scholarly journals A Major Limitation of the Direction of Causation Model: Non-Shared Environmental Confounding

2019 ◽  
Vol 22 (1) ◽  
pp. 14-26 ◽  
Author(s):  
Stig Hebbelstrup Rye Rasmussen ◽  
Steven Ludeke ◽  
Jacob V. B. Hjelmborg
Keyword(s):  
Causal Relationship ◽  
Behavioral Genetics ◽  
Causal Models ◽  
Causal Effects ◽  
Simulation Studies ◽  
Degree Correlation ◽  
Direction Of Causation ◽  
Modest Degree ◽  
Mode Of Inheritance ◽  
Doc Model

AbstractDetermining (1) the direction of causation and (2) the size of causal effects between two constructs is a central challenge of the scientific study of humans. In the early 1990s, researchers in behavioral genetics invented what was termed the direction of causation (DoC) model to address exactly these two concerns. The model claims that for any two traits whose mode of inheritance is sufficiently different, the direction of causation can be ascertained using a sufficiently large genetically informative sample. Using a series of simulation studies, we demonstrate a major challenge to the DoC model, namely that it is extremely sensitive to even tiny amounts of non-shared confounding. Even under ideal conditions for the DoC model (a large sample,N= 10,000), a large causal relationship (e.g., a causal correlation of .50) with very different modes of inheritance between the two traits (e.g., a pure AE model for one trait and a pure CE model for another trait) and a modest degree (correlation of .10) of non-shared confounding between the two traits results in the choice of the wrong causal models and estimating the wrong causal effects.

Twin Studies and Politics

2019 ◽  
Author(s):  
Levente Littvay
Keyword(s):  
Political Science ◽  
Behavioral Genetics ◽  
Twin Studies ◽  
Social Scientists ◽  
Cross Sectional ◽  
Genetic Transmission ◽  
The Social ◽  
Direction Of Causation ◽  
Social Part ◽  
The Impact

As recently as 2005, John Alford and colleagues surprised political science with their twin study that found empirical evidence of the genetic transmission of political attitudes and behaviors. Reactions in the field were mixed, but one thing is for sure: it is not time to mourn the social part of the social sciences. Genetics is not the deterministic mechanism that social scientists often assume it to be. No specific part of DNA is responsible for anything but minute, indirect effects on political orientations. Genes express themselves differently in different contexts, suggesting that the political phenomenon behavioral political scientists take for granted may be quite volatile; hence, the impact of genetics is also much less stable in its foundations than initially assumed. Twin studies can offer a unique and powerful avenue to study these behavioral processes as they are more powerful than cross-sectional (or even longitudinal) studies not only for understanding heritability but also for asserting the direction of causation, the social (and, of course, genetic) pathways that explain how political phenomena are related to each other. This chapter aims to take the reader through this journey that political science has gone through over the past decade and a half and point to the synergies behavioral political science and behavioral genetics offer to the advancement of the discipline.

scholarly journals A comprehensive evaluation of methods for Mendelian randomization using realistic simulations and an analysis of 38 biomarkers for risk of type 2 diabetes

2021 ◽  
Author(s):  
Guanghao Qi ◽  
Nilanjan Chatterjee
Keyword(s):  
Type 2 Diabetes ◽  
Mendelian Randomization ◽  
Association Studies ◽  
Real Data ◽  
Causal Effects ◽  
Type I ◽  
Genome Wide Association Studies ◽  
Simulation Studies ◽  
Sample Sizes

Abstract Background Previous studies have often evaluated methods for Mendelian randomization (MR) analysis based on simulations that do not adequately reflect the data-generating mechanisms in genome-wide association studies (GWAS) and there are often discrepancies in the performance of MR methods in simulations and real data sets. Methods We use a simulation framework that generates data on full GWAS for two traits under a realistic model for effect-size distribution coherent with the heritability, co-heritability and polygenicity typically observed for complex traits. We further use recent data generated from GWAS of 38 biomarkers in the UK Biobank and performed down sampling to investigate trends in estimates of causal effects of these biomarkers on the risk of type 2 diabetes (T2D). Results Simulation studies show that weighted mode and MRMix are the only two methods that maintain the correct type I error rate in a diverse set of scenarios. Between the two methods, MRMix tends to be more powerful for larger GWAS whereas the opposite is true for smaller sample sizes. Among the other methods, random-effect IVW (inverse-variance weighted method), MR-Robust and MR-RAPS (robust adjust profile score) tend to perform best in maintaining a low mean-squared error when the InSIDE assumption is satisfied, but can produce large bias when InSIDE is violated. In real-data analysis, some biomarkers showed major heterogeneity in estimates of their causal effects on the risk of T2D across the different methods and estimates from many methods trended in one direction with increasing sample size with patterns similar to those observed in simulation studies. Conclusion The relative performance of different MR methods depends heavily on the sample sizes of the underlying GWAS, the proportion of valid instruments and the validity of the InSIDE assumption. Down-sampling analysis can be used in large GWAS for the possible detection of bias in the MR methods.

