The performance of different propensity score methods for estimating marginal hazard ratios

Statistics in Medicine - Tập 32 Số 16 - Trang 2837-2849 - 2013
Peter C. Austin1,2,3
1Dalla Lana School of Public Health, University of Toronto, Toronto, Ontario, Canada
2Institute for Clinical Evaluative Sciences, Toronto, Ontario, Canada
3Institute of Health Management, Policy and Evaluation, University of Toronto, Toronto, Ontario, Canada

Tóm tắt

Propensity score methods are increasingly being used to reduce or minimize the effects of confounding when estimating the effects of treatments, exposures, or interventions when using observational or non‐randomized data. Under the assumption of no unmeasured confounders, previous research has shown that propensity score methods allow for unbiased estimation of linear treatment effects (e.g., differences in means or proportions). However, in biomedical research, time‐to‐event outcomes occur frequently. There is a paucity of research into the performance of different propensity score methods for estimating the effect of treatment on time‐to‐event outcomes. Furthermore, propensity score methods allow for the estimation of marginal or population‐average treatment effects. We conducted an extensive series of Monte Carlo simulations to examine the performance of propensity score matching (1:1 greedy nearest‐neighbor matching within propensity score calipers), stratification on the propensity score, inverse probability of treatment weighting (IPTW) using the propensity score, and covariate adjustment using the propensity score to estimate marginal hazard ratios. We found that both propensity score matching and IPTW using the propensity score allow for the estimation of marginal hazard ratios with minimal bias. Of these two approaches, IPTW using the propensity score resulted in estimates with lower mean squared error when estimating the effect of treatment in the treated. Stratification on the propensity score and covariate adjustment using the propensity score result in biased estimation of both marginal and conditional hazard ratios. Applied researchers are encouraged to use propensity score matching and IPTW using the propensity score when estimating the relative effect of treatment on time‐to‐event outcomes. Copyright © 2012 John Wiley & Sons, Ltd.

Từ khóa


Tài liệu tham khảo

Cook TD, 2008, Introduction to Statistical Methods for Clinical Trials

10.1093/biomet/70.1.41

10.1002/pds.969

10.1002/sim.3150

10.1016/j.jclinepi.2009.06.002

10.1002/sim.2618

10.1002/sim.2781

10.1016/j.jclinepi.2007.07.011

10.1002/sim.3854

10.1093/oxfordjournals.aje.a114593

10.1093/biomet/71.3.431

10.1002/0470011815.b2a03101

10.2307/2683903

10.1002/pst.433

10.1080/01621459.1989.10478874

10.1093/epirev/mxg002

10.1080/01621459.1984.10478078

10.2307/2528036

10.1002/sim.1903

10.1111/j.1467-9531.2008.00204.x

10.1198/000313004X5824

10.1002/sim.2059

10.1080/03610910903528301

10.1080/03610910801942430

10.1002/sim.2580

10.2307/1403572

10.1002/sim.1296

10.2202/1557‐4679.1146

10.1016/j.cmpb.2003.10.004

10.1002/sim.2174

10.1002/sim.3782

10.1002/pds.1555

10.1093/aje/kwq224

10.1002/bimj.200810488

10.1007/978-1-4757-3692-2

10.1002/sim.4200

10.1002/pst.537

Thông tin
Thông tin xuất bản

Statistics in Medicine

Tập 32 Số 16

2837-2849

Thông tin tác giả