Tags

Type your tag names separated by a space and hit enter

A comparison of approaches for stratifying on the propensity score to reduce bias.
J Eval Clin Pract. 2017 Aug; 23(4):690-696.JE

Abstract

RATIONALE, AIMS, AND OBJECTIVES

Stratification is a popular propensity score (PS) adjustment technique. It has been shown that stratifying the PS into 5 quantiles can remove over 90% of the bias due to the covariates used to generate the PS. Because of this finding, many investigators partition their data into 5 quantiles of the PS without examining whether a more robust solution (one that increases covariate balance while potentially reducing bias in the outcome analysis) can be found for their data. Two approaches (referred to herein as PSCORE and PSTRATA) obtain the optimal stratification solution by repeatedly dividing the data into strata until balance is achieved between treatment and control groups on the PS. These algorithms differ in how they partition the data, and it is not known which is better, or if either is better than a 5-quantile default approach, for reducing bias in treatment effect estimates.

METHOD

Monte Carlo simulations and empirical data are used to assess whether PS strata defined by PSCORE, PSTRATA, or 5 quantiles is best at reducing bias in treatment effect estimates, when used within a marginal mean weighting framework (MMWS). These estimates are further compared to results derived using inverse probability of treatment weights (IPTW).

RESULTS

PSTRATA was slightly better than PSCORE in balancing covariates and reducing bias, while both approaches outperformed the 5-quantile approach. Overall MMWS using any stratification method outperformed IPTW.

CONCLUSIONS

Investigators should routinely use stratification approaches that obtain the optimal stratification solution, rather than simply partitioning the data into 5 quantiles of the PS. Moreover, MMWS (in conjunction with an optimal stratification approach) should be considered as an alternative to IPTW in studies that use PS weights.

Authors+Show Affiliations

Linden Consulting Group, LLC, Ann Arbor, Michigan, USA. Division of General Medicine, Medical School, University of Michigan, Ann Arbor, Michigan, USA.

Pub Type(s)

Comparative Study
Journal Article

Language

eng

PubMed ID

28074629

Citation

Linden, Ariel. "A Comparison of Approaches for Stratifying On the Propensity Score to Reduce Bias." Journal of Evaluation in Clinical Practice, vol. 23, no. 4, 2017, pp. 690-696.
Linden A. A comparison of approaches for stratifying on the propensity score to reduce bias. J Eval Clin Pract. 2017;23(4):690-696.
Linden, A. (2017). A comparison of approaches for stratifying on the propensity score to reduce bias. Journal of Evaluation in Clinical Practice, 23(4), 690-696. https://doi.org/10.1111/jep.12701
Linden A. A Comparison of Approaches for Stratifying On the Propensity Score to Reduce Bias. J Eval Clin Pract. 2017;23(4):690-696. PubMed PMID: 28074629.
* Article titles in AMA citation format should be in sentence-case
TY - JOUR T1 - A comparison of approaches for stratifying on the propensity score to reduce bias. A1 - Linden,Ariel, Y1 - 2017/01/10/ PY - 2016/12/14/received PY - 2016/12/15/accepted PY - 2017/1/12/pubmed PY - 2018/5/8/medline PY - 2017/1/12/entrez KW - covariate balance KW - inverse probability of treatment KW - marginal mean weighting through stratification KW - propensity score KW - stratification KW - treatment effects SP - 690 EP - 696 JF - Journal of evaluation in clinical practice JO - J Eval Clin Pract VL - 23 IS - 4 N2 - RATIONALE, AIMS, AND OBJECTIVES: Stratification is a popular propensity score (PS) adjustment technique. It has been shown that stratifying the PS into 5 quantiles can remove over 90% of the bias due to the covariates used to generate the PS. Because of this finding, many investigators partition their data into 5 quantiles of the PS without examining whether a more robust solution (one that increases covariate balance while potentially reducing bias in the outcome analysis) can be found for their data. Two approaches (referred to herein as PSCORE and PSTRATA) obtain the optimal stratification solution by repeatedly dividing the data into strata until balance is achieved between treatment and control groups on the PS. These algorithms differ in how they partition the data, and it is not known which is better, or if either is better than a 5-quantile default approach, for reducing bias in treatment effect estimates. METHOD: Monte Carlo simulations and empirical data are used to assess whether PS strata defined by PSCORE, PSTRATA, or 5 quantiles is best at reducing bias in treatment effect estimates, when used within a marginal mean weighting framework (MMWS). These estimates are further compared to results derived using inverse probability of treatment weights (IPTW). RESULTS: PSTRATA was slightly better than PSCORE in balancing covariates and reducing bias, while both approaches outperformed the 5-quantile approach. Overall MMWS using any stratification method outperformed IPTW. CONCLUSIONS: Investigators should routinely use stratification approaches that obtain the optimal stratification solution, rather than simply partitioning the data into 5 quantiles of the PS. Moreover, MMWS (in conjunction with an optimal stratification approach) should be considered as an alternative to IPTW in studies that use PS weights. SN - 1365-2753 UR - https://www.unboundmedicine.com/medline/citation/28074629/A_comparison_of_approaches_for_stratifying_on_the_propensity_score_to_reduce_bias_ L2 - https://doi.org/10.1111/jep.12701 DB - PRIME DP - Unbound Medicine ER -