Applying propensity scores estimated in a full cohort to adjust for confounding in subgroup analyses
Version of Record online: 8 DEC 2011
Copyright © 2011 John Wiley & Sons, Ltd.
Pharmacoepidemiology and Drug Safety
Volume 21, Issue 7, pages 697–709, July 2012
How to Cite
Rassen, J. A., Glynn, R. J., Rothman, K. J., Setoguchi, S. and Schneeweiss, S. (2012), Applying propensity scores estimated in a full cohort to adjust for confounding in subgroup analyses. Pharmacoepidem. Drug Safe., 21: 697–709. doi: 10.1002/pds.2256
- Issue online: 4 JUL 2012
- Version of Record online: 8 DEC 2011
- Manuscript Accepted: 5 SEP 2011
- Manuscript Revised: 2 SEP 2011
- Manuscript Received: 3 FEB 2011
- National Institute on Aging. Grant Number: AG023178
- propensity scores;
- confounding factors (epidemiology);
- multicenter study (publication type);
- epidemiologic methods;
- effect modifiers (epidemiology);
- comparative effectiveness research
A correctly specified propensity score (PS) estimated in a cohort (“cohort PS”) should, in expectation, remain valid in a subgroup population.
We sought to determine whether using a cohort PS can be validly applied to subgroup analyses and, thus, add efficiency to studies with many subgroups or restricted data.
In each of three cohort studies, we estimated a cohort PS, defined five subgroups, and then estimated subgroup-specific PSs. We compared difference in treatment effect estimates for subgroup analyses adjusted by cohort PSs versus subgroup-specific PSs. Then, over 10 million times, we simulated a population with known characteristics of confounding, subgroup size, treatment interactions, and treatment effect and again assessed difference in point estimates.
We observed that point estimates in most subgroups were substantially similar with the two methods of adjustment. In simulations, the effect estimates differed by a median of 3.4% (interquartile (IQ) range 1.3–10.0%). The IQ range exceeded 10% only in cases where the subgroup had < 1000 patients or few outcome events.
Our empirical and simulation results indicated that using a cohort PS in subgroup analyses was a feasible approach, particularly in larger subgroups. Copyright © 2011 John Wiley & Sons, Ltd.