Advertisement

International Journal of Public Health

, Volume 55, Issue 6, pp 701–703 | Cite as

Using Directed Acyclic Graphs to detect limitations of traditional regression in longitudinal studies

  • Erica E. M. MoodieEmail author
  • D. A. Stephens
Hints & Kinks

Abstract

Introduction

Longitudinal data are increasingly available to health researchers; these present challenges not encountered in cross-sectional data, not the least of which is the presence of time-varying confounding variables and intermediate effects.

Objectives

We review confounding and mediation in a longitudinal setting and introduce causal graphs to explain the bias that arises from conventional analyses.

Conclusions

When both time-varying confounding and mediation are present in the data, traditional regression models result in estimates of effect coefficients that are systematically incorrect, or biased. In a companion paper (Moodie and Stephens in Int J Publ Health, 2010b, this issue), we describe a class of models that yield unbiased estimates in a longitudinal setting.

Keywords

Confounding Mediation Directed Acyclic Graphs Longitudinal data 

Notes

Acknowledgments

Both authors acknowledge funding from the Natural Sciences and Engineering Research Council of Canada (NSERC). Moodie also acknowledges funding from the Canadian Institutes of Health Research (CIHR).

References

  1. Bodnar LM, Davidian M, Siega-Riz AM, Tsiatis A (2004) Marginal structural models for analyzing causal effects of time-dependent treatments: an application in perinatal epidemiology. Am J Epidemiol 159:926–934Google Scholar
  2. Brotman RM, Klebanoff MA, Nansel TR, Andrews WW, Schwebke JR, Zhang J, Yu KF, Zenilman JM, Scharfstein DO (2008) A longitudinal study of vaginal douching and bacterial vaginosis a marginal structural modeling analysis. Am J Epidemiol 168:188–196Google Scholar
  3. Greenland S, Pearl J, Robins JM (1999) Causal diagrams for epidemiologic research. Epidemiology 10:37–48CrossRefPubMedGoogle Scholar
  4. Moodie EEM, Stephens DA (2010a) Estimation of dose-response functions for longitudinal data using the generalized propensity score. Stat Methods Med Res. doi: 10.1177/0962280209340213
  5. Moodie EEM, Stephens DA (2010b) Marginal structural models: unbiased estimation for longitudinal studies. Int J Publ Health (this issue)Google Scholar
  6. Pearl J (2009) Causality, 2nd edn. Cambridge University Press, LondonGoogle Scholar
  7. Stewart CE, Fielder AR, Stephens DA, Moseley MJ (2002) Design of the Monitored Occlusion Treatment of Amblyopia Study (MOTAS). Br J Ophthalmol 86:915–919CrossRefPubMedGoogle Scholar
  8. Stewart CE, Moseley MJ, Stephens DA, Fielder AR (2004) Treatment dose-response in amblyopia therapy: the Monitored Occlusion Treatment of Amblyopia Study (MOTAS). Investig Ophthalmol Vis Sci 45:3048–3054CrossRefGoogle Scholar

Copyright information

© Swiss School of Public Health 2010

Authors and Affiliations

  1. 1.Department of Epidemiology and BiostatisticsMcGill UniversityMontrealCanada
  2. 2.Department of Mathematics and StatisticsMcGill UniversityMontrealCanada

Personalised recommendations