scholarly journals Probabilistic Reasoning Across the Causal Hierarchy

2020 ◽  
Vol 34 (06) ◽  
pp. 10170-10177 ◽  
Author(s):  
Duligur Ibeling ◽  
Thomas Icard
Keyword(s):  
Bayesian Inference ◽  
Conditional Independence ◽  
Probabilistic Reasoning ◽  
Causal Models ◽  
Causal Effects ◽  
Polynomial Space ◽  
The Third ◽  
Probabilistic Programs

We propose a formalization of the three-tier causal hierarchy of association, intervention, and counterfactuals as a series of probabilistic logical languages. Our languages are of strictly increasing expressivity, the first capable of expressing quantitative probabilistic reasoning—including conditional independence and Bayesian inference—the second encoding do-calculus reasoning for causal effects, and the third capturing a fully expressive do-calculus for arbitrary counterfactual queries. We give a corresponding series of finitary axiomatizations complete over both structural causal models and probabilistic programs, and show that satisfiability and validity for each language are decidable in polynomial space.

scholarly journals Voice break in boys—temporal relations with other pubertal milestones and likely causal effects of BMI

2019 ◽  
Vol 34 (8) ◽  
pp. 1514-1522 ◽  
Author(s):  
A S Busch ◽  
B Hollis ◽  
F R Day ◽  
K Sørensen ◽  
L Aksglaede ◽  
...  
Keyword(s):  
Causal Relationship ◽  
Limit Of Detection ◽  
Research Training ◽  
Population Based ◽  
Male Reproduction ◽  
Causal Effects ◽  
Reproductive Hormone ◽  
Cross Sectional ◽  
Testosterone Levels ◽  
Study Population

Abstract STUDY QUESTION How is timing of voice break related to other male pubertal milestones as well as to BMI? SUMMARY ANSWER We provide a comprehensive temporal analysis of male pubertal milestones, including reproductive hormone dynamics, confirm voice break as a late milestone of male puberty and report a likely causal relationship between higher BMI and earlier age at voice break in men. WHAT IS KNOWN ALREADY Voice break represents a late pubertal milestone and recalled age at voice break is frequently used in epidemiological studies as a measure of puberty. In contrast, clinical studies use mainly testicular enlargement and/or genital tanner stage as the marker of pubertal onset. However, neither correlation of pubertal milestones nor reproductive hormone dynamics have been assessed in detail previously. Further, although BMI and puberty timing are known to be closely linked, cause and effect between these traits are not known. STUDY DESIGN, SIZE, DURATION The study included a population-based mixed cross-sectional and longitudinal cohort (2006–2014, COPENHAGEN Puberty Study) of 730 healthy Danish boys. Data for 55 871 male research participants from the 23andMe study were obtained, including genome-wide single nucleotide polymorphism data and age at voice break. PARTICIPANTS/MATERIALS, SETTING, METHODS We performed a detailed evaluation of pubertal milestones and reproductive hormone levels (study population 1). A Mendelian randomization (MR) approach was used to determine the likely causal link between BMI and timing of voice break (study population 2). MAIN RESULTS AND THE ROLE OF CHANCE Voice break occurred at mean age 13.6 (95% CI: 13.5–13.8) years. At voice break, mean (95% CI) testosterone levels, LH levels and bi-testicular volume were 10.9 (10.0–11.7) nmol/L, 2.4 (2.2–2.5) IU/L and 24 (23–25) mL, respectively. Voice break correlated moderately strongly with timing of male pubertal milestones, including testicular enlargement, gonadarche, pubarche, sweat odor, axillary hair growth and testosterone above limit of detection (r2 range: 0.43–0.61). Timing of all milestones was negatively associated with age-specific BMI (all P ≤ 0.001). MR analyses inferred likely causal effects of higher BMI on earlier voice break in males (−0.35 years/approximate SD, P < 0.001). LIMITATIONS, REASONS FOR CAUTION Participation rate of the population-based cohort was 25%. Further, boys that were followed longitudinally were examined approximately every 6 months limiting the time resolution of pubertal milestones. Using adult BMI as exposure instead of prepubertal BMI in the MR analysis and the known inaccuracies of the testosterone immunoassay at low testosterone levels may be further limitations. WIDER IMPLICATIONS OF THE FINDINGS We provide valuable normative data on the temporal relation of male pubertal milestones. Further, the likely causal relationship between BMI and puberty timing highlights the importance of preventing obesity in childhood. STUDY FUNDING/COMPETING INTEREST(S) This work was supported by Danish Agency for Science, Technology and Innovation (09-067 180); Danish Ministry of the Environment, CeHoS (MST-621-00 065); Capital Region of Denmark (R129-A3966); Ministry of Higher Education and Science (DFF-1331-00 113); Innovation Fund Denmark (InnovationsFonden, 14-2013-4); The International Center for Research and Research Training in Endocrine Disrupting Effects of Male Reproduction and Child Health. B.H., F.R.D., J.R.B.P. and K.K.O. are supported by the Medical Research Council (MC_UU_12015/2). The 23andMe study is supported by the National Human Genome Research Institute of the National Institutes of Health (R44HG006981). Members of the 23andMe Research Team are employees of 23andMe, Inc. and hold stock or stock options in 23andMe. TRIAL REGISTRATION NUMBER NCT01411527

