS Indicator Survey. The primary outcomes of interest were negative attitudes toward persons with HIV, elicited using four questions about anticipated stigma and social distance. Results Standard least squares estimates suggested a statistically significant, negative association between years of schooling and HIV stigma (each P<0.001, with t-statistics ranging from 4.9 to 14.7). We then used a natural experiment design, exploiting differences in birth cohort exposure to universal primary education as an instrumental variable. Participants who were <13 years old at the time of the policy change had 1.36 additional years of schooling compared to those who were ≥13 years old. Adjusting for linear age trends before and after the discontinuity, two-stage least squares estimates suggested no statistically significant causal effect of education on HIV stigma (P-values ranged from 0.21 to 0.69). Three of the four estimated regression coefficients were positive, and in all cases the lower confidence limits convincingly excluded the possibility of large negative effect sizes. These instrumental variables estimates have a causal interpretation and were not overturned by several robustness checks. Conclusion We conclude that, for young adults in Uganda, additional years of education in the formal schooling system driven by a universal primary school intervention have not had a causal effect on reducing negative attitudes toward persons with HIV. PMID:26282707

Semiparametric Estimation of the Impacts of Longitudinal Interventions on Adolescent Obesity using Targeted Maximum-Likelihood: Accessible Estimation with the ltmle Package

PubMed Central

Decker, Anna L.; Hubbard, Alan; Crespi, Catherine M.; Seto, Edmund Y.W.; Wang, May C.

2015-01-01

While child and adolescent obesity is a serious public health concern, few studies have utilized parameters based on the causal inference literature to examine the potential impacts of early intervention. The purpose of this analysis was to estimate the causal effects of early interventions to improve physical activity and diet during adolescence on body mass index (BMI), a measure of adiposity, using improved techniques. The most widespread statistical method in studies of child and adolescent obesity is multi-variable regression, with the parameter of interest being the coefficient on the variable of interest. This approach does not appropriately adjust for time-dependent confounding, and the modeling assumptions may not always be met. An alternative parameter to estimate is one motivated by the causal inference literature, which can be interpreted as the mean change in the outcome under interventions to set the exposure of interest. The underlying data-generating distribution, upon which the estimator is based, can be estimated via a parametric or semi-parametric approach. Using data from the National Heart, Lung, and Blood Institute Growth and Health Study, a 10-year prospective cohort study of adolescent girls, we estimated the longitudinal impact of physical activity and diet interventions on 10-year BMI z-scores via a parameter motivated by the causal inference literature, using both parametric and semi-parametric estimation approaches. The parameters of interest were estimated with a recently released R package, ltmle, for estimating means based upon general longitudinal treatment regimes. We found that early, sustained intervention on total calories had a greater impact than a physical activity intervention or non-sustained interventions. Multivariable linear regression yielded inflated effect estimates compared to estimates based on targeted maximum-likelihood estimation and data-adaptive super learning. Our analysis demonstrates that sophisticated, optimal semiparametric estimation of longitudinal treatment-specific means via ltmle provides an incredibly powerful, yet easy-to-use tool, removing impediments for putting theory into practice. PMID:26046009

Estimation of Causal Mediation Effects for a Dichotomous Outcome in Multiple-Mediator Models using the Mediation Formula

PubMed Central

Nelson, Suchitra; Albert, Jeffrey M.

2013-01-01

Mediators are intermediate variables in the causal pathway between an exposure and an outcome. Mediation analysis investigates the extent to which exposure effects occur through these variables, thus revealing causal mechanisms. In this paper, we consider the estimation of the mediation effect when the outcome is binary and multiple mediators of different types exist. We give a precise definition of the total mediation effect as well as decomposed mediation effects through individual or sets of mediators using the potential outcomes framework. We formulate a model of joint distribution (probit-normal) using continuous latent variables for any binary mediators to account for correlations among multiple mediators. A mediation formula approach is proposed to estimate the total mediation effect and decomposed mediation effects based on this parametric model. Estimation of mediation effects through individual or subsets of mediators requires an assumption involving the joint distribution of multiple counterfactuals. We conduct a simulation study that demonstrates low bias of mediation effect estimators for two-mediator models with various combinations of mediator types. The results also show that the power to detect a non-zero total mediation effect increases as the correlation coefficient between two mediators increases, while power for individual mediation effects reaches a maximum when the mediators are uncorrelated. We illustrate our approach by applying it to a retrospective cohort study of dental caries in adolescents with low and high socioeconomic status. Sensitivity analysis is performed to assess the robustness of conclusions regarding mediation effects when the assumption of no unmeasured mediator-outcome confounders is violated. PMID:23650048

Estimation of causal mediation effects for a dichotomous outcome in multiple-mediator models using the mediation formula.

PubMed

Wang, Wei; Nelson, Suchitra; Albert, Jeffrey M

2013-10-30

Mediators are intermediate variables in the causal pathway between an exposure and an outcome. Mediation analysis investigates the extent to which exposure effects occur through these variables, thus revealing causal mechanisms. In this paper, we consider the estimation of the mediation effect when the outcome is binary and multiple mediators of different types exist. We give a precise definition of the total mediation effect as well as decomposed mediation effects through individual or sets of mediators using the potential outcomes framework. We formulate a model of joint distribution (probit-normal) using continuous latent variables for any binary mediators to account for correlations among multiple mediators. A mediation formula approach is proposed to estimate the total mediation effect and decomposed mediation effects based on this parametric model. Estimation of mediation effects through individual or subsets of mediators requires an assumption involving the joint distribution of multiple counterfactuals. We conduct a simulation study that demonstrates low bias of mediation effect estimators for two-mediator models with various combinations of mediator types. The results also show that the power to detect a nonzero total mediation effect increases as the correlation coefficient between two mediators increases, whereas power for individual mediation effects reaches a maximum when the mediators are uncorrelated. We illustrate our approach by applying it to a retrospective cohort study of dental caries in adolescents with low and high socioeconomic status. Sensitivity analysis is performed to assess the robustness of conclusions regarding mediation effects when the assumption of no unmeasured mediator-outcome confounders is violated. Copyright © 2013 John Wiley & Sons, Ltd.

Analysis of longitudinal marginal structural models.

PubMed

Bryan, Jenny; Yu, Zhuo; Van Der Laan, Mark J

2004-07-01

In this article we construct and study estimators of the causal effect of a time-dependent treatment on survival in longitudinal studies. We employ a particular marginal structural model (MSM), proposed by Robins (2000), and follow a general methodology for constructing estimating functions in censored data models. The inverse probability of treatment weighted (IPTW) estimator of Robins et al. (2000) is used as an initial estimator and forms the basis for an improved, one-step estimator that is consistent and asymptotically linear when the treatment mechanism is consistently estimated. We extend these methods to handle informative censoring. The proposed methodology is employed to estimate the causal effect of exercise on mortality in a longitudinal study of seniors in Sonoma County. A simulation study demonstrates the bias of naive estimators in the presence of time-dependent confounders and also shows the efficiency gain of the IPTW estimator, even in the absence such confounding. The efficiency gain of the improved, one-step estimator is demonstrated through simulation.

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…

Negative control exposure studies in the presence of measurement error: implications for attempted effect estimate calibration

PubMed Central

Sanderson, Eleanor; Macdonald-Wallis, Corrie; Davey Smith, George

2018-01-01

Abstract Background Negative control exposure studies are increasingly being used in epidemiological studies to strengthen causal inference regarding an exposure-outcome association when unobserved confounding is thought to be present. Negative control exposure studies contrast the magnitude of association of the negative control, which has no causal effect on the outcome but is associated with the unmeasured confounders in the same way as the exposure, with the magnitude of the association of the exposure with the outcome. A markedly larger effect of the exposure on the outcome than the negative control on the outcome strengthens inference that the exposure has a causal effect on the outcome. Methods We investigate the effect of measurement error in the exposure and negative control variables on the results obtained from a negative control exposure study. We do this in models with continuous and binary exposure and negative control variables using analysis of the bias of the estimated coefficients and Monte Carlo simulations. Results Our results show that measurement error in either the exposure or negative control variables can bias the estimated results from the negative control exposure study. Conclusions Measurement error is common in the variables used in epidemiological studies; these results show that negative control exposure studies cannot be used to precisely determine the size of the effect of the exposure variable, or adequately adjust for unobserved confounding; however, they can be used as part of a body of evidence to aid inference as to whether a causal effect of the exposure on the outcome is present. PMID:29088358

Negative control exposure studies in the presence of measurement error: implications for attempted effect estimate calibration.

PubMed

Sanderson, Eleanor; Macdonald-Wallis, Corrie; Davey Smith, George

2018-04-01

Negative control exposure studies are increasingly being used in epidemiological studies to strengthen causal inference regarding an exposure-outcome association when unobserved confounding is thought to be present. Negative control exposure studies contrast the magnitude of association of the negative control, which has no causal effect on the outcome but is associated with the unmeasured confounders in the same way as the exposure, with the magnitude of the association of the exposure with the outcome. A markedly larger effect of the exposure on the outcome than the negative control on the outcome strengthens inference that the exposure has a causal effect on the outcome. We investigate the effect of measurement error in the exposure and negative control variables on the results obtained from a negative control exposure study. We do this in models with continuous and binary exposure and negative control variables using analysis of the bias of the estimated coefficients and Monte Carlo simulations. Our results show that measurement error in either the exposure or negative control variables can bias the estimated results from the negative control exposure study. Measurement error is common in the variables used in epidemiological studies; these results show that negative control exposure studies cannot be used to precisely determine the size of the effect of the exposure variable, or adequately adjust for unobserved confounding; however, they can be used as part of a body of evidence to aid inference as to whether a causal effect of the exposure on the outcome is present.

Estimating Effects with Rare Outcomes and High Dimensional Covariates: Knowledge is Power

PubMed Central

Ahern, Jennifer; Galea, Sandro; van der Laan, Mark

2016-01-01

Many of the secondary outcomes in observational studies and randomized trials are rare. Methods for estimating causal effects and associations with rare outcomes, however, are limited, and this represents a missed opportunity for investigation. In this article, we construct a new targeted minimum loss-based estimator (TMLE) for the effect or association of an exposure on a rare outcome. We focus on the causal risk difference and statistical models incorporating bounds on the conditional mean of the outcome, given the exposure and measured confounders. By construction, the proposed estimator constrains the predicted outcomes to respect this model knowledge. Theoretically, this bounding provides stability and power to estimate the exposure effect. In finite sample simulations, the proposed estimator performed as well, if not better, than alternative estimators, including a propensity score matching estimator, inverse probability of treatment weighted (IPTW) estimator, augmented-IPTW and the standard TMLE algorithm. The new estimator yielded consistent estimates if either the conditional mean outcome or the propensity score was consistently estimated. As a substitution estimator, TMLE guaranteed the point estimates were within the parameter range. We applied the estimator to investigate the association between permissive neighborhood drunkenness norms and alcohol use disorder. Our results highlight the potential for double robust, semiparametric efficient estimation with rare events and high dimensional covariates. PMID:28529839

Principal Stratification — a Goal or a Tool?

PubMed Central

Pearl, Judea

2011-01-01

Principal stratification has recently become a popular tool to address certain causal inference questions, particularly in dealing with post-randomization factors in randomized trials. Here, we analyze the conceptual basis for this framework and invite response to clarify the value of principal stratification in estimating causal effects of interest. PMID:21556288

Regression Discontinuity for Causal Effect Estimation in Epidemiology.

PubMed

Oldenburg, Catherine E; Moscoe, Ellen; Bärnighausen, Till

Regression discontinuity analyses can generate estimates of the causal effects of an exposure when a continuously measured variable is used to assign the exposure to individuals based on a threshold rule. Individuals just above the threshold are expected to be similar in their distribution of measured and unmeasured baseline covariates to individuals just below the threshold, resulting in exchangeability. At the threshold exchangeability is guaranteed if there is random variation in the continuous assignment variable, e.g., due to random measurement error. Under exchangeability, causal effects can be identified at the threshold. The regression discontinuity intention-to-treat (RD-ITT) effect on an outcome can be estimated as the difference in the outcome between individuals just above (or below) versus just below (or above) the threshold. This effect is analogous to the ITT effect in a randomized controlled trial. Instrumental variable methods can be used to estimate the effect of exposure itself utilizing the threshold as the instrument. We review the recent epidemiologic literature reporting regression discontinuity studies and find that while regression discontinuity designs are beginning to be utilized in a variety of applications in epidemiology, they are still relatively rare, and analytic and reporting practices vary. Regression discontinuity has the potential to greatly contribute to the evidence base in epidemiology, in particular on the real-life and long-term effects and side-effects of medical treatments that are provided based on threshold rules - such as treatments for low birth weight, hypertension or diabetes.

Effectiveness of state-level vaccination mandates: evidence from the varicella vaccine.

PubMed

Abrevaya, Jason; Mulligan, Karen

2011-09-01

This paper utilizes longitudinal data on varicella (chickenpox) immunizations in order to estimate the causal effects of state-level school-entry and daycare-entry immunization mandates within the United States. We find significant causal effects of mandates upon vaccination rates among preschool children aged 19-35 months; these effects appear in the year of mandate adoption, peak two years after adoption, and show a minimal difference from the aggregate trend about six years after adoption. For a mandate enacted in 2000, the model and estimates imply that roughly 20% of the short-run increase in state-level immunization rates was caused by the mandate introduction. We find no evidence of differential effects for different socioeconomic groups. Combined with previous cost-benefit analyses of the varicella vaccine, the estimates suggest that state-level mandates have been effective from an economic standpoint. Copyright © 2011 Elsevier B.V. All rights reserved.

EARLY, LATE OR NEVER? WHEN DOES PARENTAL EDUCATION IMPACT CHILD OUTCOMES?

PubMed Central

Dickson, Matt; Gregg, Paul; Robinson, Harriet

2017-01-01

We estimate the causal effect of parents’ education on their children’s education and examine the timing of the impact. We identify the causal effect by exploiting the exogenous shift in (parents’) education levels induced by the 1972 minimum school leaving age reform in England. Increasing parental education has a positive causal effect on children’s outcomes that is evident in preschool assessments at age 4 and continues to be visible up to and including high-stakes examinations taken at age 16. Children of parents affected by the reform attain results around 0.1 standard deviations higher than those whose parents were not impacted. PMID:28736454

What Can We Learn about Effective Early Mathematics Teaching? A Framework for Estimating Causal Effects Using Longitudinal Survey Data

ERIC Educational Resources Information Center

Guarino, Cassandra; Dieterle, Steven G.; Bargagliotti, Anna E.; Mason, William M.

2013-01-01

This study investigates the impact of teacher characteristics and instructional strategies on the mathematics achievement of students in kindergarten and first grade and tackles the question of how best to use longitudinal survey data to elicit causal inference in the face of potential threats to validity due to nonrandom assignment to treatment.…

New Evidence of the Causal Effect of Family Size on Child Quality in a Developing Country

ERIC Educational Resources Information Center

Ponczek, Vladimir; Souza, Andre Portela

2012-01-01

This paper presents new evidence of the causal effect of family size on child quality in a developing-country context. We estimate the impact of family size on child labor and educational outcomes among Brazilian children and young adults by exploring the exogenous variation of family size driven by the presence of twins in the family. Using the…

CAUSAL INFERENCE WITH A GRAPHICAL HIERARCHY OF INTERVENTIONS

PubMed Central

Shpitser, Ilya; Tchetgen, Eric Tchetgen

2017-01-01

Identifying causal parameters from observational data is fraught with subtleties due to the issues of selection bias and confounding. In addition, more complex questions of interest, such as effects of treatment on the treated and mediated effects may not always be identified even in data where treatment assignment is known and under investigator control, or may be identified under one causal model but not another. Increasingly complex effects of interest, coupled with a diversity of causal models in use resulted in a fragmented view of identification. This fragmentation makes it unnecessarily difficult to determine if a given parameter is identified (and in what model), and what assumptions must hold for this to be the case. This, in turn, complicates the development of estimation theory and sensitivity analysis procedures. In this paper, we give a unifying view of a large class of causal effects of interest, including novel effects not previously considered, in terms of a hierarchy of interventions, and show that identification theory for this large class reduces to an identification theory of random variables under interventions from this hierarchy. Moreover, we show that one type of intervention in the hierarchy is naturally associated with queries identified under the Finest Fully Randomized Causally Interpretable Structure Tree Graph (FFRCISTG) model of Robins (via the extended g-formula), and another is naturally associated with queries identified under the Non-Parametric Structural Equation Model with Independent Errors (NPSEM-IE) of Pearl, via a more general functional we call the edge g-formula. Our results motivate the study of estimation theory for the edge g-formula, since we show it arises both in mediation analysis, and in settings where treatment assignment has unobserved causes, such as models associated with Pearl’s front-door criterion. PMID:28919652

CAUSAL INFERENCE WITH A GRAPHICAL HIERARCHY OF INTERVENTIONS.

PubMed

Shpitser, Ilya; Tchetgen, Eric Tchetgen

2016-12-01

Identifying causal parameters from observational data is fraught with subtleties due to the issues of selection bias and confounding. In addition, more complex questions of interest, such as effects of treatment on the treated and mediated effects may not always be identified even in data where treatment assignment is known and under investigator control, or may be identified under one causal model but not another. Increasingly complex effects of interest, coupled with a diversity of causal models in use resulted in a fragmented view of identification. This fragmentation makes it unnecessarily difficult to determine if a given parameter is identified (and in what model), and what assumptions must hold for this to be the case. This, in turn, complicates the development of estimation theory and sensitivity analysis procedures. In this paper, we give a unifying view of a large class of causal effects of interest, including novel effects not previously considered, in terms of a hierarchy of interventions, and show that identification theory for this large class reduces to an identification theory of random variables under interventions from this hierarchy. Moreover, we show that one type of intervention in the hierarchy is naturally associated with queries identified under the Finest Fully Randomized Causally Interpretable Structure Tree Graph (FFRCISTG) model of Robins (via the extended g-formula), and another is naturally associated with queries identified under the Non-Parametric Structural Equation Model with Independent Errors (NPSEM-IE) of Pearl, via a more general functional we call the edge g-formula. Our results motivate the study of estimation theory for the edge g-formula, since we show it arises both in mediation analysis, and in settings where treatment assignment has unobserved causes, such as models associated with Pearl's front-door criterion.

Age- and sex-specific causal effects of adiposity on cardiovascular risk factors.

PubMed

Fall, Tove; Hägg, Sara; Ploner, Alexander; Mägi, Reedik; Fischer, Krista; Draisma, Harmen H M; Sarin, Antti-Pekka; Benyamin, Beben; Ladenvall, Claes; Åkerlund, Mikael; Kals, Mart; Esko, Tõnu; Nelson, Christopher P; Kaakinen, Marika; Huikari, Ville; Mangino, Massimo; Meirhaeghe, Aline; Kristiansson, Kati; Nuotio, Marja-Liisa; Kobl, Michael; Grallert, Harald; Dehghan, Abbas; Kuningas, Maris; de Vries, Paul S; de Bruijn, Renée F A G; Willems, Sara M; Heikkilä, Kauko; Silventoinen, Karri; Pietiläinen, Kirsi H; Legry, Vanessa; Giedraitis, Vilmantas; Goumidi, Louisa; Syvänen, Ann-Christine; Strauch, Konstantin; Koenig, Wolfgang; Lichtner, Peter; Herder, Christian; Palotie, Aarno; Menni, Cristina; Uitterlinden, André G; Kuulasmaa, Kari; Havulinna, Aki S; Moreno, Luis A; Gonzalez-Gross, Marcela; Evans, Alun; Tregouet, David-Alexandre; Yarnell, John W G; Virtamo, Jarmo; Ferrières, Jean; Veronesi, Giovanni; Perola, Markus; Arveiler, Dominique; Brambilla, Paolo; Lind, Lars; Kaprio, Jaakko; Hofman, Albert; Stricker, Bruno H; van Duijn, Cornelia M; Ikram, M Arfan; Franco, Oscar H; Cottel, Dominique; Dallongeville, Jean; Hall, Alistair S; Jula, Antti; Tobin, Martin D; Penninx, Brenda W; Peters, Annette; Gieger, Christian; Samani, Nilesh J; Montgomery, Grant W; Whitfield, John B; Martin, Nicholas G; Groop, Leif; Spector, Tim D; Magnusson, Patrik K; Amouyel, Philippe; Boomsma, Dorret I; Nilsson, Peter M; Järvelin, Marjo-Riitta; Lyssenko, Valeriya; Metspalu, Andres; Strachan, David P; Salomaa, Veikko; Ripatti, Samuli; Pedersen, Nancy L; Prokopenko, Inga; McCarthy, Mark I; Ingelsson, Erik

2015-05-01

Observational studies have reported different effects of adiposity on cardiovascular risk factors across age and sex. Since cardiovascular risk factors are enriched in obese individuals, it has not been easy to dissect the effects of adiposity from those of other risk factors. We used a Mendelian randomization approach, applying a set of 32 genetic markers to estimate the causal effect of adiposity on blood pressure, glycemic indices, circulating lipid levels, and markers of inflammation and liver disease in up to 67,553 individuals. All analyses were stratified by age (cutoff 55 years of age) and sex. The genetic score was associated with BMI in both nonstratified analysis (P = 2.8 × 10(-107)) and stratified analyses (all P < 3.3 × 10(-30)). We found evidence of a causal effect of adiposity on blood pressure, fasting levels of insulin, C-reactive protein, interleukin-6, HDL cholesterol, and triglycerides in a nonstratified analysis and in the <55-year stratum. Further, we found evidence of a smaller causal effect on total cholesterol (P for difference = 0.015) in the ≥55-year stratum than in the <55-year stratum, a finding that could be explained by biology, survival bias, or differential medication. In conclusion, this study extends previous knowledge of the effects of adiposity by providing sex- and age-specific causal estimates on cardiovascular risk factors. © 2015 by the American Diabetes Association. Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered.

Sequential causal inference: Application to randomized trials of adaptive treatment strategies

PubMed Central

Dawson, Ree; Lavori, Philip W.

2009-01-01

SUMMARY Clinical trials that randomize subjects to decision algorithms, which adapt treatments over time according to individual response, have gained considerable interest as investigators seek designs that directly inform clinical decision making. We consider designs in which subjects are randomized sequentially at decision points, among adaptive treatment options under evaluation. We present a sequential method to estimate the comparative effects of the randomized adaptive treatments, which are formalized as adaptive treatment strategies. Our causal estimators are derived using Bayesian predictive inference. We use analytical and empirical calculations to compare the predictive estimators to (i) the ‘standard’ approach that allocates the sequentially obtained data to separate strategy-specific groups as would arise from randomizing subjects at baseline; (ii) the semi-parametric approach of marginal mean models that, under appropriate experimental conditions, provides the same sequential estimator of causal differences as the proposed approach. Simulation studies demonstrate that sequential causal inference offers substantial efficiency gains over the standard approach to comparing treatments, because the predictive estimators can take advantage of the monotone structure of shared data among adaptive strategies. We further demonstrate that the semi-parametric asymptotic variances, which are marginal ‘one-step’ estimators, may exhibit significant bias, in contrast to the predictive variances. We show that the conditions under which the sequential method is attractive relative to the other two approaches are those most likely to occur in real studies. PMID:17914714

Does finance affect environmental degradation: evidence from One Belt and One Road Initiative region?

PubMed

Hafeez, Muhammad; Chunhui, Yuan; Strohmaier, David; Ahmed, Manzoor; Jie, Liu

2018-04-01

This paper explores the effects of finance on environmental degradation and investigates environmental Kuznets curve (EKC) of each country among 52 that participate in the One Belt and One Road Initiative (OBORI) using the latest long panel data span (1980-2016). We utilized panel long run econometric models (fully modified ordinary least square and dynamic ordinary least square) to explore the long-run estimates in full panel and country level. Moreover, the Dumitrescu and Hurlin (2012) causality test is applied to examine the short-run causalities among our considered variables. The empirical findings validate the EKC hypothesis; the long-run estimates point out that finance significantly enhances the environmental degradation (negatively in few cases). The short-run heterogeneous causality confirms the bi-directional causality between finance and environmental degradation. The empirical outcomes suggest that policymakers should consider the environmental degradation issue caused by financial development in the One Belt and One Road region.

Understanding bias in relationships between the food environment and diet quality: the Coronary Artery Risk Development in Young Adults (CARDIA) study.

PubMed

Rummo, Pasquale E; Guilkey, David K; Ng, Shu Wen; Meyer, Katie A; Popkin, Barry M; Reis, Jared P; Shikany, James M; Gordon-Larsen, Penny

2017-12-01

The relationship between food environment exposures and diet behaviours is unclear, possibly because the majority of studies ignore potential residual confounding. We used 20 years (1985-1986, 1992-1993 2005-2006) of data from the Coronary Artery Risk Development in Young Adults (CARDIA) study across four US cities (Birmingham, Alabama; Chicago, Illinois; Minneapolis, Minnesota; Oakland, California) and instrumental variables (IV) regression to obtain causal estimates of longitudinal associations between the percentage of neighbourhood food outlets (per total food outlets within 1 km network distance of respondent residence) and an a priori diet quality score, with higher scores indicating higher diet quality. To assess the presence and magnitude of bias related to residual confounding, we compared results from causal models (IV regression) to non-causal models, including ordinary least squares regression, which does not account for residual confounding at all and fixed-effects regression, which only controls for time-invariant unmeasured characteristics. The mean diet quality score across follow-up was 63.4 (SD=12.7). A 10% increase in fast food restaurants (relative to full-service restaurants) was associated with a lower diet quality score over time using IV regression (β=-1.01, 95% CI -1.99 to -0.04); estimates were attenuated using non-causal models. The percentage of neighbourhood convenience and grocery stores (relative to supermarkets) was not associated with diet quality in any model, but estimates from non-causal models were similarly attenuated compared with causal models. Ignoring residual confounding may generate biased estimated effects of neighbourhood food outlets on diet outcomes and may have contributed to weak findings in the food environment literature. © 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.

Estimating Causal Effects in Observational Studies using Electronic Health Data: Challenges and (Some) Solutions

PubMed Central

Stuart, Elizabeth A.; DuGoff, Eva; Abrams, Michael; Salkever, David; Steinwachs, Donald

2013-01-01

Electronic health data sets, including electronic health records (EHR) and other administrative databases, are rich data sources that have the potential to help answer important questions about the effects of clinical interventions as well as policy changes. However, analyses using such data are almost always non-experimental, leading to concerns that those who receive a particular intervention are likely different from those who do not in ways that may confound the effects of interest. This paper outlines the challenges in estimating causal effects using electronic health data and offers some solutions, with particular attention paid to propensity score methods that help ensure comparisons between similar groups. The methods are illustrated with a case study describing the design of a study using Medicare and Medicaid administrative data to estimate the effect of the Medicare Part D prescription drug program on individuals with serious mental illness. PMID:24921064

The Use of Linear Instrumental Variables Methods in Health Services Research and Health Economics: A Cautionary Note

PubMed Central

Terza, Joseph V; Bradford, W David; Dismuke, Clara E

2008-01-01

Objective To investigate potential bias in the use of the conventional linear instrumental variables (IV) method for the estimation of causal effects in inherently nonlinear regression settings. Data Sources Smoking Supplement to the 1979 National Health Interview Survey, National Longitudinal Alcohol Epidemiologic Survey, and simulated data. Study Design Potential bias from the use of the linear IV method in nonlinear models is assessed via simulation studies and real world data analyses in two commonly encountered regression setting: (1) models with a nonnegative outcome (e.g., a count) and a continuous endogenous regressor; and (2) models with a binary outcome and a binary endogenous regressor. Principle Findings The simulation analyses show that substantial bias in the estimation of causal effects can result from applying the conventional IV method in inherently nonlinear regression settings. Moreover, the bias is not attenuated as the sample size increases. This point is further illustrated in the survey data analyses in which IV-based estimates of the relevant causal effects diverge substantially from those obtained with appropriate nonlinear estimation methods. Conclusions We offer this research as a cautionary note to those who would opt for the use of linear specifications in inherently nonlinear settings involving endogeneity. PMID:18546544

A Bayesian Nonparametric Causal Model for Regression Discontinuity Designs

ERIC Educational Resources Information Center

Karabatsos, George; Walker, Stephen G.

2013-01-01

The regression discontinuity (RD) design (Thistlewaite & Campbell, 1960; Cook, 2008) provides a framework to identify and estimate causal effects from a non-randomized design. Each subject of a RD design is assigned to the treatment (versus assignment to a non-treatment) whenever her/his observed value of the assignment variable equals or…

A Note on the Usefulness of the Behavioural Rasch Selection Model for Causal Inference in the Social Sciences

NASA Astrophysics Data System (ADS)

Rabbitt, Matthew P.

2016-11-01

Social scientists are often interested in examining causal relationships where the outcome of interest is represented by an intangible concept, such as an individual's well-being or ability. Estimating causal relationships in this scenario is particularly challenging because the social scientist must rely on measurement models to measure individual's properties or attributes and then address issues related to survey data, such as omitted variables. In this paper, the usefulness of the recently proposed behavioural Rasch selection model is explored using a series of Monte Carlo experiments. The behavioural Rasch selection model is particularly useful for these types of applications because it is capable of estimating the causal effect of a binary treatment effect on an outcome that is represented by an intangible concept using cross-sectional data. Other methodology typically relies of summary measures from measurement models that require additional assumptions, some of which make these approaches less efficient. Recommendations for application of the behavioural Rasch selection model are made based on results from the Monte Carlo experiments.

Applying causal mediation analysis to personality disorder research.

PubMed

Walters, Glenn D

2018-01-01

This article is designed to address fundamental issues in the application of causal mediation analysis to research on personality disorders. Causal mediation analysis is used to identify mechanisms of effect by testing variables as putative links between the independent and dependent variables. As such, it would appear to have relevance to personality disorder research. It is argued that proper implementation of causal mediation analysis requires that investigators take several factors into account. These factors are discussed under 5 headings: variable selection, model specification, significance evaluation, effect size estimation, and sensitivity testing. First, care must be taken when selecting the independent, dependent, mediator, and control variables for a mediation analysis. Some variables make better mediators than others and all variables should be based on reasonably reliable indicators. Second, the mediation model needs to be properly specified. This requires that the data for the analysis be prospectively or historically ordered and possess proper causal direction. Third, it is imperative that the significance of the identified pathways be established, preferably with a nonparametric bootstrap resampling approach. Fourth, effect size estimates should be computed or competing pathways compared. Finally, investigators employing the mediation method are advised to perform a sensitivity analysis. Additional topics covered in this article include parallel and serial multiple mediation designs, moderation, and the relationship between mediation and moderation. (PsycINFO Database Record (c) 2018 APA, all rights reserved).

Spatio-temporal Granger causality: a new framework

PubMed Central

Luo, Qiang; Lu, Wenlian; Cheng, Wei; Valdes-Sosa, Pedro A.; Wen, Xiaotong; Ding, Mingzhou; Feng, Jianfeng

2015-01-01

That physiological oscillations of various frequencies are present in fMRI signals is the rule, not the exception. Herein, we propose a novel theoretical framework, spatio-temporal Granger causality, which allows us to more reliably and precisely estimate the Granger causality from experimental datasets possessing time-varying properties caused by physiological oscillations. Within this framework, Granger causality is redefined as a global index measuring the directed information flow between two time series with time-varying properties. Both theoretical analyses and numerical examples demonstrate that Granger causality is a monotonically increasing function of the temporal resolution used in the estimation. This is consistent with the general principle of coarse graining, which causes information loss by smoothing out very fine-scale details in time and space. Our results confirm that the Granger causality at the finer spatio-temporal scales considerably outperforms the traditional approach in terms of an improved consistency between two resting-state scans of the same subject. To optimally estimate the Granger causality, the proposed theoretical framework is implemented through a combination of several approaches, such as dividing the optimal time window and estimating the parameters at the fine temporal and spatial scales. Taken together, our approach provides a novel and robust framework for estimating the Granger causality from fMRI, EEG, and other related data. PMID:23643924

Use of allele scores as instrumental variables for Mendelian randomization

PubMed Central

Burgess, Stephen; Thompson, Simon G

2013-01-01

Background An allele score is a single variable summarizing multiple genetic variants associated with a risk factor. It is calculated as the total number of risk factor-increasing alleles for an individual (unweighted score), or the sum of weights for each allele corresponding to estimated genetic effect sizes (weighted score). An allele score can be used in a Mendelian randomization analysis to estimate the causal effect of the risk factor on an outcome. Methods Data were simulated to investigate the use of allele scores in Mendelian randomization where conventional instrumental variable techniques using multiple genetic variants demonstrate ‘weak instrument’ bias. The robustness of estimates using the allele score to misspecification (for example non-linearity, effect modification) and to violations of the instrumental variable assumptions was assessed. Results Causal estimates using a correctly specified allele score were unbiased with appropriate coverage levels. The estimates were generally robust to misspecification of the allele score, but not to instrumental variable violations, even if the majority of variants in the allele score were valid instruments. Using a weighted rather than an unweighted allele score increased power, but the increase was small when genetic variants had similar effect sizes. Naive use of the data under analysis to choose which variants to include in an allele score, or for deriving weights, resulted in substantial biases. Conclusions Allele scores enable valid causal estimates with large numbers of genetic variants. The stringency of criteria for genetic variants in Mendelian randomization should be maintained for all variants in an allele score. PMID:24062299

On Estimation of the Survivor Average Causal Effect in Observational Studies when Important Confounders are Missing Due to Death

PubMed Central

Egleston, Brian L.; Scharfstein, Daniel O.; MacKenzie, Ellen

2008-01-01

We focus on estimation of the causal effect of treatment on the functional status of individuals at a fixed point in time t* after they have experienced a catastrophic event, from observational data with the following features: (1) treatment is imposed shortly after the event and is non-randomized, (2) individuals who survive to t* are scheduled to be interviewed, (3) there is interview non-response, (4) individuals who die prior to t* are missing information on pre-event confounders, (5) medical records are abstracted on all individuals to obtain information on post-event, pre-treatment confounding factors. To address the issue of survivor bias, we seek to estimate the survivor average causal effect (SACE), the effect of treatment on functional status among the cohort of individuals who would survive to t* regardless of whether or not assigned to treatment. To estimate this effect from observational data, we need to impose untestable assumptions, which depend on the collection of all confounding factors. Since pre-event information is missing on those who die prior to t*, it is unlikely that these data are missing at random (MAR). We introduce a sensitivity analysis methodology to evaluate the robustness of SACE inferences to deviations from the MAR assumption. We apply our methodology to the evaluation of the effect of trauma center care on vitality outcomes using data from the National Study on Costs and Outcomes of Trauma Care. PMID:18759833

Causal inference in public health.

PubMed

Glass, Thomas A; Goodman, Steven N; Hernán, Miguel A; Samet, Jonathan M

2013-01-01

Causal inference has a central role in public health; the determination that an association is causal indicates the possibility for intervention. We review and comment on the long-used guidelines for interpreting evidence as supporting a causal association and contrast them with the potential outcomes framework that encourages thinking in terms of causes that are interventions. We argue that in public health this framework is more suitable, providing an estimate of an action's consequences rather than the less precise notion of a risk factor's causal effect. A variety of modern statistical methods adopt this approach. When an intervention cannot be specified, causal relations can still exist, but how to intervene to change the outcome will be unclear. In application, the often-complex structure of causal processes needs to be acknowledged and appropriate data collected to study them. These newer approaches need to be brought to bear on the increasingly complex public health challenges of our globalized world.

An Empirical Comparison of Randomized Control Trials and Regression Discontinuity Estimations

ERIC Educational Resources Information Center

Barrera-Osorio, Felipe; Filmer, Deon; McIntyre, Joe

2014-01-01

Randomized controlled trials (RCTs) and regression discontinuity (RD) studies both provide estimates of causal effects. A major difference between the two is that RD only estimates local average treatment effects (LATE) near the cutoff point of the forcing variable. This has been cited as a drawback to RD designs (Cook & Wong, 2008).…

Measurement Models for Reasoned Action Theory.

PubMed

Hennessy, Michael; Bleakley, Amy; Fishbein, Martin

2012-03-01

Quantitative researchers distinguish between causal and effect indicators. What are the analytic problems when both types of measures are present in a quantitative reasoned action analysis? To answer this question, we use data from a longitudinal study to estimate the association between two constructs central to reasoned action theory: behavioral beliefs and attitudes toward the behavior. The belief items are causal indicators that define a latent variable index while the attitude items are effect indicators that reflect the operation of a latent variable scale. We identify the issues when effect and causal indicators are present in a single analysis and conclude that both types of indicators can be incorporated in the analysis of data based on the reasoned action approach.

Causal inference in economics and marketing.

PubMed

Varian, Hal R

2016-07-05

This is an elementary introduction to causal inference in economics written for readers familiar with machine learning methods. The critical step in any causal analysis is estimating the counterfactual-a prediction of what would have happened in the absence of the treatment. The powerful techniques used in machine learning may be useful for developing better estimates of the counterfactual, potentially improving causal inference.

Causal inference in economics and marketing

PubMed Central

Varian, Hal R.

2016-01-01

This is an elementary introduction to causal inference in economics written for readers familiar with machine learning methods. The critical step in any causal analysis is estimating the counterfactual—a prediction of what would have happened in the absence of the treatment. The powerful techniques used in machine learning may be useful for developing better estimates of the counterfactual, potentially improving causal inference. PMID:27382144

Identifying Seizure Onset Zone From the Causal Connectivity Inferred Using Directed Information

NASA Astrophysics Data System (ADS)

Malladi, Rakesh; Kalamangalam, Giridhar; Tandon, Nitin; Aazhang, Behnaam

2016-10-01

In this paper, we developed a model-based and a data-driven estimator for directed information (DI) to infer the causal connectivity graph between electrocorticographic (ECoG) signals recorded from brain and to identify the seizure onset zone (SOZ) in epileptic patients. Directed information, an information theoretic quantity, is a general metric to infer causal connectivity between time-series and is not restricted to a particular class of models unlike the popular metrics based on Granger causality or transfer entropy. The proposed estimators are shown to be almost surely convergent. Causal connectivity between ECoG electrodes in five epileptic patients is inferred using the proposed DI estimators, after validating their performance on simulated data. We then proposed a model-based and a data-driven SOZ identification algorithm to identify SOZ from the causal connectivity inferred using model-based and data-driven DI estimators respectively. The data-driven SOZ identification outperforms the model-based SOZ identification algorithm when benchmarked against visual analysis by neurologist, the current clinical gold standard. The causal connectivity analysis presented here is the first step towards developing novel non-surgical treatments for epilepsy.

Causal strength induction from time series data.

PubMed

Soo, Kevin W; Rottman, Benjamin M

2018-04-01

One challenge when inferring the strength of cause-effect relations from time series data is that the cause and/or effect can exhibit temporal trends. If temporal trends are not accounted for, a learner could infer that a causal relation exists when it does not, or even infer that there is a positive causal relation when the relation is negative, or vice versa. We propose that learners use a simple heuristic to control for temporal trends-that they focus not on the states of the cause and effect at a given instant, but on how the cause and effect change from one observation to the next, which we call transitions. Six experiments were conducted to understand how people infer causal strength from time series data. We found that participants indeed use transitions in addition to states, which helps them to reach more accurate causal judgments (Experiments 1A and 1B). Participants use transitions more when the stimuli are presented in a naturalistic visual format than a numerical format (Experiment 2), and the effect of transitions is not driven by primacy or recency effects (Experiment 3). Finally, we found that participants primarily use the direction in which variables change rather than the magnitude of the change for estimating causal strength (Experiments 4 and 5). Collectively, these studies provide evidence that people often use a simple yet effective heuristic for inferring causal strength from time series data. (PsycINFO Database Record (c) 2018 APA, all rights reserved).

Etiologic effects and optimal intakes of foods and nutrients for risk of cardiovascular diseases and diabetes: Systematic reviews and meta-analyses from the Nutrition and Chronic Diseases Expert Group (NutriCoDE)

PubMed Central

Peñalvo, Jose L.; Khatibzadeh, Shahab; Singh, Gitanjali M.; Rao, Mayuree; Fahimi, Saman; Powles, John; Mozaffarian, Dariush

2017-01-01

Background Dietary habits are major contributors to coronary heart disease, stroke, and diabetes. However, comprehensive evaluation of etiologic effects of dietary factors on cardiometabolic outcomes, their quantitative effects, and corresponding optimal intakes are not well-established. Objective To systematically review the evidence for effects of dietary factors on cardiometabolic diseases, including comprehensively assess evidence for causality; estimate magnitudes of etiologic effects; evaluate heterogeneity and potential for bias in these etiologic effects; and determine optimal population intake levels. Methods We utilized Bradford-Hill criteria to assess probable or convincing evidence for causal effects of multiple diet-cardiometabolic disease relationships. Etiologic effects were quantified from published or de novo meta-analyses of prospective studies or randomized clinical trials, incorporating standardized units, dose-response estimates, and heterogeneity by age and other characteristics. Potential for bias was assessed in validity analyses. Optimal intakes were determined by levels associated with lowest disease risk. Results We identified 10 foods and 7 nutrients with evidence for causal cardiometabolic effects, including protective effects of fruits, vegetables, beans/legumes, nuts/seeds, whole grains, fish, yogurt, fiber, seafood omega-3s, polyunsaturated fats, and potassium; and harms of unprocessed red meats, processed meats, sugar-sweetened beverages, glycemic load, trans-fats, and sodium. Proportional etiologic effects declined with age, but did not generally vary by sex. Established optimal population intakes were generally consistent with observed national intakes and major dietary guidelines. In validity analyses, the identified effects of individual dietary components were similar to quantified effects of dietary patterns on cardiovascular risk factors and hard endpoints. Conclusions These novel findings provide a comprehensive summary of causal evidence, quantitative etiologic effects, heterogeneity, and optimal intakes of major dietary factors for cardiometabolic diseases, informing disease impact estimation and policy planning and priorities. PMID:28448503

Estimating causal contrasts involving intermediate variables in the presence of selection bias.

PubMed

Valeri, Linda; Coull, Brent A

2016-11-20

An important goal across the biomedical and social sciences is the quantification of the role of intermediate factors in explaining how an exposure exerts an effect on an outcome. Selection bias has the potential to severely undermine the validity of inferences on direct and indirect causal effects in observational as well as in randomized studies. The phenomenon of selection may arise through several mechanisms, and we here focus on instances of missing data. We study the sign and magnitude of selection bias in the estimates of direct and indirect effects when data on any of the factors involved in the analysis is either missing at random or not missing at random. Under some simplifying assumptions, the bias formulae can lead to nonparametric sensitivity analyses. These sensitivity analyses can be applied to causal effects on the risk difference and risk-ratio scales irrespectively of the estimation approach employed. To incorporate parametric assumptions, we also develop a sensitivity analysis for selection bias in mediation analysis in the spirit of the expectation-maximization algorithm. The approaches are applied to data from a health disparities study investigating the role of stage at diagnosis on racial disparities in colorectal cancer survival. Copyright © 2016 John Wiley & Sons, Ltd. Copyright © 2016 John Wiley & Sons, Ltd.

Data Combination and Instrumental Variables in Linear Models

ERIC Educational Resources Information Center

Khawand, Christopher

2012-01-01

Instrumental variables (IV) methods allow for consistent estimation of causal effects, but suffer from poor finite-sample properties and data availability constraints. IV estimates also tend to have relatively large standard errors, often inhibiting the interpretability of differences between IV and non-IV point estimates. Lastly, instrumental…

Three Essays on Estimating Causal Treatment Effects

ERIC Educational Resources Information Center

Deutsch, Jonah

2013-01-01

This dissertation is composed of three distinct chapters, each of which addresses issues of estimating treatment effects. The first chapter empirically tests the Value-Added (VA) model using school lotteries. The second chapter, co-authored with Michael Wood, considers properties of inverse probability weighting (IPW) in simple treatment effect…

Analysis of Longitudinal Studies With Repeated Outcome Measures: Adjusting for Time-Dependent Confounding Using Conventional Methods.

PubMed

Keogh, Ruth H; Daniel, Rhian M; VanderWeele, Tyler J; Vansteelandt, Stijn

2018-05-01

Estimation of causal effects of time-varying exposures using longitudinal data is a common problem in epidemiology. When there are time-varying confounders, which may include past outcomes, affected by prior exposure, standard regression methods can lead to bias. Methods such as inverse probability weighted estimation of marginal structural models have been developed to address this problem. However, in this paper we show how standard regression methods can be used, even in the presence of time-dependent confounding, to estimate the total effect of an exposure on a subsequent outcome by controlling appropriately for prior exposures, outcomes, and time-varying covariates. We refer to the resulting estimation approach as sequential conditional mean models (SCMMs), which can be fitted using generalized estimating equations. We outline this approach and describe how including propensity score adjustment is advantageous. We compare the causal effects being estimated using SCMMs and marginal structural models, and we compare the two approaches using simulations. SCMMs enable more precise inferences, with greater robustness against model misspecification via propensity score adjustment, and easily accommodate continuous exposures and interactions. A new test for direct effects of past exposures on a subsequent outcome is described.

Causal mediation analysis with multiple causally non-ordered mediators.

PubMed

Taguri, Masataka; Featherstone, John; Cheng, Jing

2018-01-01

In many health studies, researchers are interested in estimating the treatment effects on the outcome around and through an intermediate variable. Such causal mediation analyses aim to understand the mechanisms that explain the treatment effect. Although multiple mediators are often involved in real studies, most of the literature considered mediation analyses with one mediator at a time. In this article, we consider mediation analyses when there are causally non-ordered multiple mediators. Even if the mediators do not affect each other, the sum of two indirect effects through the two mediators considered separately may diverge from the joint natural indirect effect when there are additive interactions between the effects of the two mediators on the outcome. Therefore, we derive an equation for the joint natural indirect effect based on the individual mediation effects and their interactive effect, which helps us understand how the mediation effect works through the two mediators and relative contributions of the mediators and their interaction. We also discuss an extension for three mediators. The proposed method is illustrated using data from a randomized trial on the prevention of dental caries.

Assumption Trade-Offs When Choosing Identification Strategies for Pre-Post Treatment Effect Estimation: An Illustration of a Community-Based Intervention in Madagascar.

PubMed

Weber, Ann M; van der Laan, Mark J; Petersen, Maya L

2015-03-01

Failure (or success) in finding a statistically significant effect of a large-scale intervention may be due to choices made in the evaluation. To highlight the potential limitations and pitfalls of some common identification strategies used for estimating causal effects of community-level interventions, we apply a roadmap for causal inference to a pre-post evaluation of a national nutrition program in Madagascar. Selection into the program was non-random and strongly associated with the pre-treatment (lagged) outcome. Using structural causal models (SCM), directed acyclic graphs (DAGs) and simulated data, we illustrate that an estimand with the outcome defined as the post-treatment outcome controls for confounding by the lagged outcome but not by possible unmeasured confounders. Two separate differencing estimands (of the pre- and post-treatment outcome) have the potential to adjust for a certain type of unmeasured confounding, but introduce bias if the additional identification assumptions they rely on are not met. In order to illustrate the practical impact of choice between three common identification strategies and their corresponding estimands, we used observational data from the community nutrition program in Madagascar to estimate each of these three estimands. Specifically, we estimated the average treatment effect of the program on the community mean nutritional status of children 5 years and under and found that the estimate based on the post-treatment estimand was about a quarter of the magnitude of either of the differencing estimands (0.066 SD vs. 0.26-0.27 SD increase in mean weight-for-age z-score). Choice of estimand clearly has important implications for the interpretation of the success of the program to improve nutritional status of young children. A careful appraisal of the assumptions underlying the causal model is imperative before committing to a statistical model and progressing to estimation. However, knowledge about the data-generating process must be sufficient in order to choose the identification strategy that gets us closest to the truth.

Synchrony dynamics underlying effective connectivity reconstruction of neuronal circuits

NASA Astrophysics Data System (ADS)

Yu, Haitao; Guo, Xinmeng; Qin, Qing; Deng, Yun; Wang, Jiang; Liu, Jing; Cao, Yibin

2017-04-01

Reconstruction of effective connectivity between neurons is essential for neural systems with function-related significance, characterizing directionally causal influences among neurons. In this work, causal interactions between neurons in spinal dorsal root ganglion, activated by manual acupuncture at Zusanli acupoint of experimental rats, are estimated using Granger causality (GC) method. Different patterns of effective connectivity are obtained for different frequencies and types of acupuncture. Combined with synchrony analysis between neurons, we show a dependence of effective connection on the synchronization dynamics. Based on the experimental findings, a neuronal circuit model with synaptic connections is constructed. The variation of neuronal effective connectivity with respect to its structural connectivity and synchronization dynamics is further explored. Simulation results show that reciprocally causal interactions with statistically significant are formed between well-synchronized neurons. The effective connectivity may be not necessarily equivalent to synaptic connections, but rather depend on the synchrony relationship. Furthermore, transitions of effective interaction between neurons are observed following the synchronization transitions induced by conduction delay and synaptic conductance. These findings are helpful to further investigate the dynamical mechanisms underlying the reconstruction of effective connectivity of neuronal population.

Measurement Models for Reasoned Action Theory

PubMed Central

Hennessy, Michael; Bleakley, Amy; Fishbein, Martin

2012-01-01

Quantitative researchers distinguish between causal and effect indicators. What are the analytic problems when both types of measures are present in a quantitative reasoned action analysis? To answer this question, we use data from a longitudinal study to estimate the association between two constructs central to reasoned action theory: behavioral beliefs and attitudes toward the behavior. The belief items are causal indicators that define a latent variable index while the attitude items are effect indicators that reflect the operation of a latent variable scale. We identify the issues when effect and causal indicators are present in a single analysis and conclude that both types of indicators can be incorporated in the analysis of data based on the reasoned action approach. PMID:23243315

Revisiting the child health-wealth nexus.

PubMed

Fakir, Adnan M S

2016-12-01

The causal link between a household's economic standing and child health is known to suffer from endogeneity. While past studies have exemplified the causal link to be small, albeit statistically significant, this paper aims to estimate the causal effect to investigate whether the effect of income after controlling for the endogeneity remains small in the long run. By correcting for the bias, and knowing the bias direction, one can also infer about the underlying backward effect. This paper uses an instrument variables two-stage-least-squares estimation on the Young Lives 2009 cross-sectional dataset from Andhra Pradesh, India, to understand the aforementioned relationship. The selected measure of household economic standing differentially affects the estimation. There is significant positive effect of both short-run household expenditure and long-run household wealth on child stunting, with the latter having a larger impact. The backward link running from child health to household income is likely an inverse association in our sample with lower child health inducing higher earnings. While higher average community education improved child health, increased community entertainment expenditure is found to have a negative effect. While policies catered towards improving household wealth will decrease child stunting in the long run, maternal education and the community play an equally reinforcing role in improving child health and are perhaps faster routes to achieving the goal of better child health in the short run.

Effects of Dual-Language Immersion Programs on Student Achievement: Evidence from Lottery Data

ERIC Educational Resources Information Center

Steele, Jennifer L.; Slater, Robert O.; Zamarro, Gema; Miller, Trey; Li, Jennifer; Burkhauser, Susan; Bacon, Michael

2017-01-01

Using data from seven cohorts of language immersion lottery applicants in a large, urban school district, we estimate the causal effects of immersion programs on students' test scores in reading, mathematics, and science and on English learners' (EL) reclassification. We estimate positive intent-to-treat (ITT) effects on reading performance in…

Effects of Dual-Language Immersion Programs on Student Achievement: Evidence from Lottery Data

ERIC Educational Resources Information Center

Steele, Jennifer L.; Slater, Robert O.; Zamarro, Gema; Miller, Trey; Li, Jennifer; Burkhauser, Susan; Bacon, Michael

2017-01-01

Using data from seven cohorts of language immersion lottery applicants in a large, urban school district, we estimate the causal effects of immersion programs on students' test scores in reading, mathematics, and science, and on English learners' (EL) reclassification. We estimate positive intent-to-treat (ITT) effects on reading performance in…

Age at Menarche and Time Spent in Education: A Mendelian Randomization Study.

PubMed

Gill, D; Del Greco M, F; Rawson, T M; Sivakumaran, P; Brown, A; Sheehan, N A; Minelli, C

2017-09-01

Menarche signifies the primary event in female puberty and is associated with changes in self-identity. It is not clear whether earlier puberty causes girls to spend less time in education. Observational studies on this topic are likely to be affected by confounding environmental factors. The Mendelian randomization (MR) approach addresses these issues by using genetic variants (such as single nucleotide polymorphisms, SNPs) as proxies for the risk factor of interest. We use this technique to explore whether there is a causal effect of age at menarche on time spent in education. Instruments and SNP-age at menarche estimates are identified from a Genome Wide Association Study (GWAS) meta-analysis of 182,416 women of European descent. The effects of instruments on time spent in education are estimated using a GWAS meta-analysis of 118,443 women performed by the Social Science Genetic Association Consortium (SSGAC). In our main analysis, we demonstrate a small but statistically significant causal effect of age at menarche on time spent in education: a 1 year increase in age at menarche is associated with 0.14 years (53 days) increase in time spent in education (95% CI 0.10-0.21 years, p = 3.5 × 10 -8 ). The causal effect is confirmed in sensitivity analyses. In identifying this positive causal effect of age at menarche on time spent in education, we offer further insight into the social effects of puberty in girls.

Investigating genetic correlations and causal effects between caffeine consumption and sleep behaviours.

PubMed

Treur, Jorien L; Gibson, Mark; Taylor, Amy E; Rogers, Peter J; Munafò, Marcus R

2018-04-22

Observationally, higher caffeine consumption is associated with poorer sleep and insomnia. We investigated whether these associations are a result of shared genetic risk factors and/or (possibly bidirectional) causal effects. Summary-level data were available from genome-wide association studies on caffeine intake (n = 91 462), plasma caffeine and caffeine metabolic rate (n = 9876), sleep duration and chronotype (being a "morning" versus an "evening" person) (n = 128 266), and insomnia complaints (n = 113 006). First, genetic correlations were calculated, reflecting the extent to which genetic variants influencing caffeine consumption and those influencing sleep overlap. Next, causal effects were estimated with bidirectional, two-sample Mendelian randomization. This approach utilizes the genetic variants most robustly associated with an exposure variable as an "instrument" to test causal effects. Estimates from individual variants were combined using inverse-variance weighted meta-analysis, weighted median regression and MR-Egger regression. We found no clear evidence for a genetic correlation between caffeine intake and sleep duration (rg = 0.000, p = .998), chronotype (rg = 0.086, p = .192) or insomnia complaints (rg = -0.034, p = .700). For plasma caffeine and caffeine metabolic rate, genetic correlations could not be calculated because of the small sample size. Mendelian randomization did not support causal effects of caffeine intake on sleep, or vice versa. There was weak evidence that higher plasma caffeine levels causally decrease the odds of being a morning person. Although caffeine may acutely affect sleep when taken shortly before bedtime, our findings suggest that a sustained pattern of high caffeine consumption is more likely to be associated with poorer sleep through shared environmental factors. Future research should identify such environments, which could aid the development of interventions to improve sleep. © 2018 The Authors. Journal of Sleep Research published by John Wiley & Sons Ltd on behalf of European Sleep Research Society.

Causal Analysis of Self-tracked Time Series Data Using a Counterfactual Framework for N-of-1 Trials.

PubMed

Daza, Eric J

2018-02-01

Many of an individual's historically recorded personal measurements vary over time, thereby forming a time series (e.g., wearable-device data, self-tracked fitness or nutrition measurements, regularly monitored clinical events or chronic conditions). Statistical analyses of such n-of-1 (i.e., single-subject) observational studies (N1OSs) can be used to discover possible cause-effect relationships to then self-test in an n-of-1 randomized trial (N1RT). However, a principled way of determining how and when to interpret an N1OS association as a causal effect (e.g., as if randomization had occurred) is needed.Our goal in this paper is to help bridge the methodological gap between risk-factor discovery and N1RT testing by introducing a basic counterfactual framework for N1OS design and personalized causal analysis.We introduce and characterize what we call the average period treatment effect (APTE), i.e., the estimand of interest in an N1RT, and build an analytical framework around it that can accommodate autocorrelation and time trends in the outcome, effect carryover from previous treatment periods, and slow onset or decay of the effect. The APTE is loosely defined as a contrast (e.g., difference, ratio) of averages of potential outcomes the individual can theoretically experience under different treatment levels during a given treatment period. To illustrate the utility of our framework for APTE discovery and estimation, two common causal inference methods are specified within the N1OS context. We then apply the framework and methods to search for estimable and interpretable APTEs using six years of the author's self-tracked weight and exercise data, and report both the preliminary findings and the challenges we faced in conducting N1OS causal discovery.Causal analysis of an individual's time series data can be facilitated by an N1RT counterfactual framework. However, for inference to be valid, the veracity of certain key assumptions must be assessed critically, and the hypothesized causal models must be interpretable and meaningful. Schattauer GmbH.

Estimators for Clustered Education RCTs Using the Neyman Model for Causal Inference

ERIC Educational Resources Information Center

Schochet, Peter Z.

2013-01-01

This article examines the estimation of two-stage clustered designs for education randomized control trials (RCTs) using the nonparametric Neyman causal inference framework that underlies experiments. The key distinction between the considered causal models is whether potential treatment and control group outcomes are considered to be fixed for…

Socioeconomic inequality in health in the British household panel: Tests of the social causation, health selection and the indirect selection hypothesis using dynamic fixed effects panel models.

PubMed

Foverskov, Else; Holm, Anders

2016-02-01

Despite social inequality in health being well documented, it is still debated which causal mechanism best explains the negative association between socioeconomic position (SEP) and health. This paper is concerned with testing the explanatory power of three widely proposed causal explanations for social inequality in health in adulthood: the social causation hypothesis (SEP determines health), the health selection hypothesis (health determines SEP) and the indirect selection hypothesis (no causal relationship). We employ dynamic data of respondents aged 30 to 60 from the last nine waves of the British Household Panel Survey. Household income and location on the Cambridge Scale is included as measures of different dimensions of SEP and health is measured as a latent factor score. The causal hypotheses are tested using a time-based Granger approach by estimating dynamic fixed effects panel regression models following the method suggested by Anderson and Hsiao. We propose using this method to estimate the associations over time since it allows one to control for all unobserved time-invariant factors and hence lower the chances of biased estimates due to unobserved heterogeneity. The results showed no proof of the social causation hypothesis over a one to five year period and limited support for the health selection hypothesis was seen only for men in relation to HH income. These findings were robust in multiple sensitivity analysis. We conclude that the indirect selection hypothesis may be the most important in explaining social inequality in health in adulthood, indicating that the well-known cross-sectional correlations between health and SEP in adulthood seem not to be driven by a causal relationship, but instead by dynamics and influences in place before the respondents turn 30 years old that affect both their health and SEP onwards. The conclusion is limited in that we do not consider the effect of specific diseases and causal relationships in adulthood may be present over a longer timespan than 5 years. Copyright © 2015 Elsevier Ltd. All rights reserved.

Understanding how pain education causes changes in pain and disability: protocol for a causal mediation analysis of the PREVENT trial.

PubMed

Lee, Hopin; Moseley, G Lorimer; Hübscher, Markus; Kamper, Steven J; Traeger, Adrian C; Skinner, Ian W; McAuley, James H

2015-07-01

Pain education is a complex intervention developed to help clinicians manage low back pain. Although complex interventions are usually evaluated by their effects on outcomes, such as pain or disability, most do not directly target these outcomes; instead, they target intermediate factors that are presumed to be associated with the outcomes. The mechanisms underlying treatment effects, or the effect of an intervention on an intermediate factor and its subsequent effect on outcome, are rarely investigated in clinical trials. This leaves a gap in the evidence for understanding how treatments exert their effects on outcomes. Mediation analysis provides a method for identifying and quantifying the mechanisms that underlie interventions. To determine whether the effect of pain education on pain and disability is mediated by changes in self-efficacy, catastrophisation and back pain beliefs. Causal mediation analysis of the PREVENT randomised controlled trial. Two hundred and two participants with acute low back pain from primary care clinics in the Sydney metropolitan area. Participants will be randomised to receive either 'pain education' (intervention group) or 'sham education' (control group). All outcome measures (including patient characteristics), primary outcome measures (pain and disability), and putative mediating variables (self-efficacy, catastrophisation and back pain beliefs) will be measured prior to randomisation. Putative mediators and primary outcome measures will be measured 1 week after the intervention, and primary outcome measures will be measured 3 months after the onset of low back pain. Causal mediation analysis under the potential outcomes framework will be used to test single and multiple mediator models. A sensitivity analysis will be conducted to evaluate the robustness of the estimated mediation effects on the influence of violating sequential ignorability--a critical assumption for causal inference. Mediation analysis of clinical trials can estimate how much the total effect of the treatment on the outcome is carried through an indirect path. Using mediation analysis to understand these mechanisms can generate evidence that can be used to tailor treatments and optimise treatment effects. In this study, the causal mediation effects of a pain education intervention for acute non-specific low back pain will be estimated. This knowledge is critical for further development and refinement of interventions for conditions such as low back pain. Copyright © 2015 Australian Physiotherapy Association. Published by Elsevier B.V. All rights reserved.

