Jump to ContentJump to Main Navigation

The International Journal of Biostatistics

Ed. by Chambaz, Antoine / Hubbard, Alan E. / van der Laan, Mark J.

2 Issues per year


IMPACT FACTOR 2014: 0.741
5-year IMPACT FACTOR: 1.475

SCImago Journal Rank (SJR) 2014: 1.247
Source Normalized Impact per Paper (SNIP) 2014: 1.078
Impact per Publication (IPP) 2014: 1.206

Mathematical Citation Quotient (MCQ) 2014: 0.07

Type I Error Rates, Coverage of Confidence Intervals, and Variance Estimation in Propensity-Score Matched Analyses

Peter C Austin1

1Institute for Clinical Evaluative Sciences

Citation Information: The International Journal of Biostatistics. Volume 5, Issue 1, ISSN (Online) 1557-4679, DOI: 10.2202/1557-4679.1146, April 2009

Publication History

Published Online:
2009-04-14

Propensity-score matching is frequently used in the medical literature to reduce or eliminate the effect of treatment selection bias when estimating the effect of treatments or exposures on outcomes using observational data. In propensity-score matching, pairs of treated and untreated subjects with similar propensity scores are formed. Recent systematic reviews of the use of propensity-score matching found that the large majority of researchers ignore the matched nature of the propensity-score matched sample when estimating the statistical significance of the treatment effect. We conducted a series of Monte Carlo simulations to examine the impact of ignoring the matched nature of the propensity-score matched sample on Type I error rates, coverage of confidence intervals, and variance estimation of the treatment effect. We examined estimating differences in means, relative risks, odds ratios, rate ratios from Poisson models, and hazard ratios from Cox regression models. We demonstrated that accounting for the matched nature of the propensity-score matched sample tended to result in type I error rates that were closer to the advertised level compared to when matching was not incorporated into the analyses. Similarly, accounting for the matched nature of the sample tended to result in confidence intervals with coverage rates that were closer to the nominal level, compared to when matching was not taken into account. Finally, accounting for the matched nature of the sample resulted in estimates of standard error that more closely reflected the sampling variability of the treatment effect compared to when matching was not taken into account.

Keywords: propensity score; matching; propensity-score matching; variance estimation; coverage; simulations; type I error; observational studies

Citing Articles

Here you can find all Crossref-listed publications in which this article is cited. If you would like to receive automatic email messages as soon as this article is cited in other publications, simply activate the “Citation Alert” on the top of this page.

[1]
Rajender R. Aparasu, Satabdi Chatterjee, Sandhya Mehta, and Hua Chen
Medical Care, 2012, Volume 50, Number 11, Page 961
[2]
Gyung-Min Park, Seon Ha Kim, Min-Woo Jo, Sung Ho Her, Seungbong Han, Jung-Min Ahn, Duk-Woo Park, Soo-Jin Kang, Seung-Whan Lee, Young-Hak Kim, Cheol Whan Lee, Beom-Jun Kim, Jung-Min Koh, Hong-Kyu Kim, Jaewon Choe, Seong-Wook Park, and Seung-Jung Park
Medicine, 2015, Volume 94, Number 21, Page e917
[3]
Masatsugu Hamaji, Fengshi Chen, Yukinori Matsuo, Atsushi Kawaguchi, Satoshi Morita, Nami Ueki, Makoto Sonobe, Yasushi Nagata, Masahiro Hiraoka, and Hiroshi Date
The Annals of Thoracic Surgery, 2015, Volume 99, Number 4, Page 1122
[4]
Victor A. Ferraris, Daniel L. Davenport, Sibu P. Saha, Alethea Bernard, Peter C. Austin, and Joseph B. Zwischenberger
The Annals of Thoracic Surgery, 2011, Volume 91, Number 6, Page 1674
[5]
Sandhya Mehta, Hua Chen, Michael Johnson, and Rajender R. Aparasu
The American Journal of Geriatric Pharmacotherapy, 2011, Volume 9, Number 2, Page 120
[7]
Sandhya Mehta, Hua Chen, Michael L. Johnson, and Rajender R. Aparasu
Drugs & Aging, 2010, Volume 27, Number 10, Page 815
[8]
Dallas P. Seitz, Sudeep S. Gill, Chaim M. Bell, Peter C. Austin, Andrea Gruneir, Geoff M. Anderson, and Paula A. Rochon
Journal of the American Geriatrics Society, 2014, Volume 62, Number 11, Page 2102
[9]
Peter C. Austin
Statistics in Medicine, 2013, Volume 32, Number 16, Page 2837
[10]
Peter C. Austin and Dylan S. Small
Statistics in Medicine, 2014, Volume 33, Number 24, Page 4306
[11]
Liang Li
Communications in Statistics - Simulation and Computation, 2014, Volume 43, Number 10, Page 2498
[12]
David J. Peters, Andy Hochstetler, Matt DeLisi, and Hui-Ju Kuo
Journal of Quantitative Criminology, 2015, Volume 31, Number 1, Page 149
[13]
Melissa M. Garrido, Amy S. Kelley, Julia Paris, Katherine Roza, Diane E. Meier, R. Sean Morrison, and Melissa D. Aldridge
Health Services Research, 2014, Volume 49, Number 5, Page 1701
[15]
Etienne Gayat, Matthieu Resche-Rigon, Jean-Yves Mary, and Raphaël Porcher
Pharmaceutical Statistics, 2012, Volume 11, Number 3, Page 222
[16]
[18]
Robin Mitra and Jerome P. Reiter
Statistics in Medicine, 2011, Volume 30, Number 6, Page 627

Comments (0)

Please log in or register to comment.