scholarly journals Estimation of causal effects using linear non-Gaussian causal models with hidden variables

2008 ◽  
Vol 49 (2) ◽  
pp. 362-378 ◽  
Author(s):  
Patrik O. Hoyer ◽  
Shohei Shimizu ◽  
Antti J. Kerminen ◽  
Markus Palviainen
Keyword(s):  
Hidden Variables ◽  
Causal Models ◽  
Causal Effects ◽  
Non Gaussian

scholarly journals Balancing Score Adjusted Targeted Minimum Loss-based Estimation

2015 ◽  
Vol 3 (2) ◽  
pp. 139-155 ◽  
Author(s):  
Samuel David Lendle ◽  
Bruce Fireman ◽  
Mark J. van der Laan
Keyword(s):  
Propensity Score ◽  
Bias Reduction ◽  
Average Treatment Effect ◽  
Causal Effects ◽  
Simulation Studies ◽  
Minimum Loss ◽  
Inverse Probability ◽  
Matching Estimator ◽  
Class Of Estimators ◽  
Doubly Robust

AbstractAdjusting 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.

scholarly journals Using genetic information to test causal relationships in cross-sectional data

2012 ◽  
Vol 24 (3) ◽  
pp. 328-344 ◽  
Author(s):  
Brad Verhulst ◽  
Ryne Estabrook
Keyword(s):  
Personality Traits ◽  
Genetic Information ◽  
Political Attitudes ◽  
Genetic Relationships ◽  
Genetic Models ◽  
Cross Sectional ◽  
Structural Pattern ◽  
Direction Of Causation ◽  
Personality And Politics ◽  
Doc Model

Cross-sectional data from twins contain information that can be used to derive a test of causality between traits. This test of directionality is based upon the fact that genetic relationships between family members conform to an established structural pattern. In this paper we examine several common methods for empirically testing causality as well as several genetic models that we build on for the Direction of Causation (DoC) model. We then discuss the mathematical components of the DoC model and highlight limitations of the model and potential solutions to these limitations. We conclude by presenting an example from the personality and politics literature that has begun to explore the question whether or not personality traits cause people to hold specific political attitudes.

scholarly journals Linear Markovian models for lag exposure assessment

2018 ◽  
Vol 55 (2) ◽  
pp. 179-195 ◽  
Author(s):  
Alessandro Magrini
Keyword(s):  
Linear Regression ◽  
Exposure Assessment ◽  
Causal Models ◽  
Causal Effects ◽  
Regression Coefficients ◽  
Standard Approach ◽  
General Context ◽  
Time Lags ◽  
Markovian Models ◽  
Distributed Lag