New Insights into Signed Path Coefficient Granger Causality Analysis.

PubMed

Zhang, Jian; Li, Chong; Jiang, Tianzi

2016-01-01

Granger causality analysis, as a time series analysis technique derived from econometrics, has been applied in an ever-increasing number of publications in the field of neuroscience, including fMRI, EEG/MEG, and fNIRS. The present study mainly focuses on the validity of "signed path coefficient Granger causality," a Granger-causality-derived analysis method that has been adopted by many fMRI researches in the last few years. This method generally estimates the causality effect among the time series by an order-1 autoregression, and defines a positive or negative coefficient as an "excitatory" or "inhibitory" influence. In the current work we conducted a series of computations from resting-state fMRI data and simulation experiments to illustrate the signed path coefficient method was flawed and untenable, due to the fact that the autoregressive coefficients were not always consistent with the real causal relationships and this would inevitablely lead to erroneous conclusions. Overall our findings suggested that the applicability of this kind of causality analysis was rather limited, hence researchers should be more cautious in applying the signed path coefficient Granger causality to fMRI data to avoid misinterpretation.

Simulation-Extrapolation for Estimating Means and Causal Effects with Mismeasured Covariates

ERIC Educational Resources Information Center

Lockwood, J. R.; McCaffrey, Daniel F.

2015-01-01

Regression, weighting and related approaches to estimating a population mean from a sample with nonrandom missing data often rely on the assumption that conditional on covariates, observed samples can be treated as random. Standard methods using this assumption generally will fail to yield consistent estimators when covariates are measured with…

Incorporating Functional Genomic Information in Genetic Association Studies Using an Empirical Bayes Approach.

PubMed

Spencer, Amy V; Cox, Angela; Lin, Wei-Yu; Easton, Douglas F; Michailidou, Kyriaki; Walters, Kevin

2016-04-01

There is a large amount of functional genetic data available, which can be used to inform fine-mapping association studies (in diseases with well-characterised disease pathways). Single nucleotide polymorphism (SNP) prioritization via Bayes factors is attractive because prior information can inform the effect size or the prior probability of causal association. This approach requires the specification of the effect size. If the information needed to estimate a priori the probability density for the effect sizes for causal SNPs in a genomic region isn't consistent or isn't available, then specifying a prior variance for the effect sizes is challenging. We propose both an empirical method to estimate this prior variance, and a coherent approach to using SNP-level functional data, to inform the prior probability of causal association. Through simulation we show that when ranking SNPs by our empirical Bayes factor in a fine-mapping study, the causal SNP rank is generally as high or higher than the rank using Bayes factors with other plausible values of the prior variance. Importantly, we also show that assigning SNP-specific prior probabilities of association based on expert prior functional knowledge of the disease mechanism can lead to improved causal SNPs ranks compared to ranking with identical prior probabilities of association. We demonstrate the use of our methods by applying the methods to the fine mapping of the CASP8 region of chromosome 2 using genotype data from the Collaborative Oncological Gene-Environment Study (COGS) Consortium. The data we analysed included approximately 46,000 breast cancer case and 43,000 healthy control samples. © 2016 The Authors. *Genetic Epidemiology published by Wiley Periodicals, Inc.

Doubly Robust Additive Hazards Models to Estimate Effects of a Continuous Exposure on Survival.

PubMed

Wang, Yan; Lee, Mihye; Liu, Pengfei; Shi, Liuhua; Yu, Zhi; Abu Awad, Yara; Zanobetti, Antonella; Schwartz, Joel D

2017-11-01

The effect of an exposure on survival can be biased when the regression model is misspecified. Hazard difference is easier to use in risk assessment than hazard ratio and has a clearer interpretation in the assessment of effect modifications. We proposed two doubly robust additive hazards models to estimate the causal hazard difference of a continuous exposure on survival. The first model is an inverse probability-weighted additive hazards regression. The second model is an extension of the doubly robust estimator for binary exposures by categorizing the continuous exposure. We compared these with the marginal structural model and outcome regression with correct and incorrect model specifications using simulations. We applied doubly robust additive hazard models to the estimation of hazard difference of long-term exposure to PM2.5 (particulate matter with an aerodynamic diameter less than or equal to 2.5 microns) on survival using a large cohort of 13 million older adults residing in seven states of the Southeastern United States. We showed that the proposed approaches are doubly robust. We found that each 1 μg m increase in annual PM2.5 exposure was associated with a causal hazard difference in mortality of 8.0 × 10 (95% confidence interval 7.4 × 10, 8.7 × 10), which was modified by age, medical history, socioeconomic status, and urbanicity. The overall hazard difference translates to approximately 5.5 (5.1, 6.0) thousand deaths per year in the study population. The proposed approaches improve the robustness of the additive hazards model and produce a novel additive causal estimate of PM2.5 on survival and several additive effect modifications, including social inequality.

Spillover effects in epidemiology: parameters, study designs and methodological considerations

PubMed Central

Benjamin-Chung, Jade; Arnold, Benjamin F; Berger, David; Luby, Stephen P; Miguel, Edward; Colford Jr, John M; Hubbard, Alan E

2018-01-01

Abstract Many public health interventions provide benefits that extend beyond their direct recipients and impact people in close physical or social proximity who did not directly receive the intervention themselves. A classic example of this phenomenon is the herd protection provided by many vaccines. If these ‘spillover effects’ (i.e. ‘herd effects’) are present in the same direction as the effects on the intended recipients, studies that only estimate direct effects on recipients will likely underestimate the full public health benefits of the intervention. Causal inference assumptions for spillover parameters have been articulated in the vaccine literature, but many studies measuring spillovers of other types of public health interventions have not drawn upon that literature. In conjunction with a systematic review we conducted of spillovers of public health interventions delivered in low- and middle-income countries, we classified the most widely used spillover parameters reported in the empirical literature into a standard notation. General classes of spillover parameters include: cluster-level spillovers; spillovers conditional on treatment or outcome density, distance or the number of treated social network links; and vaccine efficacy parameters related to spillovers. We draw on high quality empirical examples to illustrate each of these parameters. We describe study designs to estimate spillovers and assumptions required to make causal inferences about spillovers. We aim to advance and encourage methods for spillover estimation and reporting by standardizing spillover parameter nomenclature and articulating the causal inference assumptions required to estimate spillovers. PMID:29106568

An integrated specification for the nexus of water pollution and economic growth in China: Panel cointegration, long-run causality and environmental Kuznets curve.

PubMed

Zhang, Chen; Wang, Yuan; Song, Xiaowei; Kubota, Jumpei; He, Yanmin; Tojo, Junji; Zhu, Xiaodong

2017-12-31

This paper concentrates on a Chinese context and makes efforts to develop an integrated process to explicitly elucidate the relationship between economic growth and water pollution discharge-chemical oxygen demand (COD) discharge and ammonia nitrogen (NH 3 -N), using two unbalanced panel data sets covering the period separately from 1990 to 2014, and 2001 to 2014. In our present study, the panel unit root tests, cointegration tests, and Granger causality tests allowing for cross-sectional dependence, nonstationary, and heterogeneity are conducted to examine the causal effects of economic growth on COD/NH 3 -N discharge. Further, we simultaneously apply semi-parametric fixed effects estimation and parametric fixed effects estimation to investigate environmental Kuznets curve relationship for COD/NH 3 -N discharge. Our empirical results show a long-term bidirectional causality between economic growth and COD/NH 3 -N discharge in China. Within the Stochastic Impacts by Regression on Population, Affluence and Technology framework, we find evidence in support of an inverted U-shaped curved link between economic growth and COD/NH 3 -N discharge. To the best of our knowledge, there have not been any efforts made in investigating the nexus of economic growth and water pollution in such an integrated manner. Therefore, this study takes a fresh look on this topic. Copyright © 2017 Elsevier B.V. All rights reserved.

Use of Causal Diagrams to Inform the Design and Interpretation of Observational Studies: An Example from the Study of Heart and Renal Protection (SHARP).

PubMed

Staplin, Natalie; Herrington, William G; Judge, Parminder K; Reith, Christina A; Haynes, Richard; Landray, Martin J; Baigent, Colin; Emberson, Jonathan

2017-03-07

Observational studies often seek to estimate the causal relevance of an exposure to an outcome of interest. However, many possible biases can arise when estimating such relationships, in particular bias because of confounding. To control for confounding properly, careful consideration of the nature of the assumed relationships between the exposure, the outcome, and other characteristics is required. Causal diagrams provide a simple graphic means of displaying such relationships, describing the assumptions made, and allowing for the identification of a set of characteristics that should be taken into account ( i.e. , adjusted for) in any analysis. Furthermore, causal diagrams can be used to identify other possible sources of bias (such as selection bias), which if understood from the outset, can inform the planning of appropriate analyses. In this article, we review the basic theory of causal diagrams and describe some of the methods available to identify which characteristics need to be taken into account when estimating the total effect of an exposure on an outcome. In doing so, we review the concept of collider bias and show how it is inappropriate to adjust for characteristics that may be influenced, directly or indirectly, by both the exposure and the outcome of interest. A motivating example is taken from the Study of Heart and Renal Protection, in which the relevance of smoking to progression to ESRD is considered. Copyright © 2017 by the American Society of Nephrology.

Reduction of Working Time: Does It Lead to a Healthy Lifestyle?

PubMed

Ahn, Taehyun

2016-08-01

I examine whether working hours have a causal effect on the health behaviors of workers. In assessing the causal relationship, I estimate fixed-effects instrumental variable models by using exogenous variation in adopting a reduced workweek in South Korea as an instrument for work hours. The estimation results reveal that shortening work hours induces individuals to exercise regularly and decreases the likelihood of smoking, with more pronounced effects for heavy smokers. While a work-hour reduction substantially increases the probability of drinking participation, it does not significantly affect the likelihood of frequent or daily drinking habits. In addition, the effect of a work-hour reduction on regular exercise is salient among women and older groups, and the effect on smoking behaviors is more pronounced among men and middle-aged groups. Copyright © 2015 John Wiley & Sons, Ltd. Copyright © 2015 John Wiley & Sons, Ltd.

Estimating Causal Effects of Education Interventions Using a Two-Rating Regression Discontinuity Design: Lessons from a Simulation Study

ERIC Educational Resources Information Center

Porter, Kristin E.; Reardon, Sean F.; Unlu, Fatih; Bloom, Howard S.; Robinson-Cimpian, Joseph P.

2014-01-01

A valuable extension of the single-rating regression discontinuity design (RDD) is a multiple-rating RDD (MRRDD). To date, four main methods have been used to estimate average treatment effects at the multiple treatment frontiers of an MRRDD: the "surface" method, the "frontier" method, the "binding-score" method, and…

Using Propensity Scores for Estimating Causal Effects: A Study in the Development of Moral Reasoning

ERIC Educational Resources Information Center

Grunwald, Heidi E.; Mayhew, Matthew J.

2008-01-01

The purpose of this study was to illustrate the use of propensity scores for creating comparison groups, partially controlling for pretreatment course selection bias, and estimating the treatment effects of selected courses on the development of moral reasoning in undergraduate students. Specifically, we used a sample of convenience for comparing…

Does money really matter? Estimating impacts of family income on young children's achievement with data from random-assignment experiments.

PubMed

Duncan, Greg J; Morris, Pamela A; Rodrigues, Chris

2011-09-01

Social scientists do not agree on the size and nature of the causal impacts of parental income on children's achievement. We revisit this issue using a set of welfare and antipoverty experiments conducted in the 1990s. We utilize an instrumental variables strategy to leverage the variation in income and achievement that arises from random assignment to the treatment group to estimate the causal effect of income on child achievement. Our estimates suggest that a $1,000 increase in annual income increases young children's achievement by 5%-6% of a standard deviation. As such, our results suggest that family income has a policy-relevant, positive impact on the eventual school achievement of preschool children.

Causal mediation analysis with multiple mediators in the presence of treatment noncompliance.

PubMed

Park, Soojin; Kürüm, Esra

2018-05-20

Randomized experiments are often complicated because of treatment noncompliance. This challenge prevents researchers from identifying the mediated portion of the intention-to-treated (ITT) effect, which is the effect of the assigned treatment that is attributed to a mediator. One solution suggests identifying the mediated ITT effect on the basis of the average causal mediation effect among compliers when there is a single mediator. However, considering the complex nature of the mediating mechanisms, it is natural to assume that there are multiple variables that mediate through the causal path. Motivated by an empirical analysis of a data set collected in a randomized interventional study, we develop a method to estimate the mediated portion of the ITT effect when both multiple dependent mediators and treatment noncompliance exist. This enables researchers to make an informed decision on how to strengthen the intervention effect by identifying relevant mediators despite treatment noncompliance. We propose a nonparametric estimation procedure and provide a sensitivity analysis for key assumptions. We conduct a Monte Carlo simulation study to assess the finite sample performance of the proposed approach. The proposed method is illustrated by an empirical analysis of JOBS II data, in which a job training intervention was used to prevent mental health deterioration among unemployed individuals. Copyright © 2018 John Wiley & Sons, Ltd.

Generalizability of causal inference in observational studies under retrospective convenience sampling.

PubMed

Hu, Zonghui; Qin, Jing

2018-05-20

Many observational studies adopt what we call retrospective convenience sampling (RCS). With the sample size in each arm prespecified, RCS randomly selects subjects from the treatment-inclined subpopulation into the treatment arm and those from the control-inclined into the control arm. Samples in each arm are representative of the respective subpopulation, but the proportion of the 2 subpopulations is usually not preserved in the sample data. We show in this work that, under RCS, existing causal effect estimators actually estimate the treatment effect over the sample population instead of the underlying study population. We investigate how to correct existing methods for consistent estimation of the treatment effect over the underlying population. Although RCS is adopted in medical studies for ethical and cost-effective purposes, it also has a big advantage for statistical inference: When the tendency to receive treatment is low in a study population, treatment effect estimators under RCS, with proper correction, are more efficient than their parallels under random sampling. These properties are investigated both theoretically and through numerical demonstration. Published 2018. This article is a U.S. Government work and is in the public domain in the USA.

Causal mediation analysis with a latent mediator.

PubMed

Albert, Jeffrey M; Geng, Cuiyu; Nelson, Suchitra

2016-05-01

Health researchers are often interested in assessing the direct effect of a treatment or exposure on an outcome variable, as well as its indirect (or mediation) effect through an intermediate variable (or mediator). For an outcome following a nonlinear model, the mediation formula may be used to estimate causally interpretable mediation effects. This method, like others, assumes that the mediator is observed. However, as is common in structural equations modeling, we may wish to consider a latent (unobserved) mediator. We follow a potential outcomes framework and assume a generalized structural equations model (GSEM). We provide maximum-likelihood estimation of GSEM parameters using an approximate Monte Carlo EM algorithm, coupled with a mediation formula approach to estimate natural direct and indirect effects. The method relies on an untestable sequential ignorability assumption; we assess robustness to this assumption by adapting a recently proposed method for sensitivity analysis. Simulation studies show good properties of the proposed estimators in plausible scenarios. Our method is applied to a study of the effect of mother education on occurrence of adolescent dental caries, in which we examine possible mediation through latent oral health behavior. © 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

Does More Schooling Improve Health Outcomes and Health Related Behaviors? Evidence from U.K. Twins

PubMed Central

Amin, Vikesh; Behrman, Jere R.; Spector, Tim D.

2013-01-01

Several recent studies using instrumental variables based on changes in compulsory schoolleaving age laws have estimated the causal effect of schooling on health outcomes and health-related behaviors in the U.K. Despite using the same identification strategy and similar datasets, no consensus has been reached. We contribute to the literature by providing results for the U.K. using a different research design and a different dataset. Specifically, we estimate the effect of schooling on health outcomes (obesity and physical health) and health-related behaviors (smoking, alcohol consumption and exercise) for women through within-MZ twins estimates using the TwinsUK database. For physical health, alcohol consumption and exercise, the within-MZ twins estimates are uninformative about whether there is a causal effect. However, we find (1) that the significant association between schooling and smoking status is due to unobserved endowments that are correlated with schooling and smoking (2) there is some indication that more schooling reduces the body mass index for women, even once these unobserved endowments have been controlled for. PMID:24415826

Metabolic signatures of adiposity in young adults: Mendelian randomization analysis and effects of weight change.

PubMed

Würtz, Peter; Wang, Qin; Kangas, Antti J; Richmond, Rebecca C; Skarp, Joni; Tiainen, Mika; Tynkkynen, Tuulia; Soininen, Pasi; Havulinna, Aki S; Kaakinen, Marika; Viikari, Jorma S; Savolainen, Markku J; Kähönen, Mika; Lehtimäki, Terho; Männistö, Satu; Blankenberg, Stefan; Zeller, Tanja; Laitinen, Jaana; Pouta, Anneli; Mäntyselkä, Pekka; Vanhala, Mauno; Elliott, Paul; Pietiläinen, Kirsi H; Ripatti, Samuli; Salomaa, Veikko; Raitakari, Olli T; Järvelin, Marjo-Riitta; Smith, George Davey; Ala-Korpela, Mika

2014-12-01

Increased adiposity is linked with higher risk for cardiometabolic diseases. We aimed to determine to what extent elevated body mass index (BMI) within the normal weight range has causal effects on the detailed systemic metabolite profile in early adulthood. We used Mendelian randomization to estimate causal effects of BMI on 82 metabolic measures in 12,664 adolescents and young adults from four population-based cohorts in Finland (mean age 26 y, range 16-39 y; 51% women; mean ± standard deviation BMI 24 ± 4 kg/m(2)). Circulating metabolites were quantified by high-throughput nuclear magnetic resonance metabolomics and biochemical assays. In cross-sectional analyses, elevated BMI was adversely associated with cardiometabolic risk markers throughout the systemic metabolite profile, including lipoprotein subclasses, fatty acid composition, amino acids, inflammatory markers, and various hormones (p<0.0005 for 68 measures). Metabolite associations with BMI were generally stronger for men than for women (median 136%, interquartile range 125%-183%). A gene score for predisposition to elevated BMI, composed of 32 established genetic correlates, was used as the instrument to assess causality. Causal effects of elevated BMI closely matched observational estimates (correspondence 87% ± 3%; R(2)= 0.89), suggesting causative influences of adiposity on the levels of numerous metabolites (p<0.0005 for 24 measures), including lipoprotein lipid subclasses and particle size, branched-chain and aromatic amino acids, and inflammation-related glycoprotein acetyls. Causal analyses of certain metabolites and potential sex differences warrant stronger statistical power. Metabolite changes associated with change in BMI during 6 y of follow-up were examined for 1,488 individuals. Change in BMI was accompanied by widespread metabolite changes, which had an association pattern similar to that of the cross-sectional observations, yet with greater metabolic effects (correspondence 160% ± 2%; R(2) = 0.92). Mendelian randomization indicates causal adverse effects of increased adiposity with multiple cardiometabolic risk markers across the metabolite profile in adolescents and young adults within the non-obese weight range. Consistent with the causal influences of adiposity, weight changes were paralleled by extensive metabolic changes, suggesting a broadly modifiable systemic metabolite profile in early adulthood. Please see later in the article for the Editors' Summary.

An application of model-fitting procedures for marginal structural models.

PubMed

Mortimer, Kathleen M; Neugebauer, Romain; van der Laan, Mark; Tager, Ira B

2005-08-15

Marginal structural models (MSMs) are being used more frequently to obtain causal effect estimates in observational studies. Although the principal estimator of MSM coefficients has been the inverse probability of treatment weight (IPTW) estimator, there are few published examples that illustrate how to apply IPTW or discuss the impact of model selection on effect estimates. The authors applied IPTW estimation of an MSM to observational data from the Fresno Asthmatic Children's Environment Study (2000-2002) to evaluate the effect of asthma rescue medication use on pulmonary function and compared their results with those obtained through traditional regression methods. Akaike's Information Criterion and cross-validation methods were used to fit the MSM. In this paper, the influence of model selection and evaluation of key assumptions such as the experimental treatment assignment assumption are discussed in detail. Traditional analyses suggested that medication use was not associated with an improvement in pulmonary function--a finding that is counterintuitive and probably due to confounding by symptoms and asthma severity. The final MSM estimated that medication use was causally related to a 7% improvement in pulmonary function. The authors present examples that should encourage investigators who use IPTW estimation to undertake and discuss the impact of model-fitting procedures to justify the choice of the final weights.

Reflecting on explanatory ability: A mechanism for detecting gaps in causal knowledge.

PubMed

Johnson, Dan R; Murphy, Meredith P; Messer, Riley M

2016-05-01

People frequently overestimate their understanding-with a particularly large blind-spot for gaps in their causal knowledge. We introduce a metacognitive approach to reducing overestimation, termed reflecting on explanatory ability (REA), which is briefly thinking about how well one could explain something in a mechanistic, step-by-step, causally connected manner. Nine experiments demonstrated that engaging in REA just before estimating one's understanding substantially reduced overestimation. Moreover, REA reduced overestimation with nearly the same potency as generating full explanations, but did so 20 times faster (although only for high complexity objects). REA substantially reduced overestimation by inducing participants to quickly evaluate an object's inherent causal complexity (Experiments 4-7). REA reduced overestimation by also fostering step-by-step, causally connected processing (Experiments 2 and 3). Alternative explanations for REA's effects were ruled out including a general conservatism account (Experiments 4 and 5) and a covert explanation account (Experiment 8). REA's overestimation-reduction effect generalized beyond objects (Experiments 1-8) to sociopolitical policies (Experiment 9). REA efficiently detects gaps in our causal knowledge with implications for improving self-directed learning, enhancing self-insight into vocational and academic abilities, and even reducing extremist attitudes. (c) 2016 APA, all rights reserved).

Sufficiency and Necessity Assumptions in Causal Structure Induction

ERIC Educational Resources Information Center

Mayrhofer, Ralf; Waldmann, Michael R.

2016-01-01

Research on human causal induction has shown that people have general prior assumptions about causal strength and about how causes interact with the background. We propose that these prior assumptions about the parameters of causal systems do not only manifest themselves in estimations of causal strength or the selection of causes but also when…

The impact of education on the probability of receiving periodontal treatment. Causal effects measured by using the introduction of a school reform in Norway.

PubMed

Grytten, Jostein; Skau, Irene

2017-09-01

The aim of the present study was to estimate the causal effect of education on the probability of receiving periodontal treatment in the adult Norwegian population. In Norway, a substantial part of the cost of periodontal treatment is subsidized by the National Insurance Scheme. In that case, one might expect that the influence of individual resources, such as education, on receiving treatment would be reduced or eliminated. Causal effects were estimated by using data on a school reform in Norway. During the period 1960-1972, all municipalities in Norway were required to increase the number of compulsory years of schooling from seven to nine years. The education reform was used to create exogenous variation in the education variable. The education data were combined with large sets of data from the Norwegian Health Economics Administration and Statistics Norway. Since municipalities implemented the reform at different times, we have both cross-sectional and time-series variation in the reform instrument. Thus we were able to estimate the effect of education on the probability of receiving periodontal treatment by controlling for municipality fixed effects and trend variables. The probability of receiving periodontal treatment increased by 1.4-1.8 percentage points per additional year of schooling. This is a reasonably strong effect, which indicates that policies to increase the level of education in the population can be an effective tool to improve oral health, including periodontal health. Copyright © 2017 Elsevier Ltd. All rights reserved.

Quantum violation of an instrumental test

NASA Astrophysics Data System (ADS)

Chaves, Rafael; Carvacho, Gonzalo; Agresti, Iris; Di Giulio, Valerio; Aolita, Leandro; Giacomini, Sandro; Sciarrino, Fabio

2018-03-01

Inferring causal relations from experimental observations is of primal importance in science. Instrumental tests provide an essential tool for that aim, as they allow one to estimate causal dependencies even in the presence of unobserved common causes. In view of Bell's theorem, which implies that quantum mechanics is incompatible with our most basic notions of causality, it is of utmost importance to understand whether and how paradigmatic causal tools obtained in a classical setting can be carried over to the quantum realm. Here we show that quantum effects imply radically different predictions in the instrumental scenario. Among other results, we show that an instrumental test can be violated by entangled quantum states. Furthermore, we demonstrate such violation using a photonic set-up with active feed-forward of information, thus providing an experimental proof of this new form of non-classical behaviour. Our findings have fundamental implications in causal inference and may also lead to new applications of quantum technologies.

Granger causality for state-space models

NASA Astrophysics Data System (ADS)

Barnett, Lionel; Seth, Anil K.

2015-04-01

Granger causality has long been a prominent method for inferring causal interactions between stochastic variables for a broad range of complex physical systems. However, it has been recognized that a moving average (MA) component in the data presents a serious confound to Granger causal analysis, as routinely performed via autoregressive (AR) modeling. We solve this problem by demonstrating that Granger causality may be calculated simply and efficiently from the parameters of a state-space (SS) model. Since SS models are equivalent to autoregressive moving average models, Granger causality estimated in this fashion is not degraded by the presence of a MA component. This is of particular significance when the data has been filtered, downsampled, observed with noise, or is a subprocess of a higher dimensional process, since all of these operations—commonplace in application domains as diverse as climate science, econometrics, and the neurosciences—induce a MA component. We show how Granger causality, conditional and unconditional, in both time and frequency domains, may be calculated directly from SS model parameters via solution of a discrete algebraic Riccati equation. Numerical simulations demonstrate that Granger causality estimators thus derived have greater statistical power and smaller bias than AR estimators. We also discuss how the SS approach facilitates relaxation of the assumptions of linearity, stationarity, and homoscedasticity underlying current AR methods, thus opening up potentially significant new areas of research in Granger causal analysis.

Multivariate Granger causality: an estimation framework based on factorization of the spectral density matrix

PubMed Central

Wen, Xiaotong; Rangarajan, Govindan; Ding, Mingzhou

2013-01-01

Granger causality is increasingly being applied to multi-electrode neurophysiological and functional imaging data to characterize directional interactions between neurons and brain regions. For a multivariate dataset, one might be interested in different subsets of the recorded neurons or brain regions. According to the current estimation framework, for each subset, one conducts a separate autoregressive model fitting process, introducing the potential for unwanted variability and uncertainty. In this paper, we propose a multivariate framework for estimating Granger causality. It is based on spectral density matrix factorization and offers the advantage that the estimation of such a matrix needs to be done only once for the entire multivariate dataset. For any subset of recorded data, Granger causality can be calculated through factorizing the appropriate submatrix of the overall spectral density matrix. PMID:23858479

Estimating Causal Effects of Education Interventions Using a Two-Rating Regression Discontinuity Design: Lessons from a Simulation Study and an Application

ERIC Educational Resources Information Center

Porter, Kristin E.; Reardon, Sean F.; Unlu, Fatih; Bloom, Howard S.; Cimpian, Joseph R.

2017-01-01

A valuable extension of the single-rating regression discontinuity design (RDD) is a multiple-rating RDD (MRRDD). To date, four main methods have been used to estimate average treatment effects at the multiple treatment frontiers of an MRRDD: the "surface" method, the "frontier" method, the "binding-score" method, and…

The Teacher, the Physician and the Person: Exploring Causal Connections between Teaching Performance and Role Model Types Using Directed Acyclic Graphs

PubMed Central

Boerebach, Benjamin C. M.; Lombarts, Kiki M. J. M. H.; Scherpbier, Albert J. J.; Arah, Onyebuchi A.

2013-01-01

Background In fledgling areas of research, evidence supporting causal assumptions is often scarce due to the small number of empirical studies conducted. In many studies it remains unclear what impact explicit and implicit causal assumptions have on the research findings; only the primary assumptions of the researchers are often presented. This is particularly true for research on the effect of faculty’s teaching performance on their role modeling. Therefore, there is a need for robust frameworks and methods for transparent formal presentation of the underlying causal assumptions used in assessing the causal effects of teaching performance on role modeling. This study explores the effects of different (plausible) causal assumptions on research outcomes. Methods This study revisits a previously published study about the influence of faculty’s teaching performance on their role modeling (as teacher-supervisor, physician and person). We drew eight directed acyclic graphs (DAGs) to visually represent different plausible causal relationships between the variables under study. These DAGs were subsequently translated into corresponding statistical models, and regression analyses were performed to estimate the associations between teaching performance and role modeling. Results The different causal models were compatible with major differences in the magnitude of the relationship between faculty’s teaching performance and their role modeling. Odds ratios for the associations between teaching performance and the three role model types ranged from 31.1 to 73.6 for the teacher-supervisor role, from 3.7 to 15.5 for the physician role, and from 2.8 to 13.8 for the person role. Conclusions Different sets of assumptions about causal relationships in role modeling research can be visually depicted using DAGs, which are then used to guide both statistical analysis and interpretation of results. Since study conclusions can be sensitive to different causal assumptions, results should be interpreted in the light of causal assumptions made in each study. PMID:23936020

Quasi-experiments to establish causal effects of HIV care and treatment and to improve the cascade of care

PubMed Central

Bor, Jacob; Geldsetzer, Pascal; Venkataramani, Atheendar; Bärnighausen, Till

2015-01-01

Purpose of review Randomized, population-representative trials of clinical interventions are rare. Quasi-experiments have been used successfully to generate causal evidence on the cascade of HIV care in a broad range of real-world settings. Recent findings Quasi-experiments exploit exogenous, or quasi-random, variation occurring naturally in the world or because of an administrative rule or policy change to estimate causal effects. Well designed quasi-experiments have greater internal validity than typical observational research designs. At the same time, quasi-experiments may also have potential for greater external validity than experiments and can be implemented when randomized clinical trials are infeasible or unethical. Quasi-experimental studies have established the causal effects of HIV testing and initiation of antiretroviral therapy on health, economic outcomes and sexual behaviors, as well as indirect effects on other community members. Recent quasi-experiments have evaluated specific interventions to improve patient performance in the cascade of care, providing causal evidence to optimize clinical management of HIV. Summary Quasi-experiments have generated important data on the real-world impacts of HIV testing and treatment and on interventions to improve the cascade of care. With the growth in large-scale clinical and administrative data, quasi-experiments enable rigorous evaluation of policies implemented in real-world settings. PMID:26371463

Quasi-experiments to establish causal effects of HIV care and treatment and to improve the cascade of care.

PubMed

Bor, Jacob; Geldsetzer, Pascal; Venkataramani, Atheendar; Bärnighausen, Till

2015-11-01

Randomized, population-representative trials of clinical interventions are rare. Quasi-experiments have been used successfully to generate causal evidence on the cascade of HIV care in a broad range of real-world settings. Quasi-experiments exploit exogenous, or quasi-random, variation occurring naturally in the world or because of an administrative rule or policy change to estimate causal effects. Well designed quasi-experiments have greater internal validity than typical observational research designs. At the same time, quasi-experiments may also have potential for greater external validity than experiments and can be implemented when randomized clinical trials are infeasible or unethical. Quasi-experimental studies have established the causal effects of HIV testing and initiation of antiretroviral therapy on health, economic outcomes and sexual behaviors, as well as indirect effects on other community members. Recent quasi-experiments have evaluated specific interventions to improve patient performance in the cascade of care, providing causal evidence to optimize clinical management of HIV. Quasi-experiments have generated important data on the real-world impacts of HIV testing and treatment and on interventions to improve the cascade of care. With the growth in large-scale clinical and administrative data, quasi-experiments enable rigorous evaluation of policies implemented in real-world settings.

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.

New Insights into Signed Path Coefficient Granger Causality Analysis

PubMed Central

Zhang, Jian; Li, Chong; Jiang, Tianzi

2016-01-01

Granger causality analysis, as a time series analysis technique derived from econometrics, has been applied in an ever-increasing number of publications in the field of neuroscience, including fMRI, EEG/MEG, and fNIRS. The present study mainly focuses on the validity of “signed path coefficient Granger causality,” a Granger-causality-derived analysis method that has been adopted by many fMRI researches in the last few years. This method generally estimates the causality effect among the time series by an order-1 autoregression, and defines a positive or negative coefficient as an “excitatory” or “inhibitory” influence. In the current work we conducted a series of computations from resting-state fMRI data and simulation experiments to illustrate the signed path coefficient method was flawed and untenable, due to the fact that the autoregressive coefficients were not always consistent with the real causal relationships and this would inevitablely lead to erroneous conclusions. Overall our findings suggested that the applicability of this kind of causality analysis was rather limited, hence researchers should be more cautious in applying the signed path coefficient Granger causality to fMRI data to avoid misinterpretation. PMID:27833547

Relationships between infant mortality, birth spacing and fertility in Matlab, Bangladesh.

PubMed

van Soest, Arthur; Saha, Unnati Rani

2018-01-01

Although research on the fertility response to childhood mortality is widespread in demographic literature, very few studies focused on the two-way causal relationships between infant mortality and fertility. Understanding the nature of such relationships is important in order to design effective policies to reduce child mortality and improve family planning. In this study, we use dynamic panel data techniques to analyse the causal effects of infant mortality on birth intervals and fertility, as well as the causal effects of birth intervals on mortality in rural Bangladesh, accounting for unobserved heterogeneity and reverse causality. Simulations based upon the estimated model show whether (and to what extent) mortality and fertility can be reduced by breaking the causal links between short birth intervals and infant mortality. We find a replacement effect of infant mortality on total fertility of about 0.54 children for each infant death in the comparison area with standard health services. Eliminating the replacement effect would lengthen birth intervals and reduce the total number of births, resulting in a fall in mortality by 2.45 children per 1000 live births. These effects are much smaller in the treatment area with extensive health services and information on family planning, where infant mortality is smaller, birth intervals are longer, and total fertility is lower. In both areas, we find evidence of boy preference in family planning.

Relationships between infant mortality, birth spacing and fertility in Matlab, Bangladesh

PubMed Central

van Soest, Arthur

2018-01-01

Although research on the fertility response to childhood mortality is widespread in demographic literature, very few studies focused on the two-way causal relationships between infant mortality and fertility. Understanding the nature of such relationships is important in order to design effective policies to reduce child mortality and improve family planning. In this study, we use dynamic panel data techniques to analyse the causal effects of infant mortality on birth intervals and fertility, as well as the causal effects of birth intervals on mortality in rural Bangladesh, accounting for unobserved heterogeneity and reverse causality. Simulations based upon the estimated model show whether (and to what extent) mortality and fertility can be reduced by breaking the causal links between short birth intervals and infant mortality. We find a replacement effect of infant mortality on total fertility of about 0.54 children for each infant death in the comparison area with standard health services. Eliminating the replacement effect would lengthen birth intervals and reduce the total number of births, resulting in a fall in mortality by 2.45 children per 1000 live births. These effects are much smaller in the treatment area with extensive health services and information on family planning, where infant mortality is smaller, birth intervals are longer, and total fertility is lower. In both areas, we find evidence of boy preference in family planning. PMID:29702692

Assessing Methods for Generalizing Experimental Impact Estimates to Target Populations

ERIC Educational Resources Information Center

Kern, Holger L.; Stuart, Elizabeth A.; Hill, Jennifer; Green, Donald P.

2016-01-01

Randomized experiments are considered the gold standard for causal inference because they can provide unbiased estimates of treatment effects for the experimental participants. However, researchers and policymakers are often interested in using a specific experiment to inform decisions about other target populations. In education research,…

Propensity Score Weighting with Error-Prone Covariates

ERIC Educational Resources Information Center

McCaffrey, Daniel F.; Lockwood, J. R.; Setodji, Claude M.

2011-01-01

Inverse probability weighting (IPW) estimates are widely used in applications where data are missing due to nonresponse or censoring or in observational studies of causal effects where the counterfactuals cannot be observed. This extensive literature has shown the estimators to be consistent and asymptotically normal under very general conditions,…

Does Money Really Matter? Estimating Impacts of Family Income on Young Children's Achievement With Data From Random-Assignment Experiments

PubMed Central

Duncan, Greg J.; Morris, Pamela A.; Rodrigues, Chris

2011-01-01

Social scientists do not agree on the size and nature of the causal impacts of parental income on children's achievement. We revisit this issue using a set of welfare and antipoverty experiments conducted in the 1990s. We utilize an instrumental variables strategy to leverage the variation in income and achievement that arises from random assignment to the treatment group to estimate the causal effect of income on child achievement. Our estimates suggest that a $1,000 increase in annual income increases young children's achievement by 5%–6% of a standard deviation. As such, our results suggest that family income has a policy-relevant, positive impact on the eventual school achievement of preschool children. PMID:21688900

Causality links among renewable energy consumption, CO2 emissions, and economic growth in Africa: evidence from a panel ARDL-PMG approach.

PubMed

Attiaoui, Imed; Toumi, Hassen; Ammouri, Bilel; Gargouri, Ilhem

2017-05-01

This research examines the causality (For the remainder of the paper, the notion of causality refers to Granger causality.) links among renewable energy consumption (REC), CO 2 emissions (CE), non-renewable energy consumption (NREC), and economic growth (GDP) using an autoregressive distributed lag model based on the pooled mean group estimation (ARDL-PMG) and applying Granger causality tests for a panel consisting of 22 African countries for the period between 1990 and 2011. There is unidirectional and irreversible short-run causality from CE to GDP. The causal direction between CE and REC is unobservable over the short-term. Moreover, we find unidirectional, short-run causality from REC to GDP. When testing per pair of variables, there are short-run bidirectional causalities among REC, CE, and GDP. However, if we add CE to the variables REC and NREC, the causality to GDP is observable, and causality from the pair REC and NREC to economic growth is neutral. Likewise, if we add NREC to the variables GDP and REC, there is causality. There are bidirectional long-run causalities among REC, CE, and GDP, which supports the feedback assumption. Causality from GDP to REC is not strong for the panel. If we test per pair of variables, the strong causality from GDP and CE to REC is neutral. The long-run PMG estimates show that NREC and gross domestic product increase CE, whereas REC decreases CE.

A Discordance Weighting Approach Estimating Occupational and Income Returns to Education.

PubMed

Andersson, Matthew A

2018-04-23

Schooling differences between identical twins are often utilized as a natural experiment to estimate returns to education. Despite longstanding doubts about the truly random nature of within-twin-pair schooling discordance, such discordance has not yet been understood comprehensively, in terms of diverse between- and within-family peer, academic, familial, social, and health exposures. Here, a predictive analysis using national U.S. midlife twin data shows that within-pair schooling differences are endogenous to a variety of childhood exposures. Using discordance propensities, returns to education under a true natural experiment are simulated. Results for midlife occupation and income reveal differences in estimated returns to education that are statistically insignificant, suggesting that twin-based estimates of causal effects are robust. Moreover, identical and fraternal twins show similar levels of discordance endogeneity and similar responses to propensity weighting, suggesting that the identical twins may not provide demonstrably better leverage in the causal identification of educational returns.

Implementation and reporting of causal mediation analysis in 2015: a systematic review in epidemiological studies.

PubMed

Liu, Shao-Hsien; Ulbricht, Christine M; Chrysanthopoulou, Stavroula A; Lapane, Kate L

2016-07-20

Causal mediation analysis is often used to understand the impact of variables along the causal pathway of an occurrence relation. How well studies apply and report the elements of causal mediation analysis remains unknown. We systematically reviewed epidemiological studies published in 2015 that employed causal mediation analysis to estimate direct and indirect effects of observed associations between an exposure on an outcome. We identified potential epidemiological studies through conducting a citation search within Web of Science and a keyword search within PubMed. Two reviewers independently screened studies for eligibility. For eligible studies, one reviewer performed data extraction, and a senior epidemiologist confirmed the extracted information. Empirical application and methodological details of the technique were extracted and summarized. Thirteen studies were eligible for data extraction. While the majority of studies reported and identified the effects of measures, most studies lacked sufficient details on the extent to which identifiability assumptions were satisfied. Although most studies addressed issues of unmeasured confounders either from empirical approaches or sensitivity analyses, the majority did not examine the potential bias arising from the measurement error of the mediator. Some studies allowed for exposure-mediator interaction and only a few presented results from models both with and without interactions. Power calculations were scarce. Reporting of causal mediation analysis is varied and suboptimal. Given that the application of causal mediation analysis will likely continue to increase, developing standards of reporting of causal mediation analysis in epidemiological research would be prudent.

The Educational Consequences of Teen Childbearing

PubMed Central

Kane, Jennifer B.; Morgan, S. Philip; Harris, Kathleen Mullan; Guilkey, David K.

2013-01-01

A huge literature shows that teen mothers face a variety of detriments across the life course, including truncated educational attainment. To what extent is this association causal? The estimated effects of teen motherhood on schooling vary widely, ranging from no discernible difference to 2.6 fewer years among teen mothers. The magnitude of educational consequences is therefore uncertain, despite voluminous policy and prevention efforts that rest on the assumption of a negative and presumably causal effect. This study adjudicates between two potential sources of inconsistency in the literature—methodological differences or cohort differences—by using a single, high-quality data source: namely, The National Longitudinal Study of Adolescent Health. We replicate analyses across four different statistical strategies: ordinary least squares regression; propensity score matching; and parametric and semiparametric maximum likelihood estimation. Results demonstrate educational consequences of teen childbearing, with estimated effects between 0.7 and 1.9 fewer years of schooling among teen mothers. We select our preferred estimate (0.7), derived from semiparametric maximum likelihood estimation, on the basis of weighing the strengths and limitations of each approach. Based on the range of estimated effects observed in our study, we speculate that variable statistical methods are the likely source of inconsistency in the past. We conclude by discussing implications for future research and policy, and recommend that future studies employ a similar multimethod approach to evaluate findings. PMID:24078155

An Evolutionary Perspective on Epistasis and the Missing Heritability

PubMed Central

Hemani, Gibran; Knott, Sara; Haley, Chris

2013-01-01

The relative importance between additive and non-additive genetic variance has been widely argued in quantitative genetics. By approaching this question from an evolutionary perspective we show that, while additive variance can be maintained under selection at a low level for some patterns of epistasis, the majority of the genetic variance that will persist is actually non-additive. We propose that one reason that the problem of the “missing heritability” arises is because the additive genetic variation that is estimated to be contributing to the variance of a trait will most likely be an artefact of the non-additive variance that can be maintained over evolutionary time. In addition, it can be shown that even a small reduction in linkage disequilibrium between causal variants and observed SNPs rapidly erodes estimates of epistatic variance, leading to an inflation in the perceived importance of additive effects. We demonstrate that the perception of independent additive effects comprising the majority of the genetic architecture of complex traits is biased upwards and that the search for causal variants in complex traits under selection is potentially underpowered by parameterising for additive effects alone. Given dense SNP panels the detection of causal variants through genome-wide association studies may be improved by searching for epistatic effects explicitly. PMID:23509438

Generalized causal mediation and path analysis: Extensions and practical considerations.

PubMed

Albert, Jeffrey M; Cho, Jang Ik; Liu, Yiying; Nelson, Suchitra

2018-01-01

Causal mediation analysis seeks to decompose the effect of a treatment or exposure among multiple possible paths and provide casually interpretable path-specific effect estimates. Recent advances have extended causal mediation analysis to situations with a sequence of mediators or multiple contemporaneous mediators. However, available methods still have limitations, and computational and other challenges remain. The present paper provides an extended causal mediation and path analysis methodology. The new method, implemented in the new R package, gmediation (described in a companion paper), accommodates both a sequence (two stages) of mediators and multiple mediators at each stage, and allows for multiple types of outcomes following generalized linear models. The methodology can also handle unsaturated models and clustered data. Addressing other practical issues, we provide new guidelines for the choice of a decomposition, and for the choice of a reference group multiplier for the reduction of Monte Carlo error in mediation formula computations. The new method is applied to data from a cohort study to illuminate the contribution of alternative biological and behavioral paths in the effect of socioeconomic status on dental caries in adolescence.

Student Sorting and Bias in Value Added Estimation: Selection on Observables and Unobservables. NBER Working Paper No. 14666

ERIC Educational Resources Information Center

Rothstein, Jesse

2009-01-01

Non-random assignment of students to teachers can bias value added estimates of teachers' causal effects. Rothstein (2008a, b) shows that typical value added models indicate large counter-factual effects of 5th grade teachers on students' 4th grade learning, indicating that classroom assignments are far from random. This paper quantifies the…

The Impact of Family Income on Child Achievement: Evidence from the Earned Income Tax Credit. NBER Working Paper No. 14599

ERIC Educational Resources Information Center

Dahl, Gordon; Lochner, Lance

2008-01-01

Past estimates of the effect of family income on child development have often been plagued by endogeneity and measurement error. In this paper, we use two simulated instrumental variables strategies to estimate the causal effect of income on children's math and reading achievement. Our identification derives from the large, non-linear changes in…

The Impact of Family Income on Child Achievement: Evidence from the Earned Income Tax Credit. Discussion Paper No. 1361-09

ERIC Educational Resources Information Center

Dahl, Gordon; Lochner, Lance

2009-01-01

Past estimates of the effect of family income on child development have often been plagued by endogeneity and measurement error. In this paper, we use two simulated instrumental variables strategies to estimate the causal effect of income on children's math and reading achievement. Our identification derives from the large, non-linear changes…

Gang membership and substance use: guilt as a gendered causal pathway

PubMed Central

Coffman, Donna L.; Melde, Chris; Esbensen, Finn-Aage

2014-01-01

Objectives We examine whether anticipated guilt for substance use is a gendered mechanism underlying the noted enhancement effect of gang membership on illegal drug use. We also demonstrate a method for making stronger causal inferences when assessing mediation in the presence of moderation and time-varying confounding. Methods We estimate a series of inverse propensity weighted models to obtain unbiased estimates of mediation in the presence of confounding of the exposure (i.e., gang membership) and mediator (i.e., anticipated guilt) using three waves of data from a multi-site panel study of a law-related education program for youth (N=1,113). Results The onset of gang membership significantly decreased anticipated substance use guilt among both male and female respondents. This reduction was significantly associated with increased frequency of substance use only for female respondents, however, suggesting that gender moderates the mechanism through which gang membership influences substance use. Conclusions Criminologists are often concerned with identifying causal pathways for antisocial and/or delinquent behavior, but confounders of the exposure, mediator, and outcome often interfere with efforts to assess mediation. Many new approaches have been proposed for strengthening causal inference for mediation effects. After controlling for confounding using inverse propensity weighting, our results suggest that interventions aimed at reducing substance use by current and former female gang members should focus on the normative aspects of these behaviors. PMID:26190954

Gang membership and substance use: guilt as a gendered causal pathway.

PubMed

Coffman, Donna L; Melde, Chris; Esbensen, Finn-Aage

2015-03-01

We examine whether anticipated guilt for substance use is a gendered mechanism underlying the noted enhancement effect of gang membership on illegal drug use. We also demonstrate a method for making stronger causal inferences when assessing mediation in the presence of moderation and time-varying confounding. We estimate a series of inverse propensity weighted models to obtain unbiased estimates of mediation in the presence of confounding of the exposure (i.e., gang membership) and mediator (i.e., anticipated guilt) using three waves of data from a multi-site panel study of a law-related education program for youth ( N =1,113). The onset of gang membership significantly decreased anticipated substance use guilt among both male and female respondents. This reduction was significantly associated with increased frequency of substance use only for female respondents, however, suggesting that gender moderates the mechanism through which gang membership influences substance use. Criminologists are often concerned with identifying causal pathways for antisocial and/or delinquent behavior, but confounders of the exposure, mediator, and outcome often interfere with efforts to assess mediation. Many new approaches have been proposed for strengthening causal inference for mediation effects. After controlling for confounding using inverse propensity weighting, our results suggest that interventions aimed at reducing substance use by current and former female gang members should focus on the normative aspects of these behaviors.

Causal Mediation Analysis of Survival Outcome with Multiple Mediators.

PubMed

Huang, Yen-Tsung; Yang, Hwai-I

2017-05-01

Mediation analyses have been a popular approach to investigate the effect of an exposure on an outcome through a mediator. Mediation models with multiple mediators have been proposed for continuous and dichotomous outcomes. However, development of multimediator models for survival outcomes is still limited. We present methods for multimediator analyses using three survival models: Aalen additive hazard models, Cox proportional hazard models, and semiparametric probit models. Effects through mediators can be characterized by path-specific effects, for which definitions and identifiability assumptions are provided. We derive closed-form expressions for path-specific effects for the three models, which are intuitively interpreted using a causal diagram. Mediation analyses using Cox models under the rare-outcome assumption and Aalen additive hazard models consider effects on log hazard ratio and hazard difference, respectively; analyses using semiparametric probit models consider effects on difference in transformed survival time and survival probability. The three models were applied to a hepatitis study where we investigated effects of hepatitis C on liver cancer incidence mediated through baseline and/or follow-up hepatitis B viral load. The three methods show consistent results on respective effect scales, which suggest an adverse estimated effect of hepatitis C on liver cancer not mediated through hepatitis B, and a protective estimated effect mediated through the baseline (and possibly follow-up) of hepatitis B viral load. Causal mediation analyses of survival outcome with multiple mediators are developed for additive hazard and proportional hazard and probit models with utility demonstrated in a hepatitis study.

Instrumental variables as bias amplifiers with general outcome and confounding.

PubMed

Ding, P; VanderWeele, T J; Robins, J M

2017-06-01

Drawing causal inference with observational studies is the central pillar of many disciplines. One sufficient condition for identifying the causal effect is that the treatment-outcome relationship is unconfounded conditional on the observed covariates. It is often believed that the more covariates we condition on, the more plausible this unconfoundedness assumption is. This belief has had a huge impact on practical causal inference, suggesting that we should adjust for all pretreatment covariates. However, when there is unmeasured confounding between the treatment and outcome, estimators adjusting for some pretreatment covariate might have greater bias than estimators without adjusting for this covariate. This kind of covariate is called a bias amplifier, and includes instrumental variables that are independent of the confounder, and affect the outcome only through the treatment. Previously, theoretical results for this phenomenon have been established only for linear models. We fill in this gap in the literature by providing a general theory, showing that this phenomenon happens under a wide class of models satisfying certain monotonicity assumptions. We further show that when the treatment follows an additive or multiplicative model conditional on the instrumental variable and the confounder, these monotonicity assumptions can be interpreted as the signs of the arrows of the causal diagrams.

A Powerful, Potential Outcomes Method for Estimating Any Estimand across Multiple Groups

ERIC Educational Resources Information Center

Pattanayak, Cassandra W.; Rubin, Donald B.; Zell, Elizabeth R.

2013-01-01

In educational research, outcome measures are often estimated across separate studies or across schools, districts, or other subgroups to assess the overall causal effect of an active treatment versus a control treatment. Students may be partitioned into such strata or blocks by experimental design, or separated into studies within a…

Discovering graphical Granger causality using the truncating lasso penalty

PubMed Central

Shojaie, Ali; Michailidis, George

2010-01-01

Motivation: Components of biological systems interact with each other in order to carry out vital cell functions. Such information can be used to improve estimation and inference, and to obtain better insights into the underlying cellular mechanisms. Discovering regulatory interactions among genes is therefore an important problem in systems biology. Whole-genome expression data over time provides an opportunity to determine how the expression levels of genes are affected by changes in transcription levels of other genes, and can therefore be used to discover regulatory interactions among genes. Results: In this article, we propose a novel penalization method, called truncating lasso, for estimation of causal relationships from time-course gene expression data. The proposed penalty can correctly determine the order of the underlying time series, and improves the performance of the lasso-type estimators. Moreover, the resulting estimate provides information on the time lag between activation of transcription factors and their effects on regulated genes. We provide an efficient algorithm for estimation of model parameters, and show that the proposed method can consistently discover causal relationships in the large p, small n setting. The performance of the proposed model is evaluated favorably in simulated, as well as real, data examples. Availability: The proposed truncating lasso method is implemented in the R-package ‘grangerTlasso’ and is freely available at http://www.stat.lsa.umich.edu/∼shojaie/ Contact: shojaie@umich.edu PMID:20823316

The Specification of Causal Models with Tetrad IV: A Review

ERIC Educational Resources Information Center

Landsheer, J. A.

2010-01-01

Tetrad IV is a program designed for the specification of causal models. It is specifically designed to search for causal relations, but also offers the possibility to estimate the parameters of a structural equation model. It offers a remarkable graphical user interface, which facilitates building, evaluating, and searching for causal models. The…

Non-Gaussian Methods for Causal Structure Learning.

PubMed

Shimizu, Shohei

2018-05-22

Causal structure learning is one of the most exciting new topics in the fields of machine learning and statistics. In many empirical sciences including prevention science, the causal mechanisms underlying various phenomena need to be studied. Nevertheless, in many cases, classical methods for causal structure learning are not capable of estimating the causal structure of variables. This is because it explicitly or implicitly assumes Gaussianity of data and typically utilizes only the covariance structure. In many applications, however, non-Gaussian data are often obtained, which means that more information may be contained in the data distribution than the covariance matrix is capable of containing. Thus, many new methods have recently been proposed for using the non-Gaussian structure of data and inferring the causal structure of variables. This paper introduces prevention scientists to such causal structure learning methods, particularly those based on the linear, non-Gaussian, acyclic model known as LiNGAM. These non-Gaussian data analysis tools can fully estimate the underlying causal structures of variables under assumptions even in the presence of unobserved common causes. This feature is in contrast to other approaches. A simulated example is also provided.

Collinearity and Causal Diagrams: A Lesson on the Importance of Model Specification.

PubMed

Schisterman, Enrique F; Perkins, Neil J; Mumford, Sunni L; Ahrens, Katherine A; Mitchell, Emily M

2017-01-01

Correlated data are ubiquitous in epidemiologic research, particularly in nutritional and environmental epidemiology where mixtures of factors are often studied. Our objectives are to demonstrate how highly correlated data arise in epidemiologic research and provide guidance, using a directed acyclic graph approach, on how to proceed analytically when faced with highly correlated data. We identified three fundamental structural scenarios in which high correlation between a given variable and the exposure can arise: intermediates, confounders, and colliders. For each of these scenarios, we evaluated the consequences of increasing correlation between the given variable and the exposure on the bias and variance for the total effect of the exposure on the outcome using unadjusted and adjusted models. We derived closed-form solutions for continuous outcomes using linear regression and empirically present our findings for binary outcomes using logistic regression. For models properly specified, total effect estimates remained unbiased even when there was almost perfect correlation between the exposure and a given intermediate, confounder, or collider. In general, as the correlation increased, the variance of the parameter estimate for the exposure in the adjusted models increased, while in the unadjusted models, the variance increased to a lesser extent or decreased. Our findings highlight the importance of considering the causal framework under study when specifying regression models. Strategies that do not take into consideration the causal structure may lead to biased effect estimation for the original question of interest, even under high correlation.

Instrumental variables and Mendelian randomization with invalid instruments

NASA Astrophysics Data System (ADS)

Kang, Hyunseung

Instrumental variables (IV) methods have been widely used to determine the causal effect of a treatment, exposure, policy, or an intervention on an outcome of interest. The IV method relies on having a valid instrument, a variable that is (A1) associated with the exposure, (A2) has no direct effect on the outcome, and (A3) is unrelated to the unmeasured confounders associated with the exposure and the outcome. However, in practice, finding a valid instrument, especially those that satisfy (A2) and (A3), can be challenging. For example, in Mendelian randomization studies where genetic markers are used as instruments, complete knowledge about instruments' validity is equivalent to complete knowledge about the involved genes' functions. The dissertation explores the theory, methods, and application of IV methods when invalid instruments are present. First, when we have multiple candidate instruments, we establish a theoretical bound whereby causal effects are only identified as long as less than 50% of instruments are invalid, without knowing which of the instruments are invalid. We also propose a fast penalized method, called sisVIVE, to estimate the causal effect. We find that sisVIVE outperforms traditional IV methods when invalid instruments are present both in simulation studies as well as in real data analysis. Second, we propose a robust confidence interval under the multiple invalid IV setting. This work is an extension of our work on sisVIVE. However, unlike sisVIVE which is robust to violations of (A2) and (A3), our confidence interval procedure provides honest coverage even if all three assumptions, (A1)-(A3), are violated. Third, we study the single IV setting where the one IV we have may actually be invalid. We propose a nonparametric IV estimation method based on full matching, a technique popular in causal inference for observational data, that leverages observed covariates to make the instrument more valid. We propose an estimator along with inferential results that are robust to mis-specifications of the covariate-outcome model. We also provide a sensitivity analysis should the instrument turn out to be invalid, specifically violate (A3). Fourth, in application work, we study the causal effect of malaria on stunting among children in Ghana. Previous studies on the effect of malaria and stunting were observational and contained various unobserved confounders, most notably nutritional deficiencies. To infer causality, we use the sickle cell genotype, a trait that confers some protection against malaria and was randomly assigned at birth, as an IV and apply our nonparametric IV method. We find that the risk of stunting increases by 0.22 (95% CI: 0.044,1) for every malaria episode and is sensitive to unmeasured confounders.

A classical regression framework for mediation analysis: fitting one model to estimate mediation effects.

PubMed

Saunders, Christina T; Blume, Jeffrey D

2017-10-26

Mediation analysis explores the degree to which an exposure's effect on an outcome is diverted through a mediating variable. We describe a classical regression framework for conducting mediation analyses in which estimates of causal mediation effects and their variance are obtained from the fit of a single regression model. The vector of changes in exposure pathway coefficients, which we named the essential mediation components (EMCs), is used to estimate standard causal mediation effects. Because these effects are often simple functions of the EMCs, an analytical expression for their model-based variance follows directly. Given this formula, it is instructive to revisit the performance of routinely used variance approximations (e.g., delta method and resampling methods). Requiring the fit of only one model reduces the computation time required for complex mediation analyses and permits the use of a rich suite of regression tools that are not easily implemented on a system of three equations, as would be required in the Baron-Kenny framework. Using data from the BRAIN-ICU study, we provide examples to illustrate the advantages of this framework and compare it with the existing approaches. © The Author 2017. Published by Oxford University Press.

What matters most: quantifying an epidemiology of consequence

PubMed Central

Keyes, Katherine; Galea, Sandro

2015-01-01

Risk factor epidemiology has contributed to substantial public health success. In this essay, we argue, however, that the focus on risk factor epidemiology has led epidemiology to ever increasing focus on the estimation of precise causal effects of exposures on an outcome at the expense of engagement with the broader causal architecture that produces population health. To conduct an epidemiology of consequence, a systematic effort is needed to engage our science in a critical reflection both about how well and under what conditions or assumptions we can assess causal effects and also on what will truly matter most for changing population health. Such an approach changes the priorities and values of the discipline and requires reorientation of how we structure the questions we ask and the methods we use, as well as how we teach epidemiology to our emerging scholars. PMID:25749559

The Mediation Formula: A Guide to the Assessment of Causal Pathways in Nonlinear Models

DTIC Science & Technology

2011-10-27

through (25), (26) and (27), rather than going through (23) ( van der Laan and Rubin, 2006). 29 values, though disparities in parameters may not...graphs. Epidemiology 22 378–381. Petersen, M., Sinisi, S. and van der Laan, M. (2006). Estimation of direct causal effects. Epidemiology 17 276–284...and J. Halpern, eds.). College Publications, UK, 415–444. van der Laan, M. J. and Rubin, D. (2006). Targeted maximum likelihood learning. The

Instrumental variable approaches to identifying the causal effect of educational attainment on dementia risk

PubMed Central

Nguyen, Thu T.; Tchetgen Tchetgen, Eric J.; Kawachi, Ichiro; Gilman, Stephen E.; Walter, Stefan; Liu, Sze Y.; Manly, Jennifer; Glymour, M. Maria

2015-01-01

Purpose Education is an established correlate of cognitive status in older adulthood, but whether expanding educational opportunities would improve cognitive functioning remains unclear given limitations of prior studies for causal inference. Therefore, we conducted instrumental variable (IV) analyses of the association between education and dementia risk, using for the first time in this area, genetic variants as instruments as well as state-level school policies. Methods IV analyses in the Health and Retirement Study cohort (1998–2010) used two sets of instruments: 1) a genetic risk score constructed from three single nucleotide polymorphisms (SNPs) (n=8,054); and 2) compulsory schooling laws (CSLs) and state school characteristics (term length, student teacher ratios, and expenditures) (n=13,167). Results Employing the genetic risk score as an IV, there was a 1.1% reduction in dementia risk per year of schooling (95% CI: −2.4, 0.02). Leveraging compulsory schooling laws and state school characteristics as IVs, there was a substantially larger protective effect (−9.5%; 95% CI: −14.8, −4.2). Analyses evaluating the plausibility of the IV assumptions indicated estimates derived from analyses relying on CSLs provide the best estimates of the causal effect of education. Conclusion IV analyses suggest education is protective against risk of dementia in older adulthood. PMID:26633592

Pairwise measures of causal direction in the epidemiology of sleep problems and depression.

