Skip to main content

Advertisement

SpringerLink
Log in

Identification of operating mediation and mechanism in the sufficient-component cause framework

  • METHODS
  • Published:
European Journal of Epidemiology Aims and scope Submit manuscript

Abstract

The assessment of mediation and mechanism is one way to more deeply explore cause-effect relationships, providing a stronger test and explanation of the observed associations. Most previous studies have described direct and indirect effects in terms of potential outcomes and response types, exploring mediation analysis in the counterfactual (= potential-outcome) framework. A recent paper by Hafeman (Eur J Epidemiol 23(11):711–721, 2008) provided a conceptual description of mediation in the sufficient-component cause framework, and VanderWeele (Eur J Epidemiol 24(5):217–224, 2009) explored the distinctions and relationships between the concepts of mediation and mechanism. This study builds on this prior work and demonstrates that further insight can be given by elucidating the concepts of mediation and mechanism in the sufficient-component cause framework, distinguishing their operation from presence. The careful consideration of the concepts of mediation and mechanism can clarify the relationship between them. Then, the present article describes how investigators can identify mediation as well as mechanism by showing their correspondence with direct and indirect effects in the counterfactual framework. This study also demonstrates how a researcher can decompose the total effect into the effect due to mediated paths and the effect due to non-mediated paths in terms of the probabilities of background factors of sufficient causes.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1

Similar content being viewed by others

Abbreviations

PDE:

Pure direct effect

PIE:

Pure indirect effect

TDE:

Total direct effect

TIE:

Total indirect effect

References

  1. Pearl J. Causality: models, reasoning, and inference. 2nd ed. New York, NY: Cambridge University Press; 2009.

    Google Scholar 

  2. Hafeman DM, Schwartz S. Opening the black box: a motivation for the assessment of mediation. Int J Epidemiol. 2009;38(3):838–45.

    Article  PubMed  Google Scholar 

  3. Hafeman DM. A sufficient cause based approach to the assessment of mediation. Eur J Epidemiol. 2008;23(11):711–21.

    Article  PubMed  Google Scholar 

  4. VanderWeele TJ. Mediation and mechanism. Eur J Epidemiol. 2009;24(5):217–24.

    Article  PubMed  Google Scholar 

  5. Hafeman DM. “Proportion explained”: a causal interpretation for standard measures of indirect effect? Am J Epidemiol. 2009;170(11):1443–8.

    Article  PubMed  Google Scholar 

  6. Hafeman DM, VanderWeele TJ. Alternative assumptions for the identification of direct and indirect effects. Epidemiology. 2011. doi:10.1097/EDE.0b013e3181c311b2.

  7. Little RJ, Rubin DB. Causal effects in clinical and epidemiological studies via potential outcomes: concepts and analytical approaches. Annu Rev Public Health. 2000;21:121–45.

    Article  PubMed  CAS  Google Scholar 

  8. Rothman KJ, Greenland S, Poole C, et al. Causation and causal inference. In: Rothman KJ, Greenland S, Lash TL, editors. Modern Epidemiology. 3rd ed. Philadelphia, PA: Wolters Kluwer Health/Lippincott Williams & Wilkins; 2008. p. 5–31.

    Google Scholar 

  9. Greenland S, Rothman KJ, Lash TL. Measures of effect and measures of association. In: Rothman KJ, Greenland S, Lash TL, editors. Modern epidemiology. 3rd ed. Philadelphia, PA: Wolters Kluwer Health/Lippincott Williams & Wilkins; 2008. p. 51–70.

    Google Scholar 

  10. Hernán MA. A definition of causal effect for epidemiological research. J Epidemiol Community Health. 2004;58(4):265–71.

    Article  PubMed  Google Scholar 

  11. Maldonado G, Greenland S. Estimating causal effects. Int J Epidemiol. 2002;31(2):422–9.

    Article  PubMed  Google Scholar 

  12. Greenland S, Lash TL, Rothman KJ. Concepts of interaction. In: Rothman KJ, Greenland S, Lash TL, editors. Modern epidemiology. 3rd ed. Philadelphia, PA: Wolters Kluwer Health/Lippincott Williams & Wilkins; 2008. p. 71–83.

    Google Scholar 

  13. Greenland S, Poole C. Invariants and noninvariants in the concept of interdependent effects. Scand J Work Environ Health. 1988;14(2):125–9.

    PubMed  CAS  Google Scholar 

  14. Miettinen OS. Causal and preventive interdependence. Elementary principles. Scand J Work Environ Health. 1982;8(3):159–68.

    PubMed  CAS  Google Scholar 

  15. Petersen ML, Sinisi SE, van der Laan MJ. Estimation of direct causal effects. Epidemiology. 2006;17(3):276–84.

    Article  PubMed  Google Scholar 

  16. Robins JM. Semantics of causal DAG models and the identification of direct and indirect effects. In: Green PJ, Hjort NL, Richardson S, editors. Highly Structured Stocahstic Systems. Oxford, New York: Oxford statistical science series 2003;27:70–81.

  17. VanderWeele TJ, Vansteelandt S. Conceptual issues concerning mediation, interventions and composition. Stat Interface. 2009;2(4):457–68.

