Journal of Preventive Medicine and Public Health

The Korean Society for Preventive Medicine (대한예방의학회)
권호 표지
DOI QR코드

DOI QR Code

Introduction to Mediation Analysis and Examples of Its Application to Real-world Data

  • Jung, Sun Jae (Department of Preventive Medicine, Yonsei University College of Medicine)
  • Received : 2021.02.09
  • Accepted : 2021.04.29
  • Published : 2021.05.31
Download PDF
⟨ Previous Next ⟩

Abstract

Traditional epidemiological assessments, which mainly focused on evaluating the statistical association between two major components-the exposure and outcome-have recently evolved to ascertain the in-between process, which can explain the underlying causal pathway. Mediation analysis has emerged as a compelling method to disentangle the complex nature of these pathways. The statistical method of mediation analysis has evolved from simple regression analysis to causal mediation analysis, and each amendment refined the underlying mathematical theory and required assumptions. This short guide will introduce the basic statistical framework and assumptions of both traditional and modern mediation analyses, providing examples conducted with real-world data.

Keywords

Acknowledgement

This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science and ICT (2020R1C1C1003502) and a faculty research grant of Yonsei University College of Medicine for 2019 (6-2019-0114).

References

  1. Weed DL. Beyond black box epidemiology. Am J Public Health 1998;88(1):12-14. https://doi.org/10.2105/AJPH.88.1.12
  2. Valeri L, VanderWeele TJ. SAS macro for causal mediation analysis with survival data. Epidemiology 2015;26(2):e23-e24. https://doi.org/10.1097/EDE.0000000000000253
  3. VanderWeele TJ. Explanation in causal inference: methods for mediation and interaction. Oxford: Oxford University Press; 2015, p. 3-245.
  4. Valeri L, Vanderweele TJ. Mediation analysis allowing for exposure-mediator interactions and causal interpretation: theoretical assumptions and implementation with SAS and SPSS macros. Psychol Methods 2013;18(2):137-150. https://doi.org/10.1037/a0031034
  5. VanderWeele TJ. A unification of mediation and interaction: a 4-way decomposition. Epidemiology 2014;25(5):749-761. https://doi.org/10.1097/ede.0000000000000121
  6. Hayes AF. Introduction to mediation, moderation, and conditional process analysis: a regression-based approach. New York: Guilford Press; 2013, p. 3-216.
  7. Robins JM, Greenland S. Identifiability and exchangeability for direct and indirect effects. Epidemiology 1992;3(2):143-155. https://doi.org/10.1097/00001648-199203000-00013
  8. Judd CM, Kenny DA. Estimating the effects of social interventions. Cambridge: Cambridge University Press; 1981, p. 103-131.
  9. Judd CM, Kenny DA. Process analysis: estimating mediation in treatment evaluations. Eval Rev 1981;5(5):602-619. https://doi.org/10.1177/0193841X8100500502
  10. Baron RM, Kenny DA. The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. J Pers Soc Psychol 1986;51(6): 1173-1182. https://doi.org/10.1037/0022-3514.51.6.1173
  11. Jiang Z, VanderWeele TJ. When is the difference method conservative for assessing mediation? Am J Epidemiol 2015;182(2):105-108. https://doi.org/10.1093/aje/kwv059
  12. Lange T, Vansteelandt S, Bekaert M. A simple unified approach for estimating natural direct and indirect effects. Am J Epidemiol 2012;176(3):190-195. https://doi.org/10.1093/aje/kwr525
  13. Hayes AF. Introduction to mediation, moderation, and conditional process analysis: a regression-based approach. New York: Guilford Press; 2013, p. 551-632.
  14. Preacher KJ, Hayes AF. Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behav Res Methods 2008;40(3):879-891. https://doi.org/10.3758/BRM.40.3.879
  15. Kim K, Jung SJ, Baek JM, Yim HW, Jeong H, Kim DJ, et al. Associations between social network properties and metabolic syndrome and the mediating effect of physical activity: findings from the Cardiovascular and Metabolic Diseases Etiology Research Center (CMERC) Cohort. BMJ Open Diabetes Res Care 2020;8(1):e001272. https://doi.org/10.1136/bmjdrc-2020-001272
  16. Herz A, Petermann S. Beyond interviewer effects in the standardized measurement of ego-centric networks. Soc Networks 2017;50:70-82. https://doi.org/10.1016/j.socnet.2017.01.003
  17. Imai K, Keele L, Tingley D, Yamamoto T. Causal mediation analysis using R. In: Vinod H, editor. Advances in social science research using R. New York: Springer; 2010, p. 129-154.
  18. Tingley D, Yamamoto T, Hirose K, Keele L, Imai K. Mediation: R package for causal mediation analysis. J Stat Softw 2014;59(5):38.
  19. VanderWeele TJ. Simple relations between principal stratification and direct and indirect effects. Stat Probab Lett 2008; 78(17):2957-2962. https://doi.org/10.1016/j.spl.2008.05.029
  20. Pearl J. Direct and indirect effects; 2001 [cited 2021 Feb 1]. Available from: https://arxiv.org/ftp/arxiv/papers/1301/1301.2300.pdf.
  21. VanderWeele TJ. Mediation analysis: a practitioner's guide. Annu Rev Public Health 2016;37:17-32. https://doi.org/10.1146/annurev-publhealth-032315-021402
  22. Berzuini C, Dawid P, Bernardinell L. Causality: statistical perspectives and applications. Chichester: John Wiley & Sons; 2012, p. 6-11.
  23. Lee GB, Kim HC, Jeon YJ, Jung SJ. Association between socioeconomic status and longitudinal sleep quality patterns mediated by depressive symptoms. Sleep 2021:zsab044.
  24. Yung YF, Lamm M, Zhang W, SAS Institute Inc. Causal mediation analysis with the CAUSALMED procedure; 2018 [cited 2021 Feb 1]. Available from: https://www.sas.com/content/dam/SAS/support/en/sas-global-forum-proceedings/2018/1991-2018.pdf.
  25. Fiedler K, Schott M, Meiser T. What mediation analysis can (not) do. J Exp Soc Psychol 2011;47(6):1231-1236. https://doi.org/10.1016/j.jesp.2011.05.007
  26. le Cessie S, Debeij J, Rosendaal FR, Cannegieter SC, Vandenbroucke JP. Quantification of bias in direct effects estimates due to different types of measurement error in the mediator. Epidemiology 2012;23(4):551-560. https://doi.org/10.1097/EDE.0b013e318254f5de
  27. Lange T, Vansteelandt S, Bekaert M. A simple unified approach for estimating natural direct and indirect effects. Am J Epidemiol 2012;176(3):190-195. https://doi.org/10.1093/aje/kwr525