PubMed

Rosenström, Tom; Jokela, Markus; Puttonen, Sampsa; Hintsanen, Mirka; Pulkki-Råback, Laura; Viikari, Jorma S; Raitakari, Olli T; Keltikangas-Järvinen, Liisa

2012-01-01

Depressive mood is often preceded by sleep problems, suggesting that they increase the risk of depression. Sleep problems can also reflect prodromal symptom of depression, thus temporal precedence alone is insufficient to confirm causality. The authors applied recently introduced statistical causal-discovery algorithms that can estimate causality from cross-sectional samples in order to infer the direction of causality between the two sets of symptoms from a novel perspective. Two common-population samples were used; one from the Young Finns study (690 men and 997 women, average age 37.7 years, range 30-45), and another from the Wisconsin Longitudinal study (3101 men and 3539 women, average age 53.1 years, range 52-55). These included three depression questionnaires (two in Young Finns data) and two sleep problem questionnaires. Three different causality estimates were constructed for each data set, tested in a benchmark data with a (practically) known causality, and tested for assumption violations using simulated data. Causality algorithms performed well in the benchmark data and simulations, and a prediction was drawn for future empirical studies to confirm: for minor depression/dysphoria, sleep problems cause significantly more dysphoria than dysphoria causes sleep problems. The situation may change as depression becomes more severe, or more severe levels of symptoms are evaluated; also, artefacts due to severe depression being less well presented in the population data than minor depression may intervene the estimation for depression scales that emphasize severe symptoms. The findings are consistent with other emerging epidemiological and biological evidence.

Estimating the Variance of Design Parameters

ERIC Educational Resources Information Center

Hedberg, E. C.; Hedges, L. V.; Kuyper, A. M.

2015-01-01

Randomized experiments are generally considered to provide the strongest basis for causal inferences about cause and effect. Consequently randomized field trials have been increasingly used to evaluate the effects of education interventions, products, and services. Populations of interest in education are often hierarchically structured (such as…

Finite Feedback Cycling in Structural Equation Models

ERIC Educational Resources Information Center

Hayduk, Leslie A.

2009-01-01

In models containing reciprocal effects, or longer causal loops, the usual effect estimates assume that any effect touching a loop initiates an infinite cycling of effects around that loop. The real world, in contrast, might permit only finite feedback cycles. I use a simple hypothetical model to demonstrate that if the world permits only a few…

Regression Discontinuity Designs in Epidemiology

PubMed Central

Moscoe, Ellen; Mutevedzi, Portia; Newell, Marie-Louise; Bärnighausen, Till

2014-01-01

When patients receive an intervention based on whether they score below or above some threshold value on a continuously measured random variable, the intervention will be randomly assigned for patients close to the threshold. The regression discontinuity design exploits this fact to estimate causal treatment effects. In spite of its recent proliferation in economics, the regression discontinuity design has not been widely adopted in epidemiology. We describe regression discontinuity, its implementation, and the assumptions required for causal inference. We show that regression discontinuity is generalizable to the survival and nonlinear models that are mainstays of epidemiologic analysis. We then present an application of regression discontinuity to the much-debated epidemiologic question of when to start HIV patients on antiretroviral therapy. Using data from a large South African cohort (2007–2011), we estimate the causal effect of early versus deferred treatment eligibility on mortality. Patients whose first CD4 count was just below the 200 cells/μL CD4 count threshold had a 35% lower hazard of death (hazard ratio = 0.65 [95% confidence interval = 0.45–0.94]) than patients presenting with CD4 counts just above the threshold. We close by discussing the strengths and limitations of regression discontinuity designs for epidemiology. PMID:25061922

Genetic Obesity and the Risk of Atrial Fibrillation – Causal Estimates from Mendelian Randomization

PubMed Central

Chatterjee, Neal A.; Arking, Dan E.; Ellinor, Patrick T.; Heeringa, Jan; Lin, Honghuang; Lubitz, Steven A.; Soliman, Elsayed Z.; Verweij, Niek; Alonso, Alvaro; Benjamin, Emelia J.; Gudnason, Vilmundur; Stricker, Bruno H. C.; Van Der Harst, Pim; Chasman, Daniel I.; Albert, Christine M.

2017-01-01

Background Observational studies have identified an association between body mass index (BMI) and incident atrial fibrillation (AF). Inferring causality from observational studies, however, is subject to residual confounding, reverse causation, and bias. The primary objective of this study was to evaluate the causal association between BMI and AF using genetic predictors of BMI. Methods We identified 51 646 individuals of European ancestry without AF at baseline from seven prospective population-based cohorts initiated between 1987 and 2002 in the United States, Iceland, and the Netherlands with incident AF ascertained between 1987 and 2012. Cohort-specific mean follow-up ranged 7.4 to 19.2 years, over which period there were a total of 4178 cases of incident AF. We performed a Mendelian randomization with instrumental variable analysis to estimate a cohort-specific causal hazard ratio for the association between BMI and AF. Two genetic instruments for BMI were utilized: FTO genotype (rs1558902) and a BMI gene score comprised of 39 single nucleotide polymorphisms identified by genome-wide association studies to be associated with BMI. Cohort-specific estimates were combined by random-effects, inverse variance weighted meta-analysis. Results In age- and sex-adjusted meta-analysis, both genetic instruments were significantly associated with BMI (FTO: 0.43 [95% CI: 0.32 – 0.54] kg/m2 per A-allele, p<0.001); BMI gene score: 1.05 [95% CI: 0.90-1.20] kg/m2 per 1 unit increase, p<0.001) and incident AF (FTO – HR: 1.07 [1.02-1.11] per A-allele, p=0.004; BMI gene score – HR: 1.11 [1.05-1.18] per 1-unit increase, p<0.001). Age- and sex-adjusted instrumental variable estimates for the causal association between BMI and incident AF were HR 1.15 [1.04-1.26] per kg/m2, p=0.005 (FTO) and 1.11 [1.05-1.17] per kg/m2, p<0.001 (BMI gene score). Both of these estimates were consistent with the meta-analyzed estimate between observed BMI and AF (age- and sex-adjusted HR 1.05 [1.04-1.06] per kg/m2, p<0.001). Multivariable adjustment did not significantly change findings. Conclusions Our data are consistent with a causal relationship between BMI and incident AF. These data support the possibility that public health initiatives targeting primordial prevention of obesity may reduce the incidence of AF. PMID:27974350

Assessing the causal role of adiposity on disordered eating in childhood, adolescence, and adulthood: a Mendelian randomization analysis

PubMed Central

2017-01-01

Background: Observational studies have shown that higher body mass index (BMI) is associated with increased risk of developing disordered eating patterns. However, the causal direction of this relation remains ambiguous. Objective: We used Mendelian randomization (MR) to infer the direction of causality between BMI and disordered eating in childhood, adolescence, and adulthood. Design: MR analyses were conducted with a genetic score as an instrumental variable for BMI to assess the causal effect of BMI at age 7 y on disordered eating patterns at age 13 y with the use of data from the Avon Longitudinal Study of Parents and Children (ALSPAC) (n = 4473). To examine causality in the reverse direction, MR analyses were used to estimate the effect of the same disordered eating patterns at age 13 y on BMI at age 17 y via a split-sample approach in the ALSPAC. We also investigated the causal direction of the association between BMI and eating disorders (EDs) in adults via a two-sample MR approach and publically available genome-wide association study data. Results: MR results indicated that higher BMI at age 7 y likely causes higher levels of binge eating and overeating, weight and shape concerns, and weight-control behavior patterns in both males and females and food restriction in males at age 13 y. Furthermore, results suggested that higher levels of binge eating and overeating in males at age 13 y likely cause higher BMI at age 17 y. We showed no evidence of causality between BMI and EDs in adulthood in either direction. Conclusions: This study provides evidence to suggest a causal effect of higher BMI in childhood and increased risk of disordered eating at age 13 y. Furthermore, higher levels of binge eating and overeating may cause higher BMI in later life. These results encourage an exploration of the ways to break the causal chain between these complex phenotypes, which could inform and prevent disordered eating problems in adolescence. PMID:28747331

Assessing the causal role of adiposity on disordered eating in childhood, adolescence, and adulthood: a Mendelian randomization analysis.

PubMed

Reed, Zoe E; Micali, Nadia; Bulik, Cynthia M; Davey Smith, George; Wade, Kaitlin H

2017-09-01

Background: Observational studies have shown that higher body mass index (BMI) is associated with increased risk of developing disordered eating patterns. However, the causal direction of this relation remains ambiguous. Objective: We used Mendelian randomization (MR) to infer the direction of causality between BMI and disordered eating in childhood, adolescence, and adulthood. Design: MR analyses were conducted with a genetic score as an instrumental variable for BMI to assess the causal effect of BMI at age 7 y on disordered eating patterns at age 13 y with the use of data from the Avon Longitudinal Study of Parents and Children (ALSPAC) ( n = 4473). To examine causality in the reverse direction, MR analyses were used to estimate the effect of the same disordered eating patterns at age 13 y on BMI at age 17 y via a split-sample approach in the ALSPAC. We also investigated the causal direction of the association between BMI and eating disorders (EDs) in adults via a two-sample MR approach and publically available genome-wide association study data. Results: MR results indicated that higher BMI at age 7 y likely causes higher levels of binge eating and overeating, weight and shape concerns, and weight-control behavior patterns in both males and females and food restriction in males at age 13 y. Furthermore, results suggested that higher levels of binge eating and overeating in males at age 13 y likely cause higher BMI at age 17 y. We showed no evidence of causality between BMI and EDs in adulthood in either direction. Conclusions: This study provides evidence to suggest a causal effect of higher BMI in childhood and increased risk of disordered eating at age 13 y. Furthermore, higher levels of binge eating and overeating may cause higher BMI in later life. These results encourage an exploration of the ways to break the causal chain between these complex phenotypes, which could inform and prevent disordered eating problems in adolescence.

Plasma urate concentration and risk of coronary heart disease: a Mendelian randomisation analysis.

PubMed

White, Jon; Sofat, Reecha; Hemani, Gibran; Shah, Tina; Engmann, Jorgen; Dale, Caroline; Shah, Sonia; Kruger, Felix A; Giambartolomei, Claudia; Swerdlow, Daniel I; Palmer, Tom; McLachlan, Stela; Langenberg, Claudia; Zabaneh, Delilah; Lovering, Ruth; Cavadino, Alana; Jefferis, Barbara; Finan, Chris; Wong, Andrew; Amuzu, Antoinette; Ong, Ken; Gaunt, Tom R; Warren, Helen; Davies, Teri-Louise; Drenos, Fotios; Cooper, Jackie; Ebrahim, Shah; Lawlor, Debbie A; Talmud, Philippa J; Humphries, Steve E; Power, Christine; Hypponen, Elina; Richards, Marcus; Hardy, Rebecca; Kuh, Diana; Wareham, Nicholas; Ben-Shlomo, Yoav; Day, Ian N; Whincup, Peter; Morris, Richard; Strachan, Mark W J; Price, Jacqueline; Kumari, Meena; Kivimaki, Mika; Plagnol, Vincent; Whittaker, John C; Smith, George Davey; Dudbridge, Frank; Casas, Juan P; Holmes, Michael V; Hingorani, Aroon D

2016-04-01

Increased circulating plasma urate concentration is associated with an increased risk of coronary heart disease, but the extent of any causative effect of urate on risk of coronary heart disease is still unclear. In this study, we aimed to clarify any causal role of urate on coronary heart disease risk using Mendelian randomisation analysis. We first did a fixed-effects meta-analysis of the observational association of plasma urate and risk of coronary heart disease. We then used a conventional Mendelian randomisation approach to investigate the causal relevance using a genetic instrument based on 31 urate-associated single nucleotide polymorphisms (SNPs). To account for potential pleiotropic associations of certain SNPs with risk factors other than urate, we additionally did both a multivariable Mendelian randomisation analysis, in which the genetic associations of SNPs with systolic and diastolic blood pressure, HDL cholesterol, and triglycerides were included as covariates, and an Egger Mendelian randomisation (MR-Egger) analysis to estimate a causal effect accounting for unmeasured pleiotropy. In the meta-analysis of 17 prospective observational studies (166 486 individuals; 9784 coronary heart disease events) a 1 SD higher urate concentration was associated with an odds ratio (OR) for coronary heart disease of 1·07 (95% CI 1·04-1·10). The corresponding OR estimates from the conventional, multivariable adjusted, and Egger Mendelian randomisation analysis (58 studies; 198 598 individuals; 65 877 events) were 1·18 (95% CI 1·08-1·29), 1·10 (1·00-1·22), and 1·05 (0·92-1·20), respectively, per 1 SD increment in plasma urate. Conventional and multivariate Mendelian randomisation analysis implicates a causal role for urate in the development of coronary heart disease, but these estimates might be inflated by hidden pleiotropy. Egger Mendelian randomisation analysis, which accounts for pleiotropy but has less statistical power, suggests there might be no causal effect. These results might help investigators to determine the priority of trials of urate lowering for the prevention of coronary heart disease compared with other potential interventions. UK National Institute for Health Research, British Heart Foundation, and UK Medical Research Council. Copyright © 2016 White et al. Open Access article distributed under the terms of CC BY. Published by Elsevier Ltd.. All rights reserved.

Plasma urate concentration and risk of coronary heart disease: a Mendelian randomisation analysis

PubMed Central

White, Jon; Sofat, Reecha; Hemani, Gibran; Shah, Tina; Engmann, Jorgen; Dale, Caroline; Shah, Sonia; Kruger, Felix A; Giambartolomei, Claudia; Swerdlow, Daniel I; Palmer, Tom; McLachlan, Stela; Langenberg, Claudia; Zabaneh, Delilah; Lovering, Ruth; Cavadino, Alana; Jefferis, Barbara; Finan, Chris; Wong, Andrew; Amuzu, Antoinette; Ong, Ken; Gaunt, Tom R; Warren, Helen; Davies, Teri-Louise; Drenos, Fotios; Cooper, Jackie; Ebrahim, Shah; Lawlor, Debbie A; Talmud, Philippa J; Humphries, Steve E; Power, Christine; Hypponen, Elina; Richards, Marcus; Hardy, Rebecca; Kuh, Diana; Wareham, Nicholas; Ben-Shlomo, Yoav; Day, Ian N; Whincup, Peter; Morris, Richard; Strachan, Mark W J; Price, Jacqueline; Kumari, Meena; Kivimaki, Mika; Plagnol, Vincent; Whittaker, John C; Smith, George Davey; Dudbridge, Frank; Casas, Juan P; Holmes, Michael V; Hingorani, Aroon D

2016-01-01

Summary Background Increased circulating plasma urate concentration is associated with an increased risk of coronary heart disease, but the extent of any causative effect of urate on risk of coronary heart disease is still unclear. In this study, we aimed to clarify any causal role of urate on coronary heart disease risk using Mendelian randomisation analysis. Methods We first did a fixed-effects meta-analysis of the observational association of plasma urate and risk of coronary heart disease. We then used a conventional Mendelian randomisation approach to investigate the causal relevance using a genetic instrument based on 31 urate-associated single nucleotide polymorphisms (SNPs). To account for potential pleiotropic associations of certain SNPs with risk factors other than urate, we additionally did both a multivariable Mendelian randomisation analysis, in which the genetic associations of SNPs with systolic and diastolic blood pressure, HDL cholesterol, and triglycerides were included as covariates, and an Egger Mendelian randomisation (MR-Egger) analysis to estimate a causal effect accounting for unmeasured pleiotropy. Findings In the meta-analysis of 17 prospective observational studies (166 486 individuals; 9784 coronary heart disease events) a 1 SD higher urate concentration was associated with an odds ratio (OR) for coronary heart disease of 1·07 (95% CI 1·04–1·10). The corresponding OR estimates from the conventional, multivariable adjusted, and Egger Mendelian randomisation analysis (58 studies; 198 598 individuals; 65 877 events) were 1·18 (95% CI 1·08–1·29), 1·10 (1·00–1·22), and 1·05 (0·92–1·20), respectively, per 1 SD increment in plasma urate. Interpretation Conventional and multivariate Mendelian randomisation analysis implicates a causal role for urate in the development of coronary heart disease, but these estimates might be inflated by hidden pleiotropy. Egger Mendelian randomisation analysis, which accounts for pleiotropy but has less statistical power, suggests there might be no causal effect. These results might help investigators to determine the priority of trials of urate lowering for the prevention of coronary heart disease compared with other potential interventions. Funding UK National Institute for Health Research, British Heart Foundation, and UK Medical Research Council. PMID:26781229

Identifying Causal Variants at Loci with Multiple Signals of Association

PubMed Central

Hormozdiari, Farhad; Kostem, Emrah; Kang, Eun Yong; Pasaniuc, Bogdan; Eskin, Eleazar

2014-01-01

Although genome-wide association studies have successfully identified thousands of risk loci for complex traits, only a handful of the biologically causal variants, responsible for association at these loci, have been successfully identified. Current statistical methods for identifying causal variants at risk loci either use the strength of the association signal in an iterative conditioning framework or estimate probabilities for variants to be causal. A main drawback of existing methods is that they rely on the simplifying assumption of a single causal variant at each risk locus, which is typically invalid at many risk loci. In this work, we propose a new statistical framework that allows for the possibility of an arbitrary number of causal variants when estimating the posterior probability of a variant being causal. A direct benefit of our approach is that we predict a set of variants for each locus that under reasonable assumptions will contain all of the true causal variants with a high confidence level (e.g., 95%) even when the locus contains multiple causal variants. We use simulations to show that our approach provides 20–50% improvement in our ability to identify the causal variants compared to the existing methods at loci harboring multiple causal variants. We validate our approach using empirical data from an expression QTL study of CHI3L2 to identify new causal variants that affect gene expression at this locus. CAVIAR is publicly available online at http://genetics.cs.ucla.edu/caviar/. PMID:25104515

Identifying causal variants at loci with multiple signals of association.

PubMed

Hormozdiari, Farhad; Kostem, Emrah; Kang, Eun Yong; Pasaniuc, Bogdan; Eskin, Eleazar

2014-10-01

Although genome-wide association studies have successfully identified thousands of risk loci for complex traits, only a handful of the biologically causal variants, responsible for association at these loci, have been successfully identified. Current statistical methods for identifying causal variants at risk loci either use the strength of the association signal in an iterative conditioning framework or estimate probabilities for variants to be causal. A main drawback of existing methods is that they rely on the simplifying assumption of a single causal variant at each risk locus, which is typically invalid at many risk loci. In this work, we propose a new statistical framework that allows for the possibility of an arbitrary number of causal variants when estimating the posterior probability of a variant being causal. A direct benefit of our approach is that we predict a set of variants for each locus that under reasonable assumptions will contain all of the true causal variants with a high confidence level (e.g., 95%) even when the locus contains multiple causal variants. We use simulations to show that our approach provides 20-50% improvement in our ability to identify the causal variants compared to the existing methods at loci harboring multiple causal variants. We validate our approach using empirical data from an expression QTL study of CHI3L2 to identify new causal variants that affect gene expression at this locus. CAVIAR is publicly available online at http://genetics.cs.ucla.edu/caviar/. Copyright © 2014 by the Genetics Society of America.

On the role of the entorhinal cortex in the effective connectivity of the hippocampal formation.

PubMed

López-Madrona, Víctor J; Matias, Fernanda S; Pereda, Ernesto; Canals, Santiago; Mirasso, Claudio R

2017-04-01

Inferring effective connectivity from neurophysiological data is a challenging task. In particular, only a finite (and usually small) number of sites are simultaneously recorded, while the response of one of these sites can be influenced by other sites that are not being recorded. In the hippocampal formation, for instance, the connections between areas CA1-CA3, the dentate gyrus (DG), and the entorhinal cortex (EC) are well established. However, little is known about the relations within the EC layers, which might strongly affect the resulting effective connectivity estimations. In this work, we build excitatory/inhibitory neuronal populations representing the four areas CA1, CA3, the DG, and the EC and fix their connectivities. We model the EC by three layers (LII, LIII, and LV) and assume any possible connection between them. Our results, based on Granger Causality (GC) and Partial Transfer Entropy (PTE) measurements, reveal that the estimation of effective connectivity in the hippocampus strongly depends on the connectivities between EC layers. Moreover, we find, for certain EC configurations, very different results when comparing GC and PTE measurements. We further demonstrate that causal links can be robustly inferred regardless of the excitatory or inhibitory nature of the connection, adding complexity to their interpretation. Overall, our work highlights the importance of a careful analysis of the connectivity methods to prevent unrealistic conclusions when only partial information about the experimental system is available, as usually happens in brain networks. Our results suggest that the combination of causality measures with neuronal modeling based on precise neuroanatomical tracing may provide a powerful framework to disambiguate causal interactions in the brain.

ASSESSING CAUSALITY AND PERSISTENCE IN ASSOCIATIONS BETWEEN FAMILY DINNERS AND ADOLESCENT WELL-BEING

PubMed Central

Musick, Kelly; Meier, Ann

2013-01-01

Adolescents who share meals with their parents score better on a range of well-being indicators. Using three waves of the National Longitudinal Survey of Adolescent Health (N = 17,977), we assessed the causal nature of these associations and the extent to which they persist into adulthood. We examined links between family dinners and adolescent mental health, substance use, and delinquency at wave 1, accounting for detailed measures of the family environment to test whether family meals simply proxy for other family processes. As a more stringent test of causality, we estimated fixed effects models from waves 1 and 2, and we used wave 3 to explore persistence in the influence of family dinners. Associations between family dinners and adolescent well-being remained significant, net of controls, and some held up to stricter tests of causality. Beyond indirect benefits via earlier well-being, however, family dinners associations did not persist into adulthood. PMID:23794750

ASSESSING CAUSALITY AND PERSISTENCE IN ASSOCIATIONS BETWEEN FAMILY DINNERS AND ADOLESCENT WELL-BEING.

PubMed

Musick, Kelly; Meier, Ann

2012-06-01

Adolescents who share meals with their parents score better on a range of well-being indicators. Using three waves of the National Longitudinal Survey of Adolescent Health ( N = 17,977), we assessed the causal nature of these associations and the extent to which they persist into adulthood. We examined links between family dinners and adolescent mental health, substance use, and delinquency at wave 1, accounting for detailed measures of the family environment to test whether family meals simply proxy for other family processes. As a more stringent test of causality, we estimated fixed effects models from waves 1 and 2, and we used wave 3 to explore persistence in the influence of family dinners. Associations between family dinners and adolescent well-being remained significant, net of controls, and some held up to stricter tests of causality. Beyond indirect benefits via earlier well-being, however, family dinners associations did not persist into adulthood.

Confounding in statistical mediation analysis: What it is and how to address it.

PubMed

Valente, Matthew J; Pelham, William E; Smyth, Heather; MacKinnon, David P

2017-11-01

Psychology researchers are often interested in mechanisms underlying how randomized interventions affect outcomes such as substance use and mental health. Mediation analysis is a common statistical method for investigating psychological mechanisms that has benefited from exciting new methodological improvements over the last 2 decades. One of the most important new developments is methodology for estimating causal mediated effects using the potential outcomes framework for causal inference. Potential outcomes-based methods developed in epidemiology and statistics have important implications for understanding psychological mechanisms. We aim to provide a concise introduction to and illustration of these new methods and emphasize the importance of confounder adjustment. First, we review the traditional regression approach for estimating mediated effects. Second, we describe the potential outcomes framework. Third, we define what a confounder is and how the presence of a confounder can provide misleading evidence regarding mechanisms of interventions. Fourth, we describe experimental designs that can help rule out confounder bias. Fifth, we describe new statistical approaches to adjust for measured confounders of the mediator-outcome relation and sensitivity analyses to probe effects of unmeasured confounders on the mediated effect. All approaches are illustrated with application to a real counseling intervention dataset. Counseling psychologists interested in understanding the causal mechanisms of their interventions can benefit from incorporating the most up-to-date techniques into their mediation analyses. (PsycINFO Database Record (c) 2017 APA, all rights reserved).

Effect Size Measures for Mediation Models: Quantitative Strategies for Communicating Indirect Effects

ERIC Educational Resources Information Center

Preacher, Kristopher J.; Kelley, Ken

2011-01-01

The statistical analysis of mediation effects has become an indispensable tool for helping scientists investigate processes thought to be causal. Yet, in spite of many recent advances in the estimation and testing of mediation effects, little attention has been given to methods for communicating effect size and the practical importance of those…

Essays on Economics of Education

ERIC Educational Resources Information Center

Kim, Bo Min

2013-01-01

This dissertation analyzes the effects of the educational programs for students who need special care in secondary and postsecondary school. These educational programs present serious endogeneity problems to a researcher estimating their causal effects, because most of students tried to avoid these programs. I extend the current literature in the…

A Difference-in-Differences Approach to Assess the Effect of a Heat Action Plan on Heat-Related Mortality, and Differences in Effectiveness According to Sex, Age, and Socioeconomic Status (Montreal, Quebec).

PubMed

Benmarhnia, Tarik; Bailey, Zinzi; Kaiser, David; Auger, Nathalie; King, Nicholas; Kaufman, Jay S

2016-11-01

The impact of heat waves on mortality and health inequalities is well documented. Very few studies have assessed the effectiveness of heat action plans (HAPs) on health, and none has used quasi-experimental methods to estimate causal effects of such programs. We developed a quasi-experimental method to estimate the causal effects associated with HAPs that allows the identification of heterogeneity across subpopulations, and to apply this method specifically to the case of the Montreal (Quebec, Canada) HAP. A difference-in-differences approach was undertaken using Montreal death registry data for the summers of 2000-2007 to assess the effectiveness of the Montreal HAP, implemented in 2004, on mortality. To study equity in the effect of HAP implementation, we assessed whether the program effects were heterogeneous across sex (male vs. female), age (≥ 65 years vs. < 65 years), and neighborhood education levels (first vs. third tertile). We conducted sensitivity analyses to assess the validity of the estimated causal effect of the HAP program. We found evidence that the HAP contributed to reducing mortality on hot days, and that the mortality reduction attributable to the program was greater for elderly people and people living in low-education neighborhoods. These findings show promise for programs aimed at reducing the impact of extreme temperatures and health inequities. We propose a new quasi-experimental approach that can be easily applied to evaluate the impact of any program or intervention triggered when daily thresholds are reached. Citation: Benmarhnia T, Bailey Z, Kaiser D, Auger N, King N, Kaufman J. 2016. A difference-in-differences approach to assess the effect of a heat action plan on heat-related mortality, and differences in effectiveness according to sex, age, and socioeconomic status (Montreal, Quebec). Environ Health Perspect 124:1694-1699; http://dx.doi.org/10.1289/EHP203.

Methodological Considerations in Estimation of Phenotype Heritability Using Genome-Wide SNP Data, Illustrated by an Analysis of the Heritability of Height in a Large Sample of African Ancestry Adults

PubMed Central

Chen, Fang; He, Jing; Zhang, Jianqi; Chen, Gary K.; Thomas, Venetta; Ambrosone, Christine B.; Bandera, Elisa V.; Berndt, Sonja I.; Bernstein, Leslie; Blot, William J.; Cai, Qiuyin; Carpten, John; Casey, Graham; Chanock, Stephen J.; Cheng, Iona; Chu, Lisa; Deming, Sandra L.; Driver, W. Ryan; Goodman, Phyllis; Hayes, Richard B.; Hennis, Anselm J. M.; Hsing, Ann W.; Hu, Jennifer J.; Ingles, Sue A.; John, Esther M.; Kittles, Rick A.; Kolb, Suzanne; Leske, M. Cristina; Monroe, Kristine R.; Murphy, Adam; Nemesure, Barbara; Neslund-Dudas, Christine; Nyante, Sarah; Ostrander, Elaine A; Press, Michael F.; Rodriguez-Gil, Jorge L.; Rybicki, Ben A.; Schumacher, Fredrick; Stanford, Janet L.; Signorello, Lisa B.; Strom, Sara S.; Stevens, Victoria; Van Den Berg, David; Wang, Zhaoming; Witte, John S.; Wu, Suh-Yuh; Yamamura, Yuko; Zheng, Wei; Ziegler, Regina G.; Stram, Alexander H.; Kolonel, Laurence N.; Marchand, Loïc Le; Henderson, Brian E.; Haiman, Christopher A.; Stram, Daniel O.

2015-01-01

Height has an extremely polygenic pattern of inheritance. Genome-wide association studies (GWAS) have revealed hundreds of common variants that are associated with human height at genome-wide levels of significance. However, only a small fraction of phenotypic variation can be explained by the aggregate of these common variants. In a large study of African-American men and women (n = 14,419), we genotyped and analyzed 966,578 autosomal SNPs across the entire genome using a linear mixed model variance components approach implemented in the program GCTA (Yang et al Nat Genet 2010), and estimated an additive heritability of 44.7% (se: 3.7%) for this phenotype in a sample of evidently unrelated individuals. While this estimated value is similar to that given by Yang et al in their analyses, we remain concerned about two related issues: (1) whether in the complete absence of hidden relatedness, variance components methods have adequate power to estimate heritability when a very large number of SNPs are used in the analysis; and (2) whether estimation of heritability may be biased, in real studies, by low levels of residual hidden relatedness. We addressed the first question in a semi-analytic fashion by directly simulating the distribution of the score statistic for a test of zero heritability with and without low levels of relatedness. The second question was addressed by a very careful comparison of the behavior of estimated heritability for both observed (self-reported) height and simulated phenotypes compared to imputation R2 as a function of the number of SNPs used in the analysis. These simulations help to address the important question about whether today's GWAS SNPs will remain useful for imputing causal variants that are discovered using very large sample sizes in future studies of height, or whether the causal variants themselves will need to be genotyped de novo in order to build a prediction model that ultimately captures a large fraction of the variability of height, and by implication other complex phenotypes. Our overall conclusions are that when study sizes are quite large (5,000 or so) the additive heritability estimate for height is not apparently biased upwards using the linear mixed model; however there is evidence in our simulation that a very large number of causal variants (many thousands) each with very small effect on phenotypic variance will need to be discovered to fill the gap between the heritability explained by known versus unknown causal variants. We conclude that today's GWAS data will remain useful in the future for causal variant prediction, but that finding the causal variants that need to be predicted may be extremely laborious. PMID:26125186

Methodological Considerations in Estimation of Phenotype Heritability Using Genome-Wide SNP Data, Illustrated by an Analysis of the Heritability of Height in a Large Sample of African Ancestry Adults.

PubMed

Chen, Fang; He, Jing; Zhang, Jianqi; Chen, Gary K; Thomas, Venetta; Ambrosone, Christine B; Bandera, Elisa V; Berndt, Sonja I; Bernstein, Leslie; Blot, William J; Cai, Qiuyin; Carpten, John; Casey, Graham; Chanock, Stephen J; Cheng, Iona; Chu, Lisa; Deming, Sandra L; Driver, W Ryan; Goodman, Phyllis; Hayes, Richard B; Hennis, Anselm J M; Hsing, Ann W; Hu, Jennifer J; Ingles, Sue A; John, Esther M; Kittles, Rick A; Kolb, Suzanne; Leske, M Cristina; Millikan, Robert C; Monroe, Kristine R; Murphy, Adam; Nemesure, Barbara; Neslund-Dudas, Christine; Nyante, Sarah; Ostrander, Elaine A; Press, Michael F; Rodriguez-Gil, Jorge L; Rybicki, Ben A; Schumacher, Fredrick; Stanford, Janet L; Signorello, Lisa B; Strom, Sara S; Stevens, Victoria; Van Den Berg, David; Wang, Zhaoming; Witte, John S; Wu, Suh-Yuh; Yamamura, Yuko; Zheng, Wei; Ziegler, Regina G; Stram, Alexander H; Kolonel, Laurence N; Le Marchand, Loïc; Henderson, Brian E; Haiman, Christopher A; Stram, Daniel O

2015-01-01

Height has an extremely polygenic pattern of inheritance. Genome-wide association studies (GWAS) have revealed hundreds of common variants that are associated with human height at genome-wide levels of significance. However, only a small fraction of phenotypic variation can be explained by the aggregate of these common variants. In a large study of African-American men and women (n = 14,419), we genotyped and analyzed 966,578 autosomal SNPs across the entire genome using a linear mixed model variance components approach implemented in the program GCTA (Yang et al Nat Genet 2010), and estimated an additive heritability of 44.7% (se: 3.7%) for this phenotype in a sample of evidently unrelated individuals. While this estimated value is similar to that given by Yang et al in their analyses, we remain concerned about two related issues: (1) whether in the complete absence of hidden relatedness, variance components methods have adequate power to estimate heritability when a very large number of SNPs are used in the analysis; and (2) whether estimation of heritability may be biased, in real studies, by low levels of residual hidden relatedness. We addressed the first question in a semi-analytic fashion by directly simulating the distribution of the score statistic for a test of zero heritability with and without low levels of relatedness. The second question was addressed by a very careful comparison of the behavior of estimated heritability for both observed (self-reported) height and simulated phenotypes compared to imputation R2 as a function of the number of SNPs used in the analysis. These simulations help to address the important question about whether today's GWAS SNPs will remain useful for imputing causal variants that are discovered using very large sample sizes in future studies of height, or whether the causal variants themselves will need to be genotyped de novo in order to build a prediction model that ultimately captures a large fraction of the variability of height, and by implication other complex phenotypes. Our overall conclusions are that when study sizes are quite large (5,000 or so) the additive heritability estimate for height is not apparently biased upwards using the linear mixed model; however there is evidence in our simulation that a very large number of causal variants (many thousands) each with very small effect on phenotypic variance will need to be discovered to fill the gap between the heritability explained by known versus unknown causal variants. We conclude that today's GWAS data will remain useful in the future for causal variant prediction, but that finding the causal variants that need to be predicted may be extremely laborious.

The search for causal inferences: using propensity scores post hoc to reduce estimation error with nonexperimental research.

PubMed

Tumlinson, Samuel E; Sass, Daniel A; Cano, Stephanie M

2014-03-01

While experimental designs are regarded as the gold standard for establishing causal relationships, such designs are usually impractical owing to common methodological limitations. The objective of this article is to illustrate how propensity score matching (PSM) and using propensity scores (PS) as a covariate are viable alternatives to reduce estimation error when experimental designs cannot be implemented. To mimic common pediatric research practices, data from 140 simulated participants were used to resemble an experimental and nonexperimental design that assessed the effect of treatment status on participant weight loss for diabetes. Pretreatment participant characteristics (age, gender, physical activity, etc.) were then used to generate PS for use in the various statistical approaches. Results demonstrate how PSM and using the PS as a covariate can be used to reduce estimation error and improve statistical inferences. References for issues related to the implementation of these procedures are provided to assist researchers.

Estimation of treatment efficacy with complier average causal effects (CACE) in a randomized stepped wedge trial.

PubMed

Gruber, Joshua S; Arnold, Benjamin F; Reygadas, Fermin; Hubbard, Alan E; Colford, John M

2014-05-01

Complier average causal effects (CACE) estimate the impact of an intervention among treatment compliers in randomized trials. Methods used to estimate CACE have been outlined for parallel-arm trials (e.g., using an instrumental variables (IV) estimator) but not for other randomized study designs. Here, we propose a method for estimating CACE in randomized stepped wedge trials, where experimental units cross over from control conditions to intervention conditions in a randomized sequence. We illustrate the approach with a cluster-randomized drinking water trial conducted in rural Mexico from 2009 to 2011. Additionally, we evaluated the plausibility of assumptions required to estimate CACE using the IV approach, which are testable in stepped wedge trials but not in parallel-arm trials. We observed small increases in the magnitude of CACE risk differences compared with intention-to-treat estimates for drinking water contamination (risk difference (RD) = -22% (95% confidence interval (CI): -33, -11) vs. RD = -19% (95% CI: -26, -12)) and diarrhea (RD = -0.8% (95% CI: -2.1, 0.4) vs. RD = -0.1% (95% CI: -1.1, 0.9)). Assumptions required for IV analysis were probably violated. Stepped wedge trials allow investigators to estimate CACE with an approach that avoids the stronger assumptions required for CACE estimation in parallel-arm trials. Inclusion of CACE estimates in stepped wedge trials with imperfect compliance could enhance reporting and interpretation of the results of such trials.

What matters most: quantifying an epidemiology of consequence.

PubMed

Keyes, Katherine; Galea, Sandro

2015-05-01

Risk factor epidemiology has contributed to substantial public health success. In this essay, we argue, however, that the focus on risk factor epidemiology has led epidemiology to ever increasing focus on the estimation of precise causal effects of exposures on an outcome at the expense of engagement with the broader causal architecture that produces population health. To conduct an epidemiology of consequence, a systematic effort is needed to engage our science in a critical reflection both about how well and under what conditions or assumptions we can assess causal effects and also on what will truly matter most for changing population health. Such an approach changes the priorities and values of the discipline and requires reorientation of how we structure the questions we ask and the methods we use, as well as how we teach epidemiology to our emerging scholars. Copyright © 2015 Elsevier Inc. All rights reserved.

A Causal Model for Joint Evaluation of Placebo and Treatment-Specific Effects in Clinical Trials

PubMed Central

Zhang, Zhiwei; Kotz, Richard M.; Wang, Chenguang; Ruan, Shiling; Ho, Martin

2014-01-01

Summary Evaluation of medical treatments is frequently complicated by the presence of substantial placebo effects, especially on relatively subjective endpoints, and the standard solution to this problem is a randomized, double-blinded, placebo-controlled clinical trial. However, effective blinding does not guarantee that all patients have the same belief or mentality about which treatment they have received (or treatmentality, for brevity), making it difficult to interpret the usual intent-to-treat effect as a causal effect. We discuss the causal relationships among treatment, treatmentality and the clinical outcome of interest, and propose a causal model for joint evaluation of placebo and treatment-specific effects. The model highlights the importance of measuring and incorporating patient treatmentality and suggests that each treatment group should be considered a separate observational study with a patient's treatmentality playing the role of an uncontrolled exposure. This perspective allows us to adapt existing methods for dealing with confounding to joint estimation of placebo and treatment-specific effects using measured treatmentality data, commonly known as blinding assessment data. We first apply this approach to the most common type of blinding assessment data, which is categorical, and illustrate the methods using an example from asthma. We then propose that blinding assessment data can be collected as a continuous variable, specifically when a patient's treatmentality is measured as a subjective probability, and describe analytic methods for that case. PMID:23432119

Causal Inference in Retrospective Studies.

ERIC Educational Resources Information Center

Holland, Paul W.; Rubin, Donald B.

1988-01-01

The problem of drawing causal inferences from retrospective case-controlled studies is considered. A model for causal inference in prospective studies is applied to retrospective studies. Limitations of case-controlled studies are formulated concerning relevant parameters that can be estimated in such studies. A coffee-drinking/myocardial…

Modeling the Perception of Audiovisual Distance: Bayesian Causal Inference and Other Models

PubMed Central

2016-01-01

Studies of audiovisual perception of distance are rare. Here, visual and auditory cue interactions in distance are tested against several multisensory models, including a modified causal inference model. In this causal inference model predictions of estimate distributions are included. In our study, the audiovisual perception of distance was overall better explained by Bayesian causal inference than by other traditional models, such as sensory dominance and mandatory integration, and no interaction. Causal inference resolved with probability matching yielded the best fit to the data. Finally, we propose that sensory weights can also be estimated from causal inference. The analysis of the sensory weights allows us to obtain windows within which there is an interaction between the audiovisual stimuli. We find that the visual stimulus always contributes by more than 80% to the perception of visual distance. The visual stimulus also contributes by more than 50% to the perception of auditory distance, but only within a mobile window of interaction, which ranges from 1 to 4 m. PMID:27959919

How Financial Literacy Affects Household Wealth Accumulation.

PubMed

Behrman, Jere R; Mitchell, Olivia S; Soo, Cindy K; Bravo, David

2012-05-01

This study isolates the causal effects of financial literacy and schooling on wealth accumulation using a new household dataset and an instrumental variables (IV) approach. Financial literacy and schooling attainment are both strongly positively associated with wealth outcomes in linear regression models, whereas the IV estimates reveal even more potent effects of financial literacy. They also indicate that the schooling effect only becomes positive when interacted with financial literacy. Estimated impacts are substantial enough to imply that investments in financial literacy could have large wealth payoffs.

How Financial Literacy Affects Household Wealth Accumulation

PubMed Central

Behrman, Jere R.; Mitchell, Olivia S.; Soo, Cindy K.; Bravo, David

2012-01-01

This study isolates the causal effects of financial literacy and schooling on wealth accumulation using a new household dataset and an instrumental variables (IV) approach. Financial literacy and schooling attainment are both strongly positively associated with wealth outcomes in linear regression models, whereas the IV estimates reveal even more potent effects of financial literacy. They also indicate that the schooling effect only becomes positive when interacted with financial literacy. Estimated impacts are substantial enough to imply that investments in financial literacy could have large wealth payoffs. PMID:23355747

Pairwise Measures of Causal Direction in the Epidemiology of Sleep Problems and Depression

PubMed Central

Rosenström, Tom; Jokela, Markus; Puttonen, Sampsa; Hintsanen, Mirka; Pulkki-Råback, Laura; Viikari, Jorma S.; Raitakari, Olli T.; Keltikangas-Järvinen, Liisa

2012-01-01

Depressive mood is often preceded by sleep problems, suggesting that they increase the risk of depression. Sleep problems can also reflect prodromal symptom of depression, thus temporal precedence alone is insufficient to confirm causality. The authors applied recently introduced statistical causal-discovery algorithms that can estimate causality from cross-sectional samples in order to infer the direction of causality between the two sets of symptoms from a novel perspective. Two common-population samples were used; one from the Young Finns study (690 men and 997 women, average age 37.7 years, range 30–45), and another from the Wisconsin Longitudinal study (3101 men and 3539 women, average age 53.1 years, range 52–55). These included three depression questionnaires (two in Young Finns data) and two sleep problem questionnaires. Three different causality estimates were constructed for each data set, tested in a benchmark data with a (practically) known causality, and tested for assumption violations using simulated data. Causality algorithms performed well in the benchmark data and simulations, and a prediction was drawn for future empirical studies to confirm: for minor depression/dysphoria, sleep problems cause significantly more dysphoria than dysphoria causes sleep problems. The situation may change as depression becomes more severe, or more severe levels of symptoms are evaluated; also, artefacts due to severe depression being less well presented in the population data than minor depression may intervene the estimation for depression scales that emphasize severe symptoms. The findings are consistent with other emerging epidemiological and biological evidence. PMID:23226400

Assessing Causality in the Association between Child Adiposity and Physical Activity Levels: A Mendelian Randomization Analysis

PubMed Central

Richmond, Rebecca C.; Davey Smith, George; Ness, Andy R.; den Hoed, Marcel; McMahon, George; Timpson, Nicholas J.

2014-01-01

Background Cross-sectional studies have shown that objectively measured physical activity is associated with childhood adiposity, and a strong inverse dose–response association with body mass index (BMI) has been found. However, few studies have explored the extent to which this association reflects reverse causation. We aimed to determine whether childhood adiposity causally influences levels of physical activity using genetic variants reliably associated with adiposity to estimate causal effects. Methods and Findings The Avon Longitudinal Study of Parents and Children collected data on objectively assessed activity levels of 4,296 children at age 11 y with recorded BMI and genotypic data. We used 32 established genetic correlates of BMI combined in a weighted allelic score as an instrumental variable for adiposity to estimate the causal effect of adiposity on activity. In observational analysis, a 3.3 kg/m2 (one standard deviation) higher BMI was associated with 22.3 (95% CI, 17.0, 27.6) movement counts/min less total physical activity (p = 1.6×10−16), 2.6 (2.1, 3.1) min/d less moderate-to-vigorous-intensity activity (p = 3.7×10−29), and 3.5 (1.5, 5.5) min/d more sedentary time (p = 5.0×10−4). In Mendelian randomization analyses, the same difference in BMI was associated with 32.4 (0.9, 63.9) movement counts/min less total physical activity (p = 0.04) (∼5.3% of the mean counts/minute), 2.8 (0.1, 5.5) min/d less moderate-to-vigorous-intensity activity (p = 0.04), and 13.2 (1.3, 25.2) min/d more sedentary time (p = 0.03). There was no strong evidence for a difference between variable estimates from observational estimates. Similar results were obtained using fat mass index. Low power and poor instrumentation of activity limited causal analysis of the influence of physical activity on BMI. Conclusions Our results suggest that increased adiposity causes a reduction in physical activity in children and support research into the targeting of BMI in efforts to increase childhood activity levels. Importantly, this does not exclude lower physical activity also leading to increased adiposity, i.e., bidirectional causation. Please see later in the article for the Editors' Summary PMID:24642734

Dynamic Granger-Geweke causality modeling with application to interictal spike propagation

PubMed Central

Lin, Fa-Hsuan; Hara, Keiko; Solo, Victor; Vangel, Mark; Belliveau, John W.; Stufflebeam, Steven M.; Hamalainen, Matti S.

2010-01-01

A persistent problem in developing plausible neurophysiological models of perception, cognition, and action is the difficulty of characterizing the interactions between different neural systems. Previous studies have approached this problem by estimating causal influences across brain areas activated during cognitive processing using Structural Equation Modeling and, more recently, with Granger-Geweke causality. While SEM is complicated by the need for a priori directional connectivity information, the temporal resolution of dynamic Granger-Geweke estimates is limited because the underlying autoregressive (AR) models assume stationarity over the period of analysis. We have developed a novel optimal method for obtaining data-driven directional causality estimates with high temporal resolution in both time and frequency domains. This is achieved by simultaneously optimizing the length of the analysis window and the chosen AR model order using the SURE criterion. Dynamic Granger-Geweke causality in time and frequency domains is subsequently calculated within a moving analysis window. We tested our algorithm by calculating the Granger-Geweke causality of epileptic spike propagation from the right frontal lobe to the left frontal lobe. The results quantitatively suggested the epileptic activity at the left frontal lobe was propagated from the right frontal lobe, in agreement with the clinical diagnosis. Our novel computational tool can be used to help elucidate complex directional interactions in the human brain. PMID:19378280

An alternative empirical likelihood method in missing response problems and causal inference.

PubMed

Ren, Kaili; Drummond, Christopher A; Brewster, Pamela S; Haller, Steven T; Tian, Jiang; Cooper, Christopher J; Zhang, Biao

2016-11-30

Missing responses are common problems in medical, social, and economic studies. When responses are missing at random, a complete case data analysis may result in biases. A popular debias method is inverse probability weighting proposed by Horvitz and Thompson. To improve efficiency, Robins et al. proposed an augmented inverse probability weighting method. The augmented inverse probability weighting estimator has a double-robustness property and achieves the semiparametric efficiency lower bound when the regression model and propensity score model are both correctly specified. In this paper, we introduce an empirical likelihood-based estimator as an alternative to Qin and Zhang (2007). Our proposed estimator is also doubly robust and locally efficient. Simulation results show that the proposed estimator has better performance when the propensity score is correctly modeled. Moreover, the proposed method can be applied in the estimation of average treatment effect in observational causal inferences. Finally, we apply our method to an observational study of smoking, using data from the Cardiovascular Outcomes in Renal Atherosclerotic Lesions clinical trial. Copyright © 2016 John Wiley & Sons, Ltd. Copyright © 2016 John Wiley & Sons, Ltd.

Do material, psychosocial and behavioural factors mediate the relationship between disability acquisition and mental health? A sequential causal mediation analysis.

PubMed

Aitken, Zoe; Simpson, Julie Anne; Gurrin, Lyle; Bentley, Rebecca; Kavanagh, Anne Marie

2018-01-29

There is evidence of a causal relationship between disability acquisition and poor mental health; however, the mechanism by which disability affects mental health is poorly understood. This gap in understanding limits the development of effective interventions to improve the mental health of people with disabilities. We used four waves of data from the Household, Income and Labour Dynamics in Australia Survey (2011-14) to compare self-reported mental health between individuals who acquired any disability (n=387) and those who remained disability-free (n=7936). We tested three possible pathways from disability acquisition to mental health, examining the effect of material, psychosocial and behavioural mediators. The effect was partitioned into natural direct and indirect effects through the mediators using a sequential causal mediation analysis approach. Multiple imputation using chained equations was used to assess the impact of missing data. Disability acquisition was estimated to cause a five-point decline in mental health [estimated mean difference: -5.3, 95% confidence interval (CI) -6.8, -3.7]. The indirect effect through material factors was estimated to be a 1.7-point difference (-1.7, 95% CI -2.8, -0.6), explaining 32% of the total effect, with a negligible proportion of the effect explained by the addition of psychosocial characteristics (material and psychosocial: -1.7, 95% CI -3.0, -0.5) and a further 5% by behavioural factors (material-psychosocial-behavioural: -2.0, 95% CI -3.4, -0.6). The finding that the effect of disability acquisition on mental health operates predominantly through material rather than psychosocial and behavioural factors has important implications. The results highlight the need for better social protection, including income support, employment and education opportunities, and affordable housing for people who acquire a disability. © The Author(s) 2018; all rights reserved. Published by Oxford University Press on behalf of the International Epidemiological Association

Quantifying the improvement in sepsis diagnosis, documentation, and coding: the marginal causal effect of year of hospitalization on sepsis diagnosis.

PubMed

Jafarzadeh, S Reza; Thomas, Benjamin S; Marschall, Jonas; Fraser, Victoria J; Gill, Jeff; Warren, David K

2016-01-01

To quantify the coinciding improvement in the clinical diagnosis of sepsis, its documentation in the electronic health records, and subsequent medical coding of sepsis for billing purposes in recent years. We examined 98,267 hospitalizations in 66,208 patients who met systemic inflammatory response syndrome criteria at a tertiary care center from 2008 to 2012. We used g-computation to estimate the causal effect of the year of hospitalization on receiving an International Classification of Diseases, Ninth Revision, Clinical Modification discharge diagnosis code for sepsis by estimating changes in the probability of getting diagnosed and coded for sepsis during the study period. When adjusted for demographics, Charlson-Deyo comorbidity index, blood culture frequency per hospitalization, and intensive care unit admission, the causal risk difference for receiving a discharge code for sepsis per 100 hospitalizations with systemic inflammatory response syndrome, had the hospitalization occurred in 2012, was estimated to be 3.9% (95% confidence interval [CI], 3.8%-4.0%), 3.4% (95% CI, 3.3%-3.5%), 2.2% (95% CI, 2.1%-2.3%), and 0.9% (95% CI, 0.8%-1.1%) from 2008 to 2011, respectively. Patients with similar characteristics and risk factors had a higher of probability of getting diagnosed, documented, and coded for sepsis in 2012 than in previous years, which contributed to an apparent increase in sepsis incidence. Copyright © 2016 Elsevier Inc. All rights reserved.

Dimension reduction of frequency-based direct Granger causality measures on short time series.

PubMed

Siggiridou, Elsa; Kimiskidis, Vasilios K; Kugiumtzis, Dimitris

2017-09-01

The mainstream in the estimation of effective brain connectivity relies on Granger causality measures in the frequency domain. If the measure is meant to capture direct causal effects accounting for the presence of other observed variables, as in multi-channel electroencephalograms (EEG), typically the fit of a vector autoregressive (VAR) model on the multivariate time series is required. For short time series of many variables, the estimation of VAR may not be stable requiring dimension reduction resulting in restricted or sparse VAR models. The restricted VAR obtained by the modified backward-in-time selection method (mBTS) is adapted to the generalized partial directed coherence (GPDC), termed restricted GPDC (RGPDC). Dimension reduction on other frequency based measures, such the direct directed transfer function (dDTF), is straightforward. First, a simulation study using linear stochastic multivariate systems is conducted and RGPDC is favorably compared to GPDC on short time series in terms of sensitivity and specificity. Then the two measures are tested for their ability to detect changes in brain connectivity during an epileptiform discharge (ED) from multi-channel scalp EEG. It is shown that RGPDC identifies better than GPDC the connectivity structure of the simulated systems, as well as changes in the brain connectivity, and is less dependent on the free parameter of VAR order. The proposed dimension reduction in frequency measures based on VAR constitutes an appropriate strategy to estimate reliably brain networks within short-time windows. Copyright © 2017 Elsevier B.V. All rights reserved.

Mediation Analysis with Survival Outcomes: Accelerated Failure Time vs. Proportional Hazards Models.

PubMed

Gelfand, Lois A; MacKinnon, David P; DeRubeis, Robert J; Baraldi, Amanda N

2016-01-01

Survival time is an important type of outcome variable in treatment research. Currently, limited guidance is available regarding performing mediation analyses with survival outcomes, which generally do not have normally distributed errors, and contain unobserved (censored) events. We present considerations for choosing an approach, using a comparison of semi-parametric proportional hazards (PH) and fully parametric accelerated failure time (AFT) approaches for illustration. We compare PH and AFT models and procedures in their integration into mediation models and review their ability to produce coefficients that estimate causal effects. Using simulation studies modeling Weibull-distributed survival times, we compare statistical properties of mediation analyses incorporating PH and AFT approaches (employing SAS procedures PHREG and LIFEREG, respectively) under varied data conditions, some including censoring. A simulated data set illustrates the findings. AFT models integrate more easily than PH models into mediation models. Furthermore, mediation analyses incorporating LIFEREG produce coefficients that can estimate causal effects, and demonstrate superior statistical properties. Censoring introduces bias in the coefficient estimate representing the treatment effect on outcome-underestimation in LIFEREG, and overestimation in PHREG. With LIFEREG, this bias can be addressed using an alternative estimate obtained from combining other coefficients, whereas this is not possible with PHREG. When Weibull assumptions are not violated, there are compelling advantages to using LIFEREG over PHREG for mediation analyses involving survival-time outcomes. Irrespective of the procedures used, the interpretation of coefficients, effects of censoring on coefficient estimates, and statistical properties should be taken into account when reporting results.

Packet Randomized Experiments for Eliminating Classes of Confounders

PubMed Central

Pavela, Greg; Wiener, Howard; Fontaine, Kevin R.; Fields, David A.; Voss, Jameson D.; Allison, David B.

2014-01-01

Background Although randomization is considered essential for causal inference, it is often not possible to randomize in nutrition and obesity research. To address this, we develop a framework for an experimental design—packet randomized experiments (PREs), which improves causal inferences when randomization on a single treatment variable is not possible. This situation arises when subjects are randomly assigned to a condition (such as a new roommate) which varies in one characteristic of interest (such as weight), but also varies across many others. There has been no general discussion of this experimental design, including its strengths, limitations, and statistical properties. As such, researchers are left to develop and apply PREs on an ad hoc basis, limiting its potential to improve causal inferences among nutrition and obesity researchers. Methods We introduce PREs as an intermediary design between randomized controlled trials and observational studies. We review previous research that used the PRE design and describe its application in obesity-related research, including random roommate assignments, heterochronic parabiosis, and the quasi-random assignment of subjects to geographic areas. We then provide a statistical framework to control for potential packet-level confounders not accounted for by randomization. Results PREs have successfully been used to improve causal estimates of the effect of roommates, altitude, and breastfeeding on weight outcomes. When certain assumptions are met, PREs can asymptotically control for packet-level characteristics. This has the potential to statistically estimate the effect of a single treatment even when randomization to a single treatment did not occur. Conclusions Applying PREs to obesity-related research will improve decisions about clinical, public health, and policy actions insofar as it offers researchers new insight into cause and effect relationships among variables. PMID:25444088

Empirical evaluation of the conceptual model underpinning a regional aquatic long-term monitoring program using causal modelling

USGS Publications Warehouse

Irvine, Kathryn M.; Miller, Scott; Al-Chokhachy, Robert K.; Archer, Erik; Roper, Brett B.; Kershner, Jeffrey L.

2015-01-01

Conceptual models are an integral facet of long-term monitoring programs. Proposed linkages between drivers, stressors, and ecological indicators are identified within the conceptual model of most mandated programs. We empirically evaluate a conceptual model developed for a regional aquatic and riparian monitoring program using causal models (i.e., Bayesian path analysis). We assess whether data gathered for regional status and trend estimation can also provide insights on why a stream may deviate from reference conditions. We target the hypothesized causal pathways for how anthropogenic drivers of road density, percent grazing, and percent forest within a catchment affect instream biological condition. We found instream temperature and fine sediments in arid sites and only fine sediments in mesic sites accounted for a significant portion of the maximum possible variation explainable in biological condition among managed sites. However, the biological significance of the direct effects of anthropogenic drivers on instream temperature and fine sediments were minimal or not detected. Consequently, there was weak to no biological support for causal pathways related to anthropogenic drivers’ impact on biological condition. With weak biological and statistical effect sizes, ignoring environmental contextual variables and covariates that explain natural heterogeneity would have resulted in no evidence of human impacts on biological integrity in some instances. For programs targeting the effects of anthropogenic activities, it is imperative to identify both land use practices and mechanisms that have led to degraded conditions (i.e., moving beyond simple status and trend estimation). Our empirical evaluation of the conceptual model underpinning the long-term monitoring program provided an opportunity for learning and, consequently, we discuss survey design elements that require modification to achieve question driven monitoring, a necessary step in the practice of adaptive monitoring. We suspect our situation is not unique and many programs may suffer from the same inferential disconnect. Commonly, the survey design is optimized for robust estimates of regional status and trend detection and not necessarily to provide statistical inferences on the causal mechanisms outlined in the conceptual model, even though these relationships are typically used to justify and promote the long-term monitoring of a chosen ecological indicator. Our application demonstrates a process for empirical evaluation of conceptual models and exemplifies the need for such interim assessments in order for programs to evolve and persist.

The Causal Effect of Class Size on Academic Achievement: Multivariate Instrumental Variable Estimators with Data Missing at Random

ERIC Educational Resources Information Center

Shin, Yongyun; Raudenbush, Stephen W.

2011-01-01

This article addresses three questions: Does reduced class size cause higher academic achievement in reading, mathematics, listening, and word recognition skills? If it does, how large are these effects? Does the magnitude of such effects vary significantly across schools? The authors analyze data from Tennessee's Student/Teacher Achievement Ratio…

Education and Entrepreneurship in Canada: Evidence from (Repeated) Cross-Sectional Data

ERIC Educational Resources Information Center

Masakure, Oliver

2015-01-01

This paper estimates the causal effect of education on entrepreneurship choice in Canada taking into account the endogeneity of education. The data come from the General and Social Surveys (2000-2009). We consider the effect of two extreme education levels: university and some/no education. Regressions are based on fixed effects with two-stage…

Quantifying and Testing Indirect Effects in Simple Mediation Models when the Constituent Paths Are Nonlinear

ERIC Educational Resources Information Center

Hayes, Andrew F.; Preacher, Kristopher J.

2010-01-01

Most treatments of indirect effects and mediation in the statistical methods literature and the corresponding methods used by behavioral scientists have assumed linear relationships between variables in the causal system. Here we describe and extend a method first introduced by Stolzenberg (1980) for estimating indirect effects in models of…

Balancing Score Adjusted Targeted Minimum Loss-based Estimation

PubMed Central

Lendle, Samuel David; Fireman, Bruce; van der Laan, Mark J.

