Display options
Cite
VanderWeele TJ, Chiba Y. Sensitivity analysis for direct and indirect effects in the presence of exposure-induced mediator-outcome confounders. Epidemiol Biostat Public Health. 2014;11(2)doi: 10.2427/9027.
VanderWeele, T. J., & Chiba, Y. (2014). Sensitivity analysis for direct and indirect effects in the presence of exposure-induced mediator-outcome confounders. Epidemiology, biostatistics and public health, 11(2), . https://doi.org/10.2427/9027
VanderWeele, Tyler J, and Chiba, Yasutaka. "Sensitivity analysis for direct and indirect effects in the presence of exposure-induced mediator-outcome confounders." Epidemiology, biostatistics and public health vol. 11,2 (2014). doi: https://doi.org/10.2427/9027
VanderWeele TJ, Chiba Y. Sensitivity analysis for direct and indirect effects in the presence of exposure-induced mediator-outcome confounders. Epidemiol Biostat Public Health. 2014;11(2). doi: 10.2427/9027. PMID: 25580387; PMCID: PMC4287391.
Copy Download .nbib Format: NLM
Share it on

Epidemiol Biostat Public Health. 2014;11(2). doi: 10.2427/9027.

Sensitivity analysis for direct and indirect effects in the presence of exposure-induced mediator-outcome confounders.

Epidemiology, biostatistics and public health

Tyler J VanderWeele, Yasutaka Chiba

Affiliations

  1. Departments of Epidemiology and biostatistics, Harvard School of Public Health, Boston, MA, USA.
  2. Division of Biostatistics, Clinical Research Center, Kinki University School of Medicine, Osaka, Japan.

PMID: 25580387 PMCID: PMC4287391 DOI: 10.2427/9027

Abstract

Questions of mediation are often of interest in reasoning about mechanisms, and methods have been developed to address these questions. However, these methods make strong assumptions about the absence of confounding. Even if exposure is randomized, there may be mediator-outcome confounding variables. Inference about direct and indirect effects is particularly challenging if these mediator-outcome confounders are affected by the exposure because in this case these effects are not identified irrespective of whether data is available on these exposure-induced mediator-outcome confounders. In this paper, we provide a sensitivity analysis technique for natural direct and indirect effects that is applicable even if there are mediator-outcome confounders affected by the exposure. We give techniques for both the difference and risk ratio scales and compare the technique to other possible approaches.

Keywords: Confounding; direct and indirect effects; mediation; sensitivity analysis

References

  1. Int J Epidemiol. 2002 Feb;31(1):163-5 - PubMed
  2. Epidemiol Perspect Innov. 2004 Oct 08;1(1):4 - PubMed
  3. Epidemiology. 1992 Mar;3(2):143-55 - PubMed
  4. Stat Med. 2009 Feb 15;28(4):558-71 - PubMed
  5. Stat Methods Med Res. 2010 Jun;19(3):237-70 - PubMed
  6. J Stat Plan Inference. 2009 Oct 1;139(10):3473-3487 - PubMed
  7. Epidemiology. 2010 Jul;21(4):540-51 - PubMed
  8. Epidemiology. 2011 Nov;22(6):753-64 - PubMed
  9. Psychol Methods. 2010 Dec;15(4):309-34 - PubMed
  10. Am J Epidemiol. 2010 Dec 15;172(12):1339-48 - PubMed
  11. Paediatr Perinat Epidemiol. 2011 Mar;25(2):111-5 - PubMed
  12. Am J Epidemiol. 2011 Apr 1;173(7):745-51 - PubMed
  13. Am J Epidemiol. 2011 Jul 1;174(1):99-108 - PubMed
  14. Int J Biostat. 2011 Mar 04;7(1):16 - PubMed
  15. Int J Biostat. 2008;4(1):Article 23 - PubMed
  16. Biometrics. 2012 Dec;68(4):1019-27 - PubMed
  17. Psychol Methods. 2013 Jun;18(2):137-50 - PubMed
  18. Ann Stat. 2012 Jun;40(3):1816-1845 - PubMed
  19. J Pers Soc Psychol. 1986 Dec;51(6):1173-82 - PubMed

Publication Types

Grant support