Studying noncollapsibility of the odds ratio with marginal structural and logistic regression models

Abstract

One approach to quantifying the magnitude of confounding in observational studies is to compare estimates with and without adjustment for a covariate, but this strategy is known to be defective for noncollapsible measures such as the odds ratio. Comparing estimates from marginal structural and standard logistic regression models, the total difference between crude and conditional effects can be decomposed into the sum of a noncollapsibility effect and confounding bias. We provide an analytic approach to assess the noncollapsibility effect in a point-exposure study and provide a general formula for expressing the noncollapsibility effect. Next, we provide a graphical approach that illustrates the relationship between the noncollapsibility effect and the baseline risk, and reveals the behavior of the noncollapsibility effect for a range of different exposure and covariate effects. Various observations about noncollapsibility can be made from the different scenarios with or without confounding; for example, the magnitude of effect of the covariate plays a more important role in the noncollapsibility effect than does that of the effect of the exposure. In order to explore the noncollapsibility effect of the odds ratio in the presence of time-varying confounding, we simulated an observational cohort study. The magnitude of noncollapsibility was generally comparable to the effect in the point-exposure study in our simulation settings. Finally, in an applied example we demonstrate that collapsibility can have an important impact on estimation in practice.

Keywords

noncollapsibility odds ratio marginal structural model logistic regression model confounding bias

Get full access to this article

View all access options for this article.

References

Whittemore

. Collapsibility of multidimensional contingency tables. J R Stat Soc Series B Stat Methodol 1978; 40: 328–340.

Ducharme

LePage

. Testing collapsibility in contingency tables. J R Stat Soc Series B Stat Methodol 1986; 48: 197–205.

Greenland

Mickey

. Closed-form and dually consistent methods for inference on collapsibility in 2 * 2*K and 2* J* K tables. J R Stat Soc Series C Appl Stat 1988; 37: 335–343.

Greenland

Robins

Pearl

. Confounding and collapsibility in causal inference. Stat Sci 1999; 14: 29–46.

Newman

. Biostatistical methods in epidemiology, New York: Wiley, 2001, pp. 49–50.

Clogg

Petkova

Shihadeh

. Statistical methods for analyzing collapsibility in regression models. J Educ Behav Stat 1992; 17: 51–74.

Greenland

Morgenstern

. Confounding in health research. Annu Rev Public Health 2001; 22: 189–212.

Groenwold

RHH

Moons

KGM

Peelen

. Reporting of treatment effects from randomized trials: A plea for multivariable risk ratios. Contemp Clin Trials 2011; 32: 399–402.

Gail

Wieand

Piantadosi

. Biased estimates of treatment effect in randomized experiments with nonlinear regressions and omitted covariates. Biometrika 1984; 71: 431–444.

10.

Wermuth

. Parametric collapsibility and lack of moderating effects in contingency tables with a dichotomous response variable. J R Stat Soc Series B Stat Methodol 1987; 49: 353–364.

11.

Shapiro

. Collapsing contingency tables—a geometric approach. Am Stat 1982; 36: 43–46.

12.

Geng

. Conditions for non-confounding and collapsibility without knowledge of completely constructed causal diagrams. Scand J Stat 2002; 29: 169–181.

13.

Hernán

Clayton

Keiding

. The Simpson’s paradox unraveled. Int J Epidemiol 2011; 40: 780–785.

14.

Greenland

Pearl

. Adjustments and their consequences—collapsibility analysis using graphical models. Int Stat Rev 2011; 79: 401–426.

15.

Grayson

. Confounding confounding. Am J Epidemiol 1987; 126: 546–553.

16.

Greenland

Morgenstern

Poole

. RE: Confounding confounding (letter). Am J Epidemiol 1989; 129: 1086–1089.

17.

Becher

. The concept of residual confounding in regression models and some applications. Stat Med 1992; 11: 1747–1758.

18.

Yanagawa

. Case-control studies: Assessing the effect of a confounding factor. Biometrika 1984; 71: 191–194.

19.

Guo

Geng

. Collapsibility of logistic regression coefficients. J R Stat Soc Series B Stat Methodol 1995; 57: 263–267.

20.

Newman

. Commonalities in the classical, collapsibility and counterfactual concepts of confounding. J Clin Epidemiol 2004; 57: 325–329.

21.

Greenland

. Interpretation and choice of effect measures in epidemiologic analyses. Am J Epidemiol 1987; 125: 761–768.

22.

Samuels

. Matching and design efficiency in epidemiological studies. Biometrika 1981; 68: 577–588.

23.

Janes

Dominici

Zeger

. On quantifying the magnitude of confounding. Biostatistics 2010; 11: 572–582.

24.

Neuhaus

Kalbfleisch

Hauck

. A comparison of cluster-specific and population-averaged approaches for analyzing correlated binary data. Int Stat Rev 1991; 59: 25–35.

25.

Flanders

Khoury

. Indirect assessment of confounding: Graphic description and limits on effect of adjusting for covariates. Epidemiology 1990; 1: 239–246.

26.

Hernán

Brumback

Robins

. Marginal structural models to estimate the causal effect of zidovudine on the survival of HIV-positive men. Epidemiology 2000; 11: 561–570.

27.

Kramer

Chalmers

Hodnett

. Promotion of Breastfeeding Intervention Trial (PROBIT): A randomized trial in the Republic of Belarus. JAMA 2001; 285: 413–420.

28.

Kramer

Guo

Platt

. Breastfeeding and infant growth: Biology or bias? Pediatrics 2002; 110: 343–347.

29.

Kramer

Moodie

Dahhou

. Breastfeeding and infant size: Evidence of reverse causality. Am J Epidemiol 2011; 173: 978–983.

30.

Maldonado

Greenland

. Estimating causal effects. Int J Epidemiol 2002; 31: 422–429.

31.

Miettinen

Cook

. Confounding: Essence and detection. Am J Epidemiol 1981; 114: 593–603.

32.

Greenland

Robins

. Identifiability, exchangeability, and epidemiological confounding. Int J Epidemiol 1986; 15: 413–419.

33.

Pang

Kaufman

Platt

. Mixing of confounding and non-collapsibility: A notable deficiency of the odds ratio. Am J Cardiol 2013; 111: 302–303.

34.

Thomas R. Measure of dependence, confounding and collapsibility. SISMID 2011; Causal Model Lecture 3 (unpublished course notes obtained from the author).

35.

Prentice

. A case-cohort design for epidemiologic cohort studies and disease prevention trials. Biometrika 1986; 73: 1–11.

36.

Norton EC. Log odds and ends. NBER working paper series. Working Paper 18252. http://www.nber.org/papers/w18252.

37.

Greenland

. Model-based estimation of relative risks and other epidemiologic measures in studies of common outcomes and in case-control studies. Am J Epidemiol 2004; 160: 301–305.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

1.65 MB