SummaryLinear regression with temporally delayed covariates (distributed-lag linear regression) is a standard approach to lag exposure assessment, but it is limited to a single biomarker of interest and cannot provide insights on the relationships holding among the pathogen exposures, thus precluding the assessment of causal effects in a general context. In this paper, to overcome these limitations, distributed-lag linear regression is applied to Markovian structural causal models. Dynamic causal effects are defined as a function of regression coefficients at different time lags. The proposed methodology is illustrated using a simple lag exposure assessment problem.

scholarly journals Causal inference under over-simplified longitudinal causal models

2021 ◽  
Vol 0 (0) ◽  
Author(s):  
Lola Étiévant ◽  
Vivian Viallon
Keyword(s):  
Causal Inference ◽  
Causal Effect ◽  
Sufficient Conditions ◽  
Repeated Measurements ◽  
Causal Models ◽  
Causal Effects ◽  
Sensitivity Analyses ◽  
Time Varying ◽  
Weighted Averages

Abstract Many causal models of interest in epidemiology involve longitudinal exposures, confounders and mediators. However, repeated measurements are not always available or used in practice, leading analysts to overlook the time-varying nature of exposures and work under over-simplified causal models. Our objective is to assess whether – and how – causal effects identified under such misspecified causal models relates to true causal effects of interest. We derive sufficient conditions ensuring that the quantities estimated in practice under over-simplified causal models can be expressed as weighted averages of longitudinal causal effects of interest. Unsurprisingly, these sufficient conditions are very restrictive, and our results state that the quantities estimated in practice should be interpreted with caution in general, as they usually do not relate to any longitudinal causal effect of interest. Our simulations further illustrate that the bias between the quantities estimated in practice and the weighted averages of longitudinal causal effects of interest can be substantial. Overall, our results confirm the need for repeated measurements to conduct proper analyses and/or the development of sensitivity analyses when they are not available.

Heterogeneous Causal Effect of Polysubstance Usage on Drug Overdose (Preprint)

2021 ◽  
Author(s):  
Mohammad Arif Ul Alam
Keyword(s):  
Drug Overdose ◽  
Causal Effect ◽  
Association Studies ◽  
Causal Models ◽  
Causal Effects ◽  
Opioid Overdose ◽  
Critical Question ◽  
Electronic Health Record Data ◽  
Drug Usage ◽  
Health Crisis

BACKGROUND Drug overdose has become a public health crisis in United States with devastating consequences. However, most of the drug overdose incidences are the consequence of recitative polysubstance usage over a defined period of time which can be happened by either the intentional usage of required drug with other drugs or by accident. Thus, predicting the effects of polysubstance usage is extremely important for clinicians to decide which combination of drugs should be prescribed. Although, machine learning community has made great progress toward using such rich models for supervised prediction, precision medicine problem such as polysubstance usage effects on drug overdose requires heterogeneous causal models, for which there is significantly less theoretical and practical guidance available. Recent advancement of structural causal models can provide ample insights of causal effects from observational data via identifiable causal directed graphs. OBJECTIVE Develop a system to identify heterogeneous causal effect of polysubstance usage from large electronic health record data METHODS We propose a system to estimate heterogeneous concurrent drug usage effects on overdose estimation, that consists of efficient co-variate selection, subgroup selection, generation of and heterogeneous causal effect estimation. Although, there has been several association studies have been proposed in the state-of-art methods, heterogeneous causal effects have never been studied in concurrent drug usage and drug overdose problem. We apply our framework to answer a critical question, ”can concurrent usage of benzodiazepines and opioids has heterogeneous causal effects on opioid overdose epidemic?” RESULTS Using Truven MarketScan claim data collected from 2001 to 2013 have shown significant promise of our proposed framework’s efficacy. Our efficient causal inference model estimated that the causal effect is higher (19%) than the regression studies (15%) to estimate the risks associated with the concurrent usage of opioid and benzodiazepines on opioid overdose. CONCLUSIONS Our generic framework can be a foundation of investigating concurrent events’ causal effects on any outcome that involves heterogeneity

Sign in / Sign up

Export Citation Format

Share Document