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

被引:689
作者
Austin, Peter C. [1 ,2 ,3 ]
机构
[1] Inst Clin Evaluat Sci, Toronto, ON M4N 3M5, Canada
[2] Univ Toronto, Inst Hlth Management Policy & Evaluat, Toronto, ON, Canada
[3] Univ Toronto, Dalla Lana Sch Publ Hlth, Toronto, ON, Canada
基金
加拿大健康研究院;
关键词
propensity score; survival analysis; inverse probability of treatment weighting (IPTW); Monte Carlo simulations; observational study; time-to-event outcomes; INVERSE PROBABILITY; UNTREATED SUBJECTS; RISK DIFFERENCES; MODELS; SUBCLASSIFICATION; ADJUSTMENT; BIAS;
D O I
10.1002/sim.5705
中图分类号
Q [生物科学];
学科分类号
07 ; 0710 ; 09 ;
摘要
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 (c) 2012 John Wiley & Sons, Ltd.
引用
收藏
页码:2837 / 2849
页数:13
相关论文
共 37 条
[11]   The performance of different propensity-score methods for estimating differences in proportions (risk differences or absolute risk reductions) in observational studies [J].
Austin, Peter C. .
STATISTICS IN MEDICINE, 2010, 29 (20) :2137-2148
[12]   A Data-Generation Process for Data with Specified Risk Differences or Numbers Needed to Treat [J].
Austin, Peter C. .
COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION, 2010, 39 (03) :563-577
[13]   A substantial and confusing variation exists in handling of baseline covariates in randomized controlled trials: a review of trials published in leading medical journals [J].
Austin, Peter C. ;
Manca, Andrea ;
Zwarenstein, Merrick ;
Juurlink, David N. ;
Stanbrook, Matthew B. .
JOURNAL OF CLINICAL EPIDEMIOLOGY, 2010, 63 (02) :142-153
[14]   Type I Error Rates, Coverage of Confidence Intervals, and Variance Estimation in Propensity-Score Matched Analyses [J].
Austin, Peter C. .
INTERNATIONAL JOURNAL OF BIOSTATISTICS, 2009, 5 (01)
[15]   Some Methods of Propensity-Score Matching had Superior Performance to Others: Results of an Empirical Investigation and Monte Carlo simulations [J].
Austin, Peter C. .
BIOMETRICAL JOURNAL, 2009, 51 (01) :171-184
[16]   Generating survival times to simulate Cox proportional hazards models [J].
Bender, R ;
Augustin, T ;
Blettner, M .
STATISTICS IN MEDICINE, 2005, 24 (11) :1713-1723
[17]   EFFECTIVENESS OF ADJUSTMENT BY SUBCLASSIFICATION IN REMOVING BIAS IN OBSERVATIONAL STUDIES [J].
COCHRAN, WG .
BIOMETRICS, 1968, 24 (02) :295-&
[18]   Adjusted survival curves with inverse probability weights [J].
Cole, SR ;
Hernán, MA .
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE, 2004, 75 (01) :45-49
[19]  
Cook T., 2008, INTRO STAT METHODS C
[20]   Matched cohort methods for injury research [J].
Cummings, P ;
McKnight, B ;
Greenland, S .
EPIDEMIOLOGIC REVIEWS, 2003, 25 :43-50