2015-01-01

Adjusting for a balancing score is sufficient for bias reduction when estimating causal effects including the average treatment effect and effect among the treated. Estimators that adjust for the propensity score in a nonparametric way, such as matching on an estimate of the propensity score, can be consistent when the estimated propensity score is not consistent for the true propensity score but converges to some other balancing score. We call this property the balancing score property, and discuss a class of estimators that have this property. We introduce a targeted minimum loss-based estimator (TMLE) for a treatment-specific mean with the balancing score property that is additionally locally efficient and doubly robust. We investigate the new estimator’s performance relative to other estimators, including another TMLE, a propensity score matching estimator, an inverse probability of treatment weighted estimator, and a regression-based estimator in simulation studies. PMID:26561539

Semiparametric estimation of treatment effect in a pretest-posttest study.

PubMed

Leon, Selene; Tsiatis, Anastasios A; Davidian, Marie

2003-12-01

Inference on treatment effects in a pretest-posttest study is a routine objective in medicine, public health, and other fields. A number of approaches have been advocated. We take a semiparametric perspective, making no assumptions about the distributions of baseline and posttest responses. By representing the situation in terms of counterfactual random variables, we exploit recent developments in the literature on missing data and causal inference, to derive the class of all consistent treatment effect estimators, identify the most efficient such estimator, and outline strategies for implementation of estimators that may improve on popular methods. We demonstrate the methods and their properties via simulation and by application to a data set from an HIV clinical trial.

Correntropy-based partial directed coherence for testing multivariate Granger causality in nonlinear processes

NASA Astrophysics Data System (ADS)

Kannan, Rohit; Tangirala, Arun K.

2014-06-01

Identification of directional influences in multivariate systems is of prime importance in several applications of engineering and sciences such as plant topology reconstruction, fault detection and diagnosis, and neurosciences. A spectrum of related directionality measures, ranging from linear measures such as partial directed coherence (PDC) to nonlinear measures such as transfer entropy, have emerged over the past two decades. The PDC-based technique is simple and effective, but being a linear directionality measure has limited applicability. On the other hand, transfer entropy, despite being a robust nonlinear measure, is computationally intensive and practically implementable only for bivariate processes. The objective of this work is to develop a nonlinear directionality measure, termed as KPDC, that possesses the simplicity of PDC but is still applicable to nonlinear processes. The technique is founded on a nonlinear measure called correntropy, a recently proposed generalized correlation measure. The proposed method is equivalent to constructing PDC in a kernel space where the PDC is estimated using a vector autoregressive model built on correntropy. A consistent estimator of the KPDC is developed and important theoretical results are established. A permutation scheme combined with the sequential Bonferroni procedure is proposed for testing hypothesis on absence of causality. It is demonstrated through several case studies that the proposed methodology effectively detects Granger causality in nonlinear processes.

The Effects of Resources across School Phases: A Summary of Recent Evidence. CEP Discussion Paper No. 1226

ERIC Educational Resources Information Center

Gibbons, Stephen; McNally, Sandra

2013-01-01

This report provides an overview and discussion of the past decade of academic evidence on the causal effects of resources in schooling on students' outcomes. Early evidence lacked good strategies for estimating the effects of schools resources, leading many people to conclude that spending more on schools had no effect. More recent evidence using…

The mutual causality analysis between the stock and futures markets

NASA Astrophysics Data System (ADS)

Yao, Can-Zhong; Lin, Qing-Wen

2017-07-01

In this paper we employ the conditional Granger causality model to estimate the information flow, and find that the improved model outperforms the Granger causality model in revealing the asymmetric correlation between stocks and futures in the Chinese market. First, we find that information flows estimated by Granger causality tests from futures to stocks are greater than those from stocks to futures. Additionally, average correlation coefficients capture some important characteristics between stock prices and information flows over time. Further, we find that direct information flows estimated by conditional Granger causality tests from stocks to futures are greater than those from futures to stocks. Besides, the substantial increases of information flows and direct information flows exhibit a certain degree of synchronism with the occurrences of important events. Finally, the comparative analysis with the asymmetric ratio and the bootstrap technique demonstrates the slight asymmetry of information flows and the significant asymmetry of direct information flows. It reveals that the information flows from futures to stocks are slightly greater than those in the reverse direction, while the direct information flows from stocks to futures are significantly greater than those in the reverse direction.

Aircraft noise, health, and residential sorting: evidence from two quasi-experiments.

PubMed

Boes, Stefan; Nüesch, Stephan; Stillman, Steven

2013-09-01

We explore two unexpected changes in flight regulations to estimate the causal effect of aircraft noise on health. Detailed measures of noise are linked with longitudinal data on individual health outcomes based on the exact address information. Controlling for individual heterogeneity and spatial sorting into different neighborhoods, we find that aircraft noise significantly increases sleeping problems and headaches. Models that do not control for such heterogeneity and sorting substantially underestimate the negative health effects, which suggests that individuals self-select into residence based on their unobserved sensitivity to noise. Our study demonstrates that the combination of quasi-experimental variation and panel data is very powerful for identifying causal effects in epidemiological field studies. Copyright © 2013 John Wiley & Sons, Ltd.

Estimating short-run and long-run interaction mechanisms in interictal state.

PubMed

Ozkaya, Ata; Korürek, Mehmet

2010-04-01

We address the issue of analyzing electroencephalogram (EEG) from seizure patients in order to test, model and determine the statistical properties that distinguish between EEG states (interictal, pre-ictal, ictal) by introducing a new class of time series analysis methods. In the present study: firstly, we employ statistical methods to determine the non-stationary behavior of focal interictal epileptiform series within very short time intervals; secondly, for such intervals that are deemed non-stationary we suggest the concept of Autoregressive Integrated Moving Average (ARIMA) process modelling, well known in time series analysis. We finally address the queries of causal relationships between epileptic states and between brain areas during epileptiform activity. We estimate the interaction between different EEG series (channels) in short time intervals by performing Granger-causality analysis and also estimate such interaction in long time intervals by employing Cointegration analysis, both analysis methods are well-known in econometrics. Here we find: first, that the causal relationship between neuronal assemblies can be identified according to the duration and the direction of their possible mutual influences; second, that although the estimated bidirectional causality in short time intervals yields that the neuronal ensembles positively affect each other, in long time intervals neither of them is affected (increasing amplitudes) from this relationship. Moreover, Cointegration analysis of the EEG series enables us to identify whether there is a causal link from the interictal state to ictal state.

Correlations between cannabis use and IQ change in the Dunedin cohort are consistent with confounding from socioeconomic status

PubMed Central

Rogeberg, Ole

2013-01-01

Does cannabis use have substantial and permanent effects on neuropsychological functioning? Renewed and intense attention to the issue has followed recent research on the Dunedin cohort, which found a positive association between, on the one hand, adolescent-onset cannabis use and dependence and, on the other hand, a decline in IQ from childhood to adulthood [Meier et al. (2012) Proc Natl Acad Sci USA 109(40):E2657–E2664]. The association is given a causal interpretation by the authors, but existing research suggests an alternative confounding model based on time-varying effects of socioeconomic status on IQ. A simulation of the confounding model reproduces the reported associations from the Dunedin cohort, suggesting that the causal effects estimated in Meier et al. are likely to be overestimates, and that the true effect could be zero. Further analyses of the Dunedin cohort are proposed to distinguish between the competing interpretations. Although it would be too strong to say that the results have been discredited, the methodology is flawed and the causal inference drawn from the results premature. PMID:23319626

A Critical Look at Methodologies Used to Evaluate Charter School Effectiveness

ERIC Educational Resources Information Center

Ackerman, Matthew; Egalite, Anna J.

2017-01-01

There is no consensus among researchers on charter school effectiveness in the USA, in part because of discrepancies in the research methods employed across various studies. Causal impact estimates from experimental studies demonstrate large positive impacts, but concerns about the generalizability of these results have prompted the development of…

The Role of Motivation-Related Variables in Assessment of Intelligence in Severely Involved Quadraplegic Children.

ERIC Educational Resources Information Center

Eagle, Rita Simon

1985-01-01

Five measures of intellectual level (IQ, MA, teachers' estimates of intelligence, level of object permanence, and representation of causality) were correlated with effectance variables and with each other for 42 quadraplegics (one to 12 years). Results reveal some relationship between intellectual level and effectance, although cognitive ability…

Effects of Causal Attributions Following Mathematics Tasks on Student Cognitions about a Subsequent Task

ERIC Educational Resources Information Center

Seegers, Gerard; Van Putten, Cornelis M.; Vermeer, Harriet J.

2004-01-01

The authors investigated the effects of former learning experiences on how students adapt to challenging mathematics tasks. A distinction has been made between domain-specific variables (goal orientation, self-concept of mathematics ability) and task- (or context-) specific appraisals (estimated competence for, attractiveness and relevance of the…

Does poor school performance cause later psychosocial problems among children in foster care? Evidence from national longitudinal registry data.

PubMed

Forsman, Hilma; Brännström, Lars; Vinnerljung, Bo; Hjern, Anders

2016-07-01

Research has shown that children in foster care are a high-risk group for adverse economic, social and health related outcomes in young adulthood. Children's poor school performance has been identified as a major risk factor for these poor later life outcomes. Aiming to support the design of effective intervention strategies, this study examines the hypothesized causal effect of foster children's poor school performance on subsequent psychosocial problems, here conceptualized as economic hardship, illicit drug use, and mental health problems, in young adulthood. Using the potential outcomes approach, longitudinal register data on more than 7500 Swedish foster children born 1973-1978 were analyzed by means of doubly robust treatment-effect estimators. The results show that poor school performance has a negative impact on later psychosocial problems net of observed background attributes and potential selection on unobservables, suggesting that the estimated effects allow for causal interpretations. Promotion of school performance may thus be a viable intervention path for policymakers and practitioners interested in improving foster children's overall life chances. Copyright © 2016 Elsevier Ltd. All rights reserved.

Why High-Order Polynomials Should Not Be Used in Regression Discontinuity Designs. NBER Working Paper No. 20405

ERIC Educational Resources Information Center

Gelman, Andrew; Imbens, Guido

2014-01-01

It is common in regression discontinuity analysis to control for high order (third, fourth, or higher) polynomials of the forcing variable. We argue that estimators for causal effects based on such methods can be misleading, and we recommend researchers do not use them, and instead use estimators based on local linear or quadratic polynomials or…

The Labor Market Returns to a For-Profit College Education. NBER Working Paper No. 18343

ERIC Educational Resources Information Center

Cellini, Stephanie Riegg; Chaudhary, Latika

2012-01-01

A lengthy literature estimating the returns to education has largely ignored the for-profit sector. In this paper, we offer some of the first causal estimates of the earnings gains to for-profit colleges. We rely on restricted-use data from the 1997 National Longitudinal Survey of Youth (NLSY97) to implement an individual fixed effects estimation…

Comparison of causality analysis on simultaneously measured fMRI and NIRS signals during motor tasks.

PubMed

Anwar, Abdul Rauf; Muthalib, Makii; Perrey, Stephane; Galka, Andreas; Granert, Oliver; Wolff, Stephan; Deuschl, Guenther; Raethjen, Jan; Heute, Ulrich; Muthuraman, Muthuraman

2013-01-01

Brain activity can be measured using different modalities. Since most of the modalities tend to complement each other, it seems promising to measure them simultaneously. In to be presented research, the data recorded from Functional Magnetic Resonance Imaging (fMRI) and Near Infrared Spectroscopy (NIRS), simultaneously, are subjected to causality analysis using time-resolved partial directed coherence (tPDC). Time-resolved partial directed coherence uses the principle of state space modelling to estimate Multivariate Autoregressive (MVAR) coefficients. This method is useful to visualize both frequency and time dynamics of causality between the time series. Afterwards, causality results from different modalities are compared by estimating the Spearman correlation. In to be presented study, we used directionality vectors to analyze correlation, rather than actual signal vectors. Results show that causality analysis of the fMRI correlates more closely to causality results of oxy-NIRS as compared to deoxy-NIRS in case of a finger sequencing task. However, in case of simple finger tapping, no clear difference between oxy-fMRI and deoxy-fMRI correlation is identified.

Life-threatening complications of ibogaine: three case reports.

PubMed

Paling, F P; Andrews, L M; Valk, G D; Blom, H J

2012-11-01

Ibogaine is a naturally occurring psychoactive alkaloid extracted from the roots of the Tabernanthe iboga plant, which in alternative medicine is used to treat drug dependency. However, this upcoming, online advocated therapy can be dangerous due to its potentially lethal adverse effects. We present three cases in which toxic side effects were noted. We used the Naranjo scale to estimate the probability of a causal relationship between these effects and ibogaine. Findings in these three cases are suggestive of a causal relationship between the use of ibogaine and serious respiratory and cardiac problems (including lengthening of the QT interval). In our opinion it is of great importance that clinicians are aware of these potentially serious side effects and realise that widespread online marketing practices will give many more people access to ibogaine.

The stochastic system approach for estimating dynamic treatments effect.

PubMed

Commenges, Daniel; Gégout-Petit, Anne

2015-10-01

The problem of assessing the effect of a treatment on a marker in observational studies raises the difficulty that attribution of the treatment may depend on the observed marker values. As an example, we focus on the analysis of the effect of a HAART on CD4 counts, where attribution of the treatment may depend on the observed marker values. This problem has been treated using marginal structural models relying on the counterfactual/potential response formalism. Another approach to causality is based on dynamical models, and causal influence has been formalized in the framework of the Doob-Meyer decomposition of stochastic processes. Causal inference however needs assumptions that we detail in this paper and we call this approach to causality the "stochastic system" approach. First we treat this problem in discrete time, then in continuous time. This approach allows incorporating biological knowledge naturally. When working in continuous time, the mechanistic approach involves distinguishing the model for the system and the model for the observations. Indeed, biological systems live in continuous time, and mechanisms can be expressed in the form of a system of differential equations, while observations are taken at discrete times. Inference in mechanistic models is challenging, particularly from a numerical point of view, but these models can yield much richer and reliable results.

Causal inference with missing exposure information: Methods and applications to an obstetric study.

PubMed

Zhang, Zhiwei; Liu, Wei; Zhang, Bo; Tang, Li; Zhang, Jun

2016-10-01

Causal inference in observational studies is frequently challenged by the occurrence of missing data, in addition to confounding. Motivated by the Consortium on Safe Labor, a large observational study of obstetric labor practice and birth outcomes, this article focuses on the problem of missing exposure information in a causal analysis of observational data. This problem can be approached from different angles (i.e. missing covariates and causal inference), and useful methods can be obtained by drawing upon the available techniques and insights in both areas. In this article, we describe and compare a collection of methods based on different modeling assumptions, under standard assumptions for missing data (i.e. missing-at-random and positivity) and for causal inference with complete data (i.e. no unmeasured confounding and another positivity assumption). These methods involve three models: one for treatment assignment, one for the dependence of outcome on treatment and covariates, and one for the missing data mechanism. In general, consistent estimation of causal quantities requires correct specification of at least two of the three models, although there may be some flexibility as to which two models need to be correct. Such flexibility is afforded by doubly robust estimators adapted from the missing covariates literature and the literature on causal inference with complete data, and by a newly developed triply robust estimator that is consistent if any two of the three models are correct. The methods are applied to the Consortium on Safe Labor data and compared in a simulation study mimicking the Consortium on Safe Labor. © The Author(s) 2013.

Nonparametric methods for doubly robust estimation of continuous treatment effects.

PubMed

Kennedy, Edward H; Ma, Zongming; McHugh, Matthew D; Small, Dylan S

2017-09-01

Continuous treatments (e.g., doses) arise often in practice, but many available causal effect estimators are limited by either requiring parametric models for the effect curve, or by not allowing doubly robust covariate adjustment. We develop a novel kernel smoothing approach that requires only mild smoothness assumptions on the effect curve, and still allows for misspecification of either the treatment density or outcome regression. We derive asymptotic properties and give a procedure for data-driven bandwidth selection. The methods are illustrated via simulation and in a study of the effect of nurse staffing on hospital readmissions penalties.

Estimating the Effect of Targeted Screening Strategies: An Application to Colonoscopy and Colorectal Cancer.

PubMed

Thomas, Duncan C

2017-07-01

Screening behavior depends on previous screening history and family members' behaviors, which can act as both confounders and intermediate variables on a causal pathway from screening to disease risk. Conventional analyses that adjust for these variables can lead to incorrect inferences about the causal effect of screening if high-risk individuals are more likely to be screened. Analyzing the data in a manner that treats screening as randomized conditional on covariates allows causal parameters to be estimated; inverse probability weighting based on propensity of exposure scores is one such method considered here. I simulated family data under plausible models for the underlying disease process and for screening behavior to assess the performance of alternative methods of analysis and whether a targeted screening approach based on individuals' risk factors would lead to a greater reduction in cancer incidence in the population than a uniform screening policy. Simulation results indicate that there can be a substantial underestimation of the effect of screening on subsequent cancer risk when using conventional analysis approaches, which is avoided by using inverse probability weighting. A large case-control study of colonoscopy and colorectal cancer from Germany shows a strong protective effect of screening, but inverse probability weighting makes this effect even stronger. Targeted screening approaches based on either fixed risk factors or family history yield somewhat greater reductions in cancer incidence with fewer screens needed to prevent one cancer than population-wide approaches, but the differences may not be large enough to justify the additional effort required. See video abstract at, http://links.lww.com/EDE/B207.

Contrasting Causal Effects of Workplace Interventions.

PubMed

Izano, Monika A; Brown, Daniel M; Neophytou, Andreas M; Garcia, Erika; Eisen, Ellen A

2018-07-01

Occupational exposure guidelines are ideally based on estimated effects of static interventions that assign constant exposure over a working lifetime. Static effects are difficult to estimate when follow-up extends beyond employment because their identifiability requires additional assumptions. Effects of dynamic interventions that assign exposure while at work, allowing subjects to leave and become unexposed thereafter, are more easily identifiable but result in different estimates. Given the practical implications of exposure limits, we explored the drivers of the differences between static and dynamic interventions in a simulation study where workers could terminate employment because of an intermediate adverse health event that functions as a time-varying confounder. The two effect estimates became more similar with increasing strength of the health event and outcome relationship and with increasing time between health event and employment termination. Estimates were most dissimilar when the intermediate health event occurred early in employment, providing an effective screening mechanism.

Stress increases the risk of type 2 diabetes onset in women: A 12-year longitudinal study using causal modelling

PubMed Central

Oldmeadow, Christopher; Hure, Alexis; Luu, Judy; Loxton, Deborah

2017-01-01

Background Type 2 diabetes is associated with significant morbidity and mortality. Modifiable risk factors have been found to contribute up to 60% of type 2 diabetes risk. However, type 2 diabetes continues to rise despite implementation of interventions based on traditional risk factors. There is a clear need to identify additional risk factors for chronic disease prevention. The aim of this study was to examine the relationship between perceived stress and type 2 diabetes onset, and partition the estimates into direct and indirect effects. Methods and findings Women born in 1946–1951 (n = 12,844) completed surveys for the Australian Longitudinal Study on Women’s Health in 1998, 2001, 2004, 2007 and 2010. The total causal effect was estimated using logistic regression and marginal structural modelling. Controlled direct effects were estimated through conditioning in the regression model. A graded association was found between perceived stress and all mediators in the multivariate time lag analyses. A significant association was found between hypertension, as well as physical activity and body mass index, and diabetes, but not smoking or diet quality. Moderate/high stress levels were associated with a 2.3-fold increase in the odds of diabetes three years later, for the total estimated effect. Results were only slightly attenuated when the direct and indirect effects of perceived stress on diabetes were partitioned, with the mediators only explaining 10–20% of the excess variation in diabetes. Conclusions Perceived stress is a strong risk factor for type 2 diabetes. The majority of the effect estimate of stress on diabetes risk is not mediated by the traditional risk factors of hypertension, physical activity, smoking, diet quality, and body mass index. This gives a new pathway for diabetes prevention trials and clinical practice. PMID:28222165

Bayesian regression discontinuity designs: incorporating clinical knowledge in the causal analysis of primary care data.

PubMed

Geneletti, Sara; O'Keeffe, Aidan G; Sharples, Linda D; Richardson, Sylvia; Baio, Gianluca

2015-07-10

The regression discontinuity (RD) design is a quasi-experimental design that estimates the causal effects of a treatment by exploiting naturally occurring treatment rules. It can be applied in any context where a particular treatment or intervention is administered according to a pre-specified rule linked to a continuous variable. Such thresholds are common in primary care drug prescription where the RD design can be used to estimate the causal effect of medication in the general population. Such results can then be contrasted to those obtained from randomised controlled trials (RCTs) and inform prescription policy and guidelines based on a more realistic and less expensive context. In this paper, we focus on statins, a class of cholesterol-lowering drugs, however, the methodology can be applied to many other drugs provided these are prescribed in accordance to pre-determined guidelines. Current guidelines in the UK state that statins should be prescribed to patients with 10-year cardiovascular disease risk scores in excess of 20%. If we consider patients whose risk scores are close to the 20%  risk score threshold, we find that there is an element of random variation in both the risk score itself and its measurement. We can therefore consider the threshold as a randomising device that assigns statin prescription to individuals just above the threshold and withholds it from those just below. Thus, we are effectively replicating the conditions of an RCT in the area around the threshold, removing or at least mitigating confounding. We frame the RD design in the language of conditional independence, which clarifies the assumptions necessary to apply an RD design to data, and which makes the links with instrumental variables clear. We also have context-specific knowledge about the expected sizes of the effects of statin prescription and are thus able to incorporate this into Bayesian models by formulating informative priors on our causal parameters. © 2015 The Authors. Statistics in Medicine Published by John Wiley & Sons Ltd.

Mediation Analysis with Survival Outcomes: Accelerated Failure Time vs. Proportional Hazards Models

PubMed Central

Gelfand, Lois A.; MacKinnon, David P.; DeRubeis, Robert J.; Baraldi, Amanda N.

2016-01-01

Objective: Survival time is an important type of outcome variable in treatment research. Currently, limited guidance is available regarding performing mediation analyses with survival outcomes, which generally do not have normally distributed errors, and contain unobserved (censored) events. We present considerations for choosing an approach, using a comparison of semi-parametric proportional hazards (PH) and fully parametric accelerated failure time (AFT) approaches for illustration. Method: We compare PH and AFT models and procedures in their integration into mediation models and review their ability to produce coefficients that estimate causal effects. Using simulation studies modeling Weibull-distributed survival times, we compare statistical properties of mediation analyses incorporating PH and AFT approaches (employing SAS procedures PHREG and LIFEREG, respectively) under varied data conditions, some including censoring. A simulated data set illustrates the findings. Results: AFT models integrate more easily than PH models into mediation models. Furthermore, mediation analyses incorporating LIFEREG produce coefficients that can estimate causal effects, and demonstrate superior statistical properties. Censoring introduces bias in the coefficient estimate representing the treatment effect on outcome—underestimation in LIFEREG, and overestimation in PHREG. With LIFEREG, this bias can be addressed using an alternative estimate obtained from combining other coefficients, whereas this is not possible with PHREG. Conclusions: When Weibull assumptions are not violated, there are compelling advantages to using LIFEREG over PHREG for mediation analyses involving survival-time outcomes. Irrespective of the procedures used, the interpretation of coefficients, effects of censoring on coefficient estimates, and statistical properties should be taken into account when reporting results. PMID:27065906

The Effects of Poverty on the Mental, Emotional, and Behavioral Health of Children and Youth: Implications for Prevention

ERIC Educational Resources Information Center

Yoshikawa, Hirokazu; Aber, J. Lawrence; Beardslee, William R.

2012-01-01

This article considers the implications for prevention science of recent advances in research on family poverty and children's mental, emotional, and behavioral health. First, we describe definitions of poverty and the conceptual and empirical challenges to estimating the causal effects of poverty on children's mental, emotional, and behavioral…

ICT, Literacy and Teacher Change: The Effectiveness of ICT Options in Kenya

ERIC Educational Resources Information Center

Piper, Benjamin

2014-01-01

There is a dearth of literature that use research design for causal inference that estimate the effect of information and communications technology (ICT) programs on literacy outcomes in early primary, particularly in Sub-Saharan Africa. There are several programs that have used ICT at a large scale, including Los Angeles, Peru, Nicaragua, Rwanda…

Survival analysis using inverse probability of treatment weighted methods based on the generalized propensity score.

PubMed

Sugihara, Masahiro

2010-01-01

In survival analysis, treatment effects are commonly evaluated based on survival curves and hazard ratios as causal treatment effects. In observational studies, these estimates may be biased due to confounding factors. The inverse probability of treatment weighted (IPTW) method based on the propensity score is one of the approaches utilized to adjust for confounding factors between binary treatment groups. As a generalization of this methodology, we developed an exact formula for an IPTW log-rank test based on the generalized propensity score for survival data. This makes it possible to compare the group differences of IPTW Kaplan-Meier estimators of survival curves using an IPTW log-rank test for multi-valued treatments. As causal treatment effects, the hazard ratio can be estimated using the IPTW approach. If the treatments correspond to ordered levels of a treatment, the proposed method can be easily extended to the analysis of treatment effect patterns with contrast statistics. In this paper, the proposed method is illustrated with data from the Kyushu Lipid Intervention Study (KLIS), which investigated the primary preventive effects of pravastatin on coronary heart disease (CHD). The results of the proposed method suggested that pravastatin treatment reduces the risk of CHD and that compliance to pravastatin treatment is important for the prevention of CHD. (c) 2009 John Wiley & Sons, Ltd.

Explaining the Relationship between Employment and Juvenile Delinquency.

PubMed

Staff, Jeremy; Osgood, D Wayne; Schulenberg, John E; Bachman, Jerald G; Messersmith, Emily E

2010-11-28

Most criminological theories predict an inverse relationship between employment and crime, but teenagers' involvement in paid work during the school year is positively correlated with delinquency and substance use. Whether the work-delinquency association is causal or spurious has long been debated. This study estimates the effect of paid work on juvenile delinquency using longitudinal data from the national Monitoring the Future project. We address issues of spuriousness by using a two-level hierarchical model to estimate the relationships of within-individual changes in juvenile delinquency and substance use to those in paid work and other explanatory variables. We also disentangle effects of actual employment from preferences for employment to provide insight about the likely role of time-varying selection factors tied to employment, delinquency, school engagement, and leisure activities. Whereas causal effects of employment would produce differences based on whether and how many hours respondents worked, we found significantly higher rates of crime and substance use among non-employed youth who preferred intensive versus moderate work. Our findings suggest the relationship between high-intensity work and delinquency results from preexisting factors that lead youth to desire varying levels of employment.

Systems GMM estimates of the health care spending and GDP relationship: a note.

PubMed

Kumar, Saten

2013-06-01

This paper utilizes the systems generalized method of moments (GMM) [Arellano and Bover (1995) J Econometrics 68:29-51; Blundell and Bond (1998) J Econometrics 87:115-143], and panel Granger causality [Hurlin and Venet (2001) Granger Causality tests in panel data models with fixed coefficients. Mime'o, University Paris IX], to investigate the health care spending and gross domestic product (GDP) relationship for organisation for economic co-operation and development countries over the period 1960-2007. The system GMM estimates confirm that the contribution of real GDP to health spending is significant and positive. The panel Granger causality tests imply that a bi-directional causality exists between health spending and GDP. To this end, policies aimed at raising health spending will eventually improve the well-being of the population in the long run.

Serum iron level and kidney function: a Mendelian randomization study.

PubMed

Del Greco M, Fabiola; Foco, Luisa; Pichler, Irene; Eller, Philipp; Eller, Kathrin; Benyamin, Beben; Whitfield, John B; Pramstaller, Peter P; Thompson, John R; Pattaro, Cristian; Minelli, Cosetta

2017-02-01

Iron depletion is a known consequence of chronic kidney disease (CKD), but there is contradicting epidemiological evidence on whether iron itself affects kidney function and whether its effect is protective or detrimental in the general population. While epidemiological studies tend to be affected by confounding and reverse causation, Mendelian randomization (MR) can provide unconfounded estimates of causal effects by using genes as instruments. We performed an MR study of the effect of serum iron levels on estimated glomerular filtration rate (eGFR), using genetic variants known to be associated with iron. MR estimates of the effect of iron on eGFR were derived based on the association of each variant with iron and eGFR from two large genome-wide meta-analyses on 48 978 and 74 354 individuals. We performed a similar MR analysis for ferritin, which measures iron stored in the body, using variants associated with ferritin. A combined MR estimate across all variants showed a 1.3% increase in eGFR per standard deviation increase in iron (95% confidence interval 0.4–2.1%; P = 0.004). The results for ferritin were consistent with those for iron. Secondary MR analyses of the effects of iron and ferritin on CKD did not show significant associations but had very low statistical power. Our study suggests a protective effect of iron on kidney function in the general population. Further research is required to confirm this causal association, investigate it in study populations at higher risk of CKD and explore its underlying mechanism of action.

Principal Score Methods: Assumptions, Extensions, and Practical Considerations

ERIC Educational Resources Information Center

Feller, Avi; Mealli, Fabrizia; Miratrix, Luke

2017-01-01

Researchers addressing posttreatment complications in randomized trials often turn to principal stratification to define relevant assumptions and quantities of interest. One approach for the subsequent estimation of causal effects in this framework is to use methods based on the "principal score," the conditional probability of belonging…

Graphical Models for Quasi-Experimental Designs

ERIC Educational Resources Information Center

Kim, Yongnam; Steiner, Peter M.; Hall, Courtney E.; Su, Dan

2016-01-01

Experimental and quasi-experimental designs play a central role in estimating cause-effect relationships in education, psychology, and many other fields of the social and behavioral sciences. This paper presents and discusses the causal graphs of experimental and quasi-experimental designs. For quasi-experimental designs the authors demonstrate…

Redundant variables and Granger causality

NASA Astrophysics Data System (ADS)

Angelini, L.; de Tommaso, M.; Marinazzo, D.; Nitti, L.; Pellicoro, M.; Stramaglia, S.

2010-03-01

We discuss the use of multivariate Granger causality in presence of redundant variables: the application of the standard analysis, in this case, leads to under estimation of causalities. Using the un-normalized version of the causality index, we quantitatively develop the notions of redundancy and synergy in the frame of causality and propose two approaches to group redundant variables: (i) for a given target, the remaining variables are grouped so as to maximize the total causality and (ii) the whole set of variables is partitioned to maximize the sum of the causalities between subsets. We show the application to a real neurological experiment, aiming to a deeper understanding of the physiological basis of abnormal neuronal oscillations in the migraine brain. The outcome by our approach reveals the change in the informational pattern due to repetitive transcranial magnetic stimulations.

Exploring the link between environmental pollution and economic growth in EU-28 countries: Is there an environmental Kuznets curve?

PubMed Central

Armeanu, Daniel; Vintilă, Georgeta; Gherghina, Ştefan Cristian; Drăgoi, Mihaela Cristina; Teodor, Cristian

2018-01-01

This study examines the Environmental Kuznets Curve hypothesis (EKC), considering the primary energy consumption among other country-specific variables, for a panel of the EU-28 countries during the period 1990–2014. By estimating pooled OLS regressions with Driscoll-Kraay standard errors in order to account for cross-sectional dependence, the results confirm the EKC hypothesis in the case of emissions of sulfur oxides and emissions of non-methane volatile organic compounds. In addition to pooled estimations, the output of fixed-effects regressions with Driscoll-Kraay standard errors support the EKC hypothesis for greenhouse gas emissions, greenhouse gas emissions intensity of energy consumption, emissions of nitrogen oxides, emissions of non-methane volatile organic compounds and emissions of ammonia. Additionally, the empirical findings from panel vector error correction model reveal a short-run unidirectional causality from GDP per capita growth to greenhouse gas emissions, as well as a bidirectional causal link between primary energy consumption and greenhouse gas emissions. Furthermore, since there occurred no causal link between economic growth and primary energy consumption, the neo-classical view was confirmed, namely the neutrality hypothesis. PMID:29742169

Exploring the link between environmental pollution and economic growth in EU-28 countries: Is there an environmental Kuznets curve?

PubMed

Armeanu, Daniel; Vintilă, Georgeta; Andrei, Jean Vasile; Gherghina, Ştefan Cristian; Drăgoi, Mihaela Cristina; Teodor, Cristian

2018-01-01

This study examines the Environmental Kuznets Curve hypothesis (EKC), considering the primary energy consumption among other country-specific variables, for a panel of the EU-28 countries during the period 1990-2014. By estimating pooled OLS regressions with Driscoll-Kraay standard errors in order to account for cross-sectional dependence, the results confirm the EKC hypothesis in the case of emissions of sulfur oxides and emissions of non-methane volatile organic compounds. In addition to pooled estimations, the output of fixed-effects regressions with Driscoll-Kraay standard errors support the EKC hypothesis for greenhouse gas emissions, greenhouse gas emissions intensity of energy consumption, emissions of nitrogen oxides, emissions of non-methane volatile organic compounds and emissions of ammonia. Additionally, the empirical findings from panel vector error correction model reveal a short-run unidirectional causality from GDP per capita growth to greenhouse gas emissions, as well as a bidirectional causal link between primary energy consumption and greenhouse gas emissions. Furthermore, since there occurred no causal link between economic growth and primary energy consumption, the neo-classical view was confirmed, namely the neutrality hypothesis.

Assessing the role of insulin-like growth factors and binding proteins in prostate cancer using Mendelian randomization: Genetic variants as instruments for circulating levels.

PubMed

Bonilla, Carolina; Lewis, Sarah J; Rowlands, Mari-Anne; Gaunt, Tom R; Davey Smith, George; Gunnell, David; Palmer, Tom; Donovan, Jenny L; Hamdy, Freddie C; Neal, David E; Eeles, Rosalind; Easton, Doug; Kote-Jarai, Zsofia; Al Olama, Ali Amin; Benlloch, Sara; Muir, Kenneth; Giles, Graham G; Wiklund, Fredrik; Grönberg, Henrik; Haiman, Christopher A; Schleutker, Johanna; Nordestgaard, Børge G; Travis, Ruth C; Pashayan, Nora; Khaw, Kay-Tee; Stanford, Janet L; Blot, William J; Thibodeau, Stephen; Maier, Christiane; Kibel, Adam S; Cybulski, Cezary; Cannon-Albright, Lisa; Brenner, Hermann; Park, Jong; Kaneva, Radka; Batra, Jyotsna; Teixeira, Manuel R; Pandha, Hardev; Lathrop, Mark; Martin, Richard M; Holly, Jeff M P

2016-10-01

Circulating insulin-like growth factors (IGFs) and their binding proteins (IGFBPs) are associated with prostate cancer. Using genetic variants as instruments for IGF peptides, we investigated whether these associations are likely to be causal. We identified from the literature 56 single nucleotide polymorphisms (SNPs) in the IGF axis previously associated with biomarker levels (8 from a genome-wide association study [GWAS] and 48 in reported candidate genes). In ∼700 men without prostate cancer and two replication cohorts (N ∼ 900 and ∼9,000), we examined the properties of these SNPS as instrumental variables (IVs) for IGF-I, IGF-II, IGFBP-2 and IGFBP-3. Those confirmed as strong IVs were tested for association with prostate cancer risk, low (< 7) vs. high (≥ 7) Gleason grade, localised vs. advanced stage, and mortality, in 22,936 controls and 22,992 cases. IV analysis was used in an attempt to estimate the causal effect of circulating IGF peptides on prostate cancer. Published SNPs in the IGFBP1/IGFBP3 gene region, particularly rs11977526, were strong instruments for IGF-II and IGFBP-3, less so for IGF-I. Rs11977526 was associated with high (vs. low) Gleason grade (OR per IGF-II/IGFBP-3 level-raising allele 1.05; 95% CI: 1.00, 1.10). Using rs11977526 as an IV we estimated the causal effect of a one SD increase in IGF-II (∼265 ng/mL) on risk of high vs. low grade disease as 1.14 (95% CI: 1.00, 1.31). Because of the potential for pleiotropy of the genetic instruments, these findings can only causally implicate the IGF pathway in general, not any one specific biomarker. © 2016 UICC.

Understanding the impact of career academy attendance: an application of the principal stratification framework for causal effects accounting for partial compliance.

PubMed

Page, Lindsay C

2012-04-01

Results from MDRC's longitudinal, random-assignment evaluation of career-academy high schools reveal that several years after high-school completion, those randomized to receive the academy opportunity realized a $175 (11%) increase in monthly earnings, on average. In this paper, I investigate the impact of duration of actual academy enrollment, as nearly half of treatment group students either never enrolled or participated for only a portion of high school. I capitalize on data from this experimental evaluation and utilize a principal stratification framework and Bayesian inference to investigate the causal impact of academy participation. This analysis focuses on a sample of 1,306 students across seven sites in the MDRC evaluation. Participation is measured by number of years of academy enrollment, and the outcome of interest is average monthly earnings in the period of four to eight years after high school graduation. I estimate an average causal effect of treatment assignment on subsequent monthly earnings of approximately $588 among males who remained enrolled in an academy throughout high school and more modest impacts among those who participated only partially. Different from an instrumental variables approach to treatment non-compliance, which allows for the estimation of linear returns to treatment take-up, the more general framework of principal stratification allows for the consideration of non-linear returns, although at the expense of additional model-based assumptions.

Causal Effect of Parental Schooling on Early Childhood Undernutrition: Quasi-Experimental Evidence From Zimbabwe.

PubMed

De Neve, Jan-Walter; Subramanian, S V

2018-01-01

An estimated 3.1 million children die each year because of undernutrition. Although cross-sectional and longitudinal studies have found a protective association between greater parental education and undernutrition in their children, no randomized trial has identified a causal effect, to our knowledge. Using the 1980 education reform in Zimbabwe as a natural experiment, we estimated the causal effect of additional parental schooling on the probability of anthropometric failure in their children under 5 years of age (ages 3 through 59 months). Analyzing data on 8,243 children from the 1988, 1999, 2005-2006, and 2010-2011 Demographic and Health Surveys, we found no effect of parental schooling on early childhood undernutrition at the national level in Zimbabwe. Among households in the urban and high-wealth-index subsamples, each additional year of maternal schooling led to absolute reductions in the probability of a child's being wasted of 5.2 percentage points (95% confidence interval (CI): -9.3, -1.2) and 3.6 percentage points (95% CI: -6.9, -0.4), respectively. In the subsample of children between the ages of 3 and 23 months, each additional year of paternal schooling increased the probability of a child's being stunted by 9.6 percentage points (95% CI: 1.4, 17.9). Secondary schooling alone may not be enough to improve early childhood nutrition in low-resource settings. © The Author(s) 2017. Published by Oxford University Press on behalf of the Johns Hopkins Bloomberg School of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Bilirubin as a potential causal factor in type 2 diabetes risk: a Mendelian randomization study

PubMed Central

Abbasi, Ali; Deetman, Petronella E.; Corpeleijn, Eva; Gansevoort, Ron T.; Gans, Rijk O.B.; Hillege, Hans L.; van der Harst, Pim; Stolk, Ronald P.; Navis, Gerjan; Alizadeh, Behrooz Z.; Bakker, Stephan J.L.

2014-01-01

Circulating bilirubin, a natural antioxidant, is associated with decreased risk of type 2 diabetes (T2D), but the nature of the relationship remains unknown. We performed Mendelian randomization in a prospective cohort of 3,381 participants free of diabetes at baseline (aged 28-75 years; women, 52.6%). We used rs6742078 located in UDP-glucuronosyltransferase (UGT1A1) locus as instrumental variable (IV) to study a potential causal effect of serum total bilirubin on T2D risk. T2D developed in a total of 210 (6.2%) participants during a median follow-up of 7.8 years. In adjusted analyses, rs6742078, which explained 19.5% of bilirubin variation, was strongly associated with total bilirubin (a 0.68-SD increase in bilirubin levels per T allele; P<1×10−122) and was also associated with T2D risk (OR 0.69 [95%CI, 0.54-0.90]; P=0.006). Per 1-SD increase in log-transformed bilirubin levels, we observed a 25% (OR 0.75 [95%CI, 0.62-0.92]; P=0.004) lower risk of T2D. In Mendelian randomization analysis, the causal risk reduction for T2D was estimated to be 42% (causal ORIVestimation per 1-SD increase in log-transformed bilirubin 0.58 [95%CI, 0.39-0.84]; P=0.005), which was comparable to the observational estimate (Durbin-Wu-Hausman chi-square test Pfor difference =0.19). These novel results provide evidence that elevated bilirubin is causally associated with risk of T2D and support its role as a protective determinant. PMID:25368098

Contours of a causal feedback mechanism between adaptive personality and psychosocial function in patients with personality disorders: a secondary analysis from a randomized clinical trial.

PubMed

Klungsøyr, Ole; Antonsen, Bjørnar; Wilberg, Theresa

2017-06-05

Patients with personality disorders commonly exhibit impairment in psychosocial function that persists over time even with diagnostic remission. Further causal knowledge may help to identify and assess factors with a potential to alleviate this impairment. Psychosocial function is associated with personality functioning which describes personality disorder severity in DSM-5 (section III) and which can reportedly be improved by therapy. The reciprocal association between personality functioning and psychosocial function was assessed, in 113 patients with different personality disorders, in a secondary longitudinal analysis of data from a randomized clinical trial, over six years. Personality functioning was represented by three domains of the Severity Indices of Personality Problems: Relational Capacity, Identity Integration, and Self-control. Psychosocial function was measured by Global Assessment of Functioning. The marginal structural model was used for estimation of causal effects of the three personality functioning domains on psychosocial function, and vice versa. The attractiveness of this model lies in the ability to assess an effect of a time - varying exposure on an outcome, while adjusting for time - varying confounding. Strong causal effects were found. A hypothetical intervention to increase Relational Capacity by one standard deviation, both at one and two time-points prior to assessment of psychosocial function, would increase psychosocial function by 3.5 standard deviations (95% CI: 2.0, 4.96). Significant effects of Identity Integration and Self-control on psychosocial function, and from psychosocial function on all three domains of personality functioning, although weaker, were also found. This study indicates that persistent impairment in psychosocial function can be addressed through a causal pathway of personality functioning, with interventions of at least 18 months duration.

Global mortality consequences of climate change accounting for adaptation costs and benefits

NASA Astrophysics Data System (ADS)

Rising, J. A.; Jina, A.; Carleton, T.; Hsiang, S. M.; Greenstone, M.

2017-12-01

Empirically-based and plausibly causal estimates of the damages of climate change are greatly needed to inform rapidly developing global and local climate policies. To accurately reflect the costs of climate change, it is essential to estimate how much populations will adapt to a changing climate, yet adaptation remains one of the least understood aspects of social responses to climate. In this paper, we develop and implement a novel methodology to estimate climate impacts on mortality rates. We assemble comprehensive sub-national panel data in 41 countries that account for 56% of the world's population, and combine them with high resolution daily climate data to flexibly estimate the causal effect of temperature on mortality. We find the impacts of temperature on mortality have a U-shaped response; both hot days and cold days cause excess mortality. However, this average response obscures substantial heterogeneity, as populations are differentially adapted to extreme temperatures. Our empirical model allows us to extrapolate response functions across the entire globe, as well as across time, using a range of economic, population, and climate change scenarios. We also develop a methodology to capture not only the benefits of adaptation, but also its costs. We combine these innovations to produce the first causal, micro-founded, global, empirically-derived climate damage function for human health. We project that by 2100, business-as-usual climate change is likely to incur mortality-only costs that amount to approximately 5% of global GDP for 5°C degrees of warming above pre-industrial levels. On average across model runs, we estimate that the upper bound on adaptation costs amounts to 55% of the total damages.

Estimating the causal influence of body mass index on risk of Parkinson disease: A Mendelian randomisation study

PubMed Central

Price, T. Ryan; De Pablo-Fernandez, Eduardo; Haycock, Philip C.; Schrag, Anette; Lees, Andrew J.; Hardy, John; Singleton, Andrew; Nalls, Mike A.; Pearce, Neil; Wood, Nicholas W.

2017-01-01

Background Both positive and negative associations between higher body mass index (BMI) and Parkinson disease (PD) have been reported in observational studies, but it has been difficult to establish causality because of the possibility of residual confounding or reverse causation. To our knowledge, Mendelian randomisation (MR)—the use of genetic instrumental variables (IVs) to explore causal effects—has not previously been used to test the effect of BMI on PD. Methods and findings Two-sample MR was undertaken using genome-wide association (GWA) study data. The associations between the genetic instruments and BMI were obtained from the GIANT consortium and consisted of the per-allele difference in mean BMI for 77 independent variants that reached genome-wide significance. The per-allele difference in log-odds of PD for each of these variants was estimated from a recent meta-analysis, which included 13,708 cases of PD and 95,282 controls. The inverse-variance weighted method was used to estimate a pooled odds ratio (OR) for the effect of a 5-kg/m2 higher BMI on PD. Evidence of directional pleiotropy averaged across all variants was sought using MR–Egger regression. Frailty simulations were used to assess whether causal associations were affected by mortality selection. A combined genetic IV expected to confer a lifetime exposure of 5-kg/m2 higher BMI was associated with a lower risk of PD (OR 0.82, 95% CI 0.69–0.98). MR–Egger regression gave similar results, suggesting that directional pleiotropy was unlikely to be biasing the result (intercept 0.002; p = 0.654). However, the apparent protective influence of higher BMI could be at least partially induced by survival bias in the PD GWA study, as demonstrated by frailty simulations. Other important limitations of this application of MR include the inability to analyse non-linear associations, to undertake subgroup analyses, and to gain mechanistic insights. Conclusions In this large study using two-sample MR, we found that variants known to influence BMI had effects on PD in a manner consistent with higher BMI leading to lower risk of PD. The mechanism underlying this apparent protective effect warrants further study. PMID:28609445

Welfare Reform in California: Early Results from the Impact Analysis.

ERIC Educational Resources Information Center

Klerman, Jacob Alex; Hotz, V. Joseph; Reardon, Elaine; Cox, Amy G.; Farley, Donna O.; Haider, Steven J.; Imbens, Guido; Schoeni, Robert

The impact of California Work Opportunity and Responsibility to Kids (CalWORKS), which was passed to increase California welfare recipients' participation in welfare-to-work (WTW) activities, was examined. The impact study consisted of a nonexperimental program evaluation that used statistical models to estimate causal effects and a simulation…

Estimating Causal Effects in Mediation Analysis Using Propensity Scores

ERIC Educational Resources Information Center

Coffman, Donna L.

2011-01-01

Mediation is usually assessed by a regression-based or structural equation modeling (SEM) approach that we refer to as the classical approach. This approach relies on the assumption that there are no confounders that influence both the mediator, "M", and the outcome, "Y". This assumption holds if individuals are randomly…

The Implications of "Contamination" for Experimental Design in Education

ERIC Educational Resources Information Center

Rhoads, Christopher H.

2011-01-01

Experimental designs that randomly assign entire clusters of individuals (e.g., schools and classrooms) to treatments are frequently advocated as a way of guarding against contamination of the estimated average causal effect of treatment. However, in the absence of contamination, experimental designs that randomly assign intact clusters to…

Within-Cluster and Across-Cluster Matching with Observational Multilevel Data

ERIC Educational Resources Information Center

Kim, Jee-Seon; Steiner, Peter M.; Hall, Courtney; Thoemmes, Felix

2013-01-01

When randomized experiments cannot be conducted in practice, propensity score (PS) techniques for matching treated and control units are frequently used for estimating causal treatment effects from observational data. Despite the popularity of PS techniques, they are not yet well studied for matching multilevel data where selection into treatment…

Training the East German Labour Force. Microeconometric Evaluations of Continuous Vocational Training after Unification. Studies in Contemporary Economics.

ERIC Educational Resources Information Center

Lechner, Michael

This book presents empirical evaluations of the effects of different types of training programs in East Germany. Chapter 1 presents an overview of labor, the study objectives and results, and discussion of causality and the identification problem in evaluation studies. Chapter 2 examines point estimates of the effects of two types of continuous…

The Effect of Private School Vouchers on Political Participation: Experimental Evidence from New York City

ERIC Educational Resources Information Center

Carlson, Deven; Chingos, Matthew M.; Campbell, David E.

2017-01-01

In 1997, the New York School Choice Scholarships Foundation Program (SCSF) randomly offered three-year scholarships to attend private schools to approximately 1,000 low-income families in New York City. In this paper we leverage exogenous variation generated by the SCSF to estimate the causal effect of the private school voucher offer--and the…

Does Cultural Capital Really Affect Academic Achievement? New Evidence from Combined Sibling and Panel Data

ERIC Educational Resources Information Center

Jaeger, Mads Meier

2011-01-01

This article provides new estimates of the causal effect of cultural capital on academic achievement. The author analyzes data from the National Longitudinal Survey of Youth--Children and Young Adults and uses a fixed effect design to address the problem of omitted variable bias, which has resulted in too optimistic results in previous research.…

The Effect of Teacher-Family Communication on Student Engagement: Evidence from a Randomized Field Experiment

ERIC Educational Resources Information Center

Kraft, Matthew A.; Dougherty, Shaun M.

2013-01-01

In this study, we evaluate the efficacy of teacher communication with parents and students as a means of increasing student engagement. We estimate the causal effect of teacher communication by conducting a randomized field experiment in which sixth- and ninth-grade students were assigned to receive a daily phone call home and a text/written…

Assessing Causality in a Complex Security Environment

DTIC Science & Technology

2015-01-01

social sciences that could genuinely benefit those students. Causality is one of these critical issues. Causality has many definitions, but we might...protests (Ivan Bandura ) Report Documentation Page Form ApprovedOMB No. 0704-0188 Public reporting burden for the collection of information is estimated...relatively simple theory of what leads to a stable deter- rent relationship between two states. Mearsheimer argued that when State A fields a

Identification and estimation of survivor average causal effects.

PubMed

Tchetgen Tchetgen, Eric J

2014-09-20

In longitudinal studies, outcomes ascertained at follow-up are typically undefined for individuals who die prior to the follow-up visit. In such settings, outcomes are said to be truncated by death and inference about the effects of a point treatment or exposure, restricted to individuals alive at the follow-up visit, could be biased even if as in experimental studies, treatment assignment were randomized. To account for truncation by death, the survivor average causal effect (SACE) defines the effect of treatment on the outcome for the subset of individuals who would have survived regardless of exposure status. In this paper, the author nonparametrically identifies SACE by leveraging post-exposure longitudinal correlates of survival and outcome that may also mediate the exposure effects on survival and outcome. Nonparametric identification is achieved by supposing that the longitudinal data arise from a certain nonparametric structural equations model and by making the monotonicity assumption that the effect of exposure on survival agrees in its direction across individuals. A novel weighted analysis involving a consistent estimate of the survival process is shown to produce consistent estimates of SACE. A data illustration is given, and the methods are extended to the context of time-varying exposures. We discuss a sensitivity analysis framework that relaxes assumptions about independent errors in the nonparametric structural equations model and may be used to assess the extent to which inference may be altered by a violation of key identifying assumptions. © 2014 The Authors. Statistics in Medicine published by John Wiley & Sons, Ltd.

Identification and estimation of survivor average causal effects

PubMed Central

Tchetgen, Eric J Tchetgen

2014-01-01

In longitudinal studies, outcomes ascertained at follow-up are typically undefined for individuals who die prior to the follow-up visit. In such settings, outcomes are said to be truncated by death and inference about the effects of a point treatment or exposure, restricted to individuals alive at the follow-up visit, could be biased even if as in experimental studies, treatment assignment were randomized. To account for truncation by death, the survivor average causal effect (SACE) defines the effect of treatment on the outcome for the subset of individuals who would have survived regardless of exposure status. In this paper, the author nonparametrically identifies SACE by leveraging post-exposure longitudinal correlates of survival and outcome that may also mediate the exposure effects on survival and outcome. Nonparametric identification is achieved by supposing that the longitudinal data arise from a certain nonparametric structural equations model and by making the monotonicity assumption that the effect of exposure on survival agrees in its direction across individuals. A novel weighted analysis involving a consistent estimate of the survival process is shown to produce consistent estimates of SACE. A data illustration is given, and the methods are extended to the context of time-varying exposures. We discuss a sensitivity analysis framework that relaxes assumptions about independent errors in the nonparametric structural equations model and may be used to assess the extent to which inference may be altered by a violation of key identifying assumptions. © 2014 The Authors. Statistics in Medicine published by John Wiley & Sons, Ltd. PMID:24889022

A causal analysis framework for land-use change and the potential role of bioenergy policy

DOE PAGES

Efroymson, Rebecca A.; Kline, Keith L.; Angelsen, Arild; ...

2016-10-05

Here we propose a causal analysis framework to increase the reliability of land-use change (LUC) models and the accuracy of net greenhouse gas (GHG) emissions calculations for biofuels. The health-sciences-inspired framework is used here to determine probable causes of LUC, with an emphasis on bioenergy and deforestation. Calculations of net GHG emissions for LUC are critical in determining whether a fuel qualifies as a biofuel or advanced biofuel category under national (U.S., U.K.), state (California), and European Union regulations. Biofuel policymakers and scientists continue to discuss whether presumed indirect land-use change (ILUC) estimates, which often involve deforestation, should be includedmore » in GHG accounting for biofuel pathways. Current estimates of ILUC for bioenergy rely largely on economic simulation models that focus on causal pathways involving global commodity trade and use coarse land cover data with simple land classification systems. ILUC estimates are highly uncertain, partly because changes are not clearly defined and key causal links are not sufficiently included in the models. The proposed causal analysis framework begins with a definition of the change that has occurred and proceeds to a strength-of-evidence approach based on types of epidemiological evidence including plausibility of the relationship, completeness of the causal pathway, spatial co-occurrence, time order, analogous agents, simulation model results, and quantitative agent response relationships.Lastly, we discuss how LUC may be allocated among probable causes for policy purposes and how the application of the framework has the potential to increase the validity of LUC models and resolve ILUC and biofuel controversies.« less

A causal analysis framework for land-use change and the potential role of bioenergy policy

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

Efroymson, Rebecca A.; Kline, Keith L.; Angelsen, Arild

Here we propose a causal analysis framework to increase the reliability of land-use change (LUC) models and the accuracy of net greenhouse gas (GHG) emissions calculations for biofuels. The health-sciences-inspired framework is used here to determine probable causes of LUC, with an emphasis on bioenergy and deforestation. Calculations of net GHG emissions for LUC are critical in determining whether a fuel qualifies as a biofuel or advanced biofuel category under national (U.S., U.K.), state (California), and European Union regulations. Biofuel policymakers and scientists continue to discuss whether presumed indirect land-use change (ILUC) estimates, which often involve deforestation, should be includedmore » in GHG accounting for biofuel pathways. Current estimates of ILUC for bioenergy rely largely on economic simulation models that focus on causal pathways involving global commodity trade and use coarse land cover data with simple land classification systems. ILUC estimates are highly uncertain, partly because changes are not clearly defined and key causal links are not sufficiently included in the models. The proposed causal analysis framework begins with a definition of the change that has occurred and proceeds to a strength-of-evidence approach based on types of epidemiological evidence including plausibility of the relationship, completeness of the causal pathway, spatial co-occurrence, time order, analogous agents, simulation model results, and quantitative agent response relationships.Lastly, we discuss how LUC may be allocated among probable causes for policy purposes and how the application of the framework has the potential to increase the validity of LUC models and resolve ILUC and biofuel controversies.« less

Causal inference as an emerging statistical approach in neurology: an example for epilepsy in the elderly.

PubMed

Moura, Lidia Mvr; Westover, M Brandon; Kwasnik, David; Cole, Andrew J; Hsu, John

2017-01-01

The elderly population faces an increasing number of cases of chronic neurological conditions, such as epilepsy and Alzheimer's disease. Because the elderly with epilepsy are commonly excluded from randomized controlled clinical trials, there are few rigorous studies to guide clinical practice. When the elderly are eligible for trials, they either rarely participate or frequently have poor adherence to therapy, thus limiting both generalizability and validity. In contrast, large observational data sets are increasingly available, but are susceptible to bias when using common analytic approaches. Recent developments in causal inference-analytic approaches also introduce the possibility of emulating randomized controlled trials to yield valid estimates. We provide a practical example of the application of the principles of causal inference to a large observational data set of patients with epilepsy. This review also provides a framework for comparative-effectiveness research in chronic neurological conditions.

Assessing the Genetic Predisposition of Education on Myopia: a Mendelian Randomization Study

PubMed Central

Cuellar-Partida, Gabriel; Lu, Yi; Kho, Pik Fang; Hewitt, Alex W.; Wichmann, H.-Erich; Yazar, Seyhan; Stambolian, Dwight; Bailey-Wilson, Joan E.; Wojciechowski, Robert; Wang, Jie Jin; Mitchell, Paul; Mackey, David A.; MacGregor, Stuart

2016-01-01

Myopia is the largest cause of uncorrected visual impairments globally and its recent dramatic increase in the population has made it a major public health problem. In observational studies, educational attainment has been consistently reported to be correlated to myopia. Nonetheless, correlation does not imply causation. Observational studies do not tell us if education causes myopia or if instead there are confounding factors underlying the association. In this work, we use a two-step least squares instrumental-variable (IV) approach to estimate the causal effect of education on refractive error, specifically myopia. We used the results from the educational attainment GWAS from the Social Science Genetic Association Consortium to define a polygenic risk score (PGRS) in three cohorts of late middle age and elderly Caucasian individuals (N=5,649). In a meta-analysis of the three cohorts, using the PGRS as an IV, we estimated that each z-score increase in education (approximately 2 years of education) results in a reduction of 0.92 ± 0.29 diopters (P=1.04×10−3). Our estimate of the effect of education on myopia was higher (P=0.01) than the observed estimate (0.25 ± 0.03 diopters reduction per education z-score [~2 years] increase). This suggests that observational studies may actually underestimate the true effect. Our Mendelian Randomization (MR) analysis provides new evidence for a causal role of educational attainment on refractive error. PMID:26497973

Assessing the Genetic Predisposition of Education on Myopia: A Mendelian Randomization Study.

PubMed

Cuellar-Partida, Gabriel; Lu, Yi; Kho, Pik Fang; Hewitt, Alex W; Wichmann, H-Erich; Yazar, Seyhan; Stambolian, Dwight; Bailey-Wilson, Joan E; Wojciechowski, Robert; Wang, Jie Jin; Mitchell, Paul; Mackey, David A; MacGregor, Stuart

2016-01-01

Myopia is the largest cause of uncorrected visual impairments globally and its recent dramatic increase in the population has made it a major public health problem. In observational studies, educational attainment has been consistently reported to be correlated to myopia. Nonetheless, correlation does not imply causation. Observational studies do not tell us if education causes myopia or if instead there are confounding factors underlying the association. In this work, we use a two-step least squares instrumental-variable (IV) approach to estimate the causal effect of education on refractive error, specifically myopia. We used the results from the educational attainment GWAS from the Social Science Genetic Association Consortium to define a polygenic risk score (PGRS) in three cohorts of late middle age and elderly Caucasian individuals (N = 5,649). In a meta-analysis of the three cohorts, using the PGRS as an IV, we estimated that each z-score increase in education (approximately 2 years of education) results in a reduction of 0.92 ± 0.29 diopters (P = 1.04 × 10(-3) ). Our estimate of the effect of education on myopia was higher (P = 0.01) than the observed estimate (0.25 ± 0.03 diopters reduction per education z-score [∼2 years] increase). This suggests that observational studies may actually underestimate the true effect. Our Mendelian Randomization (MR) analysis provides new evidence for a causal role of educational attainment on refractive error. © 2015 WILEY PERIODICALS, INC.

Causality as a Rigorous Notion and Quantitative Causality Analysis with Time Series

NASA Astrophysics Data System (ADS)

Liang, X. S.