    Google Scholar 

  18. Robins JM, Greenland S. Identifiability and exchangeability for direct and indirect effects. Epidemiology. 1992;3(2):143–55.

    Article  PubMed  CAS  Google Scholar 

  19. VanderWeele TJ. Bias formulas for sensitivity analysis for direct and indirect effects. Epidemiology. 2010;21(4):540–51.

    Article  PubMed  Google Scholar 

  20. Rothman KJ. Causes. Am J Epidemiol. 1976;104(6):587–92.

    PubMed  CAS  Google Scholar 

  21. Khoury MJ, Flanders WD, Greenland S, et al. On the measurement of susceptibility in epidemiologic studies. Am J Epidemiol. 1989;129(1):183–90.

    PubMed  CAS  Google Scholar 

  22. Gatto NM, Campbell UB. Redundant causation from a sufficient cause perspective. Epidemiol Perspect Innov. 2010;7(1):5.

    Article  PubMed  Google Scholar 

  23. VanderWeele TJ. Attributable fractions for sufficient cause interactions. Int J Biostat. 2010;6(2):5.

    Google Scholar 

  24. Greenland S, Robins JM. Conceptual problems in the definition and interpretation of attributable fractions. Am J Epidemiol. 1988;128(6):1185–97.

    PubMed  CAS  Google Scholar 

  25. Robins JM, Greenland S. Estimability and estimation of excess and etiologic fractions. Stat Med. 1989;8(7):845–59.

    Article  PubMed  CAS  Google Scholar 

  26. Robins JM, Greenland S. The probability of causation under a stochastic model for individual risk. Biometrics. 1989;45(4):1125–38.

    Article  PubMed  CAS  Google Scholar 

  27. VanderWeele TJ. Marginal structural models for the estimation of direct and indirect effects. Epidemiology. 2009;20(1):18–26.

    Article  PubMed  Google Scholar 

  28. Vansteelandt S. Estimating direct effects in cohort and case-control studies. Epidemiology. 2009;20(6):851–60.

    Article  PubMed  Google Scholar 

  29. VanderWeele TJ, Vansteelandt S. Odds ratios for mediation analysis for a dichotomous outcome. Am J Epidemiol. 2010;172(12):1339–48.

    Article  PubMed  Google Scholar 

  30. Imai K, Keele L, Tingley D. A general approach to causal mediation analysis. Psychol Methods. 2010;15(4):309–34.

    Article  PubMed  Google Scholar 

  31. Imai K, Keele L, Yamamoto T. Identification, inference and sensitivity analysis for causal mediation effects. Stat Sci. 2010;25(1):51–71.

    Article  Google Scholar 

  32. Flanders WD. On the relationship of sufficient component cause models with potential outcome (counterfactual) models. Eur J Epidemiol. 2006;21(12):847–53.

    Article  PubMed  Google Scholar 

  33. VanderWeele TJ, Hernán MA. From counterfactuals to sufficient component causes and vice versa. Eur J Epidemiol. 2006;21(12):855–8.

    Article  PubMed  Google Scholar 

  34. Suzuki E, Yamamoto E, Tsuda T. On the link between sufficient-cause model and potential-outcome model. Epidemiology. 2011;22(1):131–2.

    Article  PubMed  Google Scholar 

  35. Greenland S, Brumback B. An overview of relations among causal modelling methods. Int J Epidemiol. 2002;31(5):1030–7.

    Article  PubMed  Google Scholar 

Download references

Acknowledgments

We thank Tyler J. VanderWeele for his helpful comments on the earlier version of this manuscript.

Author information

Authors and Affiliations

  1. Department of Epidemiology, Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, 2-5-1 Shikata-cho, Kita-ku, Okayama, 700-8558, Japan

    Etsuji Suzuki

  2. Department of Information Science, Faculty of Informatics, Okayama University of Science, Okayama, Japan

    Eiji Yamamoto

  3. Department of Environmental Epidemiology, Okayama University Graduate School of Environmental Science, Okayama, Japan

    Toshihide Tsuda

Authors
  1. Etsuji Suzuki
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Eiji Yamamoto
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Toshihide Tsuda
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Etsuji Suzuki.

Appendices

Appendix 1

See Tables 4 and 5.

Table 4 Decomposition of total effect into effects due to non-M-mediated paths and M-mediated paths
Full size table
Table 5 Enumeration of 64 MY response types and corresponding potential outcomes
Full size table

Appendix 2

See Fig. 2.

Fig. 2
figure 2

The Relationship between mediation and mechanism. We consider 3 binary variables as follows; exposure X, mediator M, and outcome Y. We illustrate the relationship between mediation of M between X and Y and mechanism involving M for Y if X were set to 1. This Venn diagram shows inclusion relations between mediation and mechanism by distinguishing their presence and operation. This diagram also describes sufficient conditions for them. Note that each condition is described in a rectangle shape

Full size image

Rights and permissions

Reprints and permissions

About this article

Cite this article

Suzuki, E., Yamamoto, E. & Tsuda, T. Identification of operating mediation and mechanism in the sufficient-component cause framework. Eur J Epidemiol 26, 347–357 (2011). https://doi.org/10.1007/s10654-011-9568-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10654-011-9568-3

Keywords

Search

Navigation