Propensity score weighting under limited overlap and model misspecification.

Loading...
Thumbnail Image

Date

2020-07-21

Journal Title

Journal ISSN

Volume Title

Repository Usage Stats

124
views
90
downloads

Citation Stats

Abstract

Propensity score weighting methods are often used in non-randomized studies to adjust for confounding and assess treatment effects. The most popular among them, the inverse probability weighting, assigns weights that are proportional to the inverse of the conditional probability of a specific treatment assignment, given observed covariates. A key requirement for inverse probability weighting estimation is the positivity assumption, i.e. the propensity score must be bounded away from 0 and 1. In practice, violations of the positivity assumption often manifest by the presence of limited overlap in the propensity score distributions between treatment groups. When these practical violations occur, a small number of highly influential inverse probability weights may lead to unstable inverse probability weighting estimators, with biased estimates and large variances. To mitigate these issues, a number of alternative methods have been proposed, including inverse probability weighting trimming, overlap weights, matching weights, and entropy weights. Because overlap weights, matching weights, and entropy weights target the population for whom there is equipoise (and with adequate overlap) and their estimands depend on the true propensity score, a common criticism is that these estimators may be more sensitive to misspecifications of the propensity score model. In this paper, we conduct extensive simulation studies to compare the performances of inverse probability weighting and inverse probability weighting trimming against those of overlap weights, matching weights, and entropy weights under limited overlap and misspecified propensity score models. Across the wide range of scenarios we considered, overlap weights, matching weights, and entropy weights consistently outperform inverse probability weighting in terms of bias, root mean squared error, and coverage probability.

Department

Description

Provenance

Citation

Published Version (Please cite this version)

10.1177/0962280220940334

Publication Info

Zhou, Yunji, Roland A Matsouaka and Laine Thomas (2020). Propensity score weighting under limited overlap and model misspecification. Statistical methods in medical research. p. 962280220940334. 10.1177/0962280220940334 Retrieved from https://hdl.handle.net/10161/21404.

This is constructed from limited available data and may be imprecise. To cite this article, please review & use the official citation provided by the journal.

Scholars@Duke

Matsouaka

Roland Albert Matsouaka

Associate Professor of Biostatistics & Bioinformatics
Thomas

Laine Elliott Thomas

Professor of Biostatistics & Bioinformatics

Laine Thomas, PhD, joined the Department of Biostatistics and Bioinformatics and DCRI in 2009.  She serves as Associate Chair for Equity, Diversity and Inclusion within the Department of Biostatistics and Bioinformatics and Deputy Director of Data Science and Biostatistics at the Duke Clinical Research Institute.  She is a leader in study design and development of methods for observational and pragmatic studies, with over 240 peer reviewed clinical and methodological publications arising from scientific collaboration in the therapeutic areas of cardiovascular disease, diabetes, uterine fibroids and SARS-CoV-2 virus. She led the statistical teams on the HERO COVID-19, ORBIT-AF I & II, ACTION-CMS, CHAMP-HF, and COMPARE-UF clinical registries and secondary analyses of the NAVIGATOR and ARISTOTLE clinical trials. She has served as a primary investigator and co-investigator on numerous methodological studies with funding from NIH, AHRQ, PCORI and Burroughs Wellcome Fund, addressing observational treatment comparisons, time-varying treatments, heterogeneity of treatment effects, and randomized trials augmented by synthetic controls from real world data.      


Unless otherwise indicated, scholarly articles published by Duke faculty members are made available here with a CC-BY-NC (Creative Commons Attribution Non-Commercial) license, as enabled by the Duke Open Access Policy. If you wish to use the materials in ways not already permitted under CC-BY-NC, please consult the copyright owner. Other materials are made available here through the author’s grant of a non-exclusive license to make their work openly accessible.