2017-12-01

Given two time series, can one faithfully tell, in a rigorous and quantitative way, the cause and effect between them? Here we show that this important and challenging question (one of the major challenges in the science of big data), which is of interest in a wide variety of disciplines, has a positive answer. Particularly, for linear systems, the maximal likelihood estimator of the causality from a series X2 to another series X1, written T2→1, turns out to be concise in form: T2→1 = [C11 C12 C2,d1 — C112 C1,d1] / [C112 C22 — C11C122] where Cij (i,j=1,2) is the sample covariance between Xi and Xj, and Ci,dj the covariance between Xi and ΔXj/Δt, the difference approximation of dXj/dt using the Euler forward scheme. An immediate corollary is that causation implies correlation, but not vice versa, resolving the long-standing debate over causation versus correlation. The above formula has been validated with touchstone series purportedly generated with one-way causality that evades the classical approaches such as Granger causality test and transfer entropy analysis. It has also been applied successfully to the investigation of many real problems. Through a simple analysis with the stock series of IBM and GE, an unusually strong one-way causality is identified from the former to the latter in their early era, revealing to us an old story, which has almost faded into oblivion, about "Seven Dwarfs" competing with a "Giant" for the computer market. Another example presented here regards the cause-effect relation between the two climate modes, El Niño and Indian Ocean Dipole (IOD). In general, these modes are mutually causal, but the causality is asymmetric. To El Niño, the information flowing from IOD manifests itself as a propagation of uncertainty from the Indian Ocean. In the third example, an unambiguous one-way causality is found between CO2 and the global mean temperature anomaly. While it is confirmed that CO2 indeed drives the recent global warming, on paleoclimate scales the cause-effect relation may be completely reversed. Key words: Causation, Information flow, Uncertainty Generation, El Niño, IOD, CO2/Global warming Reference : Liang, 2014: Unraveling the cause-effect relation between time series. PRE 90, 052150 News Report: http://scitation.aip.org/content/aip/magazine/physicstoday/news/10.1063/PT.5.7124

OVERWEIGHT/OBESITY AND HUMAN CAPITAL FORMATION FROM INFANCY TO ADOLESCENCE: EVIDENCE FROM TWO LARGE US COHORTS.

PubMed

Murasko, Jason E

2015-03-01

Body mass index (BMI) levels in US children and adolescents have increased over the past several decades. The negative health effects of this trend are well-documented. Recent work has evaluated the potential effects on skills formation. Studies are mixed on whether there is an association between high BMI and skills outcomes, and those that estimate causal effects find none. This paper offers estimates on the causal effect of BMI-defined overweight and obesity on skills formation using two large cohorts of contemporary US children followed from infancy to 5 years and from kindergarten (6 years) to the eighth grade (14 years). Significant negative associations were observed in the random effects models for males in early life with respect to a mental skills assessment, for females during the pre-school years for reading and maths assessments, for both males and females during the schooling years for reading assessments and for females during the schooling years for maths assessments. Fixed effects models yielded a significant negative association only with respect to the latter. The implication of these findings is that any improvement in skills outcomes that may accompany reductions in obesity prevalence may depend on whether interventions are general to overall health productivity or whether they are specific to dietary and exercise behaviours.

The causal effects of home care use on institutional long-term care utilization and expenditures.

PubMed

Guo, Jing; Konetzka, R Tamara; Manning, Willard G

2015-03-01

Limited evidence exists on whether expanding home care saves money overall or how much institutional long-term care can be reduced. This paper estimates the causal effect of Medicaid-financed home care services on the costs and utilization of institutional long-term care using Medicaid claims data. A unique instrumental variable was applied to address the potential bias caused by omitted variables or reverse effect of institutional care use. We find that the use of Medicaid-financed home care services significantly reduced but only partially offset utilization and Medicaid expenditures on nursing facility services. A $1000 increase in Medicaid home care expenditures avoided 2.75 days in nursing facilities and reduced annual Medicaid nursing facility costs by $351 among people over age 65 when selection bias is addressed. Failure to address selection biases would misestimate the substitution and offset effects. Copyright © 2015 John Wiley & Sons, Ltd.

Does health affect portfolio choice?

PubMed

Love, David A; Smith, Paul A

2010-12-01

A number of recent studies find that poor health is empirically associated with a safer portfolio allocation. It is difficult to say, however, whether this relationship is truly causal. Both health status and portfolio choice are influenced by unobserved characteristics such as risk attitudes, impatience, information, and motivation, and these unobserved factors, if not adequately controlled for, can induce significant bias in the estimates of asset demand equations. Using the 1992-2006 waves of the Health and Retirement Study, we investigate how much of the connection between health and portfolio choice is causal and how much is due to the effects of unobserved heterogeneity. Accounting for unobserved heterogeneity with fixed effects and correlated random effects models, we find that health does not appear to significantly affect portfolio choice among single households. For married households, we find a small effect (about 2-3 percentage points) from being in the lowest of five self-reported health categories. Copyright © 2009 John Wiley & Sons, Ltd.

Globally conditioned Granger causality in brain–brain and brain–heart interactions: a combined heart rate variability/ultra-high-field (7 T) functional magnetic resonance imaging study

PubMed Central

Passamonti, Luca; Wald, Lawrence L.; Barbieri, Riccardo

2016-01-01

The causal, directed interactions between brain regions at rest (brain–brain networks) and between resting-state brain activity and autonomic nervous system (ANS) outflow (brain–heart links) have not been completely elucidated. We collected 7 T resting-state functional magnetic resonance imaging (fMRI) data with simultaneous respiration and heartbeat recordings in nine healthy volunteers to investigate (i) the causal interactions between cortical and subcortical brain regions at rest and (ii) the causal interactions between resting-state brain activity and the ANS as quantified through a probabilistic, point-process-based heartbeat model which generates dynamical estimates for sympathetic and parasympathetic activity as well as sympathovagal balance. Given the high amount of information shared between brain-derived signals, we compared the results of traditional bivariate Granger causality (GC) with a globally conditioned approach which evaluated the additional influence of each brain region on the causal target while factoring out effects concomitantly mediated by other brain regions. The bivariate approach resulted in a large number of possibly spurious causal brain–brain links, while, using the globally conditioned approach, we demonstrated the existence of significant selective causal links between cortical/subcortical brain regions and sympathetic and parasympathetic modulation as well as sympathovagal balance. In particular, we demonstrated a causal role of the amygdala, hypothalamus, brainstem and, among others, medial, middle and superior frontal gyri, superior temporal pole, paracentral lobule and cerebellar regions in modulating the so-called central autonomic network (CAN). In summary, we show that, provided proper conditioning is employed to eliminate spurious causalities, ultra-high-field functional imaging coupled with physiological signal acquisition and GC analysis is able to quantify directed brain–brain and brain–heart interactions reflecting central modulation of ANS outflow. PMID:27044985

Revisiting the emissions-energy-trade nexus: evidence from the newly industrializing countries.

PubMed

Ahmed, Khalid; Shahbaz, Muhammad; Kyophilavong, Phouphet

2016-04-01

This paper applies Pedroni's panel cointegration approach to explore the causal relationship between trade openness, carbon dioxide emissions, energy consumption, and economic growth for the panel of newly industrialized economies (i.e., Brazil, India, China, and South Africa) over the period of 1970-2013. Our panel cointegration estimation results found majority of the variables cointegrated and confirm the long-run association among the variables. The Granger causality test indicates bidirectional causality between carbon dioxide emissions and energy consumption. A unidirectional causality is found running from trade openness to carbon dioxide emission and energy consumption and economic growth to carbon dioxide emissions. The results of causality analysis suggest that the trade liberalization in newly industrialized economies induces higher energy consumption and carbon dioxide emissions. Furthermore, the causality results are checked using an innovative accounting approach which includes forecast-error variance decomposition test and impulse response function. The long-run coefficients are estimated using fully modified ordinary least square (FMOLS) method, and results conclude that the trade openness and economic growth reduce carbon dioxide emissions in the long run. The results of FMOLS test sound the existence of environmental Kuznets curve hypothesis. It means that trade liberalization induces carbon dioxide emission with increased national output, but it offsets that impact in the long run with reduced level of carbon dioxide emissions.

Evaluating data-driven causal inference techniques in noisy physical and ecological systems

NASA Astrophysics Data System (ADS)

Tennant, C.; Larsen, L.

2016-12-01

Causal inference from observational time series challenges traditional approaches for understanding processes and offers exciting opportunities to gain new understanding of complex systems where nonlinearity, delayed forcing, and emergent behavior are common. We present a formal evaluation of the performance of convergent cross-mapping (CCM) and transfer entropy (TE) for data-driven causal inference under real-world conditions. CCM is based on nonlinear state-space reconstruction, and causality is determined by the convergence of prediction skill with an increasing number of observations of the system. TE is the uncertainty reduction based on transition probabilities of a pair of time-lagged variables. With TE, causal inference is based on asymmetry in information flow between the variables. Observational data and numerical simulations from a number of classical physical and ecological systems: atmospheric convection (the Lorenz system), species competition (patch-tournaments), and long-term climate change (Vostok ice core) were used to evaluate the ability of CCM and TE to infer causal-relationships as data series become increasingly corrupted by observational (instrument-driven) or process (model-or -stochastic-driven) noise. While both techniques show promise for causal inference, TE appears to be applicable to a wider range of systems, especially when the data series are of sufficient length to reliably estimate transition probabilities of system components. Both techniques also show a clear effect of observational noise on causal inference. For example, CCM exhibits a negative logarithmic decline in prediction skill as the noise level of the system increases. Changes in TE strongly depend on noise type and which variable the noise was added to. The ability of CCM and TE to detect driving influences suggest that their application to physical and ecological systems could be transformative for understanding driving mechanisms as Earth systems undergo change.

Estimating the Effects of Parental Divorce and Death With Fixed Effects Models.

PubMed

Amato, Paul R; Anthony, Christopher J

2014-04-01

The authors used child fixed effects models to estimate the effects of parental divorce and death on a variety of outcomes using 2 large national data sets: (a) the Early Childhood Longitudinal Study, Kindergarten Cohort (kindergarten through the 5th grade) and (b) the National Educational Longitudinal Study (8th grade to the senior year of high school). In both data sets, divorce and death were associated with multiple negative outcomes among children. Although evidence for a causal effect of divorce on children was reasonably strong, effect sizes were small in magnitude. A second analysis revealed a substantial degree of variability in children's outcomes following parental divorce, with some children declining, others improving, and most not changing at all. The estimated effects of divorce appeared to be strongest among children with the highest propensity to experience parental divorce.

Investigating Causality Between Interacting Brain Areas with Multivariate Autoregressive Models of MEG Sensor Data

PubMed Central

Michalareas, George; Schoffelen, Jan-Mathijs; Paterson, Gavin; Gross, Joachim

2013-01-01

Abstract In this work, we investigate the feasibility to estimating causal interactions between brain regions based on multivariate autoregressive models (MAR models) fitted to magnetoencephalographic (MEG) sensor measurements. We first demonstrate the theoretical feasibility of estimating source level causal interactions after projection of the sensor-level model coefficients onto the locations of the neural sources. Next, we show with simulated MEG data that causality, as measured by partial directed coherence (PDC), can be correctly reconstructed if the locations of the interacting brain areas are known. We further demonstrate, if a very large number of brain voxels is considered as potential activation sources, that PDC as a measure to reconstruct causal interactions is less accurate. In such case the MAR model coefficients alone contain meaningful causality information. The proposed method overcomes the problems of model nonrobustness and large computation times encountered during causality analysis by existing methods. These methods first project MEG sensor time-series onto a large number of brain locations after which the MAR model is built on this large number of source-level time-series. Instead, through this work, we demonstrate that by building the MAR model on the sensor-level and then projecting only the MAR coefficients in source space, the true casual pathways are recovered even when a very large number of locations are considered as sources. The main contribution of this work is that by this methodology entire brain causality maps can be efficiently derived without any a priori selection of regions of interest. Hum Brain Mapp, 2013. © 2012 Wiley Periodicals, Inc. PMID:22328419

Causal gene identification using combinatorial V-structure search.

PubMed

Cai, Ruichu; Zhang, Zhenjie; Hao, Zhifeng

2013-07-01

With the advances of biomedical techniques in the last decade, the costs of human genomic sequencing and genomic activity monitoring are coming down rapidly. To support the huge genome-based business in the near future, researchers are eager to find killer applications based on human genome information. Causal gene identification is one of the most promising applications, which may help the potential patients to estimate the risk of certain genetic diseases and locate the target gene for further genetic therapy. Unfortunately, existing pattern recognition techniques, such as Bayesian networks, cannot be directly applied to find the accurate causal relationship between genes and diseases. This is mainly due to the insufficient number of samples and the extremely high dimensionality of the gene space. In this paper, we present the first practical solution to causal gene identification, utilizing a new combinatorial formulation over V-Structures commonly used in conventional Bayesian networks, by exploring the combinations of significant V-Structures. We prove the NP-hardness of the combinatorial search problem under a general settings on the significance measure on the V-Structures, and present a greedy algorithm to find sub-optimal results. Extensive experiments show that our proposal is both scalable and effective, particularly with interesting findings on the causal genes over real human genome data. Copyright © 2013 Elsevier Ltd. All rights reserved.

THE RELATION BETWEEN DIFFERENT DIMENSIONS OF ALCOHOL CONSUMPTION AND BURDEN OF DISEASE - AN OVERVIEW

PubMed Central

Rehm, Jürgen; Baliunas, Dolly; Borges, Guilherme L. G.; Graham, Kathryn; lrving, Hyacinth; Kehoe, Tara; Parry, Charles D.; Patra, Jayadeep; Popova, Svetlana; Poznyak, Vladimir; Roerecke, Michael; Room, Robin; Samokhvalov, Andriy V.; Taylor, Benjamin

2012-01-01

AIMS As part of a larger study to estimate the global burden of disease and injury attributable to alcohol: To evaluate the evidence for a causal impact of average volume of alcohol consumption and pattern of drinking on diseases and injuries;To quantify relationships identified as causal based on published meta-analyses;To separate the impact on mortality vs. morbidity where possible; andTo assess the impact of the quality of alcohol on burden of disease. METHODS Systematic literature reviews were used to identify alcohol-related diseases, birth complications and injuries using standard epidemiologic criteria to determine causality. The extent of the risk relations was taken from meta-analyses. RESULTS Evidence of a causal impact of average volume of alcohol consumption was found for the following major diseases: tuberculosis, mouth, nasopharynx, other pharynx and oropharynx cancer, oesophageal cancer, colon and rectum cancer, liver cancer, female breast cancer, diabetes mellitus, alcohol use disorders, unipolar depressive disorders, epilepsy, hypertensive heart disease, ischaemic heart disease (IHD), ischaemic and haemorrhagic stroke, conduction disorders and other dysrhythmias, lower respiratory infections (pneumonia), cirrhosis of the liver, preterm birth complications, foetal alcohol syndrome. Dose-response relationships could be quantified for all disease categories except for depressive disorders, with the relative risk increasing with increased level of alcohol consumption for most diseases. Both average volume and drinking pattern were causally linked to IHD, foetal alcohol syndrome, and unintentional and intentional injuries. For IHD, ischaemic stroke and diabetes mellitus beneficial effects were observed for patterns of light to moderate drinking without heavy drinking occasions (as defined by 60+ grams pure alcohol per day). For several disease and injury categories, the effects were stronger on mortality compared to morbidity. There was insufficient evidence to establish whether quality of alcohol had a major impact on disease burden. CONCLUSIONS Overall, these findings indicate that alcohol causally impacts many disease outcomes, both chronic and acute, and injuries. In addition, a pattern of heavy episodic drinking increases risk for some disease and all injury outcomes. Future studies need to address a number of methodological issues, especially the differential role of average volume versus drinking pattern, in order to obtain more accurate risk estimates and to better understand the nature of alcohol-disease relationships. PMID:20331573

The balanced survivor average causal effect.

PubMed

Greene, Tom; Joffe, Marshall; Hu, Bo; Li, Liang; Boucher, Ken

2013-05-07

Statistical analysis of longitudinal outcomes is often complicated by the absence of observable values in patients who die prior to their scheduled measurement. In such cases, the longitudinal data are said to be "truncated by death" to emphasize that the longitudinal measurements are not simply missing, but are undefined after death. Recently, the truncation by death problem has been investigated using the framework of principal stratification to define the target estimand as the survivor average causal effect (SACE), which in the context of a two-group randomized clinical trial is the mean difference in the longitudinal outcome between the treatment and control groups for the principal stratum of always-survivors. The SACE is not identified without untestable assumptions. These assumptions have often been formulated in terms of a monotonicity constraint requiring that the treatment does not reduce survival in any patient, in conjunction with assumed values for mean differences in the longitudinal outcome between certain principal strata. In this paper, we introduce an alternative estimand, the balanced-SACE, which is defined as the average causal effect on the longitudinal outcome in a particular subset of the always-survivors that is balanced with respect to the potential survival times under the treatment and control. We propose a simple estimator of the balanced-SACE that compares the longitudinal outcomes between equivalent fractions of the longest surviving patients between the treatment and control groups and does not require a monotonicity assumption. We provide expressions for the large sample bias of the estimator, along with sensitivity analyses and strategies to minimize this bias. We consider statistical inference under a bootstrap resampling procedure.

The Effect of Birth Weight on Academic Performance: Instrumental Variable Analysis.

PubMed

Lin, Shi Lin; Leung, Gabriel Matthew; Schooling, C Mary

2017-05-01

Observationally, lower birth weight is usually associated with poorer academic performance; whether this association is causal or the result of confounding is unknown. To investigate this question, we obtained an effect estimate, which can have a causal interpretation under specific assumptions, of birth weight on educational attainment using instrumental variable analysis based on single nucleotide polymorphisms determining birth weight combined with results from the Social Science Genetic Association Consortium study of 126,559 Caucasians. We similarly obtained an estimate of the effect of birth weight on academic performance in 4,067 adolescents from Hong Kong's (Chinese) Children of 1997 birth cohort (1997-2016), using twin status as an instrumental variable. Birth weight was not associated with years of schooling (per 100-g increase in birth weight, -0.006 years, 95% confidence interval (CI): -0.02, 0.01) or college completion (odds ratio = 1.00, 95% CI: 0.96, 1.03). Birth weight was also unrelated to academic performance in adolescents (per 100-g increase in birth weight, -0.004 grade, 95% CI: -0.04, 0.04) using instrumental variable analysis, although conventional regression gave a small positive association (0.02 higher grade, 95% CI: 0.01, 0.03). Observed associations of birth weight with academic performance may not be causal, suggesting that interventions should focus on the contextual factors generating this correlation. © The Author 2017. Published by Oxford University Press on behalf of the Johns Hopkins Bloomberg School of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

A causal model for longitudinal randomised trials with time-dependent non-compliance

PubMed Central

Becque, Taeko; White, Ian R; Haggard, Mark

2015-01-01

In the presence of non-compliance, conventional analysis by intention-to-treat provides an unbiased comparison of treatment policies but typically under-estimates treatment efficacy. With all-or-nothing compliance, efficacy may be specified as the complier-average causal effect (CACE), where compliers are those who receive intervention if and only if randomised to it. We extend the CACE approach to model longitudinal data with time-dependent non-compliance, focusing on the situation in which those randomised to control may receive treatment and allowing treatment effects to vary arbitrarily over time. Defining compliance type to be the time of surgical intervention if randomised to control, so that compliers are patients who would not have received treatment at all if they had been randomised to control, we construct a causal model for the multivariate outcome conditional on compliance type and randomised arm. This model is applied to the trial of alternative regimens for glue ear treatment evaluating surgical interventions in childhood ear disease, where outcomes are measured over five time points, and receipt of surgical intervention in the control arm may occur at any time. We fit the models using Markov chain Monte Carlo methods to obtain estimates of the CACE at successive times after receiving the intervention. In this trial, over a half of those randomised to control eventually receive intervention. We find that surgery is more beneficial than control at 6months, with a small but non-significant beneficial effect at 12months. © 2015 The Authors. Statistics in Medicine Published by JohnWiley & Sons Ltd. PMID:25778798

Fair Inference on Outcomes

PubMed Central

Nabi, Razieh; Shpitser, Ilya

2017-01-01

In this paper, we consider the problem of fair statistical inference involving outcome variables. Examples include classification and regression problems, and estimating treatment effects in randomized trials or observational data. The issue of fairness arises in such problems where some covariates or treatments are “sensitive,” in the sense of having potential of creating discrimination. In this paper, we argue that the presence of discrimination can be formalized in a sensible way as the presence of an effect of a sensitive covariate on the outcome along certain causal pathways, a view which generalizes (Pearl 2009). A fair outcome model can then be learned by solving a constrained optimization problem. We discuss a number of complications that arise in classical statistical inference due to this view and provide workarounds based on recent work in causal and semi-parametric inference.

Invited commentary: G-computation--lost in translation?

PubMed

Vansteelandt, Stijn; Keiding, Niels

2011-04-01

In this issue of the Journal, Snowden et al. (Am J Epidemiol. 2011;173(7):731-738) give a didactic explanation of G-computation as an approach for estimating the causal effect of a point exposure. The authors of the present commentary reinforce the idea that their use of G-computation is equivalent to a particular form of model-based standardization, whereby reference is made to the observed study population, a technique that epidemiologists have been applying for several decades. They comment on the use of standardized versus conditional effect measures and on the relative predominance of the inverse probability-of-treatment weighting approach as opposed to G-computation. They further propose a compromise approach, doubly robust standardization, that combines the benefits of both of these causal inference techniques and is not more difficult to implement.

Using Propensity Score Methods to Approximate Factorial Experimental Designs to Analyze the Relationship between Two Variables and an Outcome

ERIC Educational Resources Information Center

Dong, Nianbo

2015-01-01

Researchers have become increasingly interested in programs' main and interaction effects of two variables (A and B, e.g., two treatment variables or one treatment variable and one moderator) on outcomes. A challenge for estimating main and interaction effects is to eliminate selection bias across A-by-B groups. I introduce Rubin's causal model to…

Estimation of Treatment Effects without an Exclusion Restriction: With an Application to the Analysis of the School Breakfast Program. NBER Working Paper No. 15539

ERIC Educational Resources Information Center

Millimet, Daniel L.; Tchernis, Rusty

2009-01-01

While the rise in childhood obesity is clear, the policy ramifications are not. School nutrition programs such as the School Breakfast Program (SBP) have come under much scrutiny. However, the lack of experimental evidence, combined with non-random selection into these programs, makes identification of the causal effects of such programs…

Does Remediation Work for All Students? How the Effects of Postsecondary Remedial and Developmental Courses Vary by Level of Academic Preparation

ERIC Educational Resources Information Center

Boatman, Angela; Long, Bridget Terry

2018-01-01

We examine the impact of remedial and developmental courses on college students with varying levels of academic preparedness, thus focusing on a wider range of students than previous studies. Using a regression discontinuity design, we provide causal estimates of the effects of placement in different levels of remedial courses on short-,…

A Careful Look at Modern Case Selection Methods

ERIC Educational Resources Information Center

Herron, Michael C.; Quinn, Kevin M.

2016-01-01

Case studies appear prominently in political science, sociology, and other social science fields. A scholar employing a case study research design in an effort to estimate causal effects must confront the question, how should cases be selected for analysis? This question is important because the results derived from a case study research program…

Regression Discontinuity Design in Gifted and Talented Education Research

ERIC Educational Resources Information Center

Matthews, Michael S.; Peters, Scott J.; Housand, Angela M.

2012-01-01

This Methodological Brief introduces the reader to the regression discontinuity design (RDD), which is a method that when used correctly can yield estimates of research treatment effects that are equivalent to those obtained through randomized control trials and can therefore be used to infer causality. However, RDD does not require the random…

Community Destruction and Traumatic Stress in Post-Tsunami Indonesia

ERIC Educational Resources Information Center

Frankenberg, Elizabeth; Nobles, Jenna; Sumantri, Cecep

2012-01-01

How are individuals affected when the communities they live in change for the worse? This question is central to understanding neighborhood effects, but few study designs generate estimates that can be interpreted causally. We address issues of inference through a natural experiment, examining post-traumatic stress at multiple time points in a…

Unbiased Causal Inference from an Observational Study: Results of a Within-Study Comparison

ERIC Educational Resources Information Center

Pohl, Steffi; Steiner, Peter M.; Eisermann, Jens; Soellner, Renate; Cook, Thomas D.

2009-01-01

Adjustment methods such as propensity scores and analysis of covariance are often used for estimating treatment effects in nonexperimental data. Shadish, Clark, and Steiner used a within-study comparison to test how well these adjustments work in practice. They randomly assigned participating students to a randomized or nonrandomized experiment.…

Computerized Simulation in the Social Sciences: A Survey and Evaluation

ERIC Educational Resources Information Center

Garson, G. David

2009-01-01

After years at the periphery of the social sciences, simulation is now emerging as an important and widely used tool for understanding social phenomena. Through simulation, researchers can identify causal effects, specify critical parameter estimates, and clarify the state of the art with respect to what is understood about how processes evolve…

Matching with Multiple Control Groups with Adjustment for Group Differences

ERIC Educational Resources Information Center

Stuart, Elizabeth A.; Rubin, Donald B.

2008-01-01

When estimating causal effects from observational data, it is desirable to approximate a randomized experiment as closely as possible. This goal can often be achieved by choosing a subsample from the original control group that matches the treatment group on the distribution of the observed covariates. However, sometimes the original control group…

Did School Finance Equalization Increase Revenue Instability for School Districts?

ERIC Educational Resources Information Center

Balu, Rekha

2011-01-01

This study uses an Interrupted Time Series analysis with a non-equivalent comparison group to estimate the causal effect of school finance equalization on district revenue instability. The author applies a microeconomic framework to an understudied problem in education finance and policy. In so doing, she illustrates how policies can sometimes…

Narrow-sense heritability estimation of complex traits using identity-by-descent information.

PubMed

Evans, Luke M; Tahmasbi, Rasool; Jones, Matt; Vrieze, Scott I; Abecasis, Gonçalo R; Das, Sayantan; Bjelland, Douglas W; de Candia, Teresa R; Yang, Jian; Goddard, Michael E; Visscher, Peter M; Keller, Matthew C

2018-03-28

Heritability is a fundamental parameter in genetics. Traditional estimates based on family or twin studies can be biased due to shared environmental or non-additive genetic variance. Alternatively, those based on genotyped or imputed variants typically underestimate narrow-sense heritability contributed by rare or otherwise poorly tagged causal variants. Identical-by-descent (IBD) segments of the genome share all variants between pairs of chromosomes except new mutations that have arisen since the last common ancestor. Therefore, relating phenotypic similarity to degree of IBD sharing among classically unrelated individuals is an appealing approach to estimating the near full additive genetic variance while possibly avoiding biases that can occur when modeling close relatives. We applied an IBD-based approach (GREML-IBD) to estimate heritability in unrelated individuals using phenotypic simulation with thousands of whole-genome sequences across a range of stratification, polygenicity levels, and the minor allele frequencies of causal variants (CVs). In simulations, the IBD-based approach produced unbiased heritability estimates, even when CVs were extremely rare, although precision was low. However, population stratification and non-genetic familial environmental effects shared across generations led to strong biases in IBD-based heritability. We used data on two traits in ~120,000 people from the UK Biobank to demonstrate that, depending on the trait and possible confounding environmental effects, GREML-IBD can be applied to very large genetic datasets to infer the contribution of very rare variants lost using other methods. However, we observed apparent biases in these real data, suggesting that more work may be required to understand and mitigate factors that influence IBD-based heritability estimates.

Estimating the intensity of ward admission and its effect on emergency department access block.

PubMed

Luo, Wei; Cao, Jiguo; Gallagher, Marcus; Wiles, Janet

2013-07-10

Emergency department access block is an urgent problem faced by many public hospitals today. When access block occurs, patients in need of acute care cannot access inpatient wards within an optimal time frame. A widely held belief is that access block is the end product of a long causal chain, which involves poor discharge planning, insufficient bed capacity, and inadequate admission intensity to the wards. This paper studies the last link of the causal chain-the effect of admission intensity on access block, using data from a metropolitan hospital in Australia. We applied several modern statistical methods to analyze the data. First, we modeled the admission events as a nonhomogeneous Poisson process and estimated time-varying admission intensity with penalized regression splines. Next, we established a functional linear model to investigate the effect of the time-varying admission intensity on emergency department access block. Finally, we used functional principal component analysis to explore the variation in the daily time-varying admission intensities. The analyses suggest that improving admission practice during off-peak hours may have most impact on reducing the number of ED access blocks. Copyright © 2012 John Wiley & Sons, Ltd.

Explaining the Relationship between Employment and Juvenile Delinquency*

PubMed Central

Staff, Jeremy; Osgood, D. Wayne; Schulenberg, John E.; Bachman, Jerald G.; Messersmith, Emily E.

2011-01-01

Most criminological theories predict an inverse relationship between employment and crime, but teenagers' involvement in paid work during the school year is positively correlated with delinquency and substance use. Whether the work-delinquency association is causal or spurious has long been debated. This study estimates the effect of paid work on juvenile delinquency using longitudinal data from the national Monitoring the Future project. We address issues of spuriousness by using a two-level hierarchical model to estimate the relationships of within-individual changes in juvenile delinquency and substance use to those in paid work and other explanatory variables. We also disentangle effects of actual employment from preferences for employment to provide insight about the likely role of time-varying selection factors tied to employment, delinquency, school engagement, and leisure activities. Whereas causal effects of employment would produce differences based on whether and how many hours respondents worked, we found significantly higher rates of crime and substance use among non-employed youth who preferred intensive versus moderate work. Our findings suggest the relationship between high-intensity work and delinquency results from preexisting factors that lead youth to desire varying levels of employment. PMID:21442045

A method for deriving water-quality benchmarks using field data.

PubMed

Cormier, Susan M; Suter, Glenn W

2013-02-01

The authors describe a methodology that characterizes effects to individual genera observed in the field and estimate the concentration at which 5% of genera are adversely affected. Ionic strength, measured as specific conductance, is used to illustrate the methodology. Assuming some resilience in the population, 95% of the genera are afforded protection. The authors selected an unambiguous effect, the presence or absence of a genus from sampling locations. The absence of a genus, extirpation, is operationally defined as the point above which only 5% of the observations of a genus occurs. The concentrations that cause extirpation of each genus are rank-ordered from least to greatest, and the benchmark is estimated at the 5th percentile of the distribution using two-point interpolation. When a full range of exposures and many taxa are included in the model of taxonomic sensitivity, the model broadly characterizes how species in general respond to a concentration gradient of the causal agent. This recognized U.S. Environmental Protection Agency methodology has many advantages. Observations from field studies include the full range of conditions, effects, species, and interactions that occur in the environment and can be used to model some causal relationships that laboratory studies cannot. Copyright © 2012 SETAC.

Bayesian propensity scores for high-dimensional causal inference: A comparison of drug-eluting to bare-metal coronary stents.

PubMed

Spertus, Jacob V; Normand, Sharon-Lise T

2018-04-23

High-dimensional data provide many potential confounders that may bolster the plausibility of the ignorability assumption in causal inference problems. Propensity score methods are powerful causal inference tools, which are popular in health care research and are particularly useful for high-dimensional data. Recent interest has surrounded a Bayesian treatment of propensity scores in order to flexibly model the treatment assignment mechanism and summarize posterior quantities while incorporating variance from the treatment model. We discuss methods for Bayesian propensity score analysis of binary treatments, focusing on modern methods for high-dimensional Bayesian regression and the propagation of uncertainty. We introduce a novel and simple estimator for the average treatment effect that capitalizes on conjugacy of the beta and binomial distributions. Through simulations, we show the utility of horseshoe priors and Bayesian additive regression trees paired with our new estimator, while demonstrating the importance of including variance from the treatment regression model. An application to cardiac stent data with almost 500 confounders and 9000 patients illustrates approaches and facilitates comparison with existing alternatives. As measured by a falsifiability endpoint, we improved confounder adjustment compared with past observational research of the same problem. © 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

Granger causality revisited

PubMed Central

Friston, Karl J.; Bastos, André M.; Oswal, Ashwini; van Wijk, Bernadette; Richter, Craig; Litvak, Vladimir

2014-01-01

This technical paper offers a critical re-evaluation of (spectral) Granger causality measures in the analysis of biological timeseries. Using realistic (neural mass) models of coupled neuronal dynamics, we evaluate the robustness of parametric and nonparametric Granger causality. Starting from a broad class of generative (state-space) models of neuronal dynamics, we show how their Volterra kernels prescribe the second-order statistics of their response to random fluctuations; characterised in terms of cross-spectral density, cross-covariance, autoregressive coefficients and directed transfer functions. These quantities in turn specify Granger causality — providing a direct (analytic) link between the parameters of a generative model and the expected Granger causality. We use this link to show that Granger causality measures based upon autoregressive models can become unreliable when the underlying dynamics is dominated by slow (unstable) modes — as quantified by the principal Lyapunov exponent. However, nonparametric measures based on causal spectral factors are robust to dynamical instability. We then demonstrate how both parametric and nonparametric spectral causality measures can become unreliable in the presence of measurement noise. Finally, we show that this problem can be finessed by deriving spectral causality measures from Volterra kernels, estimated using dynamic causal modelling. PMID:25003817

Smoking and subsequent human papillomavirus infection: a mediation analysis.

PubMed

Eldridge, Ronald C; Pawlita, Michael; Wilson, Lauren; Castle, Philip E; Waterboer, Tim; Gravitt, Patti E; Schiffman, Mark; Wentzensen, Nicolas

2017-11-01

Smoking is an established risk factor for a human papillomavirus (HPV) infection advancing to cervical precancer and cancer, but its role earlier in the natural history is less clear. Smoking is inversely associated with possessing HPV antibodies from a past infection suggesting that smoking may influence acquiring subsequent infections. In a cohort of 1976 U.S. women, we evaluate whether reduced antibodies to HPV-16 is a mechanism for smoking's role on acquiring a subsequent HPV-16 infection, through the analytic technique of causal mediation analysis. We posit a causal model and estimate two counterfactually defined effects: a smoking impaired antibody-mediated indirect effect and a nonmediated direct effect representing all other potential mechanisms of smoking. Compared to never smokers, current smokers had increased odds of HPV-16 infection by the antibody-mediated indirect effect (odds ratio [OR] = 1.29; 95% confidence interval [CI]: 1.11, 1.73); the estimated direct effect was very imprecise (OR = 0.57; 95% CI, 0.26-1.13). We observed a stronger estimated indirect effect among women who smoked at least half a pack of cigarettes daily (OR = 1.61, 95% CI, 1.27-2.15) than among women who smoked less than that threshold (OR = 1.09; 95% CI, 0.94-1.44). This is the first study to directly test the mechanism underlying smoking as an HPV cofactor. The results support current smoking as a risk factor earlier in the natural history of HPV and are consistent with the hypothesis that smoking increases the risk of a subsequent infection by reducing immunity. Published by Elsevier Inc.

EMG-Torque Dynamics Change With Contraction Bandwidth.

PubMed

Golkar, Mahsa A; Jalaleddini, Kian; Kearney, Robert E

2018-04-01

An accurate model for ElectroMyoGram (EMG)-torque dynamics has many uses. One of its applications which has gained high attention among researchers is its use, in estimating the muscle contraction level for the efficient control of prosthesis. In this paper, the dynamic relationship between the surface EMG and torque during isometric contractions at the human ankle was studied using system identification techniques. Subjects voluntarily modulated their ankle torque in dorsiflexion direction, by activating their tibialis anterior muscle, while tracking a pseudo-random binary sequence in a torque matching task. The effects of contraction bandwidth, described by torque spectrum, on EMG-torque dynamics were evaluated by varying the visual command switching time. Nonparametric impulse response functions (IRF) were estimated between the processed surface EMG and torque. It was demonstrated that: 1) at low contraction bandwidths, the identified IRFs had unphysiological anticipatory (i.e., non-causal) components, whose amplitude decreased as the contraction bandwidth increased. We hypothesized that this non-causal behavior arose, because the EMG input contained a component due to feedback from the output torque, i.e., it was recorded from within a closed-loop. Vision was not the feedback source since the non-causal behavior persisted when visual feedback was removed. Repeating the identification using a nonparametric closed-loop identification algorithm yielded causal IRFs at all bandwidths, supporting this hypothesis. 2) EMG-torque dynamics became faster and the bandwidth of system increased as contraction modulation rate increased. Thus, accurate prediction of torque from EMG signals must take into account the contraction bandwidth sensitivity of this system.

Compared to What? Estimating Causal Effects for Latent Subgroups to Understand Variation in the Impacts of Head Start by Alternate Child Care Setting

ERIC Educational Resources Information Center

Feller, Avi; Grindal, Todd; Miratrix, Luke; Page, Lindsay C.

2014-01-01

Head Start programs currently provide early childhood education and family support services to more than 900,000 low-income children and their families across the United States with an annual budget of around $8 billion in state and federal funds. Researchers and policy makers have debated the program's effectiveness since its inception in 1964.…

Estimating the Effect of Income on Health and Mortality Using Lottery Prizes as an Exogenous Source of Income

ERIC Educational Resources Information Center

Lindahl, Mikael

2005-01-01

A new approach is presented to analyze if there is a causal effect or relationship between income and measures of good health and life expectancy. One of the findings is that winning monetary lotteries could improve general health by 3 percent and decreased probability of death within five years by 2-3 percentage points. Higher income by 10…

Estimating the Effects of Parental Divorce and Death With Fixed Effects Models

PubMed Central

Amato, Paul R.; Anthony, Christopher J.

2014-01-01

The authors used child fixed effects models to estimate the effects of parental divorce and death on a variety of outcomes using 2 large national data sets: (a) the Early Childhood Longitudinal Study, Kindergarten Cohort (kindergarten through the 5th grade) and (b) the National Educational Longitudinal Study (8th grade to the senior year of high school). In both data sets, divorce and death were associated with multiple negative outcomes among children. Although evidence for a causal effect of divorce on children was reasonably strong, effect sizes were small in magnitude. A second analysis revealed a substantial degree of variability in children’s outcomes following parental divorce, with some children declining, others improving, and most not changing at all. The estimated effects of divorce appeared to be strongest among children with the highest propensity to experience parental divorce. PMID:24659827

Causal Effects of Single-Sex Schools on College Entrance Exams and College Attendance: Random Assignment in Seoul High Schools

PubMed Central

Park, Hyunjoon; Behrman, Jere R.; Choi, Jaesung

2012-01-01

Despite the voluminous literature on the potentials of single-sex schools, there is no consensus on the effects of single-sex schools because of student selection of school types. We exploit a unique feature of schooling in Seoul—the random assignment of students into single-sex versus coeducational high schools—to assess causal effects of single-sex schools on college entrance exam scores and college attendance. Our validation of the random assignment shows comparable socioeconomic backgrounds and prior academic achievement of students attending single-sex schools and coeducational schools, which increases the credibility of our causal estimates of single-sex school effects. The three-level hierarchical model shows that attending all-boys schools or all-girls schools, rather than coeducational schools, is significantly associated with higher average scores on Korean and English test scores. Applying the school district fixed-effects models, we find that single-sex schools produce a higher percentage of graduates who attended four-year colleges and a lower percentage of graduates who attended two-year junior colleges than do coeducational schools. The positive effects of single-sex schools remain substantial, even after we take into account various school-level variables, such as teacher quality, the student-teacher ratio, the proportion of students receiving lunch support, and whether the schools are public or private. PMID:23073751

Causal effects of single-sex schools on college entrance exams and college attendance: random assignment in Seoul high schools.

PubMed

Park, Hyunjoon; Behrman, Jere R; Choi, Jaesung

2013-04-01

Despite the voluminous literature on the potentials of single-sex schools, there is no consensus on the effects of single-sex schools because of student selection of school types. We exploit a unique feature of schooling in Seoul-the random assignment of students into single-sex versus coeducational high schools-to assess causal effects of single-sex schools on college entrance exam scores and college attendance. Our validation of the random assignment shows comparable socioeconomic backgrounds and prior academic achievement of students attending single-sex schools and coeducational schools, which increases the credibility of our causal estimates of single-sex school effects. The three-level hierarchical model shows that attending all-boys schools or all-girls schools, rather than coeducational schools, is significantly associated with higher average scores on Korean and English test scores. Applying the school district fixed-effects models, we find that single-sex schools produce a higher percentage of graduates who attended four-year colleges and a lower percentage of graduates who attended two-year junior colleges than do coeducational schools. The positive effects of single-sex schools remain substantial, even after we take into account various school-level variables, such as teacher quality, the student-teacher ratio, the proportion of students receiving lunch support, and whether the schools are public or private.

Causal Inference and the Comparative Interrupted Time Series Design: Findings from Within-Study Comparisons

ERIC Educational Resources Information Center

St. Clair, Travis; Hallberg, Kelly; Cook, Thomas D.

2014-01-01

Researchers are increasingly using comparative interrupted time series (CITS) designs to estimate the effects of programs and policies when randomized controlled trials are not feasible. In a simple interrupted time series design, researchers compare the pre-treatment values of a treatment group time series to post-treatment values in order to…

Required, Practical, or Unnecessary? An Examination and Demonstration of Propensity Score Matching Using Longitudinal Secondary Data

ERIC Educational Resources Information Center

Padgett, Ryan D.; Salisbury, Mark H.; An, Brian P.; Pascarella, Ernest T.

2010-01-01

The sophisticated analytical techniques available to institutional researchers give them an array of procedures to estimate a causal effect using observational data. But as many quantitative researchers have discovered, access to a wider selection of statistical tools does not necessarily ensure construction of a better analytical model. Moreover,…

The Impact of College on Migration: Evidence from the Vietnam Generation

ERIC Educational Resources Information Center

Malamud, Ofer; Wozniak, Abigail

2012-01-01

We examine the causal effect of education on migration using variation in college attainment due to draft-avoidance behavior during the Vietnam War. We use national and state-level induction risk to identify both college attainment and veteran status for men observed in the 1980 Census. 2SLS estimates imply that additional years of college…

A Causal Inference Analysis of the Effect of Wildland Fire Smoke on Ambient Air Pollution Levels and Health Burden

EPA Science Inventory

Wildfire smoke is a major contributor to ambient air pollution levels. In this talk, we develop a spatio-temporal model to estimate the contribution of fire smoke to overall air pollution in different regions of the country. We combine numerical model output with observational da...

New Estimates of Design Parameters for Clustered Randomization Studies: Findings from North Carolina and Florida. Working Paper 43

ERIC Educational Resources Information Center

Xu, Zeyu; Nichols, Austin

2010-01-01

The gold standard in making causal inference on program effects is a randomized trial. Most randomization designs in education randomize classrooms or schools rather than individual students. Such "clustered randomization" designs have one principal drawback: They tend to have limited statistical power or precision. This study aims to…

A Comparison of Model-Based and Design-Based Impact Evaluations of Interventions in Developing Countries

ERIC Educational Resources Information Center

Hansen, Henrik; Klejnstrup, Ninja Ritter; Andersen, Ole Winckler

2013-01-01

There is a long-standing debate as to whether nonexperimental estimators of causal effects of social programs can overcome selection bias. Most existing reviews either are inconclusive or point to significant selection biases in nonexperimental studies. However, many of the reviews, the so-called "between-studies," do not make direct…

The Impact of Employment during School on College Student Academic Performance. NBER Working Paper No. 14006

ERIC Educational Resources Information Center

DeSimone, Jeffrey S.

2008-01-01

This paper estimates the effect of paid employment on grades of full-time, four-year students from four nationally representative cross sections of the Harvard College Alcohol Study administered during 1993-2001. The relationship could be causal in either direction and is likely contaminated by unobserved heterogeneity. Two-stage GMM regressions…

A Tripartite Model of EFL Teacher Attributions, Burnout, and Self-Regulation: Toward the Prospects of Effective Teaching

ERIC Educational Resources Information Center

Ghanizadeh, Afsaneh; Ghonsooly, Behzad

2014-01-01

The present study aims at delving into English as foreign language teachers' attributions by investigating the role of teacher attributions in teacher burnout and teacher self-regulation. This is accomplished by building a causal structural model through which the associations among these constructs are estimated. The results demonstrate that…

Learning More from Educational Intervention Studies: Estimating Complier Average Causal Effects in a Relevance Intervention

ERIC Educational Resources Information Center

Nagengast, Benjamin; Brisson, Brigitte M.; Hulleman, Chris S.; Gaspard, Hanna; Häfner, Isabelle; Trautwein, Ulrich

2018-01-01

An emerging literature demonstrates that relevance interventions, which ask students to produce written reflections on how what they are learning relates to their lives, improve student learning outcomes. As part of a randomized evaluation of a relevance intervention (N = 1,978 students from 82 ninth-grade classes), we used Complier Average Causal…

Effect decomposition in the presence of an exposure-induced mediator-outcome confounder

PubMed Central

VanderWeele, Tyler J.; Vansteelandt, Stijn; Robins, James M.

2014-01-01

Methods from causal mediation analysis have generalized the traditional approach to direct and indirect effects in the epidemiologic and social science literature by allowing for interaction and non-linearities. However, the methods from the causal inference literature have themselves been subject to a major limitation in that the so-called natural direct and indirect effects that are employed are not identified from data whenever there is a variable that is affected by the exposure, which also confounds the relationship between the mediator and the outcome. In this paper we describe three alternative approaches to effect decomposition that give quantities that can be interpreted as direct and indirect effects, and that can be identified from data even in the presence of an exposure-induced mediator-outcome confounder. We describe a simple weighting-based estimation method for each of these three approaches, illustrated with data from perinatal epidemiology. The methods described here can shed insight into pathways and questions of mediation even when an exposure-induced mediator-outcome confounder is present. PMID:24487213

Impact of military on biofuels consumption and GHG emissions: the evidence from G7 countries.

PubMed

Bildirici, Melike

2018-05-01

It was aimed to test the relation among the greenhouse gases emissions, economic growth, biofuels consumption, and militarization in G7 countries during the 1985-2015 period by Pedroni 1995 and panel Johansen tests and two long-run estimators-dynamic OLS and fully modified OLS. Long-run estimators found that economic growth and militarization have statistically significant positive impact on CO 2 emission of G7 countries. Furthermore, the panel causality tests were applied: Dumitrescu and Hurlin (Econ Model 29(4):1450-1460, 2012) and panel Granger causality. These tests determined the causal relationship between the variables. The results of this paper implied that economic growth and biofuels consumption depend on militarization, and economic growth and militarization are granger causes of the greenhouse gases emissions.

Initial Validation for the Estimation of Resting-State fMRI Effective Connectivity by a Generalization of the Correlation Approach.

PubMed

Xu, Nan; Spreng, R Nathan; Doerschuk, Peter C

2017-01-01

Resting-state functional MRI (rs-fMRI) is widely used to noninvasively study human brain networks. Network functional connectivity is often estimated by calculating the timeseries correlation between blood-oxygen-level dependent (BOLD) signal from different regions of interest (ROIs). However, standard correlation cannot characterize the direction of information flow between regions. In this paper, we introduce and test a new concept, prediction correlation, to estimate effective connectivity in functional brain networks from rs-fMRI. In this approach, the correlation between two BOLD signals is replaced by a correlation between one BOLD signal and a prediction of this signal via a causal system driven by another BOLD signal. Three validations are described: (1) Prediction correlation performed well on simulated data where the ground truth was known, and outperformed four other methods. (2) On simulated data designed to display the "common driver" problem, prediction correlation did not introduce false connections between non-interacting driven ROIs. (3) On experimental data, prediction correlation recovered the previously identified network organization of human brain. Prediction correlation scales well to work with hundreds of ROIs, enabling it to assess whole brain interregional connectivity at the single subject level. These results provide an initial validation that prediction correlation can capture the direction of information flow and estimate the duration of extended temporal delays in information flow between regions of interest ROIs based on BOLD signal. This approach not only maintains the high sensitivity to network connectivity provided by the correlation analysis, but also performs well in the estimation of causal information flow in the brain.

A validation of dynamic causal modelling for 7T fMRI.

PubMed

Tak, S; Noh, J; Cheong, C; Zeidman, P; Razi, A; Penny, W D; Friston, K J

2018-07-15

There is growing interest in ultra-high field magnetic resonance imaging (MRI) in cognitive and clinical neuroscience studies. However, the benefits offered by higher field strength have not been evaluated in terms of effective connectivity and dynamic causal modelling (DCM). In this study, we address the validity of DCM for 7T functional MRI data at two levels. First, we evaluate the predictive validity of DCM estimates based upon 3T and 7T in terms of reproducibility. Second, we assess improvements in the efficiency of DCM estimates at 7T, in terms of the entropy of the posterior distribution over model parameters (i.e., information gain). Using empirical data recorded during fist-closing movements with 3T and 7T fMRI, we found a high reproducibility of average connectivity and condition-specific changes in connectivity - as quantified by the intra-class correlation coefficient (ICC = 0.862 and 0.936, respectively). Furthermore, we found that the posterior entropy of 7T parameter estimates was substantially less than that of 3T parameter estimates; suggesting the 7T data are more informative - and furnish more efficient estimates. In the framework of DCM, we treated field-dependent parameters for the BOLD signal model as free parameters, to accommodate fMRI data at 3T and 7T. In addition, we made the resting blood volume fraction a free parameter, because different brain regions can differ in their vascularization. In this paper, we showed DCM enables one to infer changes in effective connectivity from 7T data reliably and efficiently. Copyright © 2018 Elsevier B.V. All rights reserved.

Does Parental Employment Affect Children's Educational Attainment?

ERIC Educational Resources Information Center

Schildberg-Hoerisch, Hannah

2011-01-01

This paper analyzes whether there exists a causal relationship between parental employment and children's educational attainment. We address potential endogeneity problems due to (i) selection of parents in the labor market by estimating a model on sibling differences and (ii) reverse causality by focusing on parents' employment when children are…

Intimate Relationships and Depression: Is There a Causal Connection?

ERIC Educational Resources Information Center

Burns, David D.; And Others

1994-01-01

Estimated causal pathways that link depression and dissatisfaction in intimate relationships in 115 depressed patients during first 12 weeks of treatment. Depression severity, as measured by Beck Depression Inventory, was negatively correlated with relationship satisfaction at intake and at 12 weeks. Structural equation modeling was not consistent…

Context and Time in Causal Learning: Contingency and Mood Dependent Effects

PubMed Central

Msetfi, Rachel M.; Wade, Caroline; Murphy, Robin A.

2013-01-01

Defining cues for instrumental causality are the temporal, spatial and contingency relationships between actions and their effects. In this study, we carried out a series of causal learning experiments that systematically manipulated time and context in positive and negative contingency conditions. In addition, we tested participants categorized as non-dysphoric and mildly dysphoric because depressed mood has been shown to affect the processing of all these causal cues. Findings showed that causal judgements made by non-dysphoric participants were contextualized at baseline and were affected by the temporal spacing of actions and effects only with generative, but not preventative, contingency relationships. Participants categorized as dysphoric made less contextualized causal ratings at baseline but were more sensitive than others to temporal manipulations across the contingencies. These effects were consistent with depression affecting causal learning through the effects of slowed time experience on accrued exposure to the context in which causal events took place. Taken together, these findings are consistent with associative approaches to causal judgement. PMID:23691147

Smoking and caffeine consumption: a genetic analysis of their association.

PubMed

Treur, Jorien L; Taylor, Amy E; Ware, Jennifer J; Nivard, Michel G; Neale, Michael C; McMahon, George; Hottenga, Jouke-Jan; Baselmans, Bart M L; Boomsma, Dorret I; Munafò, Marcus R; Vink, Jacqueline M

2017-07-01

Smoking and caffeine consumption show a strong positive correlation, but the mechanism underlying this association is unclear. Explanations include shared genetic/environmental factors or causal effects. This study employed three methods to investigate the association between smoking and caffeine. First, bivariate genetic models were applied to data of 10 368 twins from the Netherlands Twin Register in order to estimate genetic and environmental correlations between smoking and caffeine use. Second, from the summary statistics of meta-analyses of genome-wide association studies on smoking and caffeine, the genetic correlation was calculated by LD-score regression. Third, causal effects were tested using Mendelian randomization analysis in 6605 Netherlands Twin Register participants and 5714 women from the Avon Longitudinal Study of Parents and Children. Through twin modelling, a genetic correlation of r0.47 and an environmental correlation of r0.30 were estimated between current smoking (yes/no) and coffee use (high/low). Between current smoking and total caffeine use, this was r0.44 and r0.00, respectively. LD-score regression also indicated sizeable genetic correlations between smoking and coffee use (r0.44 between smoking heaviness and cups of coffee per day, r0.28 between smoking initiation and coffee use and r0.25 between smoking persistence and coffee use). Consistent with the relatively high genetic correlations and lower environmental correlations, Mendelian randomization provided no evidence for causal effects of smoking on caffeine or vice versa. Genetic factors thus explain most of the association between smoking and caffeine consumption. These findings suggest that quitting smoking may be more difficult for heavy caffeine consumers, given their genetic susceptibility. © 2016 The Authors.Addiction Biology published by John Wiley & Sons Ltd on behalf of Society for the Study of Addiction.

Maternal smoking during pregnancy and offspring antisocial behaviour: findings from a longitudinal investigation of discordant siblings.

PubMed

Paradis, Angela D; Shenassa, Edmond D; Papandonatos, George D; Rogers, Michelle L; Buka, Stephen L

2017-09-01

Although many observational studies have found a strong association between maternal smoking during pregnancy (MSP) and offspring antisocial behaviour, the likelihood that this relationship is causal remains unclear. To comment on the potential causality of this association, the current investigation used a between-within decomposition approach to examine the association between MSP and multiple indices of adolescent and adult antisocial behaviour. Study participants were offspring of women enrolled in the Providence and Boston sites of the Collaborative Perinatal Project. Information on MSP was collected prospectively. Antisocial behaviour was assessed via self-report and through official records searches. A subset of the adult offspring (average age: 39.6 years) were enrolled in a follow-up study oversampling families with multiple siblings. Participants in this follow-up study self-reported on juvenile and adult antisocial behaviours during a structured interview (n=1684). Official records of juvenile (n=3447) and adult (n=3433) criminal behaviour were obtained for participants in the Providence cohort. Statistical models allowed between-family effects of MSP exposure to differ from within-family effects. In the absence of heterogeneity in between-family versus within-family estimates, a combined estimate was calculated. MSP was associated with a range of antisocial behaviours, measured by self-report and official records. For example, MSP was associated with increased odds of elevated levels of antisocial behaviours during adolescence and adulthood, as well as violent and non-violent outcomes during both developmental periods. Findings are consistent with a small-to-moderate causal effect of MSP on adolescent and adult antisocial behaviour. 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/.

Examining the Causal Role of Leptin in Alzheimer Disease: A Mendelian Randomization Study.

PubMed

Romo, Matthew L; Schooling, C Mary

2017-01-01

Observational evidence regarding the role of leptin in Alzheimer disease (AD) is conflicting. We sought to determine the causal role of circulating leptin and soluble plasma leptin receptor (sOB-R) levels in AD using a separate-sample Mendelian randomization study. Single nucleotide polymorphisms (SNPs) independently and solely predictive of log-transformed leptin (rs10487505 [LEP], rs780093 [GCKR], rs900400 [CCNL1], rs6071166 [SLC32A1], and rs6738627 [COBLL1]) and of sOB-R (rs1137101 [LEPR], rs2767485 [LEPR], and rs1751492 [LEPR]) levels (ng/mL) were obtained from 2 previously reported genome-wide association studies. We obtained associations of leptin and sOB-R levels with AD using inverse variance weighting with fixed effects by combining Wald estimates for each SNP. Sensitivity analyses included using weighted median and MR-Egger methods and repeating the analyses using only SNPs of genome-wide significance. Using inverse variance weighting, genetically predicted circulating leptin levels were not associated with AD, albeit with wide confidence intervals (CIs): odds ratio (OR) 0.99 per log-transformed ng/mL; 95% CI 0.55-1.78. Similarly, the association of sOB-R with AD was null using inverse variance weighting (OR 1.08 per log-transformed ng/mL; 95% CI 0.83-1.41). Results from our sensitivity analyses confirmed our findings. In this first Mendelian randomization study estimating the causal effect of leptin on AD, we did not find an effect of genetically predicted circulating leptin and sOB-R levels on AD. As such, this study suggests that leptin is unlikely to be a major contributor to AD, although the wide CIs preclude a definitive assessment. © 2017 S. Karger AG, Basel.

Smoking and caffeine consumption: a genetic analysis of their association

PubMed Central

Taylor, Amy E.; Ware, Jennifer J.; Nivard, Michel G.; Neale, Michael C.; McMahon, George; Hottenga, Jouke‐Jan; Baselmans, Bart M. L.; Boomsma, Dorret I.; Munafò, Marcus R.; Vink, Jacqueline M.

2016-01-01

Abstract Smoking and caffeine consumption show a strong positive correlation, but the mechanism underlying this association is unclear. Explanations include shared genetic/environmental factors or causal effects. This study employed three methods to investigate the association between smoking and caffeine. First, bivariate genetic models were applied to data of 10 368 twins from the Netherlands Twin Register in order to estimate genetic and environmental correlations between smoking and caffeine use. Second, from the summary statistics of meta‐analyses of genome‐wide association studies on smoking and caffeine, the genetic correlation was calculated by LD‐score regression. Third, causal effects were tested using Mendelian randomization analysis in 6605 Netherlands Twin Register participants and 5714 women from the Avon Longitudinal Study of Parents and Children. Through twin modelling, a genetic correlation of r0.47 and an environmental correlation of r0.30 were estimated between current smoking (yes/no) and coffee use (high/low). Between current smoking and total caffeine use, this was r0.44 and r0.00, respectively. LD‐score regression also indicated sizeable genetic correlations between smoking and coffee use (r0.44 between smoking heaviness and cups of coffee per day, r0.28 between smoking initiation and coffee use and r0.25 between smoking persistence and coffee use). Consistent with the relatively high genetic correlations and lower environmental correlations, Mendelian randomization provided no evidence for causal effects of smoking on caffeine or vice versa. Genetic factors thus explain most of the association between smoking and caffeine consumption. These findings suggest that quitting smoking may be more difficult for heavy caffeine consumers, given their genetic susceptibility. PMID:27027469

A Bayesian Theory of Sequential Causal Learning and Abstract Transfer.

PubMed

Lu, Hongjing; Rojas, Randall R; Beckers, Tom; Yuille, Alan L

2016-03-01

Two key research issues in the field of causal learning are how people acquire causal knowledge when observing data that are presented sequentially, and the level of abstraction at which learning takes place. Does sequential causal learning solely involve the acquisition of specific cause-effect links, or do learners also acquire knowledge about abstract causal constraints? Recent empirical studies have revealed that experience with one set of causal cues can dramatically alter subsequent learning and performance with entirely different cues, suggesting that learning involves abstract transfer, and such transfer effects involve sequential presentation of distinct sets of causal cues. It has been demonstrated that pre-training (or even post-training) can modulate classic causal learning phenomena such as forward and backward blocking. To account for these effects, we propose a Bayesian theory of sequential causal learning. The theory assumes that humans are able to consider and use several alternative causal generative models, each instantiating a different causal integration rule. Model selection is used to decide which integration rule to use in a given learning environment in order to infer causal knowledge from sequential data. Detailed computer simulations demonstrate that humans rely on the abstract characteristics of outcome variables (e.g., binary vs. continuous) to select a causal integration rule, which in turn alters causal learning in a variety of blocking and overshadowing paradigms. When the nature of the outcome variable is ambiguous, humans select the model that yields the best fit with the recent environment, and then apply it to subsequent learning tasks. Based on sequential patterns of cue-outcome co-occurrence, the theory can account for a range of phenomena in sequential causal learning, including various blocking effects, primacy effects in some experimental conditions, and apparently abstract transfer of causal knowledge. Copyright © 2015 Cognitive Science Society, Inc.

Exploring the relationship between child physical abuse and adult dating violence using a causal inference approach in an emerging adult population in South Korea.

PubMed

Jennings, Wesley G; Park, MiRang; Richards, Tara N; Tomsich, Elizabeth; Gover, Angela; Powers, Ráchael A

2014-12-01

Child maltreatment is one of the most commonly examined risk factors for violence in dating relationships. Often referred to as the intergenerational transmission of violence or cycle of violence, a fair amount of research suggests that experiencing abuse during childhood significantly increases the likelihood of involvement in violent relationships later, but these conclusions are primarily based on correlational research designs. Furthermore, the majority of research linking childhood maltreatment and dating violence has focused on samples of young people from the United States. Considering these limitations, the current study uses a rigorous, propensity score matching approach to estimate the causal effect of experiencing child physical abuse on adult dating violence among a large sample of South Korean emerging adults. Results indicate that the link between child physical abuse and adult dating violence is spurious rather than causal. Study limitations and implications are discussed. Copyright © 2014 Elsevier Ltd. All rights reserved.

Making stronger causal inferences: Accounting for selection bias in associations between high performance work systems, leadership, and employee and customer satisfaction.

