Optimal caliper widths for propensity‐score matching when estimating differences in means and differences in proportions in observational studies

Pharmaceutical Statistics - Tập 10 Số 2 - Trang 150-161 - 2011
Peter C. Austin1,2,3
1Dalla Lana School of Public Health Sciences, University of Toronto, Ont., Canada
2Department of Health Management, Policy and Evaluation, University of Toronto, Ont., Canada
3Institute for Clinical Evaluative Sciences, Toronto, Ont, Canada

Tóm tắt

AbstractIn a study comparing the effects of two treatments, the propensity score is the probability of assignment to one treatment conditional on a subject's measured baseline covariates. Propensity‐score matching is increasingly being used to estimate the effects of exposures using observational data. In the most common implementation of propensity‐score matching, pairs of treated and untreated subjects are formed whose propensity scores differ by at most a pre‐specified amount (the caliper width). There has been a little research into the optimal caliper width. We conducted an extensive series of Monte Carlo simulations to determine the optimal caliper width for estimating differences in means (for continuous outcomes) and risk differences (for binary outcomes). When estimating differences in means or risk differences, we recommend that researchers match on the logit of the propensity score using calipers of width equal to 0.2 of the standard deviation of the logit of the propensity score. When at least some of the covariates were continuous, then either this value, or one close to it, minimized the mean square error of the resultant estimated treatment effect. It also eliminated at least 98% of the bias in the crude estimator, and it resulted in confidence intervals with approximately the correct coverage rates. Furthermore, the empirical type I error rate was approximately correct. When all of the covariates were binary, then the choice of caliper width had a much smaller impact on the performance of estimation of risk differences and differences in means. Copyright © 2010 John Wiley & Sons, Ltd.

Từ khóa


Tài liệu tham khảo

10.1093/biomet/70.1.41

10.1080/01621459.1984.10478078

10.1002/sim.2328

10.2307/2683903

10.1002/sim.3150

10.1016/j.jtcvs.2007.07.021

10.1161/CIRCOUTCOMES.108.790634

10.1162/003465304323023651

10.2202/1557‐4679.1146

10.1016/j.jclinepi.2005.07.004

10.1046/J.1524-4733.2002.55150.x

10.1136/bmj.310.6977.452

Jaeschke R, 1995, Basis statistics for clinicians 3: assessing the effects of treatment: measures of association, Canadian Medical Association Journal, 152, 351

10.1016/0895-4356(94)90191-0

10.1002/sim.2618

10.1002/sim.2781

10.1093/oxfordjournals.aje.a114593

10.1080/03610910903528301

10.1002/sim.3854

10.1080/03610910801942430

Cohen J, 1988, Statistical Power Analysis for the Behavioral Sciences

Cochran WG, 1973, Controlling bias in observational studies: a review, Sankhya: The Indian Journal of Statistics, 35, 416

10.1002/sim.1589

10.1001/jama.290.19.2581

Tu JV, 2004, Quality of Cardiac Care in Ontario–Phase 1. Report 1

10.2307/2684560

10.1002/bimj.200810488

10.1002/sim.3697

10.1177/0272989X09341755

Thông tin
Thông tin xuất bản

Pharmaceutical Statistics

Tập 10 Số 2

150-161

Thông tin tác giả