An Introduction to Causal Mediation Analysis With a Comparison of 2 R Packages
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Traditional mediation analysis, which relies on linear regression models, has faced criticism due to its limited suitability for cases involving different types of variables and complex covariates, such as interactions. This can result in unclear definitions of direct and indirect effects. As an alternative, causal mediation analysis using the counterfactual framework has been introduced to provide clearer definitions of direct and indirect effects while allowing for more flexible modeling methods. However, the conceptual understanding of this approach based on the counterfactual framework remains challenging for applied researchers. To address this issue, the present article was written to highlight and illustrate the definitions of causal estimands, including controlled direct effect, natural direct effect, and natural indirect effect, based on the key concept of nested counterfactuals. Furthermore, we recommend using 2 R packages, ‘medflex’ and ‘mediation’, to perform causal mediation analysis and provide public health examples. The article also offers caveats and guidelines for accurate interpretation of the results.
Abstract
Traditional mediation analysis, which relies on linear regression models, has faced criticism due to its limited suitability for cases involving different types of variables and complex covariates, such as interactions. This can result in unclear definitions of direct and indirect effects. As an alternative, causal mediation analysis using the counterfactual framework has been introduced to provide clearer definitions of direct and indirect effects while allowing for more flexible modeling methods. However, the conceptual understanding of this approach based on the counterfactual framework remains challenging for applied researchers. To address this issue, the present article was written to highlight and illustrate the definitions of causal estimands, including controlled direct effect, natural direct effect, and natural indirect effect, based on the key concept of nested counterfactuals. Furthermore, we recommend using 2 R packages, ‘medflex’ and ‘mediation’, to perform causal mediation analysis and provide public health examples. The article also offers caveats and guidelines for accurate interpretation of the results.
참고문헌 (37)
- Hafeman DM, "Opening the black box : a motivation for the assessment of mediation", Int J Epidemiol, 38(3), pp.838~845 (2009)
- Valeri L, "Mediation analysis allowing for exposure-mediator interactions and causal interpretation : theoretical assumptions and implementation with SAS and SPSS macros", Psychol Methods, 18(2), pp.137~150 (2013)
- VanderWeele TJ, "Mediation analysis : a practitioner’s guide", Annu Rev Public Health, 37, pp.17~32 (2016)
- Wright S, "The method of path coefficients", Ann Math Stat, 5(3), pp.161~215 (1934)
- Baron RM, "The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations", J Pers Soc Psychol, 51(6), pp.1173~1182 (1986)
- Imai K, "Identification, inference and sensitivity analysis for causal mediation effects", Stat Sci, 25(1), pp.51~71 (2010)
- Pearl J, "The causal mediation formula--a guide to the assessment of pathways and mechanisms", Prev Sci, 13(4), pp.426~436 (2012)
- Vansteelandt S, "Imputation strategies for the estimation of natural direct and indirect effects", Epidemiol Methods, 1(1), pp.131~158 (2012)
- Vansteelandt S, "Natural direct and indirect effects on the exposed : effect decomposition under weaker assumptions", Biometrics, 68(4), pp.1019~1027 (2012)
- Steen J, "Medflex : an R package for flexible mediation analysis using natural effect models", J Stat Softw, 76, pp.1~46 (2017)
- Tingley D, "Mediation : R package for causal mediation analysis", J Stat Softw, 59(5), pp.1~38 (2014)
- Pearl J. Direct, "Direct and indirect effects", Proceedings of the Seventeenth Conference on Uncertainty in Artificial Intelligence, pp.411~420 (2001), Morgan Kaufmann
- Lange T, "A simple unified approach for estimating natural direct and indirect effects", Am J Epidemiol, 176(3), pp.190~195 (2012)
- Theis Lange, "Applied mediation analyses: a review and tutorial", Epidemiology and Health, 39, pp.1~13 (2017), 한국역학회[DOI]
- Neyman J, "On the application of probability theory to agricultural experiments. Essay on principles", Ann Agric Sci, pp.1~51 (1923)
- Splawa-Neyman J, "On the application of probability theory to agricultural experiments. Essay on principles. Section 9", Stat Sci, 5(4), pp.465~472 (1990)
- Rosenbaum PR, "The central role of the propensity score in observational studies for causal effects", Biometrika, 70(1), pp.41~55 (1983)
- Holland PW, "Causal inference, path analysis and recursive structural equations models", Sociol Methodol, 18, pp.449~484 (1988)
- Rubin DB, "Estimating causal effects of treatments in randomized and nonrandomized studies", J Educ Psychol, 66(5), pp.688~701 (1974)
- Hernan MA, "Causal inference : what if", pp.2020, Chapman & Hall
- Holland PW, "Statistics and causal inference", J Am Stat Assoc, 81(396), pp.945~960 (1986)
- Rubin DB, "For objective causal inference, design trumps analysis", Ann Appl Stat, 2(3), pp.808~840 (2008)
- Hirano K, "Efficient estimation of average treatment effects using the estimated propensity score", Econometrica, 71(4), pp.1161~1189 (2003)
- VanderWeele TJ, "Explanation in causal inference : methods for mediation and interaction", pp.20~64 (2015), Oxford University Press
- Vanderweele TJ, "Odds ratios for mediation analysis for a dichotomous outcome", Am J Epidemiol, 172(12), pp.1339~1348 (2010)
- Robins JM, "Identifiability and exchangeability for direct and indirect effects", Epidemiology, 3(2), pp.143~155 (1992)
- Robins JM, "Highly structured stochastic systems", pp.70~82 (2003), Oxford University Press
- Pearl J, "Causality: statistical perspectives and applications", pp.151~179 (2012), John Wiley & Sons
- Pearl J, "Interpretation and identification of causal mediation", Psychol Methods, 19(4), pp.459~481 (2014)
- Imai K, "A general approach to causal mediation analysis", Psychol Methods, 15(4), pp.309~334 (2010)
- Imai K, "Unpacking the black box of causality : learning about causal mechanisms from experimental and observational studies", Am Polit Sci Rev, 105(4), pp.765~789 (2011)
- Patel L, "Educational inequality in the Dietary Approach to Stop Hypertension diet in the UK : evaluating the mediating role of income", Br J Nutr, 126(12), pp.1897~1903 (2021)
- "Comparison of five software solutions to mediation analysis" (2017)[URL]
- Robins JM, "Causality and psychopathology: finding the determinants of disorders and their cures", pp.103~158 (2011), Oxford University Press
- Tchetgen Tchetgen EJ, "Bounds for pure direct effect", Epidemiology, 25(5), pp.775~776 (2014)
- Díaz I, "Causal mediation analysis for stochastic interventions", J R Stat Soc Series B Stat Methodol, 82(3), pp.661~683 (2020)
- Díaz I, "Nonparametric efficient causal mediation with intermediate confounders", Biometrika, 108(3), pp.627~641 (2021)