PubMed

Schmidt, Joseph A; Pohler, Dionne M

2018-05-17

We develop competing hypotheses about the relationship between high performance work systems (HPWS) with employee and customer satisfaction. Drawing on 8 years of employee and customer survey data from a financial services firm, we used a recently developed empirical technique-covariate balanced propensity score (CBPS) weighting-to examine if the proposed relationships between HPWS and satisfaction outcomes can be explained by reverse causality, selection effects, or commonly omitted variables such as leadership behavior. The results provide support for leader behaviors as a primary driver of customer satisfaction, rather than HPWS, and also suggest that the problem of reverse causality requires additional attention in future human resource (HR) systems research. Model comparisons suggest that the estimates and conclusions vary across CBPS, meta-analytic, cross-sectional, and time-lagged models (with and without a lagged dependent variable as a control). We highlight the theoretical and methodological implications of the findings for HR systems research. (PsycINFO Database Record (c) 2018 APA, all rights reserved).

Integrated traffic conflict model for estimating crash modification factors.

PubMed

Shahdah, Usama; Saccomanno, Frank; Persaud, Bhagwant

2014-10-01

Crash modification factors (CMFs) for road safety treatments are usually obtained through observational models based on reported crashes. Observational Bayesian before-and-after methods have been applied to obtain more precise estimates of CMFs by accounting for the regression-to-the-mean bias inherent in naive methods. However, sufficient crash data reported over an extended period of time are needed to provide reliable estimates of treatment effects, a requirement that can be a challenge for certain types of treatment. In addition, these studies require that sites analyzed actually receive the treatment to which the CMF pertains. Another key issue with observational approaches is that they are not causal in nature, and as such, cannot provide a sound "behavioral" rationale for the treatment effect. Surrogate safety measures based on high risk vehicle interactions and traffic conflicts have been proposed to address this issue by providing a more "causal perspective" on lack of safety for different road and traffic conditions. The traffic conflict approach has been criticized, however, for lacking a formal link to observed and verified crashes, a difficulty that this paper attempts to resolve by presenting and investigating an alternative approach for estimating CMFs using simulated conflicts that are linked formally to observed crashes. The integrated CMF estimates are compared to estimates from an empirical Bayes (EB) crash-based before-and-after analysis for the same sample of treatment sites. The treatment considered involves changing left turn signal priority at Toronto signalized intersections from permissive to protected-permissive. The results are promising in that the proposed integrated method yields CMFs that closely match those obtained from the crash-based EB before-and-after analysis. Copyright © 2014 Elsevier Ltd. All rights reserved.

Effective brain network analysis with resting-state EEG data: a comparison between heroin abstinent and non-addicted subjects

NASA Astrophysics Data System (ADS)

Hu, Bin; Dong, Qunxi; Hao, Yanrong; Zhao, Qinglin; Shen, Jian; Zheng, Fang

2017-08-01

Objective. Neuro-electrophysiological tools have been widely used in heroin addiction studies. Previous studies indicated that chronic heroin abuse would result in abnormal functional organization of the brain, while few heroin addiction studies have applied the effective connectivity tool to analyze the brain functional system (BFS) alterations induced by heroin abuse. The present study aims to identify the abnormality of resting-state heroin abstinent BFS using source decomposition and effective connectivity tools. Approach. The resting-state electroencephalograph (EEG) signals were acquired from 15 male heroin abstinent (HA) subjects and 14 male non-addicted (NA) controls. Multivariate autoregressive models combined independent component analysis (MVARICA) was applied for blind source decomposition. Generalized partial directed coherence (GPDC) was applied for effective brain connectivity analysis. Effective brain networks of both HA and NA groups were constructed. The two groups of effective cortical networks were compared by the bootstrap method. Abnormal causal interactions between decomposed source regions were estimated in the 1-45 Hz frequency domain. Main results. This work suggested: (a) there were clear effective network alterations in heroin abstinent subject groups; (b) the parietal region was a dominant hub of the abnormally weaker causal pathways, and the left occipital region was a dominant hub of the abnormally stronger causal pathways. Significance. These findings provide direct evidence that chronic heroin abuse induces brain functional abnormalities. The potential value of combining effective connectivity analysis and brain source decomposition methods in exploring brain alterations of heroin addicts is also implied.

Effective brain network analysis with resting-state EEG data: a comparison between heroin abstinent and non-addicted subjects.

PubMed

Hu, Bin; Dong, Qunxi; Hao, Yanrong; Zhao, Qinglin; Shen, Jian; Zheng, Fang

2017-08-01

Neuro-electrophysiological tools have been widely used in heroin addiction studies. Previous studies indicated that chronic heroin abuse would result in abnormal functional organization of the brain, while few heroin addiction studies have applied the effective connectivity tool to analyze the brain functional system (BFS) alterations induced by heroin abuse. The present study aims to identify the abnormality of resting-state heroin abstinent BFS using source decomposition and effective connectivity tools. The resting-state electroencephalograph (EEG) signals were acquired from 15 male heroin abstinent (HA) subjects and 14 male non-addicted (NA) controls. Multivariate autoregressive models combined independent component analysis (MVARICA) was applied for blind source decomposition. Generalized partial directed coherence (GPDC) was applied for effective brain connectivity analysis. Effective brain networks of both HA and NA groups were constructed. The two groups of effective cortical networks were compared by the bootstrap method. Abnormal causal interactions between decomposed source regions were estimated in the 1-45 Hz frequency domain. This work suggested: (a) there were clear effective network alterations in heroin abstinent subject groups; (b) the parietal region was a dominant hub of the abnormally weaker causal pathways, and the left occipital region was a dominant hub of the abnormally stronger causal pathways. These findings provide direct evidence that chronic heroin abuse induces brain functional abnormalities. The potential value of combining effective connectivity analysis and brain source decomposition methods in exploring brain alterations of heroin addicts is also implied.

Can Women Have Children and a Career? IV Evidence from IVF Treatments.

PubMed

Lundborg, Petter; Plug, Erik; Rasmussen, Astrid Wurtz

2017-06-01

This paper introduces a new IV strategy based on IVF (in vitro fertilization) induced fertility variation among childless women to estimate the causal effect of having children on their career. For this purpose, we use administrative data on IVF treated women in Denmark. Because observed chances of IVF success do not depend on labor market histories, IVF treatment success provides a plausible instrument for childbearing. Our IV estimates indicate that fertility effects on earnings are: (i) negative, large, and long-lasting; (ii) driven by fertility effects on hourly earnings and not so much on labor supply; and (iii) much stronger at the extensive margin than at the intensive margin.

The Effect of Private School Vouchers on Political Participation: Experimental Evidence from New York City. Program on Education Policy and Governance Working Paper Series. PEPG 16-01

ERIC Educational Resources Information Center

Carlson, Deven; Chingos, Matthew M.; Campbell, David E.

2016-01-01

In 1997, the New York School Choice Scholarships Foundation Program (SCSF) randomly offered three-year scholarships to attend private schools to approximately 1,000 low-income families in New York City. In this paper we leverage exogenous variation generated by the SCSF to estimate the causal effect of the private school voucher offer--and the…

Exposure to traffic-related air pollution during pregnancy and term low birth weight: estimation of causal associations in a semiparametric model.

PubMed

Padula, Amy M; Mortimer, Kathleen; Hubbard, Alan; Lurmann, Frederick; Jerrett, Michael; Tager, Ira B

2012-11-01

Traffic-related air pollution is recognized as an important contributor to health problems. Epidemiologic analyses suggest that prenatal exposure to traffic-related air pollutants may be associated with adverse birth outcomes; however, there is insufficient evidence to conclude that the relation is causal. The Study of Air Pollution, Genetics and Early Life Events comprises all births to women living in 4 counties in California's San Joaquin Valley during the years 2000-2006. The probability of low birth weight among full-term infants in the population was estimated using machine learning and targeted maximum likelihood estimation for each quartile of traffic exposure during pregnancy. If everyone lived near high-volume freeways (approximated as the fourth quartile of traffic density), the estimated probability of term low birth weight would be 2.27% (95% confidence interval: 2.16, 2.38) as compared with 2.02% (95% confidence interval: 1.90, 2.12) if everyone lived near smaller local roads (first quartile of traffic density). Assessment of potentially causal associations, in the absence of arbitrary model assumptions applied to the data, should result in relatively unbiased estimates. The current results support findings from previous studies that prenatal exposure to traffic-related air pollution may adversely affect birth weight among full-term infants.

Surrogacy Assessment Using Principal Stratification and a Gaussian Copula Model

PubMed Central

Taylor, J.M.G.; Elliott, M.R.

2014-01-01

In clinical trials, a surrogate outcome (S) can be measured before the outcome of interest (T) and may provide early information regarding the treatment (Z) effect on T. Many methods of surrogacy validation rely on models for the conditional distribution of T given Z and S. However, S is a post-randomization variable, and unobserved, simultaneous predictors of S and T may exist, resulting in a non-causal interpretation. Frangakis and Rubin1 developed the concept of principal surrogacy, stratifying on the joint distribution of the surrogate marker under treatment and control to assess the association between the causal effects of treatment on the marker and the causal effects of treatment on the clinical outcome. Working within the principal surrogacy framework, we address the scenario of an ordinal categorical variable as a surrogate for a censored failure time true endpoint. A Gaussian copula model is used to model the joint distribution of the potential outcomes of T, given the potential outcomes of S. Because the proposed model cannot be fully identified from the data, we use a Bayesian estimation approach with prior distributions consistent with reasonable assumptions in the surrogacy assessment setting. The method is applied to data from a colorectal cancer clinical trial, previously analyzed by Burzykowski et al..2 PMID:24947559

A counterfactual impact evaluation of a bilingual program on students' grade point average at a spanish university.

PubMed

Arco-Tirado, J L; Fernández-Martín, F; Ramos-García, A M; Littvay, L; Villoria, J; Naranjo, J A

2018-06-01

This observational study intends to estimate the causal effects of an English as a Medium of Instruction (EMI) program (as predictor) on students Grade Point Average (GPA) (as outcome) at a particular University in Spain by using a Counterfactual Impact Evaluation (CIE). The need to address the crucial question of causal inferences in EMI programs to produce credible evidences of successful interventions contrasts, however, with the absence of experimental or quasi-experimental research and evaluation designs in the field. CIE approach is emerging as a methodologically viable solution to bridge that gap. The program evaluated here consisted in delivering an EMI program in a Primary Education Teacher Training Degree group. After achieving balance on the observed covariates and recreating a situation that would have been expected in a randomized experiment, three matching approaches such as genetic matching, nearest neighbor matching and Coarsened Exact Matching were used to analyze observational data from a total of 1288 undergraduate students, including both treatment and control group. Results show unfavorable effects of the bilingual group treatment condition. Potential interpretations and recommendations are provided in order to strengthen future causal evidences of bilingual education programs' effectiveness in Higher Education. Copyright © 2018 The Authors. Published by Elsevier Ltd.. All rights reserved.

Surrogacy assessment using principal stratification and a Gaussian copula model.

PubMed

Conlon, Asc; Taylor, Jmg; Elliott, M R

2017-02-01

In clinical trials, a surrogate outcome ( S) can be measured before the outcome of interest ( T) and may provide early information regarding the treatment ( Z) effect on T. Many methods of surrogacy validation rely on models for the conditional distribution of T given Z and S. However, S is a post-randomization variable, and unobserved, simultaneous predictors of S and T may exist, resulting in a non-causal interpretation. Frangakis and Rubin developed the concept of principal surrogacy, stratifying on the joint distribution of the surrogate marker under treatment and control to assess the association between the causal effects of treatment on the marker and the causal effects of treatment on the clinical outcome. Working within the principal surrogacy framework, we address the scenario of an ordinal categorical variable as a surrogate for a censored failure time true endpoint. A Gaussian copula model is used to model the joint distribution of the potential outcomes of T, given the potential outcomes of S. Because the proposed model cannot be fully identified from the data, we use a Bayesian estimation approach with prior distributions consistent with reasonable assumptions in the surrogacy assessment setting. The method is applied to data from a colorectal cancer clinical trial, previously analyzed by Burzykowski et al.

Dynamic Granger causality based on Kalman filter for evaluation of functional network connectivity in fMRI data

PubMed Central

Havlicek, Martin; Jan, Jiri; Brazdil, Milan; Calhoun, Vince D.

2015-01-01

Increasing interest in understanding dynamic interactions of brain neural networks leads to formulation of sophisticated connectivity analysis methods. Recent studies have applied Granger causality based on standard multivariate autoregressive (MAR) modeling to assess the brain connectivity. Nevertheless, one important flaw of this commonly proposed method is that it requires the analyzed time series to be stationary, whereas such assumption is mostly violated due to the weakly nonstationary nature of functional magnetic resonance imaging (fMRI) time series. Therefore, we propose an approach to dynamic Granger causality in the frequency domain for evaluating functional network connectivity in fMRI data. The effectiveness and robustness of the dynamic approach was significantly improved by combining a forward and backward Kalman filter that improved estimates compared to the standard time-invariant MAR modeling. In our method, the functional networks were first detected by independent component analysis (ICA), a computational method for separating a multivariate signal into maximally independent components. Then the measure of Granger causality was evaluated using generalized partial directed coherence that is suitable for bivariate as well as multivariate data. Moreover, this metric provides identification of causal relation in frequency domain, which allows one to distinguish the frequency components related to the experimental paradigm. The procedure of evaluating Granger causality via dynamic MAR was demonstrated on simulated time series as well as on two sets of group fMRI data collected during an auditory sensorimotor (SM) or auditory oddball discrimination (AOD) tasks. Finally, a comparison with the results obtained from a standard time-invariant MAR model was provided. PMID:20561919

Modelling the impact of causal and non-causal factors on disruption duration for Toronto's subway system: An exploratory investigation using hazard modelling.

PubMed

Louie, Jacob; Shalaby, Amer; Habib, Khandker Nurul

2017-01-01

Most investigations of incident-related delay duration in the transportation context are restricted to highway traffic, with little attention given to delays due to transit service disruptions. Studies of transit-based delay duration are also considerably less comprehensive than their highway counterparts with respect to examining the effects of non-causal variables on the delay duration. However, delays due to incidents in public transit service can have serious consequences on the overall urban transportation system due to the pivotal and vital role of public transit. The ability to predict the durations of various types of transit system incidents is indispensable for better management and mitigation of service disruptions. This paper presents a detailed investigation on incident delay durations in Toronto's subway system over the year 2013, focusing on the effects of the incidents' location and time, the train-type involved, and the non-adherence to proper recovery procedures. Accelerated Failure Time (AFT) hazard models are estimated to investigate the relationship between these factors and the resulting delay duration. The empirical investigation reveals that incident types that impact both safety and operations simultaneously generally have longer expected delays than incident types that impact either safety or operations alone. Incidents at interchange stations are cleared faster than incidents at non-interchange stations. Incidents during peak periods have nearly the same delay durations as off-peak incidents. The estimated models are believed to be useful tools in predicting the relative magnitude of incident delay duration for better management of subway operations. Copyright © 2016 Elsevier Ltd. All rights reserved.

Inferences about unobserved causes in human contingency learning.

PubMed

Hagmayer, York; Waldmann, Michael R

2007-03-01

Estimates of the causal efficacy of an event need to take into account the possible presence and influence of other unobserved causes that might have contributed to the occurrence of the effect. Current theoretical approaches deal differently with this problem. Associative theories assume that at least one unobserved cause is always present. In contrast, causal Bayes net theories (including Power PC theory) hypothesize that unobserved causes may be present or absent. These theories generally assume independence of different causes of the same event, which greatly simplifies modelling learning and inference. In two experiments participants were requested to learn about the causal relation between a single cause and an effect by observing their co-occurrence (Experiment 1) or by actively intervening in the cause (Experiment 2). Participants' assumptions about the presence of an unobserved cause were assessed either after each learning trial or at the end of the learning phase. The results show an interesting dissociation. Whereas there was a tendency to assume interdependence of the causes in the online judgements during learning, the final judgements tended to be more in the direction of an independence assumption. Possible explanations and implications of these findings are discussed.

Distinguishing anticipation from causality: anticipatory bias in the estimation of information flow.

PubMed

Hahs, Daniel W; Pethel, Shawn D

2011-09-16

We report that transfer entropy estimates obtained from low-resolution and/or small data sets show net information flow away from a purely anticipatory element whereas transfer entropy calculated using exact distributions show the flow towards it. This means that for real-world data sets anticipatory elements can appear to be strongly driving the network dynamics even when there is no possibility of such an influence. Furthermore, we show that in the low-resolution limit there is no statistic that can distinguish anticipatory elements from causal ones.

The role of education in the production of health: an empirical analysis of smoking behavior.

PubMed

Tenn, Steven; Herman, Douglas A; Wendling, Brett

2010-05-01

We estimate the effect of education on smoking. Our estimation strategy "differences out" the impact of unobserved characteristics correlated with education by exploiting education differences between similarly selected groups 1 year apart in their life cycle. Individuals with a given age, education, and student status in the current and previous year are compared to their counterparts born 1 year later with the same age, education, and student status in the following and current year. We find that an additional year of education does not have a causal effect on smoking. Unobserved factors correlated with education entirely explain their cross-sectional relationship.

Visceral fat estimation method by bioelectrical impedance analysis and causal analysis

NASA Astrophysics Data System (ADS)

Nakajima, Hiroshi; Tasaki, Hiroshi; Tsuchiya, Naoki; Hamaguchi, Takehiro; Shiga, Toshikazu

2011-06-01

It has been clarified that abdominal visceral fat accumulation is closely associated to the lifestyle disease and metabolic syndrome. The gold standard in medical fields is visceral fat area measured by an X-ray computer tomography (CT) scan or magnetic resonance imaging. However, their measurements are high invasive and high cost; especially a CT scan causes X-ray exposure. They are the reasons why medical fields need an instrument for viscera fat measurement with low invasive, ease of use, and low cost. The article proposes a simple and practical method of visceral fat estimation by employing bioelectrical impedance analysis and causal analysis. In the method, abdominal shape and dual impedances of abdominal surface and body total are measured to estimate a visceral fat area based on the cause-effect structure. The structure is designed according to the nature of abdominal body composition to be fine-tuned by statistical analysis. The experiments were conducted to investigate the proposed model. 180 subjects were hired to be measured by both a CT scan and the proposed method. The acquired model explained the measurement principle well and the correlation coefficient is 0.88 with the CT scan measurements.

Transfer Entropy as a Log-Likelihood Ratio

NASA Astrophysics Data System (ADS)

Barnett, Lionel; Bossomaier, Terry

2012-09-01

Transfer entropy, an information-theoretic measure of time-directed information transfer between joint processes, has steadily gained popularity in the analysis of complex stochastic dynamics in diverse fields, including the neurosciences, ecology, climatology, and econometrics. We show that for a broad class of predictive models, the log-likelihood ratio test statistic for the null hypothesis of zero transfer entropy is a consistent estimator for the transfer entropy itself. For finite Markov chains, furthermore, no explicit model is required. In the general case, an asymptotic χ2 distribution is established for the transfer entropy estimator. The result generalizes the equivalence in the Gaussian case of transfer entropy and Granger causality, a statistical notion of causal influence based on prediction via vector autoregression, and establishes a fundamental connection between directed information transfer and causality in the Wiener-Granger sense.

Transfer entropy as a log-likelihood ratio.

PubMed

Barnett, Lionel; Bossomaier, Terry

2012-09-28

Transfer entropy, an information-theoretic measure of time-directed information transfer between joint processes, has steadily gained popularity in the analysis of complex stochastic dynamics in diverse fields, including the neurosciences, ecology, climatology, and econometrics. We show that for a broad class of predictive models, the log-likelihood ratio test statistic for the null hypothesis of zero transfer entropy is a consistent estimator for the transfer entropy itself. For finite Markov chains, furthermore, no explicit model is required. In the general case, an asymptotic χ2 distribution is established for the transfer entropy estimator. The result generalizes the equivalence in the Gaussian case of transfer entropy and Granger causality, a statistical notion of causal influence based on prediction via vector autoregression, and establishes a fundamental connection between directed information transfer and causality in the Wiener-Granger sense.

Is High-Density Lipoprotein Cholesterol Causally Related to Kidney Function? Evidence From Genetic Epidemiological Studies.

PubMed

Coassin, Stefan; Friedel, Salome; Köttgen, Anna; Lamina, Claudia; Kronenberg, Florian

2016-11-01

A recent observational study with almost 2 million men reported an association between low high-density lipoprotein (HDL) cholesterol and worse kidney function. The causality of this association would be strongly supported if genetic variants associated with HDL cholesterol were also associated with kidney function. We used 68 genetic variants (single-nucleotide polymorphisms [SNPs]) associated with HDL cholesterol in genome-wide association studies including >188 000 subjects and tested their association with estimated glomerular filtration rate (eGFR) using summary statistics from another genome-wide association studies meta-analysis of kidney function including ≤133 413 subjects. Fourteen of the 68 SNPs (21%) had a P value <0.05 compared with the 5% expected by chance (Binomial test P=5.8×10 - 6 ). After Bonferroni correction, 6 SNPs were still significantly associated with eGFR. The genetic variants with the strongest associations with HDL cholesterol concentrations were not the same as those with the strongest association with kidney function and vice versa. An evaluation of pleiotropy indicated that the effects of the HDL-associated SNPs on eGFR were not mediated by HDL cholesterol. In addition, we performed a Mendelian randomization analysis. This analysis revealed a positive but nonsignificant causal effect of HDL cholesterol-increasing variants on eGFR. In summary, our findings indicate that HDL cholesterol does not causally influence eGFR and propose pleiotropic effects on eGFR for some HDL cholesterol-associated SNPs. This may cause the observed association by mechanisms other than the mere HDL cholesterol concentration. © 2016 The Authors.

Empirical Benchmarks of Hidden Bias in Educational Research: Implication for Assessing How well Propensity Score Methods Approximate Experiments and Conducting Sensitivity Analysis

ERIC Educational Resources Information Center

Dong, Nianbo; Lipsey, Mark

2014-01-01

When randomized control trials (RCT) are not feasible, researchers seek other methods to make causal inference, e.g., propensity score methods. One of the underlined assumptions for the propensity score methods to obtain unbiased treatment effect estimates is the ignorability assumption, that is, conditional on the propensity score, treatment…

Education for All: Quasi-Experimental Estimates of the Impacts of Compulsory Primary Education in Hong Kong

ERIC Educational Resources Information Center

Ou, Dongshu

2013-01-01

This paper examines the causal impacts of Hong Kong's 1971 policy of free compulsory education on students' educational attainment. Using a regression discontinuity method and Hong Kong Census data, this study compares children born just before and just after the month in which the compulsory-education law came into effect. The results show that…

Building Better Bridges to Life after High School: Experimental Evidence on Contemporary Career Academies. Working Paper 176

ERIC Educational Resources Information Center

Hemelt, Steven W.; Lenard, Matthew A.; Paeplow, Colleen G.

2017-01-01

Modern career academies aim to prepare students for college and the labor market. This paper examines the profile of students entering such academies in one school district and estimates causal effects of participation in one of the district's well-regarded academies on a range of high school and college outcomes. Using rich administrative data…

Implementation of Instrumental Variable Bounds for Data Missing Not at Random.

PubMed

Marden, Jessica R; Wang, Linbo; Tchetgen, Eric J Tchetgen; Walter, Stefan; Glymour, M Maria; Wirth, Kathleen E

2018-05-01

Instrumental variables are routinely used to recover a consistent estimator of an exposure causal effect in the presence of unmeasured confounding. Instrumental variable approaches to account for nonignorable missing data also exist but are less familiar to epidemiologists. Like instrumental variables for exposure causal effects, instrumental variables for missing data rely on exclusion restriction and instrumental variable relevance assumptions. Yet these two conditions alone are insufficient for point identification. For estimation, researchers have invoked a third assumption, typically involving fairly restrictive parametric constraints. Inferences can be sensitive to these parametric assumptions, which are typically not empirically testable. The purpose of our article is to discuss another approach for leveraging a valid instrumental variable. Although the approach is insufficient for nonparametric identification, it can nonetheless provide informative inferences about the presence, direction, and magnitude of selection bias, without invoking a third untestable parametric assumption. An important contribution of this article is an Excel spreadsheet tool that can be used to obtain empirical evidence of selection bias and calculate bounds and corresponding Bayesian 95% credible intervals for a nonidentifiable population proportion. For illustrative purposes, we used the spreadsheet tool to analyze HIV prevalence data collected by the 2007 Zambia Demographic and Health Survey (DHS).

The Use of Field and Mesocosm Experiments to Quantify Effects of Physical and Chemical Stressors in Mining-Contaminated Streams.

PubMed

Cadmus, Pete; Clements, William H; Williamson, Jacob L; Ranville, James F; Meyer, Joseph S; Gutiérrez Ginés, María Jesús

2016-07-19

Identifying causal relationships between acid mine drainage (AMD) and ecological responses in the field is challenging. In addition to the direct toxicological effects of elevated metals and reduced pH, mining activities influence aquatic organisms indirectly through physical alterations of habitat. The primary goal of this research was to quantify the relative importance of physical (metal-oxide deposition) and chemical (elevated metal concentrations) stressors on benthic macroinvertebrate communities. Mesocosm experiments conducted with natural assemblages of benthic macroinvertebrates established concentration-response relationships between metals and community structure. Field experiments quantified effects of metal-oxide contaminated substrate and showed significant differences in sensitivity among taxa. To predict the recovery of dominant taxa in the field, we integrated our measures of metal tolerance and substrate tolerance with estimates of drift propensity obtained from the literature. Our estimates of recovery were consistent with patterns observed at downstream recovery sites in the NFCC, which were dominated by caddisflies and baetid mayflies. We conclude that mesocosm and small-scale field experiments, particularly those conducted with natural communities, provide an ecologically realistic complement to laboratory toxicity tests. These experiments also control for the confounding variables associated with field-based approaches, thereby supporting causal relationships between AMD stressors and responses.

Public beliefs about and attitudes towards bipolar disorder: testing theory based models of stigma.

PubMed

Ellison, Nell; Mason, Oliver; Scior, Katrina

2015-04-01

Given the vast literature into public beliefs and attitudes towards schizophrenia and depression, there is paucity of research on attitudes towards bipolar disorder despite its similar prevalence to schizophrenia. This study explored public beliefs and attitudes towards bipolar disorder and examined the relationship between these different components of stigma. Using an online questionnaire distributed via email, social networking sites and public institutions, 753 members of the UK population were presented with a vignette depicting someone who met DSM-IV criteria for bipolar disorder. Causal beliefs, beliefs about prognosis, emotional reactions, stereotypes, and social distance were assessed in response to the vignette. Preacher and Hayes procedure for estimating direct and indirect effects of multiple mediators was used to examine the relationship between these components of stigma. Bipolar disorder was primarily associated with positive beliefs and attitudes and elicited a relatively low desire for social distance. Fear partially mediated the relationship between stereotypes and social distance. Biomedical causal beliefs reduced desire for social distance by increasing compassion, whereas fate causal beliefs increased it through eliciting fear. Psychosocial causal beliefs had mixed effects. The measurement of stigma using vignettes and self-report questionnaires has implications for ecological validity and participants may have been reluctant to reveal the true extent of their negative attitudes. Dissemination of these findings to people with bipolar disorder has implications for the reduction of internalised stigma in this population. Anti-stigma campaigns should attend to causal beliefs, stereotypes and emotional reactions as these all play a vital role in discriminatory behaviour towards people with bipolar disorder. Copyright © 2015 Elsevier B.V. All rights reserved.

Effects of urban sprawl and vehicle miles traveled on traffic fatalities.

PubMed

Yeo, Jiho; Park, Sungjin; Jang, Kitae

2015-01-01

Previous research suggests that urban sprawl increases auto-dependency and that excessive auto use increases the risk of traffic fatalities. This indirect effect of urban sprawl on traffic fatalities is compared to non-vehicle miles traveled (VMT)-related direct effect of sprawl on fatalities. We conducted a path analysis to examine the causal linkages among urban sprawl, VMT, traffic fatalities, income, and fuel cost. The path diagram includes 2 major linkages: the direct relationship between urban sprawl and traffic fatalities and the indirect effect on fatalities through increased VMT in sprawling urban areas. To measure the relative strength of these causal linkages, path coefficients are estimated using data collected nationally from 147 urbanized areas in the United States. Through both direct and indirect paths, urban sprawl is associated with greater numbers of traffic fatalities, but the direct effect of sprawl on fatalities is more influential than the indirect effect. Enhancing traffic safety can be achieved by impeding urban sprawl and encouraging compact development. On the other hand, policy tools reducing VMT may be less effective than anticipated for traffic safety.

Operational estimates of areal evapotranspiration and their significance to the science and practice of hydrology

NASA Astrophysics Data System (ADS)

Morton, F. I.

1983-10-01

Reliable estimates of areal evapotranspiration are essential to significant improvements in the science and practice of hydrology. Direct measurements, such as those provided by lysimeters, eddy flux instrumentation or Bowen-ratio instrumentation, give point values, require constant attendance by skilled personnel and are based on unverified assumptions. A critical review of the methods used for estimating areal evapotranspiration indicates that the conventional conceptual techniques, such as those used in current watershed models, are based on assumptions that are completely divorced from reality; and that causal techniques based on processes and interactions in the soil-plant-atmosphere system are not likely to prove useful for another generation. However, the complementary relationship can do much to fill the gap until such time as causal techniques become practicable because it provides the basis for models that permit areal evapotranspiration to be estimated from its effects on the routine climatological observations needed to estimate potential evapotranspiration. Such models have a realistic conceptual and empirical basis, by-pass the complexity of the soil-plant system and require no local calibration of coefficients. Therefore, they are falsifiable (i.e. can be tested rigorously) so that errors in the associated assumptions and relationships can be detected and corrected by progressive testing over an ever-widening range of environments. Such a methodology uses the entire world as a laboratory and requires that a correction made to obtain agreement between model and river-basin water budget estimates in one environment must be applicable without modification in all other environment. The most recent version of the complementary relationship areal evapotranspiration (CRAE) models is formulated and documented. The reliability of the independent operational estimates of areal evapotranspiration is tested with comparable long-term water-budget estimates for 143 river basins in North America, Africa, Ireland, Australia and New Zealand. The practicality and potential impact of such estimates are demonstrated with examples which show how the availability of such estimates can revitalize the science and practice of hydrology by providing a reliable basis for detailed water-balance studies; for further research on the development of causal models; for hydrological, agricultural and fire hazard forecasts; for detecting the development of errors in hydrometeorological records; for detecting and monitoring the effects of land-use changes; for explaining hydrologic anomalies; and for other better known applications. It is suggested that the collection of the required climatological data by hydrometric agencies could be justified on the grounds that the agencies would gain a technique for quality control and the users would gain by a significant expansion in the information content of the hydrometric data, all at minimal additional expense.

Causal mediation analysis with a binary outcome and multiple continuous or ordinal mediators: Simulations and application to an alcohol intervention.

PubMed

Nguyen, Trang Quynh; Webb-Vargas, Yenny; Koning, Ina M; Stuart, Elizabeth A

We investigate a method to estimate the combined effect of multiple continuous/ordinal mediators on a binary outcome: 1) fit a structural equation model with probit link for the outcome and identity/probit link for continuous/ordinal mediators, 2) predict potential outcome probabilities, and 3) compute natural direct and indirect effects. Step 2 involves rescaling the latent continuous variable underlying the outcome to address residual mediator variance/covariance. We evaluate the estimation of risk-difference- and risk-ratio-based effects (RDs, RRs) using the ML, WLSMV and Bayes estimators in Mplus. Across most variations in path-coefficient and mediator-residual-correlation signs and strengths, and confounding situations investigated, the method performs well with all estimators, but favors ML/WLSMV for RDs with continuous mediators, and Bayes for RRs with ordinal mediators. Bayes outperforms WLSMV/ML regardless of mediator type when estimating RRs with small potential outcome probabilities and in two other special cases. An adolescent alcohol prevention study is used for illustration.

Sensitivity analysis for mistakenly adjusting for mediators in estimating total effect in observational studies.

PubMed

Wang, Tingting; Li, Hongkai; Su, Ping; Yu, Yuanyuan; Sun, Xiaoru; Liu, Yi; Yuan, Zhongshang; Xue, Fuzhong

2017-11-20

In observational studies, epidemiologists often attempt to estimate the total effect of an exposure on an outcome of interest. However, when the underlying diagram is unknown and limited knowledge is available, dissecting bias performances is essential to estimating the total effect of an exposure on an outcome when mistakenly adjusting for mediators under logistic regression. Through simulation, we focused on six causal diagrams concerning different roles of mediators. Sensitivity analysis was conducted to assess the bias performances of varying across exposure-mediator effects and mediator-outcome effects when adjusting for the mediator. Based on the causal relationships in the real world, we compared the biases of varying across the effects of exposure-mediator with those of varying across the effects of mediator-outcome when adjusting for the mediator. The magnitude of the bias was defined by the difference between the estimated effect (using logistic regression) and the total effect of the exposure on the outcome. In four scenarios (a single mediator, two series mediators, two independent parallel mediators or two correlated parallel mediators), the biases of varying across the effects of exposure-mediator were greater than those of varying across the effects of mediator-outcome when adjusting for the mediator. In contrast, in two other scenarios (a single mediator or two independent parallel mediators in the presence of unobserved confounders), the biases of varying across the effects of exposure-mediator were less than those of varying across the effects of mediator-outcome when adjusting for the mediator. The biases were more sensitive to the variation of effects of exposure-mediator than the effects of mediator-outcome when adjusting for the mediator in the absence of unobserved confounders, while the biases were more sensitive to the variation of effects of mediator-outcome than those of exposure-mediator in the presence of an unobserved confounder. © 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.

Health Insurance and Health Status: Exploring the Causal Effect from a Policy Intervention.

PubMed

Pan, Jay; Lei, Xiaoyan; Liu, Gordon G

2016-11-01

Whether health insurance matters for health has long been a central issue for debate when assessing the full value of health insurance coverage in both developed and developing countries. In 2007, the government-led Urban Resident Basic Medical Insurance (URBMI) program was piloted in China, followed by a nationwide implementation in 2009. Different premium subsidies by government across cities and groups provide a unique opportunity to employ the instrumental variables estimation approach to identify the causal effects of health insurance on health. Using a national panel survey of the URBMI, we find that URBMI beneficiaries experience statistically better health than the uninsured. Furthermore, the insurance health benefit appears to be stronger for groups with disadvantaged education and income than for their counterparts. In addition, the insured receive more and better inpatient care, without paying more for services. Copyright © 2015 John Wiley & Sons, Ltd. Copyright © 2015 John Wiley & Sons, Ltd.

Mediation Analysis: A Practitioner's Guide.

PubMed

VanderWeele, Tyler J

2016-01-01

This article provides an overview of recent developments in mediation analysis, that is, analyses used to assess the relative magnitude of different pathways and mechanisms by which an exposure may affect an outcome. Traditional approaches to mediation in the biomedical and social sciences are described. Attention is given to the confounding assumptions required for a causal interpretation of direct and indirect effect estimates. Methods from the causal inference literature to conduct mediation in the presence of exposure-mediator interactions, binary outcomes, binary mediators, and case-control study designs are presented. Sensitivity analysis techniques for unmeasured confounding and measurement error are introduced. Discussion is given to extensions to time-to-event outcomes and multiple mediators. Further flexible modeling strategies arising from the precise counterfactual definitions of direct and indirect effects are also described. The focus throughout is on methodology that is easily implementable in practice across a broad range of potential applications.

The psychophysical law of speed estimation in Michotte's causal events.

PubMed

Parovel, Giulia; Casco, Clara

2006-11-01

Observers saw an event in which a computer-animated square moved up to and made contact with another, which after a short delay moved off, its motion appearing to be caused by launch by the first square. Observers chose whether the second (launched) square was faster in this causal event than when presented following a long delay (non-causal event). The speed of the second object in causal events was overestimated for a wide range of speeds of the first object (launcher), but accurately assessed in non-causal events. Experiments 2 and 3 showed that overestimation occurred also in other causal displays in which the trajectories were overlapping, successive, spatially separated or inverted but did not occurred with consecutive speeds that did not produce causal percepts. We also found that if the first object in a causal event was faster, then Weber's law holds and overestimation of the launched object speed was proportional to the speed of the launcher. In contrast, if the second object was faster, overestimation was constant, i.e. independent of the launcher. We propose that the particular speed integration of causal display results in overestimation and that the way overestimation depends on V1 phenomenally affects the attribution of the source of V2 motion: either in V1 (in launching) or in V2 (in triggering).

Mediation Analysis with Multiple Mediators

PubMed Central

VanderWeele, T.J.; Vansteelandt, S.

2014-01-01

Recent advances in the causal inference literature on mediation have extended traditional approaches to direct and indirect effects to settings that allow for interactions and non-linearities. In this paper, these approaches from causal inference are further extended to settings in which multiple mediators may be of interest. Two analytic approaches, one based on regression and one based on weighting are proposed to estimate the effect mediated through multiple mediators and the effects through other pathways. The approaches proposed here accommodate exposure-mediator interactions and, to a certain extent, mediator-mediator interactions as well. The methods handle binary or continuous mediators and binary, continuous or count outcomes. When the mediators affect one another, the strategy of trying to assess direct and indirect effects one mediator at a time will in general fail; the approach given in this paper can still be used. A characterization is moreover given as to when the sum of the mediated effects for multiple mediators considered separately will be equal to the mediated effect of all of the mediators considered jointly. The approach proposed in this paper is robust to unmeasured common causes of two or more mediators. PMID:25580377

Mediation Analysis with Multiple Mediators.

PubMed

VanderWeele, T J; Vansteelandt, S

2014-01-01

Recent advances in the causal inference literature on mediation have extended traditional approaches to direct and indirect effects to settings that allow for interactions and non-linearities. In this paper, these approaches from causal inference are further extended to settings in which multiple mediators may be of interest. Two analytic approaches, one based on regression and one based on weighting are proposed to estimate the effect mediated through multiple mediators and the effects through other pathways. The approaches proposed here accommodate exposure-mediator interactions and, to a certain extent, mediator-mediator interactions as well. The methods handle binary or continuous mediators and binary, continuous or count outcomes. When the mediators affect one another, the strategy of trying to assess direct and indirect effects one mediator at a time will in general fail; the approach given in this paper can still be used. A characterization is moreover given as to when the sum of the mediated effects for multiple mediators considered separately will be equal to the mediated effect of all of the mediators considered jointly. The approach proposed in this paper is robust to unmeasured common causes of two or more mediators.

FamLBL: detecting rare haplotype disease association based on common SNPs using case-parent triads.

PubMed

Wang, Meng; Lin, Shili

2014-09-15

In recent years, there has been an increasing interest in using common single-nucleotide polymorphisms (SNPs) amassed in genome-wide association studies to investigate rare haplotype effects on complex diseases. Evidence has suggested that rare haplotypes may tag rare causal single-nucleotide variants, making SNP-based rare haplotype analysis not only cost effective, but also more valuable for detecting causal variants. Although a number of methods for detecting rare haplotype association have been proposed in recent years, they are population based and thus susceptible to population stratification. We propose family-triad-based logistic Bayesian Lasso (famLBL) for estimating effects of haplotypes on complex diseases using SNP data. By choosing appropriate prior distribution, effect sizes of unassociated haplotypes can be shrunk toward zero, allowing for more precise estimation of associated haplotypes, especially those that are rare, thereby achieving greater detection power. We evaluate famLBL using simulation to gauge its type I error and power. Compared with its population counterpart, LBL, highlights famLBL's robustness property in the presence of population substructure. Further investigation by comparing famLBL with Family-Based Association Test (FBAT) reveals its advantage for detecting rare haplotype association. famLBL is implemented as an R-package available at http://www.stat.osu.edu/∼statgen/SOFTWARE/LBL/. © The Author 2014. Published by Oxford University Press. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com.

Initial Validation for the Estimation of Resting-State fMRI Effective Connectivity by a Generalization of the Correlation Approach

PubMed Central

Xu, Nan; Spreng, R. Nathan; Doerschuk, Peter C.

2017-01-01

Resting-state functional MRI (rs-fMRI) is widely used to noninvasively study human brain networks. Network functional connectivity is often estimated by calculating the timeseries correlation between blood-oxygen-level dependent (BOLD) signal from different regions of interest (ROIs). However, standard correlation cannot characterize the direction of information flow between regions. In this paper, we introduce and test a new concept, prediction correlation, to estimate effective connectivity in functional brain networks from rs-fMRI. In this approach, the correlation between two BOLD signals is replaced by a correlation between one BOLD signal and a prediction of this signal via a causal system driven by another BOLD signal. Three validations are described: (1) Prediction correlation performed well on simulated data where the ground truth was known, and outperformed four other methods. (2) On simulated data designed to display the “common driver” problem, prediction correlation did not introduce false connections between non-interacting driven ROIs. (3) On experimental data, prediction correlation recovered the previously identified network organization of human brain. Prediction correlation scales well to work with hundreds of ROIs, enabling it to assess whole brain interregional connectivity at the single subject level. These results provide an initial validation that prediction correlation can capture the direction of information flow and estimate the duration of extended temporal delays in information flow between regions of interest ROIs based on BOLD signal. This approach not only maintains the high sensitivity to network connectivity provided by the correlation analysis, but also performs well in the estimation of causal information flow in the brain. PMID:28559793

Propensity Score-Based Methods versus MTE-Based Methods in Causal Inference: Identification, Estimation, and Application

ERIC Educational Resources Information Center

Zhou, Xiang; Xie, Yu

2016-01-01

Since the seminal introduction of the propensity score (PS) by Rosenbaum and Rubin, PS-based methods have been widely used for drawing causal inferences in the behavioral and social sciences. However, the PS approach depends on the ignorability assumption: there are no unobserved confounders once observed covariates are taken into account. For…

MOVING TO INEQUALITY: NEIGHBORHOOD EFFECTS AND EXPERIMENTS MEET STRUCTURE1

PubMed Central

Sampson, Robert J.

2014-01-01

The Moving to Opportunity (MTO) housing experiment has proven to be an important intervention not just in the lives of the poor, but in social science theories of neighborhood effects. Competing causal claims have been the subject of considerable disagreement, culminating in the debate between Clampet-Lundquist and Massey (2008) and Ludwig et al. (2008). This paper assesses the debate by clarifying analytically distinct questions posed by neighborhood-level theories, reconceptualizing selection bias as a fundamental social process worthy of study in its own right rather than as a statistical nuisance, and reconsidering the scientific method of experimentation, and hence causality, in the social world of the city. I also analyze MTO and independent survey data from Chicago to examine trajectories of residential attainment. Although MTO provides crucial leverage for estimating neighborhood effects on individuals, as proponents rightly claim, I demonstrate the implications imposed by a stratified urban structure and how MTO simultaneously provides a new window on the social reproduction of concentrated inequality. PMID:25360053

The diversity effect in diagnostic reasoning.

PubMed

Rebitschek, Felix G; Krems, Josef F; Jahn, Georg

2016-07-01

Diagnostic reasoning draws on knowledge about effects and their potential causes. The causal-diversity effect in diagnostic reasoning normatively depends on the distribution of effects in causal structures, and thus, a psychological diversity effect could indicate whether causally structured knowledge is used in evaluating the probability of a diagnosis, if the effect were to covary with manipulations of causal structures. In four experiments, participants dealt with a quasi-medical scenario presenting symptom sets (effects) that consistently suggested a specified diagnosis (cause). The probability that the diagnosis was correct had to be rated for two opposed symptom sets that differed with regard to the symptoms' positions (proximal or diverse) in the causal structure that was initially acquired. The causal structure linking the diagnosis to the symptoms and the base rate of the diagnosis were manipulated to explore whether the diagnosis was rated as more probable for diverse than for proximal symptoms when alternative causations were more plausible (e.g., because of a lower base rate of the diagnosis in question). The results replicated the causal diversity effect in diagnostic reasoning across these conditions, but no consistent effects of structure and base rate variations were observed. Diversity effects computed in causal Bayesian networks are presented, illustrating the consequences of the structure manipulations and corroborating that a diversity effect across the different experimental manipulations is normatively justified. The observed diversity effects presumably resulted from shortcut reasoning about the possibilities of alternative causation.

Strengthening Causal Estimates for Links Between Spanking and Children's Externalizing Behavior Problems.

PubMed

Gershoff, Elizabeth T; Sattler, Kierra M P; Ansari, Arya

2018-01-01

Establishing causal links when experiments are not feasible is an important challenge for psychology researchers. The question of whether parents' spanking causes children's externalizing behavior problems poses such a challenge because randomized experiments of spanking are unethical, and correlational studies cannot rule out potential selection factors. This study used propensity score matching based on the lifetime prevalence and recent incidence of spanking in a large and nationally representative sample ( N = 12,112) as well as lagged dependent variables to get as close to causal estimates outside an experiment as possible. Whether children were spanked at the age of 5 years predicted increases in externalizing behavior problems by ages 6 and 8, even after the groups based on spanking prevalence or incidence were matched on a range of sociodemographic, family, and cultural characteristics and children's initial behavior problems. These statistically rigorous methods yield the conclusion that spanking predicts a deterioration of children's externalizing behavior over time.

Using instrumental variables to estimate a Cox's proportional hazards regression subject to additive confounding

PubMed Central

Tosteson, Tor D.; Morden, Nancy E.; Stukel, Therese A.; O'Malley, A. James

2014-01-01

The estimation of treatment effects is one of the primary goals of statistics in medicine. Estimation based on observational studies is subject to confounding. Statistical methods for controlling bias due to confounding include regression adjustment, propensity scores and inverse probability weighted estimators. These methods require that all confounders are recorded in the data. The method of instrumental variables (IVs) can eliminate bias in observational studies even in the absence of information on confounders. We propose a method for integrating IVs within the framework of Cox's proportional hazards model and demonstrate the conditions under which it recovers the causal effect of treatment. The methodology is based on the approximate orthogonality of an instrument with unobserved confounders among those at risk. We derive an estimator as the solution to an estimating equation that resembles the score equation of the partial likelihood in much the same way as the traditional IV estimator resembles the normal equations. To justify this IV estimator for a Cox model we perform simulations to evaluate its operating characteristics. Finally, we apply the estimator to an observational study of the effect of coronary catheterization on survival. PMID:25506259

Using instrumental variables to estimate a Cox's proportional hazards regression subject to additive confounding.

PubMed

MacKenzie, Todd A; Tosteson, Tor D; Morden, Nancy E; Stukel, Therese A; O'Malley, A James

2014-06-01

The estimation of treatment effects is one of the primary goals of statistics in medicine. Estimation based on observational studies is subject to confounding. Statistical methods for controlling bias due to confounding include regression adjustment, propensity scores and inverse probability weighted estimators. These methods require that all confounders are recorded in the data. The method of instrumental variables (IVs) can eliminate bias in observational studies even in the absence of information on confounders. We propose a method for integrating IVs within the framework of Cox's proportional hazards model and demonstrate the conditions under which it recovers the causal effect of treatment. The methodology is based on the approximate orthogonality of an instrument with unobserved confounders among those at risk. We derive an estimator as the solution to an estimating equation that resembles the score equation of the partial likelihood in much the same way as the traditional IV estimator resembles the normal equations. To justify this IV estimator for a Cox model we perform simulations to evaluate its operating characteristics. Finally, we apply the estimator to an observational study of the effect of coronary catheterization on survival.

Examining the Error of Mis-Specifying Nonlinear Confounding Effect With Application on Accelerometer-Measured Physical Activity.

PubMed

Lee, Paul H

2017-06-01

Some confounders are nonlinearly associated with dependent variables, but they are often adjusted using a linear term. The purpose of this study was to examine the error of mis-specifying the nonlinear confounding effect. We carried out a simulation study to investigate the effect of adjusting for a nonlinear confounder in the estimation of a causal relationship between the exposure and outcome in 3 ways: using a linear term, binning into 5 equal-size categories, or using a restricted cubic spline of the confounder. Continuous, binary, and survival outcomes were simulated. We examined the confounder across varying measurement error. In addition, we performed a real data analysis examining the 3 strategies to handle the nonlinear effects of accelerometer-measured physical activity in the National Health and Nutrition Examination Survey 2003-2006 data. The mis-specification of a nonlinear confounder had little impact on causal effect estimation for continuous outcomes. For binary and survival outcomes, this mis-specification introduced bias, which could be eliminated using spline adjustment only when there is small measurement error of the confounder. Real data analysis showed that the associations between high blood pressure, high cholesterol, and diabetes and mortality adjusted for physical activity with restricted cubic spline were about 3% to 11% larger than their counterparts adjusted with a linear term. For continuous outcomes, confounders with nonlinear effects can be adjusting with a linear term. Spline adjustment should be used for binary and survival outcomes on confounders with small measurement error.

Length of secondary schooling and risk of HIV infection in Botswana: evidence from a natural experiment

PubMed Central

De Neve, Jan-Walter; Fink, Günther; Subramanian, SV; Moyo, Sikhulile; Bor, Jacob

2015-01-01

Background An estimated 2·3 Million individuals are newly infected with HIV each year. Existing cross-sectional and longitudinal studies have found conflicting evidence on the association between education and HIV risk, and no randomized experiment to date has identified a causal effect of education on HIV incidence. Methods A 1996 policy reform changed the grade structure of secondary school in Botswana and increased educational attainment. We use this reform as a ‘natural experiment’ to identify the causal effect of schooling on HIV infection. Data on HIV biomarkers and demographics were obtained from the 2004 and 2008 Botswana AIDS Impact Surveys, nationally-representative household surveys (N = 7018). The association between years of schooling and HIV status was described using multivariate OLS regression models. Using exposure to the policy reform as an instrumental variable, we estimated the causal effect of years of schooling on the cumulative probability that an individual contracted HIV up to his or her age at the time of the survey. The cost-effectiveness of secondary schooling as an HIV prevention intervention was assessed in comparison to other established interventions. Findings Each additional year of secondary schooling induced by the policy change led to an absolute reduction in the cumulative risk of HIV infection of 8·1% points (p = 0·008), relative to a baseline prevalence of 25·6%. Effects were particularly large among women (11·6% points, p = 0·046). Results were robust to a wide array of sensitivity analyses. Secondary school was cost-effective as an HIV prevention intervention by standard metrics. Interpretation Additional years of secondary schooling had a large protective effect against HIV risk, particularly for women, in Botswana. Increasing progression through secondary school may be a cost-effective HIV prevention measure in HIV-endemic settings, in addition to yielding other societal benefits. Funding Takemi Program in International Health at the Harvard School of Public Health, Belgian American Educational Foundation, and Fernand Lazard Foundation. PMID:26134875

The Causal Meaning of Genomic Predictors and How It Affects Construction and Comparison of Genome-Enabled Selection Models

PubMed Central

Valente, Bruno D.; Morota, Gota; Peñagaricano, Francisco; Gianola, Daniel; Weigel, Kent; Rosa, Guilherme J. M.

2015-01-01

The term “effect” in additive genetic effect suggests a causal meaning. However, inferences of such quantities for selection purposes are typically viewed and conducted as a prediction task. Predictive ability as tested by cross-validation is currently the most acceptable criterion for comparing models and evaluating new methodologies. Nevertheless, it does not directly indicate if predictors reflect causal effects. Such evaluations would require causal inference methods that are not typical in genomic prediction for selection. This suggests that the usual approach to infer genetic effects contradicts the label of the quantity inferred. Here we investigate if genomic predictors for selection should be treated as standard predictors or if they must reflect a causal effect to be useful, requiring causal inference methods. Conducting the analysis as a prediction or as a causal inference task affects, for example, how covariates of the regression model are chosen, which may heavily affect the magnitude of genomic predictors and therefore selection decisions. We demonstrate that selection requires learning causal genetic effects. However, genomic predictors from some models might capture noncausal signal, providing good predictive ability but poorly representing true genetic effects. Simulated examples are used to show that aiming for predictive ability may lead to poor modeling decisions, while causal inference approaches may guide the construction of regression models that better infer the target genetic effect even when they underperform in cross-validation tests. In conclusion, genomic selection models should be constructed to aim primarily for identifiability of causal genetic effects, not for predictive ability. PMID:25908318

CAUSAL ANALYSIS AND PROBABILITY DATA: EXAMPLES FOR IMPAIRED AQUATIC CONDITION

EPA Science Inventory

Causal analysis is plausible reasoning applied to diagnosing observed effect(s), for example, diagnosing

cause of biological impairment in a stream. Sir Bradford Hill basically defined the application of causal

analysis when he enumerated the elements of causality f...

Mendelian randomization with fine-mapped genetic data: Choosing from large numbers of correlated instrumental variables.

PubMed

Burgess, Stephen; Zuber, Verena; Valdes-Marquez, Elsa; Sun, Benjamin B; Hopewell, Jemma C

2017-12-01

Mendelian randomization uses genetic variants to make causal inferences about the effect of a risk factor on an outcome. With fine-mapped genetic data, there may be hundreds of genetic variants in a single gene region any of which could be used to assess this causal relationship. However, using too many genetic variants in the analysis can lead to spurious estimates and inflated Type 1 error rates. But if only a few genetic variants are used, then the majority of the data is ignored and estimates are highly sensitive to the particular choice of variants. We propose an approach based on summarized data only (genetic association and correlation estimates) that uses principal components analysis to form instruments. This approach has desirable theoretical properties: it takes the totality of data into account and does not suffer from numerical instabilities. It also has good properties in simulation studies: it is not particularly sensitive to varying the genetic variants included in the analysis or the genetic correlation matrix, and it does not have greatly inflated Type 1 error rates. Overall, the method gives estimates that are less precise than those from variable selection approaches (such as using a conditional analysis or pruning approach to select variants), but are more robust to seemingly arbitrary choices in the variable selection step. Methods are illustrated by an example using genetic associations with testosterone for 320 genetic variants to assess the effect of sex hormone related pathways on coronary artery disease risk, in which variable selection approaches give inconsistent inferences. © 2017 The Authors Genetic Epidemiology Published by Wiley Periodicals, Inc.

Rising School Attendance in Rural India: An Evaluation of the Effects of Major Educational Reforms

ERIC Educational Resources Information Center

Datta Gupta, Nabanita; Dubey, Amaresh; Simonsen, Marianne

2018-01-01

We evaluate the impact of educational reforms starting from the mid-1990s in India on the school attendance rate of low-income rural children aged 6-14 compared to ineligible rural children, employing NSSO data from 1983 to 2004/2005. We estimate a triple difference model allowing for differential (linear) trends and find a positive causal effect…

Public Subsidies to Private Schools Do Make a Difference for Achievement in Mathematics: Longitudinal Evidence from Canada

ERIC Educational Resources Information Center

Lefebvre, Pierre; Merrigan, Philip; Verstraete, Matthieu

2011-01-01

Selection into private schools is the principal cause of bias when estimating the effect of private schooling on academic achievement. By exploiting the generous public subsidizing of private high schools in the province of Quebec, the second most populous province in Canada, we identify the causal impact of attendance in a private high school on…

The Needs of the Army: Using Compulsory Relocation in the Military to Estimate the Effect of Air Pollutants on Children's Health

ERIC Educational Resources Information Center

Lleras-Muney, Adriana

2010-01-01

Recent research suggests that pollution has a large impact on asthma and other respiratory and cardiovascular conditions. But this relationship and its implications are not well understood. I use changes in location due to military transfers, which occur entirely to satisfy the needs of the army, to identify the causal impact of pollution on…

Spectral factorization of wavefields and wave operators

NASA Astrophysics Data System (ADS)

Rickett, James Edward

Spectral factorization is the problem of finding a minimum-phase function with a given power spectrum. Minimum phase functions have the property that they are causal with a causal (stable) inverse. In this thesis, I factor multidimensional systems into their minimum-phase components. Helical boundary conditions resolve any ambiguities over causality, allowing me to factor multi-dimensional systems with conventional one-dimensional spectral factorization algorithms. In the first part, I factor passive seismic wavefields recorded in two-dimensional spatial arrays. The result provides an estimate of the acoustic impulse response of the medium that has higher bandwidth than autocorrelation-derived estimates. Also, the function's minimum-phase nature mimics the physics of the system better than the zero-phase autocorrelation model. I demonstrate this on helioseismic data recorded by the satellite-based Michelson Doppler Imager (MDI) instrument, and shallow seismic data recorded at Long Beach, California. In the second part of this thesis, I take advantage of the stable-inverse property of minimum-phase functions to solve wave-equation partial differential equations. By factoring multi-dimensional finite-difference stencils into minimum-phase components, I can invert them efficiently, facilitating rapid implicit extrapolation without the azimuthal anisotropy that is observed with splitting approximations. The final part of this thesis describes how to calculate diagonal weighting functions that approximate the combined operation of seismic modeling and migration. These weighting functions capture the effects of irregular subsurface illumination, which can be the result of either the surface-recording geometry, or focusing and defocusing of the seismic wavefield as it propagates through the earth. Since they are diagonal, they can be easily both factored and inverted to compensate for uneven subsurface illumination in migrated images. Experimental results show that applying these weighting functions after migration leads to significantly improved estimates of seismic reflectivity.

Exposure to Traffic-related Air Pollution During Pregnancy and Term Low Birth Weight: Estimation of Causal Associations in a Semiparametric Model

PubMed Central

Padula, Amy M.; Mortimer, Kathleen; Hubbard, Alan; Lurmann, Frederick; Jerrett, Michael; Tager, Ira B.

2012-01-01

Traffic-related air pollution is recognized as an important contributor to health problems. Epidemiologic analyses suggest that prenatal exposure to traffic-related air pollutants may be associated with adverse birth outcomes; however, there is insufficient evidence to conclude that the relation is causal. The Study of Air Pollution, Genetics and Early Life Events comprises all births to women living in 4 counties in California's San Joaquin Valley during the years 2000–2006. The probability of low birth weight among full-term infants in the population was estimated using machine learning and targeted maximum likelihood estimation for each quartile of traffic exposure during pregnancy. If everyone lived near high-volume freeways (approximated as the fourth quartile of traffic density), the estimated probability of term low birth weight would be 2.27% (95% confidence interval: 2.16, 2.38) as compared with 2.02% (95% confidence interval: 1.90, 2.12) if everyone lived near smaller local roads (first quartile of traffic density). Assessment of potentially causal associations, in the absence of arbitrary model assumptions applied to the data, should result in relatively unbiased estimates. The current results support findings from previous studies that prenatal exposure to traffic-related air pollution may adversely affect birth weight among full-term infants. PMID:23045474

Effects of breast and colorectal cancer on labour market outcomes-average effects and educational gradients.

PubMed

Heinesen, Eskil; Kolodziejczyk, Christophe

2013-12-01

We estimate causal effects of breast and colorectal cancer on labour market outcomes 1-3 years after the diagnosis. Based on Danish administrative data we estimate average treatment effects on the treated by propensity score weighting methods using persons with no cancer diagnosis as control group. We conduct robustness checks using matching, difference-in-differences methods and an alternative control group of later cancer patients. The different methods give approximately the same results. Cancer increases the risks of leaving the labour force and receiving disability pension, and the effects are larger for the less educated. Effects on income are small and mostly insignificant. We investigate some of the mechanisms which may be important in explaining the educational gradient in effects of cancer on labour market attachment. Copyright © 2013 Elsevier B.V. All rights reserved.

Economic growth, combustible renewables and waste consumption, and CO₂ emissions in North Africa.

PubMed

Ben Jebli, Mehdi; Ben Youssef, Slim

2015-10-01

This paper uses panel cointegration techniques and Granger causality tests to examine the dynamic causal link between per capita real gross domestic product (GDP), combustible renewables and waste (CRW) consumption, and CO2 emissions for a panel of five North African countries during the period 1971-2008. Granger causality test results suggest short- and long-run unidirectional causalities running from CO2 emissions and CRW consumption to real GDP and a short-run unidirectional causality running from CRW to CO2 emissions. The results from panel long-run fully modified ordinary least squares (FMOLS) and dynamic ordinary least squares (DOLS) estimates show that CO2 emissions and CRW consumption have a positive and statistically significant impact on GDP. Our policy recommendations are that these countries should use more CRW because this increases their output, reduces their energy dependency on fossil energy, and may decrease their CO2 emissions.

Prior knowledge driven Granger causality analysis on gene regulatory network discovery

DOE PAGES

Yao, Shun; Yoo, Shinjae; Yu, Dantong

2015-08-28

Our study focuses on discovering gene regulatory networks from time series gene expression data using the Granger causality (GC) model. However, the number of available time points (T) usually is much smaller than the number of target genes (n) in biological datasets. The widely applied pairwise GC model (PGC) and other regularization strategies can lead to a significant number of false identifications when n>>T. In this study, we proposed a new method, viz., CGC-2SPR (CGC using two-step prior Ridge regularization) to resolve the problem by incorporating prior biological knowledge about a target gene data set. In our simulation experiments, themore » propose new methodology CGC-2SPR showed significant performance improvement in terms of accuracy over other widely used GC modeling (PGC, Ridge and Lasso) and MI-based (MRNET and ARACNE) methods. In addition, we applied CGC-2SPR to a real biological dataset, i.e., the yeast metabolic cycle, and discovered more true positive edges with CGC-2SPR than with the other existing methods. In our research, we noticed a “ 1+1>2” effect when we combined prior knowledge and gene expression data to discover regulatory networks. Based on causality networks, we made a functional prediction that the Abm1 gene (its functions previously were unknown) might be related to the yeast’s responses to different levels of glucose. In conclusion, our research improves causality modeling by combining heterogeneous knowledge, which is well aligned with the future direction in system biology. Furthermore, we proposed a method of Monte Carlo significance estimation (MCSE) to calculate the edge significances which provide statistical meanings to the discovered causality networks. All of our data and source codes will be available under the link https://bitbucket.org/dtyu/granger-causality/wiki/Home.« less

On the causal links between health indicator, output, combustible renewables and waste consumption, rail transport, and CO2 emissions: the case of Tunisia.

PubMed

Ben Jebli, Mehdi

2016-08-01

This study employs the autoregressive distributed lag (ARDL) approach and Granger causality test to investigate the short- and long-run relationships between health indicator, real GDP, combustible renewables and waste consumption, rail transport, and carbon dioxide (CO2) emissions for the case of Tunisia, spanning the period of 1990-2011. The empirical findings suggest that the Fisher statistic of the Wald test confirm the existence of a long-run relationship between the variables. Moreover, the long-run estimated elasticities of the ARDL model provide that output and combustible renewables and waste consumption have a positive and statistically significant impact on health situation, while CO2 emissions and rail transport both contribute to the decrease of health indicator. Granger causality results affirm that, in the short-run, there is a unidirectional causality running from real GDP to health, a unidirectional causality from health to combustible renewables and waste consumption, and a unidirectional causality from all variables to CO2 emissions. In the long-run, all the computed error correction terms are significant and confirm the existence of long-run association among the variables. Our recommendations for the Tunisian policymakers are as follows: (i) exploiting wastes and renewable fuels can be a good strategy to eliminate pollution caused by emissions and subsequently improve health quality, (ii) the use of renewable energy as a main source for national rail transport is an effective strategy for public health, (iii) renewable energy investment projects are beneficial plans for the country as this contributes to the growth of its own economy and reduce energy dependence, and (iii) more renewable energy consumption leads not only to decrease pollution but also to stimulate health situation because of the increase of doctors and nurses numbers.

Learning About Causes From People: Observational Causal Learning in 24-Month-Old Infants

PubMed Central

Meltzoff, Andrew N.; Waismeyer, Anna; Gopnik, Alison

2013-01-01

How do infants and young children learn about the causal structure of the world around them? In 4 experiments we investigate whether young children initially give special weight to the outcomes of goal-directed interventions they see others perform and use this to distinguish correlations from genuine causal relations—observational causal learning. In a new 2-choice procedure, 2- to 4-year-old children saw 2 identical objects (potential causes). Activation of 1 but not the other triggered a spatially remote effect. Children systematically intervened on the causal object and predictively looked to the effect. Results fell to chance when the cause and effect were temporally reversed, so that the events were merely associated but not causally related. The youngest children (24- to 36-month-olds) were more likely to make causal inferences when covariations were the outcome of human interventions than when they were not. Observational causal learning may be a fundamental learning mechanism that enables infants to abstract the causal structure of the world. PMID:22369335

Situation models and memory: the effects of temporal and causal information on recall sequence.

PubMed

Brownstein, Aaron L; Read, Stephen J

2007-10-01

Participants watched an episode of the television show Cheers on video and then reported free recall. Recall sequence followed the sequence of events in the story; if one concept was observed immediately after another, it was recalled immediately after it. We also made a causal network of the show's story and found that recall sequence followed causal links; effects were recalled immediately after their causes. Recall sequence was more likely to follow causal links than temporal sequence, and most likely to follow causal links that were temporally sequential. Results were similar at 10-minute and 1-week delayed recall. This is the most direct and detailed evidence reported on sequential effects in recall. The causal network also predicted probability of recall; concepts with more links and concepts on the main causal chain were most likely to be recalled. This extends the causal network model to more complex materials than previous research.

Principal stratification in causal inference.

PubMed

Frangakis, Constantine E; Rubin, Donald B

2002-03-01

Many scientific problems require that treatment comparisons be adjusted for posttreatment variables, but the estimands underlying standard methods are not causal effects. To address this deficiency, we propose a general framework for comparing treatments adjusting for posttreatment variables that yields principal effects based on principal stratification. Principal stratification with respect to a posttreatment variable is a cross-classification of subjects defined by the joint potential values of that posttreatment variable tinder each of the treatments being compared. Principal effects are causal effects within a principal stratum. The key property of principal strata is that they are not affected by treatment assignment and therefore can be used just as any pretreatment covariate. such as age category. As a result, the central property of our principal effects is that they are always causal effects and do not suffer from the complications of standard posttreatment-adjusted estimands. We discuss briefly that such principal causal effects are the link between three recent applications with adjustment for posttreatment variables: (i) treatment noncompliance, (ii) missing outcomes (dropout) following treatment noncompliance. and (iii) censoring by death. We then attack the problem of surrogate or biomarker endpoints, where we show, using principal causal effects, that all current definitions of surrogacy, even when perfectly true, do not generally have the desired interpretation as causal effects of treatment on outcome. We go on to forrmulate estimands based on principal stratification and principal causal effects and show their superiority.

Association of Body Mass Index with Depression, Anxiety and Suicide—An Instrumental Variable Analysis of the HUNT Study

PubMed Central

Bjørngaard, Johan Håkon; Carslake, David; Lund Nilsen, Tom Ivar; Linthorst, Astrid C. E.; Davey Smith, George; Gunnell, David; Romundstad, Pål Richard

2015-01-01

Objective While high body mass index is associated with an increased risk of depression and anxiety, cumulative evidence indicates that it is a protective factor for suicide. The associations from conventional observational studies of body mass index with mental health outcomes are likely to be influenced by reverse causality or confounding by ill-health. In the present study, we investigated the associations between offspring body mass index and parental anxiety, depression and suicide in order to avoid problems with reverse causality and confounding by ill-health. Methods We used data from 32,457 mother-offspring and 27,753 father-offspring pairs from the Norwegian HUNT-study. Anxiety and depression were assessed using the Hospital Anxiety and Depression Scale and suicide death from national registers. Associations between offspring and own body mass index and symptoms of anxiety and depression and suicide mortality were estimated using logistic and Cox regression. Causal effect estimates were estimated with a two sample instrument variable approach using offspring body mass index as an instrument for parental body mass index. Results Both own and offspring body mass index were positively associated with depression, while the results did not indicate any substantial association between body mass index and anxiety. Although precision was low, suicide mortality was inversely associated with own body mass index and the results from the analysis using offspring body mass index supported these results. Adjusted odds ratios per standard deviation body mass index from the instrumental variable analysis were 1.22 (95% CI: 1.05, 1.43) for depression, 1.10 (95% CI: 0.95, 1.27) for anxiety, and the instrumental variable estimated hazard ratios for suicide was 0.69 (95% CI: 0.30, 1.63). Conclusion The present study’s results indicate that suicide mortality is inversely associated with body mass index. We also found support for a positive association between body mass index and depression, but not for anxiety. PMID:26167892

Association of Body Mass Index with Depression, Anxiety and Suicide-An Instrumental Variable Analysis of the HUNT Study.

PubMed

Bjørngaard, Johan Håkon; Carslake, David; Lund Nilsen, Tom Ivar; Linthorst, Astrid C E; Davey Smith, George; Gunnell, David; Romundstad, Pål Richard

2015-01-01

While high body mass index is associated with an increased risk of depression and anxiety, cumulative evidence indicates that it is a protective factor for suicide. The associations from conventional observational studies of body mass index with mental health outcomes are likely to be influenced by reverse causality or confounding by ill-health. In the present study, we investigated the associations between offspring body mass index and parental anxiety, depression and suicide in order to avoid problems with reverse causality and confounding by ill-health. We used data from 32,457 mother-offspring and 27,753 father-offspring pairs from the Norwegian HUNT-study. Anxiety and depression were assessed using the Hospital Anxiety and Depression Scale and suicide death from national registers. Associations between offspring and own body mass index and symptoms of anxiety and depression and suicide mortality were estimated using logistic and Cox regression. Causal effect estimates were estimated with a two sample instrument variable approach using offspring body mass index as an instrument for parental body mass index. Both own and offspring body mass index were positively associated with depression, while the results did not indicate any substantial association between body mass index and anxiety. Although precision was low, suicide mortality was inversely associated with own body mass index and the results from the analysis using offspring body mass index supported these results. Adjusted odds ratios per standard deviation body mass index from the instrumental variable analysis were 1.22 (95% CI: 1.05, 1.43) for depression, 1.10 (95% CI: 0.95, 1.27) for anxiety, and the instrumental variable estimated hazard ratios for suicide was 0.69 (95% CI: 0.30, 1.63). The present study's results indicate that suicide mortality is inversely associated with body mass index. We also found support for a positive association between body mass index and depression, but not for anxiety.

A self-agency bias in preschoolers' causal inferences

PubMed Central

Kushnir, Tamar; Wellman, Henry M.; Gelman, Susan A.

2013-01-01

Preschoolers' causal learning from intentional actions – causal interventions – is subject to a self-agency bias. We propose that this bias is evidence-based; it is responsive to causal uncertainty. In the current studies, two causes (one child-controlled, one experimenter-controlled) were associated with one or two effects, first independently, then simultaneously. When initial independent effects were probabilistic, and thus subsequent simultaneous actions were causally ambiguous, children showed a self-agency bias. Children showed no bias when initial effects were deterministic. Further controls establish that children's self-agency bias is not a wholesale preference but rather is influenced by uncertainty in causal evidence. These results demonstrate that children's own experience of action influences their causal learning, and suggest possible benefits in uncertain and ambiguous everyday learning contexts. PMID:19271843

Structure and Strength in Causal Induction

ERIC Educational Resources Information Center

Griffiths, Thomas L.; Tenenbaum, Joshua B.

2005-01-01

We present a framework for the rational analysis of elemental causal induction--learning about the existence of a relationship between a single cause and effect--based upon causal graphical models. This framework makes precise the distinction between causal structure and causal strength: the difference between asking whether a causal relationship…

Maternal BMI at the start of pregnancy and offspring epigenome-wide DNA methylation: findings from the pregnancy and childhood epigenetics (PACE) consortium

PubMed Central

Sharp, Gemma C.; Salas, Lucas A.; Monnereau, Claire; Allard, Catherine; Yousefi, Paul; Everson, Todd M.; Bohlin, Jon; Xu, Zongli; Huang, Rae-Chi; Reese, Sarah E.; Xu, Cheng-Jian; Baïz, Nour; Hoyo, Cathrine; Agha, Golareh; Roy, Ritu; Holloway, John W.; Ghantous, Akram; Merid, Simon K.; Bakulski, Kelly M.; Küpers, Leanne K.; Zhang, Hongmei; Richmond, Rebecca C.; Page, Christian M.; Duijts, Liesbeth; Lie, Rolv T.; Melton, Phillip E.; Vonk, Judith M.; Nohr, Ellen A.; Williams-DeVane, ClarLynda; Huen, Karen; Rifas-Shiman, Sheryl L.; Ruiz-Arenas, Carlos; Gonseth, Semira; Rezwan, Faisal I.; Herceg, Zdenko; Ekström, Sandra; Croen, Lisa; Falahi, Fahimeh; Perron, Patrice; Karagas, Margaret R.; Quraishi, Bilal M.; Suderman, Matthew; Magnus, Maria C.; Jaddoe, Vincent W.V.; Taylor, Jack A.; Anderson, Denise; Zhao, Shanshan; Smit, Henriette A.; Josey, Michele J.; Bradman, Asa; Baccarelli, Andrea A.; Bustamante, Mariona; Håberg, Siri E.; Pershagen, Göran; Hertz-Picciotto, Irva; Newschaffer, Craig; Corpeleijn, Eva; Bouchard, Luigi; Lawlor, Debbie A.; Maguire, Rachel L.; Barcellos, Lisa F.; Smith, George Davey; Eskenazi, Brenda; Karmaus, Wilfried; Marsit, Carmen J.; Hivert, Marie-France; Snieder, Harold; Fallin, M. Daniele; Melén, Erik; Munthe-Kaas, Monica C.; Arshad, Hasan; Wiemels, Joseph L.; Annesi-Maesano, Isabella; Vrijheid, Martine; Oken, Emily; Holland, Nina; Murphy, Susan K.; Sørensen, Thorkild I.A.; Koppelman, Gerard H.; Newnham, John P.; Wilcox, Allen J.; Nystad, Wenche; London, Stephanie J.; Felix, Janine F.; Relton, Caroline L.

2017-01-01

Pre-pregnancy maternal obesity is associated with adverse offspring outcomes at birth and later in life. Individual studies have shown that epigenetic modifications such as DNA methylation could contribute. Within the Pregnancy and Childhood Epigenetics (PACE) Consortium, we meta-analysed the association between pre-pregnancy maternal BMI and methylation at over 450,000 sites in newborn blood DNA, across 19 cohorts (9,340 mother-newborn pairs). We attempted to infer causality by comparing the effects of maternal versus paternal BMI and incorporating genetic variation. In four additional cohorts (1,817 mother-child pairs), we meta-analysed the association between maternal BMI at the start of pregnancy and blood methylation in adolescents. In newborns, maternal BMI was associated with small (<0.2% per BMI unit (1 kg/m2), P < 1.06 × 10−7) methylation variation at 9,044 sites throughout the genome. Adjustment for estimated cell proportions greatly attenuated the number of significant CpGs to 104, including 86 sites common to the unadjusted model. At 72/86 sites, the direction of the association was the same in newborns and adolescents, suggesting persistence of signals. However, we found evidence for a6causal intrauterine effect of maternal BMI on newborn methylation at just 8/86 sites. In conclusion, this well-powered analysis identified robust associations between maternal adiposity and variations in newborn blood DNA methylation, but these small effects may be better explained by genetic or lifestyle factors than a causal intrauterine mechanism. This highlights the need for large-scale collaborative approaches and the application of causal inference techniques in epigenetic epidemiology. PMID:29016858

Illustrating, Quantifying, and Correcting for Bias in Post-hoc Analysis of Gene-Based Rare Variant Tests of Association

PubMed Central

Grinde, Kelsey E.; Arbet, Jaron; Green, Alden; O'Connell, Michael; Valcarcel, Alessandra; Westra, Jason; Tintle, Nathan

2017-01-01

To date, gene-based rare variant testing approaches have focused on aggregating information across sets of variants to maximize statistical power in identifying genes showing significant association with diseases. Beyond identifying genes that are associated with diseases, the identification of causal variant(s) in those genes and estimation of their effect is crucial for planning replication studies and characterizing the genetic architecture of the locus. However, we illustrate that straightforward single-marker association statistics can suffer from substantial bias introduced by conditioning on gene-based test significance, due to the phenomenon often referred to as “winner's curse.” We illustrate the ramifications of this bias on variant effect size estimation and variant prioritization/ranking approaches, outline parameters of genetic architecture that affect this bias, and propose a bootstrap resampling method to correct for this bias. We find that our correction method significantly reduces the bias due to winner's curse (average two-fold decrease in bias, p < 2.2 × 10−6) and, consequently, substantially improves mean squared error and variant prioritization/ranking. The method is particularly helpful in adjustment for winner's curse effects when the initial gene-based test has low power and for relatively more common, non-causal variants. Adjustment for winner's curse is recommended for all post-hoc estimation and ranking of variants after a gene-based test. Further work is necessary to continue seeking ways to reduce bias and improve inference in post-hoc analysis of gene-based tests under a wide variety of genetic architectures. PMID:28959274

MIMICKING COUNTERFACTUAL OUTCOMES TO ESTIMATE CAUSAL EFFECTS.

PubMed

Lok, Judith J

2017-04-01

In observational studies, treatment may be adapted to covariates at several times without a fixed protocol, in continuous time. Treatment influences covariates, which influence treatment, which influences covariates, and so on. Then even time-dependent Cox-models cannot be used to estimate the net treatment effect. Structural nested models have been applied in this setting. Structural nested models are based on counterfactuals: the outcome a person would have had had treatment been withheld after a certain time. Previous work on continuous-time structural nested models assumes that counterfactuals depend deterministically on observed data, while conjecturing that this assumption can be relaxed. This article proves that one can mimic counterfactuals by constructing random variables, solutions to a differential equation, that have the same distribution as the counterfactuals, even given past observed data. These "mimicking" variables can be used to estimate the parameters of structural nested models without assuming the treatment effect to be deterministic.

Identifying Conditions That Support Causal Inference in Observational Studies in Education: Empirical Evidence from within Study Comparisons

ERIC Educational Resources Information Center

Hallberg, Kelly

2013-01-01

This dissertation is a collection of three papers that employ empirical within study comparisons (WSCs) to identify conditions that support causal inference in observational studies. WSC studies empirically estimate the extent to which a given observational study reproduces the result of a randomized clinical trial (RCT) when both share the same…

Panel estimation for renewable and non-renewable energy consumption, economic growth, CO2 emissions, the composite trade intensity, and financial openness of the commonwealth of independent states.

PubMed

Rasoulinezhad, Ehsan; Saboori, Behnaz

2018-04-13

This article investigates the long-run and causal linkages between economic growth, CO 2 emissions, renewable and non-renewable (fossil fuels) energy consumption, the Composite Trade Intensity (CTI) as a proxy for trade openness, and the Chinn-Ito index as a proxy for financial openness for a panel of the Commonwealth of Independent States (CIS) region including Armenia, Azerbaijan, Belarus, Georgia, Kazakhstan, Kyrgyzstan, Moldova, Russia, Tajikistan, Turkmenistan, Ukraine, and Uzbekistan over the period of 1992-2015. It is the first time that CTI and the Chinn-Ito indexes are used in an economic-pollution model. Employing three panel unit root tests, panel cointegration estimation methods (DOLS and FMOLS), and two panel causality tests, the main empirical results provided evidence for the bidirectional long-run relationship between all the variables in all 12 sampled countries except for economic growth-renewable energy use linkage. The findings of causality tests indicated that there is a unidirectional short-run panel causality running from economic growth, financial openness, and trade openness to CO 2 emissions and from fossil fuel energy consumption to renewable energy use.

Causal assessment of surrogacy in a meta-analysis of colorectal cancer trials

PubMed Central

Li, Yun; Taylor, Jeremy M.G.; Elliott, Michael R.; Sargent, Daniel J.

2011-01-01

When the true end points (T) are difficult or costly to measure, surrogate markers (S) are often collected in clinical trials to help predict the effect of the treatment (Z). There is great interest in understanding the relationship among S, T, and Z. A principal stratification (PS) framework has been proposed by Frangakis and Rubin (2002) to study their causal associations. In this paper, we extend the framework to a multiple trial setting and propose a Bayesian hierarchical PS model to assess surrogacy. We apply the method to data from a large collection of colon cancer trials in which S and T are binary. We obtain the trial-specific causal measures among S, T, and Z, as well as their overall population-level counterparts that are invariant across trials. The method allows for information sharing across trials and reduces the nonidentifiability problem. We examine the frequentist properties of our model estimates and the impact of the monotonicity assumption using simulations. We also illustrate the challenges in evaluating surrogacy in the counterfactual framework that result from nonidentifiability. PMID:21252079

Surrogacy assessment using principal stratification when surrogate and outcome measures are multivariate normal.

PubMed

Conlon, Anna S C; Taylor, Jeremy M G; Elliott, Michael R

2014-04-01

In clinical trials, a surrogate outcome variable (S) can be measured before the outcome of interest (T) and may provide early information regarding the treatment (Z) effect on T. Using the principal surrogacy framework introduced by Frangakis and Rubin (2002. Principal stratification in causal inference. Biometrics 58, 21-29), we consider an approach that has a causal interpretation and develop a Bayesian estimation strategy for surrogate validation when the joint distribution of potential surrogate and outcome measures is multivariate normal. From the joint conditional distribution of the potential outcomes of T, given the potential outcomes of S, we propose surrogacy validation measures from this model. As the model is not fully identifiable from the data, we propose some reasonable prior distributions and assumptions that can be placed on weakly identified parameters to aid in estimation. We explore the relationship between our surrogacy measures and the surrogacy measures proposed by Prentice (1989. Surrogate endpoints in clinical trials: definition and operational criteria. Statistics in Medicine 8, 431-440). The method is applied to data from a macular degeneration study and an ovarian cancer study.

Surrogacy assessment using principal stratification when surrogate and outcome measures are multivariate normal

PubMed Central

Conlon, Anna S. C.; Taylor, Jeremy M. G.; Elliott, Michael R.

2014-01-01

In clinical trials, a surrogate outcome variable (S) can be measured before the outcome of interest (T) and may provide early information regarding the treatment (Z) effect on T. Using the principal surrogacy framework introduced by Frangakis and Rubin (2002. Principal stratification in causal inference. Biometrics 58, 21–29), we consider an approach that has a causal interpretation and develop a Bayesian estimation strategy for surrogate validation when the joint distribution of potential surrogate and outcome measures is multivariate normal. From the joint conditional distribution of the potential outcomes of T, given the potential outcomes of S, we propose surrogacy validation measures from this model. As the model is not fully identifiable from the data, we propose some reasonable prior distributions and assumptions that can be placed on weakly identified parameters to aid in estimation. We explore the relationship between our surrogacy measures and the surrogacy measures proposed by Prentice (1989. Surrogate endpoints in clinical trials: definition and operational criteria. Statistics in Medicine 8, 431–440). The method is applied to data from a macular degeneration study and an ovarian cancer study. PMID:24285772

Does partial Granger causality really eliminate the influence of exogenous inputs and latent variables?

PubMed

Roelstraete, Bjorn; Rosseel, Yves

2012-04-30

Partial Granger causality was introduced by Guo et al. (2008) who showed that it could better eliminate the influence of latent variables and exogenous inputs than conditional G-causality. In the recent literature we can find some reviews and applications of this type of Granger causality (e.g. Smith et al., 2011; Bressler and Seth, 2010; Barrett et al., 2010). These articles apparently do not take into account a serious flaw in the original work on partial G-causality, being the negative F values that were reported and even proven to be plausible. In our opinion, this undermines the credibility of the obtained results and thus the validity of the approach. Our study is aimed to further validate partial G-causality and to find an answer why negative partial Granger causality estimates were reported. Time series were simulated from the same toy model as used in the original paper and partial and conditional causal measures were compared in the presence of confounding variables. Inference was done parametrically and using non-parametric block bootstrapping. We counter the proof that partial Granger F values can be negative, but the main conclusion of the original article remains. In the presence of unknown latent and exogenous influences, it appears that partial G-causality will better eliminate their influence than conditional G-causality, at least when non-parametric inference is used. Copyright © 2012 Elsevier B.V. All rights reserved.

Quantifying Cancer Risk from Radiation.

PubMed

Keil, Alexander P; Richardson, David B

2017-12-06

Complex statistical models fitted to data from studies of atomic bomb survivors are used to estimate the human health effects of ionizing radiation exposures. We describe and illustrate an approach to estimate population risks from ionizing radiation exposure that relaxes many assumptions about radiation-related mortality. The approach draws on developments in methods for causal inference. The results offer a different way to quantify radiation's effects and show that conventional estimates of the population burden of excess cancer at high radiation doses are driven strongly by projecting outside the range of current data. Summary results obtained using the proposed approach are similar in magnitude to those obtained using conventional methods, although estimates of radiation-related excess cancers differ for many age, sex, and dose groups. At low doses relevant to typical exposures, the strength of evidence in data is surprisingly weak. Statements regarding human health effects at low doses rely strongly on the use of modeling assumptions. © 2017 Society for Risk Analysis.

Video event classification and image segmentation based on noncausal multidimensional hidden Markov models.

PubMed

Ma, Xiang; Schonfeld, Dan; Khokhar, Ashfaq A

2009-06-01

In this paper, we propose a novel solution to an arbitrary noncausal, multidimensional hidden Markov model (HMM) for image and video classification. First, we show that the noncausal model can be solved by splitting it into multiple causal HMMs and simultaneously solving each causal HMM using a fully synchronous distributed computing framework, therefore referred to as distributed HMMs. Next we present an approximate solution to the multiple causal HMMs that is based on an alternating updating scheme and assumes a realistic sequential computing framework. The parameters of the distributed causal HMMs are estimated by extending the classical 1-D training and classification algorithms to multiple dimensions. The proposed extension to arbitrary causal, multidimensional HMMs allows state transitions that are dependent on all causal neighbors. We, thus, extend three fundamental algorithms to multidimensional causal systems, i.e., 1) expectation-maximization (EM), 2) general forward-backward (GFB), and 3) Viterbi algorithms. In the simulations, we choose to limit ourselves to a noncausal 2-D model whose noncausality is along a single dimension, in order to significantly reduce the computational complexity. Simulation results demonstrate the superior performance, higher accuracy rate, and applicability of the proposed noncausal HMM framework to image and video classification.

The Causal Effect of Student Mobility on Standardized Test Performance: A Case Study with Possible Implications for Accountability Mandates within the Elementary and Secondary Education Act.

PubMed

Selya, Arielle S; Engel-Rebitzer, Eden; Dierker, Lisa; Stephen, Eric; Rose, Jennifer; Coffman, Donna L; Otis, Mindy

2016-01-01

This paper presents a limited case study examining the causal inference of student mobility on standardized test performance, within one middle-class high school in suburban Connecticut. Administrative data were used from a district public high school enrolling 319 10th graders in 2010. Propensity score methods were used to estimate the causal effect of student mobility on Math, Science, Reading, and Writing portions of the Connecticut Academic Performance Test (CAPT), after matching mobile vs. stable students on gender, race/ethnicity, eligibility for free/reduced lunches, and special education status. Analyses showed that mobility was associated with lower performance in the CAPT Writing exam. Follow-up analyses revealed that this trend was only significant among those who were ineligible for free/reduced lunches, but not among eligible students. Additionally, mobile students who were ineligible for free/reduced lunches had lower performance in the CAPT Science exam according to some analyses. Large numbers of students transferring into a school district may adversely affect standardized test performance. This is especially relevant for policies that affect student mobility in schools, given the accountability measures in the No Child Left Behind that are currently being re-considered in the recent Every Student Succeeds Act.

Mendelian randomisation in cardiovascular research: an introduction for clinicians

PubMed Central

Bennett, Derrick A; Holmes, Michael V

2017-01-01

Understanding the causal role of biomarkers in cardiovascular and other diseases is crucial in order to find effective approaches (including pharmacological therapies) for disease treatment and prevention. Classical observational studies provide naïve estimates of the likely role of biomarkers in disease development; however, such studies are prone to bias. This has direct relevance for drug development as if drug targets track to non-causal biomarkers, this can lead to expensive failure of these drugs in phase III randomised controlled trials. In an effort to provide a more reliable indication of the likely causal role of a biomarker in the development of disease, Mendelian randomisation studies are increasingly used, and this is facilitated by the availability of large-scale genetic data. We conducted a narrative review in order to provide a description of the utility of Mendelian randomisation for clinicians engaged in cardiovascular research. We describe the rationale and provide a basic description of the methods and potential limitations of Mendelian randomisation. We give examples from the literature where Mendelian randomisation has provided pivotal information for drug discovery including predicting efficacy, informing on target-mediated adverse effects and providing potential new evidence for drug repurposing. The variety of the examples presented illustrates the importance of Mendelian randomisation in order to prioritise drug targets for cardiovascular research. PMID:28596306

The Causal Effect of Student Mobility on Standardized Test Performance: A Case Study with Possible Implications for Accountability Mandates within the Elementary and Secondary Education Act

PubMed Central

Selya, Arielle S.; Engel-Rebitzer, Eden; Dierker, Lisa; Stephen, Eric; Rose, Jennifer; Coffman, Donna L.; Otis, Mindy

2016-01-01

This paper presents a limited case study examining the causal inference of student mobility on standardized test performance, within one middle-class high school in suburban Connecticut. Administrative data were used from a district public high school enrolling 319 10th graders in 2010. Propensity score methods were used to estimate the causal effect of student mobility on Math, Science, Reading, and Writing portions of the Connecticut Academic Performance Test (CAPT), after matching mobile vs. stable students on gender, race/ethnicity, eligibility for free/reduced lunches, and special education status. Analyses showed that mobility was associated with lower performance in the CAPT Writing exam. Follow-up analyses revealed that this trend was only significant among those who were ineligible for free/reduced lunches, but not among eligible students. Additionally, mobile students who were ineligible for free/reduced lunches had lower performance in the CAPT Science exam according to some analyses. Large numbers of students transferring into a school district may adversely affect standardized test performance. This is especially relevant for policies that affect student mobility in schools, given the accountability measures in the No Child Left Behind that are currently being re-considered in the recent Every Student Succeeds Act. PMID:27486427

Detangling complex relationships in forensic data: principles and use of causal networks and their application to clinical forensic science.

PubMed

Lefèvre, Thomas; Lepresle, Aude; Chariot, Patrick

2015-09-01

The search for complex, nonlinear relationships and causality in data is hindered by the availability of techniques in many domains, including forensic science. Linear multivariable techniques are useful but present some shortcomings. In the past decade, Bayesian approaches have been introduced in forensic science. To date, authors have mainly focused on providing an alternative to classical techniques for quantifying effects and dealing with uncertainty. Causal networks, including Bayesian networks, can help detangle complex relationships in data. A Bayesian network estimates the joint probability distribution of data and graphically displays dependencies between variables and the circulation of information between these variables. In this study, we illustrate the interest in utilizing Bayesian networks for dealing with complex data through an application in clinical forensic science. Evaluating the functional impairment of assault survivors is a complex task for which few determinants are known. As routinely estimated in France, the duration of this impairment can be quantified by days of 'Total Incapacity to Work' ('Incapacité totale de travail,' ITT). In this study, we used a Bayesian network approach to identify the injury type, victim category and time to evaluation as the main determinants of the 'Total Incapacity to Work' (TIW). We computed the conditional probabilities associated with the TIW node and its parents. We compared this approach with a multivariable analysis, and the results of both techniques were converging. Thus, Bayesian networks should be considered a reliable means to detangle complex relationships in data.

Bone mineral density and risk of type 2 diabetes and coronary heart disease: A Mendelian randomization study.

PubMed

Gan, Wei; Clarke, Robert J; Mahajan, Anubha; Kulohoma, Benard; Kitajima, Hidetoshi; Robertson, Neil R; Rayner, N William; Walters, Robin G; Holmes, Michael V; Chen, Zhengming; McCarthy, Mark I

2017-01-01

Background: Observational studies have demonstrated that increased bone mineral density is associated with a higher risk of type 2 diabetes (T2D), but the relationship with risk of coronary heart disease (CHD) is less clear. Moreover, substantial uncertainty remains about the causal relevance of increased bone mineral density for T2D and CHD, which can be assessed by Mendelian randomisation studies.  Methods: We identified 235 independent single nucleotide polymorphisms (SNPs) associated at p <5×10 -8 with estimated heel bone mineral density (eBMD) in 116,501 individuals from the UK Biobank study, accounting for 13.9% of eBMD variance. For each eBMD-associated SNP, we extracted effect estimates from the largest available GWAS studies for T2D (DIAGRAM: n=26,676 T2D cases and 132,532 controls) and CHD (CARDIoGRAMplusC4D: n=60,801 CHD cases and 123,504 controls). A two-sample design using several Mendelian randomization approaches was used to investigate the causal relevance of eBMD for risk of T2D and CHD. In addition, we explored the relationship of eBMD, instrumented by the 235 SNPs, on 12 cardiovascular and metabolic risk factors. Finally, we conducted Mendelian randomization analysis in the reverse direction to investigate reverse causality. Results: Each one standard deviation increase in genetically instrumented eBMD (equivalent to 0.14 g/cm 2 ) was associated with an 8% higher risk of T2D (odds ratio [OR] 1.08; 95% confidence interval [CI]: 1.02 to 1.14; p =0.012) and 5% higher risk of CHD (OR 1.05; 95%CI: 1.00 to 1.10; p =0.034). Consistent results were obtained in sensitivity analyses using several different Mendelian randomization approaches. Equivalent increases in eBMD were also associated with lower plasma levels of HDL-cholesterol and increased insulin resistance. Mendelian randomization in the reverse direction using 94 T2D SNPs or 52 CHD SNPs showed no evidence of reverse causality with eBMD. Conclusions: These findings suggest a causal relationship between elevated bone mineral density with risks of both T2D and CHD.

Using genetics to test the causal relationship of total adiposity and periodontitis: Mendelian randomization analyses in the Gene-Lifestyle Interactions and Dental Endpoints (GLIDE) Consortium

PubMed Central

Shungin, Dmitry; Cornelis, Marilyn C; Divaris, Kimon; Holtfreter, Birte; Shaffer, John R; Yu, Yau-Hua; Barros, Silvana P; Beck, James D; Biffar, Reiner; Boerwinkle, Eric A; Crout, Richard J.; Ganna, Andrea; Hallmans, Goran; Hindy, George; Hu, Frank B; Kraft, Peter; McNeil, Daniel W; Melander, Olle; Moss, Kevin L; North, Kari E; Orho-Melander, Marju; Pedersen, Nancy L; Ridker, Paul M; Rimm, Eric B; Rose, Lynda M; Rukh, Gull; Teumer, Alexander; Weyant, Robert J; Chasman, Daniel I; Joshipura, Kaumudi; Kocher, Thomas; Magnusson, Patrik KE; Marazita, Mary L; Nilsson, Peter; Offenbacher, Steve; Davey Smith, George; Lundberg, Pernilla; Palmer, Tom M; Timpson, Nicholas J; Johansson, Ingegerd; Franks, Paul W

2015-01-01

Background: The observational relationship between obesity and periodontitis is widely known, yet causal evidence is lacking. Our objective was to investigate causal associations between periodontitis and body mass index (BMI). Methods: We performed Mendelian randomization analyses with BMI-associated loci combined in a genetic risk score (GRS) as the instrument for BMI. All analyses were conducted within the Gene-Lifestyle Interactions and Dental Endpoints (GLIDE) Consortium in 13 studies from Europe and the USA, including 49 066 participants with clinically assessed (seven studies, 42.1% of participants) and self-reported (six studies, 57.9% of participants) periodontitis and genotype data (17 672/31 394 with/without periodontitis); 68 761 participants with BMI and genotype data; and 57 871 participants (18 881/38 990 with/without periodontitis) with data on BMI and periodontitis. Results: In the observational meta-analysis of all participants, the pooled crude observational odds ratio (OR) for periodontitis was 1.13 [95% confidence interval (CI): 1.03, 1.24] per standard deviation increase of BMI. Controlling for potential confounders attenuated this estimate (OR = 1.08; 95% CI:1.03, 1.12). For clinically assessed periodontitis, corresponding ORs were 1.25 (95% CI: 1.10, 1.42) and 1.13 (95% CI: 1.10, 1.17), respectively. In the genetic association meta-analysis, the OR for periodontitis was 1.01 (95% CI: 0.99, 1.03) per GRS unit (per one effect allele) in all participants and 1.00 (95% CI: 0.97, 1.03) in participants with clinically assessed periodontitis. The instrumental variable meta-analysis of all participants yielded an OR of 1.05 (95% CI: 0.80, 1.38) per BMI standard deviation, and 0.90 (95% CI: 0.56, 1.46) in participants with clinical data. Conclusions: Our study does not support total adiposity as a causal risk factor for periodontitis, as the point estimate is very close to the null in the causal inference analysis, with wide confidence intervals. PMID:26050256

Conceptual framework for describing selected urban and community impacts of federal energy policies

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

Morris, F.A,; Marcus, A.A.; Keller, D.

1980-06-01

A conceptual framework is presented for describing selected urban and community impacts of Federal energy policies. The framework depends on a simple causal model. The outputs of the model are impacts, changes in the state of the world of particular interest to policymakers. At any given time, a set of determinants account for the state of the world with respect to an impact category. Application of the model to a particular impact category requires: establishing a definition and measure for the impact category and identifying the determinants of these impacts. Analysis of the impact of a particular policy requires themore » following: identifying the policy and its effects (as estimated by others), isolating any effects that themselves constitute an urban and community impact, identifying any effects that change the value of determinants, and describing the impact with reference to the new values of determinants. This report provides a framework for these steps. Three impacts addressed are: neighborhood stability, housing availability, and quality and availability of public services. In each chapter, a definition and measure for the impact are specified; its principal determinants are identified; how the causal model can be used to estimate impacts by applying it to three illustrative Federal policies (domestic oil price decontrol, building energy performance standards, and increased Federal aid for mass transit) is demonstrated. (MCW)« less

Mechanisms and mediation in survival analysis: towards an integrated analytical framework.

PubMed

Pratschke, Jonathan; Haase, Trutz; Comber, Harry; Sharp, Linda; de Camargo Cancela, Marianna; Johnson, Howard

2016-02-29

A wide-ranging debate has taken place in recent years on mediation analysis and causal modelling, raising profound theoretical, philosophical and methodological questions. The authors build on the results of these discussions to work towards an integrated approach to the analysis of research questions that situate survival outcomes in relation to complex causal pathways with multiple mediators. The background to this contribution is the increasingly urgent need for policy-relevant research on the nature of inequalities in health and healthcare. The authors begin by summarising debates on causal inference, mediated effects and statistical models, showing that these three strands of research have powerful synergies. They review a range of approaches which seek to extend existing survival models to obtain valid estimates of mediation effects. They then argue for an alternative strategy, which involves integrating survival outcomes within Structural Equation Models via the discrete-time survival model. This approach can provide an integrated framework for studying mediation effects in relation to survival outcomes, an issue of great relevance in applied health research. The authors provide an example of how these techniques can be used to explore whether the social class position of patients has a significant indirect effect on the hazard of death from colon cancer. The results suggest that the indirect effects of social class on survival are substantial and negative (-0.23 overall). In addition to the substantial direct effect of this variable (-0.60), its indirect effects account for more than one quarter of the total effect. The two main pathways for this indirect effect, via emergency admission (-0.12), on the one hand, and hospital caseload, on the other, (-0.10) are of similar size. The discrete-time survival model provides an attractive way of integrating time-to-event data within the field of Structural Equation Modelling. The authors demonstrate the efficacy of this approach in identifying complex causal pathways that mediate the effects of a socio-economic baseline covariate on the hazard of death from colon cancer. The results show that this approach has the potential to shed light on a class of research questions which is of particular relevance in health research.

Quantity-Quality and the One Child Policy: The Only-Child Disadvantage in School Enrollment in Rural China. NBER Working Paper No. 14973

ERIC Educational Resources Information Center

Qian, Nancy

2009-01-01

Many believe that increasing the quantity of children will lead to a decrease in their quality. This paper exploits plausibly exogenous changes in family size caused by relaxations in China's One Child Policy to estimate the causal effect of family size on school enrollment of the first child. The results show that for one-child families, an…

Causal Mediation Analysis for the Cox Proportional Hazards Model with a Smooth Baseline Hazard Estimator.

PubMed

Wang, Wei; Albert, Jeffrey M

2017-08-01

An important problem within the social, behavioral, and health sciences is how to partition an exposure effect (e.g. treatment or risk factor) among specific pathway effects and to quantify the importance of each pathway. Mediation analysis based on the potential outcomes framework is an important tool to address this problem and we consider the estimation of mediation effects for the proportional hazards model in this paper. We give precise definitions of the total effect, natural indirect effect, and natural direct effect in terms of the survival probability, hazard function, and restricted mean survival time within the standard two-stage mediation framework. To estimate the mediation effects on different scales, we propose a mediation formula approach in which simple parametric models (fractional polynomials or restricted cubic splines) are utilized to approximate the baseline log cumulative hazard function. Simulation study results demonstrate low bias of the mediation effect estimators and close-to-nominal coverage probability of the confidence intervals for a wide range of complex hazard shapes. We apply this method to the Jackson Heart Study data and conduct sensitivity analysis to assess the impact on the mediation effects inference when the no unmeasured mediator-outcome confounding assumption is violated.

Time, frequency, and time-varying Granger-causality measures in neuroscience.

PubMed

Cekic, Sezen; Grandjean, Didier; Renaud, Olivier

2018-05-20

This article proposes a systematic methodological review and an objective criticism of existing methods enabling the derivation of time, frequency, and time-varying Granger-causality statistics in neuroscience. The capacity to describe the causal links between signals recorded at different brain locations during a neuroscience experiment is indeed of primary interest for neuroscientists, who often have very precise prior hypotheses about the relationships between recorded brain signals. The increasing interest and the huge number of publications related to this topic calls for this systematic review, which describes the very complex methodological aspects underlying the derivation of these statistics. In this article, we first present a general framework that allows us to review and compare Granger-causality statistics in the time domain, and the link with transfer entropy. Then, the spectral and the time-varying extensions are exposed and discussed together with their estimation and distributional properties. Although not the focus of this article, partial and conditional Granger causality, dynamical causal modelling, directed transfer function, directed coherence, partial directed coherence, and their variant are also mentioned. Copyright © 2018 John Wiley & Sons, Ltd.

Essays on Causal Inference for Public Policy

ERIC Educational Resources Information Center

Zajonc, Tristan

2012-01-01

Effective policymaking requires understanding the causal effects of competing proposals. Relevant causal quantities include proposals' expected effect on different groups of recipients, the impact of policies over time, the potential trade-offs between competing objectives, and, ultimately, the optimal policy. This dissertation studies causal…

Causal mediation analysis with a binary outcome and multiple continuous or ordinal mediators: Simulations and application to an alcohol intervention

PubMed Central

Nguyen, Trang Quynh; Webb-Vargas, Yenny; Koning, Ina M.; Stuart, Elizabeth A.

2016-01-01

We investigate a method to estimate the combined effect of multiple continuous/ordinal mediators on a binary outcome: 1) fit a structural equation model with probit link for the outcome and identity/probit link for continuous/ordinal mediators, 2) predict potential outcome probabilities, and 3) compute natural direct and indirect effects. Step 2 involves rescaling the latent continuous variable underlying the outcome to address residual mediator variance/covariance. We evaluate the estimation of risk-difference- and risk-ratio-based effects (RDs, RRs) using the ML, WLSMV and Bayes estimators in Mplus. Across most variations in path-coefficient and mediator-residual-correlation signs and strengths, and confounding situations investigated, the method performs well with all estimators, but favors ML/WLSMV for RDs with continuous mediators, and Bayes for RRs with ordinal mediators. Bayes outperforms WLSMV/ML regardless of mediator type when estimating RRs with small potential outcome probabilities and in two other special cases. An adolescent alcohol prevention study is used for illustration. PMID:27158217

NON-MENDELIAN ETIOLOGIC FACTORS IN NEUROPSYCHIATRIC ILLNESS: PLEIOTROPY, EPIGENETICS, AND CONVERGENCE

PubMed Central

Deutsch, Curtis K; McIlvane, William J

2013-01-01

The target article by Charney on behavior genetics/genomics discusses how numerous molecular factors can inform heritability estimations and genetic association studies. These factors find application in the search for genes for behavioral phenotypes, including neuropsychiatric disorders. We elaborate upon how single causal factors can generate multiple phenotypes, and discuss how multiple causal factors may converge on common neurodevelopmental mechanisms. PMID:23095384

Accounting for perception, placebo and unmasking effects in estimating treatment effects in randomised clinical trials.

PubMed

Jamshidian, Farid; Hubbard, Alan E; Jewell, Nicholas P

2014-06-01

There is a rich literature on the role of placebos in experimental design and evaluation of therapeutic agents or interventions. The importance of masking participants, investigators and evaluators to treatment assignment (treatment or placebo) has long been stressed as a key feature of a successful trial design. Nevertheless, there is considerable variability in the technical definition of the placebo effect and the impact of treatment assignments being unmasked. We suggest a formal concept of a 'perception effect' and define unmasking and placebo effects in the context of randomised trials. We employ modern tools from causal inference to derive semi-parametric estimators of such effects. The methods are illustrated on a motivating example from a recent pain trial where the occurrence of treatment-related side effects acts as a proxy for unmasking. © The Author(s) 2011 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav.

In defense of causal-formative indicators: A minority report.

PubMed

Bollen, Kenneth A; Diamantopoulos, Adamantios

2017-09-01

Causal-formative indicators directly affect their corresponding latent variable. They run counter to the predominant view that indicators depend on latent variables and are thus often controversial. If present, such indicators have serious implications for factor analysis, reliability theory, item response theory, structural equation models, and most measurement approaches that are based on reflective or effect indicators. Psychological Methods has published a number of influential articles on causal and formative indicators as well as launching the first major backlash against them. This article examines 7 common criticisms of these indicators distilled from the literature: (a) A construct measured with "formative" indicators does not exist independently of its indicators; (b) Such indicators are causes rather than measures; (c) They imply multiple dimensions to a construct and this is a liability; (d) They are assumed to be error-free, which is unrealistic; (e) They are inherently subject to interpretational confounding; (f) They fail proportionality constraints; and (g) Their coefficients should be set in advance and not estimated. We summarize each of these criticisms and point out the flaws in the logic and evidence marshaled in their support. The most common problems are not distinguishing between what we call causal-formative and composite-formative indicators, tautological fallacies, and highlighting issues that are common to all indicators, but presenting them as special problems of causal-formative indicators. We conclude that measurement theory needs (a) to incorporate these types of indicators, and (b) to better understand their similarities to and differences from traditional indicators. (PsycINFO Database Record (c) 2017 APA, all rights reserved).

CO2 emissions, real output, energy consumption, trade, urbanization and financial development: testing the EKC hypothesis for the USA.

PubMed

Dogan, Eyup; Turkekul, Berna

2016-01-01

This study aims to investigate the relationship between carbon dioxide (CO2) emissions, energy consumption, real output (GDP), the square of real output (GDP(2)), trade openness, urbanization, and financial development in the USA for the period 1960-2010. The bounds testing for cointegration indicates that the analyzed variables are cointegrated. In the long run, energy consumption and urbanization increase environmental degradation while financial development has no effect on it, and trade leads to environmental improvements. In addition, this study does not support the validity of the environmental Kuznets curve (EKC) hypothesis for the USA because real output leads to environmental improvements while GDP(2) increases the levels of gas emissions. The results from the Granger causality test show that there is bidirectional causality between CO2 and GDP, CO2 and energy consumption, CO2 and urbanization, GDP and urbanization, and GDP and trade openness while no causality is determined between CO2 and trade openness, and gas emissions and financial development. In addition, we have enough evidence to support one-way causality running from GDP to energy consumption, from financial development to output, and from urbanization to financial development. In light of the long-run estimates and the Granger causality analysis, the US government should take into account the importance of trade openness, urbanization, and financial development in controlling for the levels of GDP and pollution. Moreover, it should be noted that the development of efficient energy policies likely contributes to lower CO2 emissions without harming real output.

Variable selection for confounder control, flexible modeling and Collaborative Targeted Minimum Loss-based Estimation in causal inference

PubMed Central

Schnitzer, Mireille E.; Lok, Judith J.; Gruber, Susan

2015-01-01

This paper investigates the appropriateness of the integration of flexible propensity score modeling (nonparametric or machine learning approaches) in semiparametric models for the estimation of a causal quantity, such as the mean outcome under treatment. We begin with an overview of some of the issues involved in knowledge-based and statistical variable selection in causal inference and the potential pitfalls of automated selection based on the fit of the propensity score. Using a simple example, we directly show the consequences of adjusting for pure causes of the exposure when using inverse probability of treatment weighting (IPTW). Such variables are likely to be selected when using a naive approach to model selection for the propensity score. We describe how the method of Collaborative Targeted minimum loss-based estimation (C-TMLE; van der Laan and Gruber, 2010) capitalizes on the collaborative double robustness property of semiparametric efficient estimators to select covariates for the propensity score based on the error in the conditional outcome model. Finally, we compare several approaches to automated variable selection in low-and high-dimensional settings through a simulation study. From this simulation study, we conclude that using IPTW with flexible prediction for the propensity score can result in inferior estimation, while Targeted minimum loss-based estimation and C-TMLE may benefit from flexible prediction and remain robust to the presence of variables that are highly correlated with treatment. However, in our study, standard influence function-based methods for the variance underestimated the standard errors, resulting in poor coverage under certain data-generating scenarios. PMID:26226129

Variable Selection for Confounder Control, Flexible Modeling and Collaborative Targeted Minimum Loss-Based Estimation in Causal Inference.

PubMed

Schnitzer, Mireille E; Lok, Judith J; Gruber, Susan

2016-05-01

This paper investigates the appropriateness of the integration of flexible propensity score modeling (nonparametric or machine learning approaches) in semiparametric models for the estimation of a causal quantity, such as the mean outcome under treatment. We begin with an overview of some of the issues involved in knowledge-based and statistical variable selection in causal inference and the potential pitfalls of automated selection based on the fit of the propensity score. Using a simple example, we directly show the consequences of adjusting for pure causes of the exposure when using inverse probability of treatment weighting (IPTW). Such variables are likely to be selected when using a naive approach to model selection for the propensity score. We describe how the method of Collaborative Targeted minimum loss-based estimation (C-TMLE; van der Laan and Gruber, 2010 [27]) capitalizes on the collaborative double robustness property of semiparametric efficient estimators to select covariates for the propensity score based on the error in the conditional outcome model. Finally, we compare several approaches to automated variable selection in low- and high-dimensional settings through a simulation study. From this simulation study, we conclude that using IPTW with flexible prediction for the propensity score can result in inferior estimation, while Targeted minimum loss-based estimation and C-TMLE may benefit from flexible prediction and remain robust to the presence of variables that are highly correlated with treatment. However, in our study, standard influence function-based methods for the variance underestimated the standard errors, resulting in poor coverage under certain data-generating scenarios.

CaSPIAN: A Causal Compressive Sensing Algorithm for Discovering Directed Interactions in Gene Networks

PubMed Central

Emad, Amin; Milenkovic, Olgica

2014-01-01

We introduce a novel algorithm for inference of causal gene interactions, termed CaSPIAN (Causal Subspace Pursuit for Inference and Analysis of Networks), which is based on coupling compressive sensing and Granger causality techniques. The core of the approach is to discover sparse linear dependencies between shifted time series of gene expressions using a sequential list-version of the subspace pursuit reconstruction algorithm and to estimate the direction of gene interactions via Granger-type elimination. The method is conceptually simple and computationally efficient, and it allows for dealing with noisy measurements. Its performance as a stand-alone platform without biological side-information was tested on simulated networks, on the synthetic IRMA network in Saccharomyces cerevisiae, and on data pertaining to the human HeLa cell network and the SOS network in E. coli. The results produced by CaSPIAN are compared to the results of several related algorithms, demonstrating significant improvements in inference accuracy of documented interactions. These findings highlight the importance of Granger causality techniques for reducing the number of false-positives, as well as the influence of noise and sampling period on the accuracy of the estimates. In addition, the performance of the method was tested in conjunction with biological side information of the form of sparse “scaffold networks”, to which new edges were added using available RNA-seq or microarray data. These biological priors aid in increasing the sensitivity and precision of the algorithm in the small sample regime. PMID:24622336

Health and Wealth of Elderly Couples: Causality Tests Using Dynamic Panel Data Models*

PubMed Central

Michaud, Pierre-Carl; van Soest, Arthur

2010-01-01

A positive relationship between socio-economic status (SES) and health, the “health-wealth gradient”, is repeatedly found in many industrialized countries. This study analyzes competing explanations for this gradient: causal effects from health to wealth (health causation) and causal effects from wealth to health (wealth or social causation). Using six biennial waves of couples aged 51–61 in 1992 from the U.S. Health and Retirement Study, we test for causality in panel data models incorporating unobserved heterogeneity and a lag structure supported by specification tests. In contrast to tests relying on models with only first order lags or without unobserved heterogeneity, these tests provide no evidence of causal wealth health effects. On the other hand, we find strong evidence of causal effects from both spouses’ health on household wealth. We also find an effect of the husband’s health on the wife’s mental health, but no other effects from one spouse’s health to health of the other spouse. PMID:18513809

Applying Propensity Score Methods in Medical Research: Pitfalls and Prospects

PubMed Central

Luo, Zhehui; Gardiner, Joseph C.; Bradley, Cathy J.

2012-01-01

The authors review experimental and nonexperimental causal inference methods, focusing on assumptions for the validity of instrumental variables and propensity score (PS) methods. They provide guidance in four areas for the analysis and reporting of PS methods in medical research and selectively evaluate mainstream medical journal articles from 2000 to 2005 in the four areas, namely, examination of balance, overlapping support description, use of estimated PS for evaluation of treatment effect, and sensitivity analyses. In spite of the many pitfalls, when appropriately evaluated and applied, PS methods can be powerful tools in assessing average treatment effects in observational studies. Appropriate PS applications can create experimental conditions using observational data when randomized controlled trials are not feasible and, thus, lead researchers to an efficient estimator of the average treatment effect. PMID:20442340

Family income, parenting styles and child behavioural-emotional outcomes.

PubMed

Dooley, Martin; Stewart, Jennifer

2007-02-01

A positive relationship between income and child outcomes has been observed in data from numerous countries. A key question concerns the extent to which this association represents a causal relationship as opposed to unobserved heterogeneity. We use data from the National Longitudinal Survey of Children and Youth to implement a series of empirical strategies for estimating the existence and size of the effect of income on behavioural-emotional outcomes. We also examine the role of parenting style. Our results indicate that there is little evidence of an effect of income on behavioural-emotional scores. The exclusion of parenting style from the models was found to not bias the estimated income effect, but parenting style was found to have a consistent impact on child outcomes. Copyright (c) 2006 John Wiley & Sons, Ltd.

Causal capture effects in chimpanzees (Pan troglodytes).

PubMed

Matsuno, Toyomi; Tomonaga, Masaki

2017-01-01

Extracting a cause-and-effect structure from the physical world is an important demand for animals living in dynamically changing environments. Human perceptual and cognitive mechanisms are known to be sensitive and tuned to detect and interpret such causal structures. In contrast to rigorous investigations of human causal perception, the phylogenetic roots of this perception are not well understood. In the present study, we aimed to investigate the susceptibility of nonhuman animals to mechanical causality by testing whether chimpanzees perceived an illusion called causal capture (Scholl & Nakayama, 2002). Causal capture is a phenomenon in which a type of bistable visual motion of objects is perceived as causal collision due to a bias from a co-occurring causal event. In our experiments, we assessed the susceptibility of perception of a bistable stream/bounce motion event to a co-occurring causal event in chimpanzees. The results show that, similar to in humans, causal "bounce" percepts were significantly increased in chimpanzees with the addition of a task-irrelevant causal bounce event that was synchronously presented. These outcomes suggest that the perceptual mechanisms behind the visual interpretation of causal structures in the environment are evolutionarily shared between human and nonhuman animals. Copyright © 2016 Elsevier B.V. All rights reserved.

Consider the alternative: The effects of causal knowledge on representing and using alternative hypotheses in judgments under uncertainty.

PubMed

Hayes, Brett K; Hawkins, Guy E; Newell, Ben R

2016-05-01

Four experiments examined the locus of impact of causal knowledge on consideration of alternative hypotheses in judgments under uncertainty. Two possible loci were examined; overcoming neglect of the alternative when developing a representation of a judgment problem and improving utilization of statistics associated with the alternative hypothesis. In Experiment 1, participants could search for information about the various components of Bayes's rule in a diagnostic problem. A majority failed to spontaneously search for information about an alternative hypothesis, but this bias was reduced when a specific alternative hypothesis was mentioned before search. No change in search patterns was found when a generic alternative cause was mentioned. Experiments 2a and 2b broadly replicated these patterns when participants rated or made binary judgments about the relevance of each of the Bayesian components. In contrast, Experiment 3 showed that when participants were given the likelihood of the data given a focal hypothesis p(D|H) and an alternative hypothesis p(D|¬H), they gave estimates of p(H|D) that were consistent with Bayesian principles. Additional causal knowledge had relatively little impact on such judgments. These results show that causal knowledge primarily affects neglect of the alternative hypothesis at the initial stage of problem representation. (PsycINFO Database Record (c) 2016 APA, all rights reserved).

Using permutation tests to enhance causal inference in interrupted time series analysis.

PubMed

Linden, Ariel

2018-06-01

Interrupted time series analysis (ITSA) is an evaluation methodology in which a single treatment unit's outcome is studied serially over time and the intervention is expected to "interrupt" the level and/or trend of that outcome. The internal validity is strengthened considerably when the treated unit is contrasted with a comparable control group. In this paper, we introduce a robustness check based on permutation tests to further improve causal inference. We evaluate the effect of California's Proposition 99 for reducing cigarette sales by iteratively casting each nontreated state into the role of "treated," creating a comparable control group using the ITSAMATCH package in Stata, and then evaluating treatment effects using ITSA regression. If statistically significant "treatment effects" are estimated for pseudotreated states, then any significant changes in the outcome of the actual treatment unit (California) cannot be attributed to the intervention. We perform these analyses setting the cutpoint significance level to P > .40 for identifying balanced matches (the highest threshold possible for which controls could still be found for California) and use the difference in differences of trends as the treatment effect estimator. Only California attained a statistically significant treatment effect, strengthening confidence in the conclusion that Proposition 99 reduced cigarette sales. The proposed permutation testing framework provides an additional robustness check to either support or refute a treatment effect identified in for the true treated unit in ITSA. Given its value and ease of implementation, this framework should be considered as a standard robustness test in all multiple group interrupted time series analyses. © 2018 John Wiley & Sons, Ltd.

Glutamatergic Signaling Drives Ketamine-Mediated Response in Depression: Evidence from Dynamic Causal Modeling.

PubMed

Gilbert, Jessica R; Yarrington, Julia S; Wills, Kathleen E; Nugent, Allison C; Zarate, Carlos A

2018-04-13

The glutamatergic modulator ketamine has rapid antidepressant effects in individuals with major depressive disorder (MDD) and bipolar depression. Thus, modulating glutamatergic transmission may be critical to effectively treating depression, though the mechanisms by which this occurs are not fully understood. This double-blind, crossover, placebo-controlled study analyzed data from 18 drug-free MDD subjects and 18 heathy controls who received a single intravenous infusion of ketamine hydrochloride (0.5 mg/kg) as well as an intravenous saline placebo. Magnetoencephalographic (MEG) recordings were collected prior to the first infusion and six to nine hours after both ketamine and placebo infusions. During scanning, participants passively received tactile stimulation to the right index finger. Antidepressant response was assessed across timepoints using the Montgomery-Asberg Depression Rating Scale (MADRS). Dynamic causal modeling (DCM) was used to measure changes in -amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA)- and N-methyl-D-aspartate (NMDA)-mediated connectivity estimates in MDD subjects and controls using a simple model of somatosensory evoked responses. Both MDD and healthy subjects showed ketamine-mediated NMDA-blockade sensitization, with MDD subjects showing enhanced NMDA connectivity estimates in backward connections, and controls showing enhanced NMDA connectivity estimates in forward connections in our model. Within our MDD subject group, ketamine efficacy-as measured by improved mood ratings-correlated with reduced NMDA and AMPA connectivity estimates in discrete extrinsic connections within the somatosensory cortical network. These findings suggest that AMPA- and NMDA-mediated glutamatergic signaling play a key role in antidepressant response to ketamine and, further, that DCM is a powerful tool for modeling AMPA- and NMDA-mediated connectivity in vivo. NCT#00088699.

[FROM STATISTICAL ASSOCIATIONS TO SCIENTIFIC CAUSALITY].

PubMed

Golan, Daniel; Linn, Shay

2015-06-01

The pathogenesis of most chronic diseases is complex and probably involves the interaction of multiple genetic and environmental risk factors. One way to learn about disease triggers is from statistically significant associations in epidemiological studies. However, associations do not necessarily prove causation. Associations can commonly result from bias, confounding and reverse causation. Several paradigms for causality inference have been developed. Henle-Koch postulates are mainly applied for infectious diseases. Austin Bradford Hill's criteria may serve as a practical tool to weigh the evidence regarding the probability that a single new risk factor for a given disease is indeed causal. These criteria are irrelevant for estimating the causal relationship between exposure to a risk factor and disease whenever biological causality has been previously established. Thus, it is highly probable that past exposure of an individual to definite carcinogens is related to his cancer, even without proving an association between this exposure and cancer in his group. For multifactorial diseases, Rothman's model of interacting sets of component causes can be applied.

The effect of early-life education on later-life mortality.

PubMed

Black, Dan A; Hsu, Yu-Chieh; Taylor, Lowell J

2015-12-01

Many studies link cross-state variation in compulsory schooling laws to early-life educational attainment, thereby providing a plausible way to investigate the causal impact of education on various lifetime outcomes. We use this strategy to estimate the effect of education on older-age mortality of individuals born in the early twentieth century U.S. Our key innovation is to combine U.S. Census data and the complete Vital Statistics records to form precise mortality estimates by sex, birth cohort, and birth state. In turn we find that virtually all of the variation in these mortality rates is captured by cohort effects and state effects alone, making it impossible to reliably tease out any additional impact due to changing educational attainment induced by state-level changes in compulsory schooling. Copyright © 2015. Published by Elsevier B.V.

Cortical Hierarchies Perform Bayesian Causal Inference in Multisensory Perception

PubMed Central

Rohe, Tim; Noppeney, Uta

2015-01-01

To form a veridical percept of the environment, the brain needs to integrate sensory signals from a common source but segregate those from independent sources. Thus, perception inherently relies on solving the “causal inference problem.” Behaviorally, humans solve this problem optimally as predicted by Bayesian Causal Inference; yet, the underlying neural mechanisms are unexplored. Combining psychophysics, Bayesian modeling, functional magnetic resonance imaging (fMRI), and multivariate decoding in an audiovisual spatial localization task, we demonstrate that Bayesian Causal Inference is performed by a hierarchy of multisensory processes in the human brain. At the bottom of the hierarchy, in auditory and visual areas, location is represented on the basis that the two signals are generated by independent sources (= segregation). At the next stage, in posterior intraparietal sulcus, location is estimated under the assumption that the two signals are from a common source (= forced fusion). Only at the top of the hierarchy, in anterior intraparietal sulcus, the uncertainty about the causal structure of the world is taken into account and sensory signals are combined as predicted by Bayesian Causal Inference. Characterizing the computational operations of signal interactions reveals the hierarchical nature of multisensory perception in human neocortex. It unravels how the brain accomplishes Bayesian Causal Inference, a statistical computation fundamental for perception and cognition. Our results demonstrate how the brain combines information in the face of uncertainty about the underlying causal structure of the world. PMID:25710328

Characterization of autoregressive processes using entropic quantifiers

NASA Astrophysics Data System (ADS)

Traversaro, Francisco; Redelico, Francisco O.

2018-01-01

The aim of the contribution is to introduce a novel information plane, the causal-amplitude informational plane. As previous works seems to indicate, Bandt and Pompe methodology for estimating entropy does not allow to distinguish between probability distributions which could be fundamental for simulation or for probability analysis purposes. Once a time series is identified as stochastic by the causal complexity-entropy informational plane, the novel causal-amplitude gives a deeper understanding of the time series, quantifying both, the autocorrelation strength and the probability distribution of the data extracted from the generating processes. Two examples are presented, one from climate change model and the other from financial markets.

Cross-section and panel estimates of peer effects in early adolescent cannabis use: With a little help from my 'friends once removed'.

PubMed

Moriarty, John; McVicar, Duncan; Higgins, Kathryn

2016-08-01

Peer effects in adolescent cannabis are difficult to estimate, due in part to the lack of appropriate data on behaviour and social ties. This paper exploits survey data that have many desirable properties and have not previously been used for this purpose. The data set, collected from teenagers in three annual waves from 2002 to 2004 contains longitudinal information about friendship networks within schools (N = 5020). We exploit these data on network structure to estimate peer effects on adolescents from their nominated friends within school using two alternative approaches to identification. First, we present a cross-sectional instrumental variable (IV) estimate of peer effects that exploits network structure at the second degree, i.e. using information on friends of friends who are not themselves ego's friends to instrument for the cannabis use of friends. Second, we present an individual fixed effects estimate of peer effects using the full longitudinal structure of the data. Both innovations allow a greater degree of control for correlated effects than is commonly the case in the substance-use peer effects literature, improving our chances of obtaining estimates of peer effects than can be plausibly interpreted as causal. Both estimates suggest positive peer effects of non-trivial magnitude, although the IV estimate is imprecise. Furthermore, when we specify identical models with behaviour and characteristics of randomly selected school peers in place of friends', we find effectively zero effect from these 'placebo' peers, lending credence to our main estimates. We conclude that cross-sectional data can be used to estimate plausible positive peer effects on cannabis use where network structure information is available and appropriately exploited. Copyright © 2016 Elsevier Ltd. All rights reserved.

Estimates of Preventability and Their Relation to Health Behavior.

ERIC Educational Resources Information Center

Poole, Gary D.

It was hypothesized that a person's estimates of the preventability of health problems would be related to health behaviors such that a person who engages in healthful behavior should make higher estimates of preventability. A study was conducted to investigate the relationship between causal attribution of health problems and health-related…

Causal network in a deafferented non-human primate brain.

PubMed

Balasubramanian, Karthikeyan; Takahashi, Kazutaka; Hatsopoulos, Nicholas G

2015-01-01

De-afferented/efferented neural ensembles can undergo causal changes when interfaced to neuroprosthetic devices. These changes occur via recruitment or isolation of neurons, alterations in functional connectivity within the ensemble and/or changes in the role of neurons, i.e., excitatory/inhibitory. In this work, emergence of a causal network and changes in the dynamics are demonstrated for a deafferented brain region exposed to BMI (brain-machine interface) learning. The BMI was controlling a robot for reach-and-grasp behavior. And, the motor cortical regions used for the BMI were deafferented due to chronic amputation, and ensembles of neurons were decoded for velocity control of the multi-DOF robot. A generalized linear model-framework based Granger causality (GLM-GC) technique was used in estimating the ensemble connectivity. Model selection was based on the AIC (Akaike Information Criterion).

Financial Barriers to STEM Study in College: Causal Effect Estimates of Need-Based Grants on the Pursuit and Completion of Courses and Degrees in STEM Fields

ERIC Educational Resources Information Center

Castleman, Benjamin L.; Long, Bridget Terry; Mabel, Zachary A.

2014-01-01

The fastest growing supply of jobs in the United States today is in Science, Technology, Engineering, and Mathematics (STEM) fields. Yet despite the availability of work in STEM, there is not a sufficient supply of workers to fill open positions. Amidst the growing demand for STEM workers, educational achievement and attainment in STEM fields in…

Learning to make things happen: Infants' observational learning of social and physical causal events.

PubMed

Waismeyer, Anna; Meltzoff, Andrew N

2017-10-01

Infants learn about cause and effect through hands-on experience; however, they also can learn about causality simply from observation. Such observational causal learning is a central mechanism by which infants learn from and about other people. Across three experiments, we tested infants' observational causal learning of both social and physical causal events. Experiment 1 assessed infants' learning of a physical event in the absence of visible spatial contact between the causes and effects. Experiment 2 developed a novel paradigm to assess whether infants could learn about a social causal event from third-party observation of a social interaction between two people. Experiment 3 compared learning of physical and social events when the outcomes occurred probabilistically (happening some, but not all, of the time). Infants demonstrated significant learning in all three experiments, although learning about probabilistic cause-effect relations was most difficult. These findings about infant observational causal learning have implications for children's rapid nonverbal learning about people, things, and their causal relations. Copyright © 2017 Elsevier Inc. All rights reserved.

Correcting for Measurement Error in Time-Varying Covariates in Marginal Structural Models.

PubMed

Kyle, Ryan P; Moodie, Erica E M; Klein, Marina B; Abrahamowicz, Michał

2016-08-01

Unbiased estimation of causal parameters from marginal structural models (MSMs) requires a fundamental assumption of no unmeasured confounding. Unfortunately, the time-varying covariates used to obtain inverse probability weights are often error-prone. Although substantial measurement error in important confounders is known to undermine control of confounders in conventional unweighted regression models, this issue has received comparatively limited attention in the MSM literature. Here we propose a novel application of the simulation-extrapolation (SIMEX) procedure to address measurement error in time-varying covariates, and we compare 2 approaches. The direct approach to SIMEX-based correction targets outcome model parameters, while the indirect approach corrects the weights estimated using the exposure model. We assess the performance of the proposed methods in simulations under different clinically plausible assumptions. The simulations demonstrate that measurement errors in time-dependent covariates may induce substantial bias in MSM estimators of causal effects of time-varying exposures, and that both proposed SIMEX approaches yield practically unbiased estimates in scenarios featuring low-to-moderate degrees of error. We illustrate the proposed approach in a simple analysis of the relationship between sustained virological response and liver fibrosis progression among persons infected with hepatitis C virus, while accounting for measurement error in γ-glutamyltransferase, using data collected in the Canadian Co-infection Cohort Study from 2003 to 2014. © The Author 2016. Published by Oxford University Press on behalf of the Johns Hopkins Bloomberg School of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Defining and estimating causal direct and indirect effects when setting the mediator to specific values is not feasible

PubMed Central

Lok, Judith J.

2016-01-01

Natural direct and indirect effects decompose the effect of a treatment into the part that is mediated by a covariate (the mediator) and the part that is not. Their definitions rely on the concept of outcomes under treatment with the mediator “set” to its value without treatment. Typically, the mechanism through which the mediator is set to this value is left unspecified, and in many applications it may be challenging to fix the mediator to particular values for each unit or patient. Moreover, how one sets the mediator may affect the distribution of the outcome. This article introduces “organic” direct and indirect effects, which can be defined and estimated without relying on setting the mediator to specific values. Organic direct and indirect effects can be applied for example to estimate how much of the effect of some treatments for HIV/AIDS on mother-to-child transmission of HIV infection is mediated by the effect of the treatment on the HIV viral load in the blood of the mother. PMID:27229743

Relationships between attitudes toward and achievement in science for rural middle school students: Patterns across gender

NASA Astrophysics Data System (ADS)

Mattern, Nancy Page Garland

Four causal models describing the relationships between attitudes and achievement have been proposed in the literature. The cross-effects, or reciprocal effects, model highlights the effects of prior attitudes on later achievement (over and above the effect of previous achievement) and of prior achievement on later attitudes (above the effect of previous attitudes). In the achievement predominant model, the effect of prior achievement on later attitudes is emphasized, controlling for the effect of previous attitudes. The effect of prior attitudes on later achievement, controlling for the effect of previous achievement, is emphasized in the attitudes predominant model. In the no cross-effects model there are no significant cross paths from prior attitudes to later achievement or from prior achievement to later attitudes. To determine the best-fitting model for rural seventh and eighth grade science girls and boys, the causal relationships over time between attitudes toward science and achievement in science were examined by gender using structural equation modeling. Data were collected in two waves, over one school year. A baseline measurement model was estimated in simultaneous two-group solutions and was a good fit to the data. Next, the four structural models were estimated and model fits compared. The three models nested within the structural cross-effects model showed significant decay of fit when compared to the fit of the cross-effects model. The cross-effects model was the best fit overall for middle school girls and boys. The cross-effects model was then tested for invariance across gender. There was significant decay of fit when model form, factor path loadings, and structural paths were constrained to be equal for girls and boys. Two structural paths, the path from prior achievement to later attitudes, and the path from prior attitudes to later attitudes, were the sources of gender non-invariance. Separate models were estimated for girls and boys, and the fits of nested models were compared. The no cross-effects model was the best-fitting model for rural middle school girls. The new no attitudes-path model was the best-fitting model for boys. Implications of these findings for teaching middle school students were discussed.

Effects of education on cognition at older ages: evidence from China's Great Famine.

PubMed

Huang, Wei; Zhou, Yi

2013-12-01

This paper explores whether educational attainment has a cognitive reserve capacity in elder life. Using pilot data from the China Health and Retirement Longitudinal Study (CHARLS), we examined the impact of education on cognitive abilities at old ages. OLS results showed that respondents who completed primary school obtained 18.2 percent higher scores on cognitive tests than those who did not. We then constructed an instrumental variable (IV) by leveraging China's Great Famine of 1959-1961 as a natural experiment to estimate the causal effect of education on cognition. Two-stage least squares (2SLS) results provided sound evidence that completing primary school significantly increases cognition scores, especially in episode memory, by almost 20 percent on average. Moreover, Regression Discontinuity (RD) analysis provides further evidence for the causal interpretation, and shows that the effects are different for the different measures of cognition we explored. Our results also show that the Great Famine can result in long-term health consequences through the pathway of losing educational opportunities other than through the pathway of nutrition deprivation. Copyright © 2013 Elsevier Ltd. All rights reserved.

High-School Exit Examinations and the Schooling Decisions of Teenagers: Evidence From Regression-Discontinuity Approaches

PubMed Central

Papay, John P.; Murnane, Richard J.; Willett, John B.

2014-01-01

We examine whether barely failing one or more state-mandated high school exit examinations in Massachusetts affects the probability that students enroll in college. We extend the exit examination literature in two ways. First, we explore longer term effects of failing these tests. We find that barely failing an exit examination, for students on the margin of passing, reduces the probability of college attendance several years after the test. Second, we explore potential interactions that arise because students must pass exit examinations in both mathematics and English language arts in order to graduate from high school. We adopt a variety of regression-discontinuity approaches to address situations where multiple variables assign individuals to a range of treatments; some of these approaches enable us to examine whether the effect of barely failing one examination depends on student performance on the other. We document the range of causal effects estimated by each approach. We argue that each approach presents opportunities and limitations for making causal inferences in such situations and that the choice of approach should match the question of interest. PMID:25606065

Social and individual determinants of adolescents' acceptance of novel healthy and cool snack products.

PubMed

Nørgaard, Maria Kümpel; Sørensen, Bjarne Taulo; Grunert, Klaus G

2014-12-01

Four new, healthy snack products, consisting of fruit, vegetables, bread, dip and topping, were tested with 600 Danish adolescents aged 9-16. Participants could view, handle, and test the products in a school setting. Acceptance was measured by overall buying intention, as well as buying intention contingent on specific substitutes and on the social situation. Price consciousness, health consciousness, snack neophobia, peer influence, social activities and word-of-mouth were measured as potential determinants of acceptance of the novel products. An exploratory analysis in TETRAD suggested that the measured constructs form three layers, with overall buying intention as the terminal causal effect, health consciousness, word of mouth, snack neophobia and peer influence as endogenous determinants, and social activities and the contingent buying intentions as mediators. Estimation of the causal relationships was conducted in LISREL. Findings show a predominance of social factors as determinants of novel snack acceptance, whereas health consciousness had only a weak and indirect effect on buying intentions and the effect of snack neophobia was partly mediated by social factors. Copyright © 2014 Elsevier Ltd. All rights reserved.

Do remittances promote financial development in Africa?

PubMed

Karikari, Nana Kwasi; Mensah, Sam; Harvey, Simon K

2016-01-01

The paper seeks to establish whether or not remittances promoted financial developments and explore the traceable causality between remittances and financial developments in some countries in Africa. We examine the association between remittances received and how they affect the availability of credit to private sector, bank deposits intermediated by financial institutions and money supply. We also question whether the development in the financial sector causes higher levels or otherwise of remittances received. This paper uses data on remittance flows to 50 developing countries in Africa from 1990 to 2011 to explore the nexus. The study uses fixed effects and random effect estimations as well as Vector Error Correction Model method on the panel data. The study shows that remittances promote certain aspects of financial development to some extent and better financial system foster receipts of remittances. The effect of causality is seen in the short run and not in the long-run. The study alludes to literature that remittances could promote financial development in the short run and the development of the financial sector helps increase the propensity to remit via formal channels.

Mediation analysis with time varying exposures and mediators

PubMed Central

VanderWeele, Tyler J.; Tchetgen Tchetgen, Eric J.

2016-01-01

Summary In this paper we consider causal mediation analysis when exposures and mediators vary over time. We give non-parametric identification results, discuss parametric implementation, and also provide a weighting approach to direct and indirect effects based on combining the results of two marginal structural models. We also discuss how our results give rise to a causal interpretation of the effect estimates produced from longitudinal structural equation models. When there are time-varying confounders affected by prior exposure and mediator, natural direct and indirect effects are not identified. However, we define a randomized interventional analogue of natural direct and indirect effects that are identified in this setting. The formula that identifies these effects we refer to as the “mediational g-formula.” When there is no mediation, the mediational g-formula reduces to Robins’ regular g-formula for longitudinal data. When there are no time-varying confounders affected by prior exposure and mediator values, then the mediational g-formula reduces to a longitudinal version of Pearl’s mediation formula. However, the mediational g-formula itself can accommodate both mediation and time-varying confounders and constitutes a general approach to mediation analysis with time-varying exposures and mediators. PMID:28824285

Causal effects of retirement timing on subjective physical and emotional health.

PubMed

Calvo, Esteban; Sarkisian, Natalia; Tamborini, Christopher R

2013-01-01

This article explores the effects of the timing of retirement on subjective physical and emotional health. Using panel data from the Health and Retirement Study (HRS), we test 4 theory-based hypotheses about these effects-that retirements maximize health when they happen earlier, later, anytime, or on time. We employ fixed and random effects regression models with instrumental variables to estimate the short- and long-term causal effects of retirement timing on self-reported health and depressive symptoms. Early retirements--those occurring prior to traditional and legal retirement age--dampen health. Workers who begin their retirement transition before cultural and institutional timetables experience the worst health outcomes; this finding offers partial support to the psychosocial-materialist approach that emphasizes the benefits of retiring later. Continued employment after traditionally expected retirement age, however, offers no health benefits. In combination, these findings offer some support for the cultural-institutional approach but suggest that we need to modify our understanding of how cultural-institutional forces operate. Retiring too early can be problematic but no disadvantages are associated with late retirements. Raising the retirement age, therefore, could potentially reduce subjective health of retirees by expanding the group of those whose retirements would be considered early.

Mediation analysis with time varying exposures and mediators.

PubMed

VanderWeele, Tyler J; Tchetgen Tchetgen, Eric J

2017-06-01

In this paper we consider causal mediation analysis when exposures and mediators vary over time. We give non-parametric identification results, discuss parametric implementation, and also provide a weighting approach to direct and indirect effects based on combining the results of two marginal structural models. We also discuss how our results give rise to a causal interpretation of the effect estimates produced from longitudinal structural equation models. When there are time-varying confounders affected by prior exposure and mediator, natural direct and indirect effects are not identified. However, we define a randomized interventional analogue of natural direct and indirect effects that are identified in this setting. The formula that identifies these effects we refer to as the "mediational g-formula." When there is no mediation, the mediational g-formula reduces to Robins' regular g-formula for longitudinal data. When there are no time-varying confounders affected by prior exposure and mediator values, then the mediational g-formula reduces to a longitudinal version of Pearl's mediation formula. However, the mediational g-formula itself can accommodate both mediation and time-varying confounders and constitutes a general approach to mediation analysis with time-varying exposures and mediators.

Does higher education protect against obesity? Evidence using Mendelian randomization.

PubMed

Böckerman, Petri; Viinikainen, Jutta; Pulkki-Råback, Laura; Hakulinen, Christian; Pitkänen, Niina; Lehtimäki, Terho; Pehkonen, Jaakko; Raitakari, Olli T

2017-08-01

The aim of this explorative study was to examine the effect of education on obesity using Mendelian randomization. Participants (N=2011) were from the on-going nationally representative Young Finns Study (YFS) that began in 1980 when six cohorts (aged 30, 33, 36, 39, 42 and 45 in 2007) were recruited. The average value of BMI (kg/m 2 ) measurements in 2007 and 2011 and genetic information were linked to comprehensive register-based information on the years of education in 2007. We first used a linear regression (Ordinary Least Squares, OLS) to estimate the relationship between education and BMI. To identify a causal relationship, we exploited Mendelian randomization and used a genetic score as an instrument for education. The genetic score was based on 74 genetic variants that genome-wide association studies (GWASs) have found to be associated with the years of education. Because the genotypes are randomly assigned at conception, the instrument causes exogenous variation in the years of education and thus enables identification of causal effects. The years of education in 2007 were associated with lower BMI in 2007/2011 (regression coefficient (b)=-0.22; 95% Confidence Intervals [CI]=-0.29, -0.14) according to the linear regression results. The results based on Mendelian randomization suggests that there may be a negative causal effect of education on BMI (b=-0.84; 95% CI=-1.77, 0.09). The findings indicate that education could be a protective factor against obesity in advanced countries. Copyright © 2017 Elsevier Inc. All rights reserved.

Bayes and blickets: Effects of knowledge on causal induction in children and adults

PubMed Central

Griffiths, Thomas L.; Sobel, David M.; Tenenbaum, Joshua B.; Gopnik, Alison

2011-01-01

People are adept at inferring novel causal relations, even from only a few observations. Prior knowledge about the probability of encountering causal relations of various types and the nature of the mechanisms relating causes and effects plays a crucial role in these inferences. We test a formal account of how this knowledge can be used and acquired, based on analyzing causal induction as Bayesian inference. Five studies explored the predictions of this account with adults and 4-year-olds, using tasks in which participants learned about the causal properties of a set of objects. The studies varied the two factors that our Bayesian approach predicted should be relevant to causal induction: the prior probability with which causal relations exist, and the assumption of a deterministic or a probabilistic relation between cause and effect. Adults’ judgments (Experiments 1, 2, and 4) were in close correspondence with the quantitative predictions of the model, and children’s judgments (Experiments 3 and 5) agreed qualitatively with this account. PMID:21972897

Targeted Maximum Likelihood Estimation for Dynamic and Static Longitudinal Marginal Structural Working Models

PubMed Central

Schwab, Joshua; Gruber, Susan; Blaser, Nello; Schomaker, Michael; van der Laan, Mark

2015-01-01

This paper describes a targeted maximum likelihood estimator (TMLE) for the parameters of longitudinal static and dynamic marginal structural models. We consider a longitudinal data structure consisting of baseline covariates, time-dependent intervention nodes, intermediate time-dependent covariates, and a possibly time-dependent outcome. The intervention nodes at each time point can include a binary treatment as well as a right-censoring indicator. Given a class of dynamic or static interventions, a marginal structural model is used to model the mean of the intervention-specific counterfactual outcome as a function of the intervention, time point, and possibly a subset of baseline covariates. Because the true shape of this function is rarely known, the marginal structural model is used as a working model. The causal quantity of interest is defined as the projection of the true function onto this working model. Iterated conditional expectation double robust estimators for marginal structural model parameters were previously proposed by Robins (2000, 2002) and Bang and Robins (2005). Here we build on this work and present a pooled TMLE for the parameters of marginal structural working models. We compare this pooled estimator to a stratified TMLE (Schnitzer et al. 2014) that is based on estimating the intervention-specific mean separately for each intervention of interest. The performance of the pooled TMLE is compared to the performance of the stratified TMLE and the performance of inverse probability weighted (IPW) estimators using simulations. Concepts are illustrated using an example in which the aim is to estimate the causal effect of delayed switch following immunological failure of first line antiretroviral therapy among HIV-infected patients. Data from the International Epidemiological Databases to Evaluate AIDS, Southern Africa are analyzed to investigate this question using both TML and IPW estimators. Our results demonstrate practical advantages of the pooled TMLE over an IPW estimator for working marginal structural models for survival, as well as cases in which the pooled TMLE is superior to its stratified counterpart. PMID:25909047

Three essays in energy consumption: Time series analyses

NASA Astrophysics Data System (ADS)

Ahn, Hee Bai

1997-10-01

Firstly, this dissertation investigates that which demand specification is an appropriate model for long-run energy demand between the conventional demand specification and the limited demand specification. In order to determine the components of a stable long-run demand for different sectors of the energy industry, I perform cointegration tests by using the Johansen test procedure. First, I test the conventional demand specification including prices and income as components. Second, I test a limited demand specification only income as a component. The reason for performing these tests is that we can determine that which demand specification is a good long-run predictor of energy consumption between the two demand specifications by using the cointegration tests. Secondly, for the purpose of planning and forecasting energy demand in case of cointegrated system, long-run elasticities are of particular interest. To retrieve the optimal level of energy demand in case of price shock, we need long-run elasticities rather than short-run elasticities. The energy demand study provides valuable information to the energy policy makers who are concerned about the long-run impact of taxes and tariffs. A long-run price elasticity is a primary barometer of the substitution effect between energy and non-energy inputs and long-run income elasticity is an important factor since we can measure the energy demand growing slowly or fast than in the past depending on the magnitude of long-run elasticity. The one other problem in estimating the total energy demand is that there exists an aggregation bias stemming from the process of summation in four different energy types for the total aggregation prices and total aggregation energy consumption. In order to measure the aggregation bias between the Btu aggregation method and the Divisia Index method, i.e., which methodology has less aggregation bias in the long-run, I compare the two estimation results with calculated results estimated on a disaggregated basis. Thus, we can confirm whether or not the theoretically superior methodology has less aggregation bias in empirical estimation. Thirdly, I investigate the causal relationships between energy use and GDP. In order to detect causal relationships both in the long-run and in the short-run, the VECM (Vector Error Correction Model) can be used if there exists cointegration relationships among the variables. I detect the causal effects between energy use and GDP by estimating the VECM based on the multivariate production function including the labor and capital variables.

Parental leave and child health.

PubMed

Ruhm, C J

2000-11-01

This study investigates whether rights to parental leave improve pediatric health. Aggregate data are used for 16 European countries over the 1969 through 1994 period. More generous paid leave is found to reduce deaths of infants and young children. The magnitudes of the estimated effects are substantial, especially where a causal effect of leave is most plausible. In particular, there is a much stronger negative relationship between leave durations and post-neonatal or child fatalities than for perinatal mortality, neonatal deaths, or low birth weight. The evidence further suggests that parental leave may be a cost-effective method of bettering child health.

The cradle of causal reasoning: newborns' preference for physical causality.

PubMed

Mascalzoni, Elena; Regolin, Lucia; Vallortigara, Giorgio; Simion, Francesca

2013-05-01

Perception of mechanical (i.e. physical) causality, in terms of a cause-effect relationship between two motion events, appears to be a powerful mechanism in our daily experience. In spite of a growing interest in the earliest causal representations, the role of experience in the origin of this sensitivity is still a matter of dispute. Here, we asked the question about the innate origin of causal perception, never tested before at birth. Three experiments were carried out to investigate sensitivity at birth to some visual spatiotemporal cues present in a launching event. Newborn babies, only a few hours old, showed that they significantly preferred a physical causality event (i.e. Michotte's Launching effect) when matched to a delay event (i.e. a delayed launching; Experiment 1) or to a non-causal event completely identical to the causal one except for the order of the displacements of the two objects involved which was swapped temporally (Experiment 3). This preference for the launching event, moreover, also depended on the continuity of the trajectory between the objects involved in the event (Experiment 2). These results support the hypothesis that the human system possesses an early available, possibly innate basic mechanism to compute causality, such a mechanism being sensitive to the additive effect of certain well-defined spatiotemporal cues present in the causal event independently of any prior visual experience. © 2013 Blackwell Publishing Ltd.

Hypothesis test of mediation effect in causal mediation model with high-dimensional continuous mediators.

PubMed

Huang, Yen-Tsung; Pan, Wen-Chi

2016-06-01

Causal mediation modeling has become a popular approach for studying the effect of an exposure on an outcome through a mediator. However, current methods are not applicable to the setting with a large number of mediators. We propose a testing procedure for mediation effects of high-dimensional continuous mediators. We characterize the marginal mediation effect, the multivariate component-wise mediation effects, and the L2 norm of the component-wise effects, and develop a Monte-Carlo procedure for evaluating their statistical significance. To accommodate the setting with a large number of mediators and a small sample size, we further propose a transformation model using the spectral decomposition. Under the transformation model, mediation effects can be estimated using a series of regression models with a univariate transformed mediator, and examined by our proposed testing procedure. Extensive simulation studies are conducted to assess the performance of our methods for continuous and dichotomous outcomes. We apply the methods to analyze genomic data investigating the effect of microRNA miR-223 on a dichotomous survival status of patients with glioblastoma multiforme (GBM). We identify nine gene ontology sets with expression values that significantly mediate the effect of miR-223 on GBM survival. © 2015, The International Biometric Society.

Causal inference with measurement error in outcomes: Bias analysis and estimation methods.

PubMed

Shu, Di; Yi, Grace Y

2017-01-01

Inverse probability weighting estimation has been popularly used to consistently estimate the average treatment effect. Its validity, however, is challenged by the presence of error-prone variables. In this paper, we explore the inverse probability weighting estimation with mismeasured outcome variables. We study the impact of measurement error for both continuous and discrete outcome variables and reveal interesting consequences of the naive analysis which ignores measurement error. When a continuous outcome variable is mismeasured under an additive measurement error model, the naive analysis may still yield a consistent estimator; when the outcome is binary, we derive the asymptotic bias in a closed-form. Furthermore, we develop consistent estimation procedures for practical scenarios where either validation data or replicates are available. With validation data, we propose an efficient method for estimation of average treatment effect; the efficiency gain is substantial relative to usual methods of using validation data. To provide protection against model misspecification, we further propose a doubly robust estimator which is consistent even when either the treatment model or the outcome model is misspecified. Simulation studies are reported to assess the performance of the proposed methods. An application to a smoking cessation dataset is presented.

Analyzing brain networks with PCA and conditional Granger causality.

PubMed

Zhou, Zhenyu; Chen, Yonghong; Ding, Mingzhou; Wright, Paul; Lu, Zuhong; Liu, Yijun

2009-07-01

Identifying directional influences in anatomical and functional circuits presents one of the greatest challenges for understanding neural computations in the brain. Granger causality mapping (GCM) derived from vector autoregressive models of data has been employed for this purpose, revealing complex temporal and spatial dynamics underlying cognitive processes. However, the traditional GCM methods are computationally expensive, as signals from thousands of voxels within selected regions of interest (ROIs) are individually processed, and being based on pairwise Granger causality, they lack the ability to distinguish direct from indirect connectivity among brain regions. In this work a new algorithm called PCA based conditional GCM is proposed to overcome these problems. The algorithm implements the following two procedures: (i) dimensionality reduction in ROIs of interest with principle component analysis (PCA), and (ii) estimation of the direct causal influences in local brain networks, using conditional Granger causality. Our results show that the proposed method achieves greater accuracy in detecting network connectivity than the commonly used pairwise Granger causality method. Furthermore, the use of PCA components in conjunction with conditional GCM greatly reduces the computational cost relative to the use of individual voxel time series. Copyright 2009 Wiley-Liss, Inc

The good, the bad, and the timely: how temporal order and moral judgment influence causal selection

PubMed Central

Reuter, Kevin; Kirfel, Lara; van Riel, Raphael; Barlassina, Luca

2014-01-01

Causal selection is the cognitive process through which one or more elements in a complex causal structure are singled out as actual causes of a certain effect. In this paper, we report on an experiment in which we investigated the role of moral and temporal factors in causal selection. Our results are as follows. First, when presented with a temporal chain in which two human agents perform the same action one after the other, subjects tend to judge the later agent to be the actual cause. Second, the impact of temporal location on causal selection is almost canceled out if the later agent did not violate a norm while the former did. We argue that this is due to the impact that judgments of norm violation have on causal selection—even if the violated norm has nothing to do with the obtaining effect. Third, moral judgments about the effect influence causal selection even in the case in which agents could not have foreseen the effect and did not intend to bring it about. We discuss our findings in connection to recent theories of the role of moral judgment in causal reasoning, on the one hand, and to probabilistic models of temporal location, on the other. PMID:25477851

Length of secondary schooling and risk of HIV infection in Botswana: evidence from a natural experiment.

PubMed

De Neve, Jan-Walter; Fink, Günther; Subramanian, S V; Moyo, Sikhulile; Bor, Jacob

2015-08-01

An estimated 2·1 million individuals are newly infected with HIV every year. Cross-sectional and longitudinal studies have reported conflicting evidence for the association between education and HIV risk, and no randomised trial has identified a causal effect for education on HIV incidence. We aimed to use a policy reform in secondary schooling in Botswana to identify the causal effect of length of schooling on new HIV infection. Data for HIV biomarkers and demographics were obtained from the nationally representative household 2004 and 2008 Botswana AIDS Impact Surveys (N=7018). In 1996, Botswana reformed the grade structure of secondary school, expanding access to grade ten and increasing educational attainment for affected cohorts. Using exposure to the policy reform as an instrumental variable, we used two-stage least squares to estimate the causal effect of years of schooling on the cumulative probability that an individual contracted HIV up to their age at the time of the survey. We also assessed the cost-effectiveness of secondary schooling as an HIV prevention intervention in comparison to other established interventions. Each additional year of secondary schooling caused by the policy change led to an absolute reduction in the cumulative risk of HIV infection of 8·1 percentage points (p=0·008), relative to a baseline prevalence of 25·5% in the pre-reform 1980 birth cohort. Effects were particularly large in women (11·6 percentage points, p=0·046). Results were robust to a wide array of sensitivity analyses. Secondary school was cost effective as an HIV prevention intervention by standard metrics (cost per HIV infection averted was US$27 753). Additional years of secondary schooling had a large protective effect against HIV risk in Botswana, particularly for women. Increasing progression through secondary school could be a cost-effective HIV prevention measure in HIV-endemic settings, in addition to yielding other societal benefits. Takemi Program in International Health at the Harvard T.H.Chan School of Public Health, Belgian American Educational Foundation, Fernand Lazard Foundation, Boston University, National Institutes of Health. Copyright © 2015 De Neve et al. Open access article distributed under the terms of CC BY. Published by Elsevier Ltd.. All rights reserved.

Effects of Maternal Work Incentives on Teen Drug Arrests

PubMed Central

Corman, Hope; Dave, Dhaval; Kalil, Ariel; Reichman, Nancy E.

2017-01-01

Purpose This study exploits differences in the implementation of welfare reform across states and over time in the United States in the attempt to identify causal effects of welfare reform on youth arrests for drug-related crimes between 1990 and 2005, the period during which welfare reform unfolded. Methodology Using monthly arrest data from the U.S. Federal Bureau of Investigation's Uniform Crime Reports, we estimate the effects of welfare reform implementation on drug-related arrests among 15–17 year olds in the United States between 1990 and 2005. We use a difference-in-differences (DD) approach that exploits the implementation of welfare reform across states and over time to estimate effects for teens exposed to welfare reform. Findings The findings, based on numerous different model specifications, suggest that welfare reform had no statistically significant effect on teen drug arrests. Most estimates were positive and suggestive of a small (3%) increase in arrests. Originality/Value This study investigated the effects of a broad-based policy change that altered maternal employment, family income, and other family characteristics on youth drug arrests. PMID:28989228

Upper limit set by causality on the tidal deformability of a neutron star

NASA Astrophysics Data System (ADS)

Van Oeveren, Eric D.; Friedman, John L.

2017-04-01

A principal goal of gravitational-wave astronomy is to constrain the neutron star equation of state (EOS) by measuring the tidal deformability of neutron stars. The tidally induced departure of the waveform from that of a point particle [or a spinless binary black hole (BBH)] increases with the stiffness of the EOS. We show that causality (the requirement that the speed of sound be less than the speed of light for a perfect fluid satisfying a one-parameter equation of state) places an upper bound on tidal deformability as a function of mass. Like the upper mass limit, the limit on deformability is obtained by using an EOS with vsound=c for high densities and matching to a low density (candidate) EOS at a matching density of order nuclear saturation density. We use these results and those of Lackey et al. [Phys. Rev. D 89, 043009 (2014), 10.1103/PhysRevD.89.043009] to estimate the resulting upper limit on the gravitational-wave phase shift of a black hole-neutron star (BHNS) binary relative to a BBH. Even for assumptions weak enough to allow a maximum mass of 4 M⊙ (a match at nuclear saturation density to an unusually stiff low-density candidate EOS), the upper limit on dimensionless tidal deformability is stringent. It leads to a still more stringent estimated upper limit on the maximum tidally induced phase shift prior to merger. We comment in an appendix on the relation among causality, the condition vsound

Generalized Recurrent Neural Network accommodating Dynamic Causal Modeling for functional MRI analysis.

PubMed

Wang, Yuan; Wang, Yao; Lui, Yvonne W

2018-05-18

Dynamic Causal Modeling (DCM) is an advanced biophysical model which explicitly describes the entire process from experimental stimuli to functional magnetic resonance imaging (fMRI) signals via neural activity and cerebral hemodynamics. To conduct a DCM study, one needs to represent the experimental stimuli as a compact vector-valued function of time, which is hard in complex tasks such as book reading and natural movie watching. Deep learning provides the state-of-the-art signal representation solution, encoding complex signals into compact dense vectors while preserving the essence of the original signals. There is growing interest in using Recurrent Neural Networks (RNNs), a major family of deep learning techniques, in fMRI modeling. However, the generic RNNs used in existing studies work as black boxes, making the interpretation of results in a neuroscience context difficult and obscure. In this paper, we propose a new biophysically interpretable RNN built on DCM, DCM-RNN. We generalize the vanilla RNN and show that DCM can be cast faithfully as a special form of the generalized RNN. DCM-RNN uses back propagation for parameter estimation. We believe DCM-RNN is a promising tool for neuroscience. It can fit seamlessly into classical DCM studies. We demonstrate face validity of DCM-RNN in two principal applications of DCM: causal brain architecture hypotheses testing and effective connectivity estimation. We also demonstrate construct validity of DCM-RNN in an attention-visual experiment. Moreover, DCM-RNN enables end-to-end training of DCM and representation learning deep neural networks, extending DCM studies to complex tasks. Copyright © 2018 Elsevier Inc. All rights reserved.

Investigating the possible causal role of coffee consumption with prostate cancer risk and progression using Mendelian randomization analysis

PubMed Central

Martin, Richard M.; Geybels, Milan S.; Stanford, Janet L.; Shui, Irene; Eeles, Rosalind; Easton, Doug; Kote‐Jarai, Zsofia; Amin Al Olama, Ali; Benlloch, Sara; Muir, Kenneth; Giles, Graham G; Wiklund, Fredrik; Gronberg, Henrik; Haiman, Christopher A; Schleutker, Johanna; Nordestgaard, Børge G.; Travis, Ruth C; Neal, David; Pashayan, Nora; Khaw, Kay‐Tee; Blot, William; Thibodeau, Stephen; Maier, Christiane; Kibel, Adam S; Cybulski, Cezary; Cannon‐Albright, Lisa; Brenner, Hermann; Park, Jong; Kaneva, Radka; Batra, Jyotsna; Teixeira, Manuel R; Pandha, Hardev; Donovan, Jenny; Munafò, Marcus R.

2016-01-01

Coffee consumption has been shown in some studies to be associated with lower risk of prostate cancer. However, it is unclear if this association is causal or due to confounding or reverse causality. We conducted a Mendelian randomisation analysis to investigate the causal effects of coffee consumption on prostate cancer risk and progression. We used two genetic variants robustly associated with caffeine intake (rs4410790 and rs2472297) as proxies for coffee consumption in a sample of 46,687 men of European ancestry from 25 studies in the PRACTICAL consortium. Associations between genetic variants and prostate cancer case status, stage and grade were assessed by logistic regression and with all‐cause and prostate cancer‐specific mortality using Cox proportional hazards regression. There was no clear evidence that a genetic risk score combining rs4410790 and rs2472297 was associated with prostate cancer risk (OR per additional coffee increasing allele: 1.01, 95% CI: 0.98,1.03) or having high‐grade compared to low‐grade disease (OR: 1.01, 95% CI: 0.97,1.04). There was some evidence that the genetic risk score was associated with higher odds of having nonlocalised compared to localised stage disease (OR: 1.03, 95% CI: 1.01, 1.06). Amongst men with prostate cancer, there was no clear association between the genetic risk score and all‐cause mortality (HR: 1.00, 95% CI: 0.97,1.04) or prostate cancer‐specific mortality (HR: 1.03, 95% CI: 0.98,1.08). These results, which should have less bias from confounding than observational estimates, are not consistent with a substantial effect of coffee consumption on reducing prostate cancer incidence or progression. PMID:27741566

Comparing Methods for Estimating Direct Costs of Adverse Drug Events.

PubMed

Gyllensten, Hanna; Jönsson, Anna K; Hakkarainen, Katja M; Svensson, Staffan; Hägg, Staffan; Rehnberg, Clas

2017-12-01

To estimate how direct health care costs resulting from adverse drug events (ADEs) and cost distribution are affected by methodological decisions regarding identification of ADEs, assigning relevant resource use to ADEs, and estimating costs for the assigned resources. ADEs were identified from medical records and diagnostic codes for a random sample of 4970 Swedish adults during a 3-month study period in 2008 and were assessed for causality. Results were compared for five cost evaluation methods, including different methods for identifying ADEs, assigning resource use to ADEs, and for estimating costs for the assigned resources (resource use method, proportion of registered cost method, unit cost method, diagnostic code method, and main diagnosis method). Different levels of causality for ADEs and ADEs' contribution to health care resource use were considered. Using the five methods, the maximum estimated overall direct health care costs resulting from ADEs ranged from Sk10,000 (Sk = Swedish krona; ~€1,500 in 2016 values) using the diagnostic code method to more than Sk3,000,000 (~€414,000) using the unit cost method in our study population. The most conservative definitions for ADEs' contribution to health care resource use and the causality of ADEs resulted in average costs per patient ranging from Sk0 using the diagnostic code method to Sk4066 (~€500) using the unit cost method. The estimated costs resulting from ADEs varied considerably depending on the methodological choices. The results indicate that costs for ADEs need to be identified through medical record review and by using detailed unit cost data. Copyright © 2017 International Society for Pharmacoeconomics and Outcomes Research (ISPOR). Published by Elsevier Inc. All rights reserved.

Near-field electromagnetic holography for high-resolution analysis of network interactions in neuronal tissue

PubMed Central

Kjeldsen, Henrik D.; Kaiser, Marcus; Whittington, Miles A.

2015-01-01

Background Brain function is dependent upon the concerted, dynamical interactions between a great many neurons distributed over many cortical subregions. Current methods of quantifying such interactions are limited by consideration only of single direct or indirect measures of a subsample of all neuronal population activity. New method Here we present a new derivation of the electromagnetic analogy to near-field acoustic holography allowing high-resolution, vectored estimates of interactions between sources of electromagnetic activity that significantly improves this situation. In vitro voltage potential recordings were used to estimate pseudo-electromagnetic energy flow vector fields, current and energy source densities and energy dissipation in reconstruction planes at depth into the neural tissue parallel to the recording plane of the microelectrode array. Results The properties of the reconstructed near-field estimate allowed both the utilization of super-resolution techniques to increase the imaging resolution beyond that of the microelectrode array, and facilitated a novel approach to estimating causal relationships between activity in neocortical subregions. Comparison with existing methods The holographic nature of the reconstruction method allowed significantly better estimation of the fine spatiotemporal detail of neuronal population activity, compared with interpolation alone, beyond the spatial resolution of the electrode arrays used. Pseudo-energy flow vector mapping was possible with high temporal precision, allowing a near-realtime estimate of causal interaction dynamics. Conclusions Basic near-field electromagnetic holography provides a powerful means to increase spatial resolution from electrode array data with careful choice of spatial filters and distance to reconstruction plane. More detailed approaches may provide the ability to volumetrically reconstruct activity patterns on neuronal tissue, but the ability to extract vectored data with the method presented already permits the study of dynamic causal interactions without bias from any prior assumptions on anatomical connectivity. PMID:26026581

Near-field electromagnetic holography for high-resolution analysis of network interactions in neuronal tissue.

PubMed

Kjeldsen, Henrik D; Kaiser, Marcus; Whittington, Miles A

2015-09-30

Brain function is dependent upon the concerted, dynamical interactions between a great many neurons distributed over many cortical subregions. Current methods of quantifying such interactions are limited by consideration only of single direct or indirect measures of a subsample of all neuronal population activity. Here we present a new derivation of the electromagnetic analogy to near-field acoustic holography allowing high-resolution, vectored estimates of interactions between sources of electromagnetic activity that significantly improves this situation. In vitro voltage potential recordings were used to estimate pseudo-electromagnetic energy flow vector fields, current and energy source densities and energy dissipation in reconstruction planes at depth into the neural tissue parallel to the recording plane of the microelectrode array. The properties of the reconstructed near-field estimate allowed both the utilization of super-resolution techniques to increase the imaging resolution beyond that of the microelectrode array, and facilitated a novel approach to estimating causal relationships between activity in neocortical subregions. The holographic nature of the reconstruction method allowed significantly better estimation of the fine spatiotemporal detail of neuronal population activity, compared with interpolation alone, beyond the spatial resolution of the electrode arrays used. Pseudo-energy flow vector mapping was possible with high temporal precision, allowing a near-realtime estimate of causal interaction dynamics. Basic near-field electromagnetic holography provides a powerful means to increase spatial resolution from electrode array data with careful choice of spatial filters and distance to reconstruction plane. More detailed approaches may provide the ability to volumetrically reconstruct activity patterns on neuronal tissue, but the ability to extract vectored data with the method presented already permits the study of dynamic causal interactions without bias from any prior assumptions on anatomical connectivity. Copyright © 2015 The Authors. Published by Elsevier B.V. All rights reserved.

Diagnostic causal reasoning with verbal information.

PubMed

Meder, Björn; Mayrhofer, Ralf

2017-08-01

In diagnostic causal reasoning, the goal is to infer the probability of causes from one or multiple observed effects. Typically, studies investigating such tasks provide subjects with precise quantitative information regarding the strength of the relations between causes and effects or sample data from which the relevant quantities can be learned. By contrast, we sought to examine people's inferences when causal information is communicated through qualitative, rather vague verbal expressions (e.g., "X occasionally causes A"). We conducted three experiments using a sequential diagnostic inference task, where multiple pieces of evidence were obtained one after the other. Quantitative predictions of different probabilistic models were derived using the numerical equivalents of the verbal terms, taken from an unrelated study with different subjects. We present a novel Bayesian model that allows for incorporating the temporal weighting of information in sequential diagnostic reasoning, which can be used to model both primacy and recency effects. On the basis of 19,848 judgments from 292 subjects, we found a remarkably close correspondence between the diagnostic inferences made by subjects who received only verbal information and those of a matched control group to whom information was presented numerically. Whether information was conveyed through verbal terms or numerical estimates, diagnostic judgments closely resembled the posterior probabilities entailed by the causes' prior probabilities and the effects' likelihoods. We observed interindividual differences regarding the temporal weighting of evidence in sequential diagnostic reasoning. Our work provides pathways for investigating judgment and decision making with verbal information within a computational modeling framework. Copyright © 2017 Elsevier Inc. All rights reserved.

The dynamic interaction between combustible renewables and waste consumption and international tourism: the case of Tunisia.

PubMed

Ben Jebli, Mehdi; Ben Youssef, Slim; Apergis, Nicholas

2015-08-01

This paper employs the autoregressive distributed lag (ARDL) bounds methodological approach to investigate the relationship between economic growth, combustible renewables and waste consumption, carbon dioxide (CO2) emissions, and international tourism for the case of Tunisia spanning the period 1990-2010. The results from the Fisher statistic of both the Wald test and the Johansen test confirm the presence of a long-run relationship among the variables under investigation. The stability of estimated parameters has been tested, while Granger causality tests recommend a short-run unidirectional causality running from economic growth and combustible renewables and waste consumption to CO2 emissions, a bidirectional causality between economic growth and combustible renewables and waste consumption and unidirectional causality running from economic growth and combustible renewables and waste consumption to international tourism. In the long-run, the error correction terms confirm the presence of bidirectional causality relationships between economic growth, CO2 emissions, combustible renewables and waste consumption, and international tourism. Our long-run estimates show that combustible renewables and waste consumption increases international tourism, and both renewables and waste consumption and international tourism increase CO2 emissions and output. We recommend that (i) Tunisia should use more combustible renewables and waste energy as this eliminates wastes from touristic zones and increases the number of tourist arrivals, leading to economic growth, and (ii) a fraction of this economic growth generated by the increase in combustible renewables and waste consumption should be invested in clean renewable energy production (i.e., solar, wind, geothermal) and energy efficiency projects.

Noninvasive Electromagnetic Source Imaging and Granger Causality Analysis: An Electrophysiological Connectome (eConnectome) Approach

PubMed Central

Sohrabpour, Abbas; Ye, Shuai; Worrell, Gregory A.; Zhang, Wenbo

2016-01-01

Objective Combined source imaging techniques and directional connectivity analysis can provide useful information about the underlying brain networks in a non-invasive fashion. Source imaging techniques have been used successfully to either determine the source of activity or to extract source time-courses for Granger causality analysis, previously. In this work, we utilize source imaging algorithms to both find the network nodes (regions of interest) and then extract the activation time series for further Granger causality analysis. The aim of this work is to find network nodes objectively from noninvasive electromagnetic signals, extract activation time-courses and apply Granger analysis on the extracted series to study brain networks under realistic conditions. Methods Source imaging methods are used to identify network nodes and extract time-courses and then Granger causality analysis is applied to delineate the directional functional connectivity of underlying brain networks. Computer simulations studies where the underlying network (nodes and connectivity pattern) is known were performed; additionally, this approach has been evaluated in partial epilepsy patients to study epilepsy networks from inter-ictal and ictal signals recorded by EEG and/or MEG. Results Localization errors of network nodes are less than 5 mm and normalized connectivity errors of ~20% in estimating underlying brain networks in simulation studies. Additionally, two focal epilepsy patients were studied and the identified nodes driving the epileptic network were concordant with clinical findings from intracranial recordings or surgical resection. Conclusion Our study indicates that combined source imaging algorithms with Granger causality analysis can identify underlying networks precisely (both in terms of network nodes location and internodal connectivity). Significance The combined source imaging and Granger analysis technique is an effective tool for studying normal or pathological brain conditions. PMID:27740473

Noninvasive Electromagnetic Source Imaging and Granger Causality Analysis: An Electrophysiological Connectome (eConnectome) Approach.

PubMed

Sohrabpour, Abbas; Ye, Shuai; Worrell, Gregory A; Zhang, Wenbo; He, Bin

2016-12-01

Combined source-imaging techniques and directional connectivity analysis can provide useful information about the underlying brain networks in a noninvasive fashion. Source-imaging techniques have been used successfully to either determine the source of activity or to extract source time-courses for Granger causality analysis, previously. In this work, we utilize source-imaging algorithms to both find the network nodes [regions of interest (ROI)] and then extract the activation time series for further Granger causality analysis. The aim of this work is to find network nodes objectively from noninvasive electromagnetic signals, extract activation time-courses, and apply Granger analysis on the extracted series to study brain networks under realistic conditions. Source-imaging methods are used to identify network nodes and extract time-courses and then Granger causality analysis is applied to delineate the directional functional connectivity of underlying brain networks. Computer simulations studies where the underlying network (nodes and connectivity pattern) is known were performed; additionally, this approach has been evaluated in partial epilepsy patients to study epilepsy networks from interictal and ictal signals recorded by EEG and/or Magnetoencephalography (MEG). Localization errors of network nodes are less than 5 mm and normalized connectivity errors of ∼20% in estimating underlying brain networks in simulation studies. Additionally, two focal epilepsy patients were studied and the identified nodes driving the epileptic network were concordant with clinical findings from intracranial recordings or surgical resection. Our study indicates that combined source-imaging algorithms with Granger causality analysis can identify underlying networks precisely (both in terms of network nodes location and internodal connectivity). The combined source imaging and Granger analysis technique is an effective tool for studying normal or pathological brain conditions.

Gene–Environment Correlation: Difficulties and a Natural Experiment–Based Strategy

PubMed Central

Li, Jiang; Liu, Hexuan; Guo, Guang

2013-01-01

Objectives. We explored how gene–environment correlations can result in endogenous models, how natural experiments can protect against this threat, and if unbiased estimates from natural experiments are generalizable to other contexts. Methods. We compared a natural experiment, the College Roommate Study, which measured genes and behaviors of college students and their randomly assigned roommates in a southern public university, with observational data from the National Longitudinal Study of Adolescent Health in 2008. We predicted exposure to exercising peers using genetic markers and estimated environmental effects on alcohol consumption. A mixed-linear model estimated an alcohol consumption variance that was attributable to genetic markers and across peer environments. Results. Peer exercise environment was associated with respondent genotype in observational data, but not in the natural experiment. The effects of peer drinking and presence of a general gene–environment interaction were similar between data sets. Conclusions. Natural experiments, like random roommate assignment, could protect against potential bias introduced by gene–environment correlations. When combined with representative observational data, unbiased and generalizable causal effects could be estimated. PMID:23927502

Altered effective connectivity within default mode network in major depression disorder

NASA Astrophysics Data System (ADS)

Li, Liang; Li, Baojuan; Bai, Yuanhan; Wang, Huaning; Zhang, Linchuan; Cui, Longbiao; Lu, Hongbing

2016-03-01

Understanding the neural basis of Major Depressive Disorder (MDD) is important for the diagnosis and treatment of this mental disorder. The default mode network (DMN) is considered to be highly involved in the MDD. To find directed interaction between DMN regions associated with the development of MDD, the effective connectivity within the DMN of the MDD patients and matched healthy controls was estimated by using a recently developed spectral dynamic causal modeling. Sixteen patients with MDD and sixteen matched healthy control subjects were included in this study. While the control group underwent the resting state fMRI scan just once, all patients underwent resting state fMRI scans before and after two months' treatment. The spectral dynamic causal modeling was used to estimate directed connections between four DMN nodes. Statistical analysis on connection strengths indicated that efferent connections from the medial frontal cortex (MFC) to posterior cingulate cortex (PCC) and to right parietal cortex (RPC) were significant higher in pretreatment MDD patients than those of the control group. After two-month treatment, the efferent connections from the MFC decreased significantly, while those from the left parietal cortex (LPC) to MFC, PCC and RPC showed a significant increase. These findings suggest that the MFC may play an important role for inhibitory conditioning of the DMN, which was disrupted in MDD patients. It also indicates that disrupted suppressive function of the MFC could be effectively restored after two-month treatment.

Effective connectivity between superior temporal gyrus and Heschl's gyrus during white noise listening: linear versus non-linear models.

PubMed

Hamid, Ka; Yusoff, An; Rahman, Mza; Mohamad, M; Hamid, Aia

2012-04-01

This fMRI study is about modelling the effective connectivity between Heschl's gyrus (HG) and the superior temporal gyrus (STG) in human primary auditory cortices. MATERIALS #ENTITYSTARTX00026; Ten healthy male participants were required to listen to white noise stimuli during functional magnetic resonance imaging (fMRI) scans. Statistical parametric mapping (SPM) was used to generate individual and group brain activation maps. For input region determination, two intrinsic connectivity models comprising bilateral HG and STG were constructed using dynamic causal modelling (DCM). The models were estimated and inferred using DCM while Bayesian Model Selection (BMS) for group studies was used for model comparison and selection. Based on the winning model, six linear and six non-linear causal models were derived and were again estimated, inferred, and compared to obtain a model that best represents the effective connectivity between HG and the STG, balancing accuracy and complexity. Group results indicated significant asymmetrical activation (p(uncorr) < 0.001) in bilateral HG and STG. Model comparison results showed strong evidence of STG as the input centre. The winning model is preferred by 6 out of 10 participants. The results were supported by BMS results for group studies with the expected posterior probability, r = 0.7830 and exceedance probability, ϕ = 0.9823. One-sample t-tests performed on connection values obtained from the winning model indicated that the valid connections for the winning model are the unidirectional parallel connections from STG to bilateral HG (p < 0.05). Subsequent model comparison between linear and non-linear models using BMS prefers non-linear connection (r = 0.9160, ϕ = 1.000) from which the connectivity between STG and the ipsi- and contralateral HG is gated by the activity in STG itself. We are able to demonstrate that the effective connectivity between HG and STG while listening to white noise for the respective participants can be explained by a non-linear dynamic causal model with the activity in STG influencing the STG-HG connectivity non-linearly.

Instrumental variables estimates of peer effects in social networks.

PubMed

An, Weihua

2015-03-01

Estimating peer effects with observational data is very difficult because of contextual confounding, peer selection, simultaneity bias, and measurement error, etc. In this paper, I show that instrumental variables (IVs) can help to address these problems in order to provide causal estimates of peer effects. Based on data collected from over 4000 students in six middle schools in China, I use the IV methods to estimate peer effects on smoking. My design-based IV approach differs from previous ones in that it helps to construct potentially strong IVs and to directly test possible violation of exogeneity of the IVs. I show that measurement error in smoking can lead to both under- and imprecise estimations of peer effects. Based on a refined measure of smoking, I find consistent evidence for peer effects on smoking. If a student's best friend smoked within the past 30 days, the student was about one fifth (as indicated by the OLS estimate) or 40 percentage points (as indicated by the IV estimate) more likely to smoke in the same time period. The findings are robust to a variety of robustness checks. I also show that sharing cigarettes may be a mechanism for peer effects on smoking. A 10% increase in the number of cigarettes smoked by a student's best friend is associated with about 4% increase in the number of cigarettes smoked by the student in the same time period. Copyright © 2014 Elsevier Inc. All rights reserved.

Quantum-coherent mixtures of causal relations

NASA Astrophysics Data System (ADS)

Maclean, Jean-Philippe W.; Ried, Katja; Spekkens, Robert W.; Resch, Kevin J.

2017-05-01

Understanding the causal influences that hold among parts of a system is critical both to explaining that system's natural behaviour and to controlling it through targeted interventions. In a quantum world, understanding causal relations is equally important, but the set of possibilities is far richer. The two basic ways in which a pair of time-ordered quantum systems may be causally related are by a cause-effect mechanism or by a common-cause acting on both. Here we show a coherent mixture of these two possibilities. We realize this nonclassical causal relation in a quantum optics experiment and derive a set of criteria for witnessing the coherence based on a quantum version of Berkson's effect, whereby two independent causes can become correlated on observation of their common effect. The interplay of causality and quantum theory lies at the heart of challenging foundational puzzles, including Bell's theorem and the search for quantum gravity.

Quantum-coherent mixtures of causal relations

PubMed Central

MacLean, Jean-Philippe W.; Ried, Katja; Spekkens, Robert W.; Resch, Kevin J.

2017-01-01

Understanding the causal influences that hold among parts of a system is critical both to explaining that system's natural behaviour and to controlling it through targeted interventions. In a quantum world, understanding causal relations is equally important, but the set of possibilities is far richer. The two basic ways in which a pair of time-ordered quantum systems may be causally related are by a cause-effect mechanism or by a common-cause acting on both. Here we show a coherent mixture of these two possibilities. We realize this nonclassical causal relation in a quantum optics experiment and derive a set of criteria for witnessing the coherence based on a quantum version of Berkson's effect, whereby two independent causes can become correlated on observation of their common effect. The interplay of causality and quantum theory lies at the heart of challenging foundational puzzles, including Bell's theorem and the search for quantum gravity. PMID:28485394

Quantum-coherent mixtures of causal relations.

PubMed

MacLean, Jean-Philippe W; Ried, Katja; Spekkens, Robert W; Resch, Kevin J

2017-05-09

Understanding the causal influences that hold among parts of a system is critical both to explaining that system's natural behaviour and to controlling it through targeted interventions. In a quantum world, understanding causal relations is equally important, but the set of possibilities is far richer. The two basic ways in which a pair of time-ordered quantum systems may be causally related are by a cause-effect mechanism or by a common-cause acting on both. Here we show a coherent mixture of these two possibilities. We realize this nonclassical causal relation in a quantum optics experiment and derive a set of criteria for witnessing the coherence based on a quantum version of Berkson's effect, whereby two independent causes can become correlated on observation of their common effect. The interplay of causality and quantum theory lies at the heart of challenging foundational puzzles, including Bell's theorem and the search for quantum gravity.