Propensity score matching in observational studies:methods and applications
10.3969/j.issn.1672-8467.2025.06.020
- VernacularTitle:观察性研究中倾向性评分匹配的方式和应用
- Author:
Ye WU
1
;
Jie TIAN
;
Wei-bing WANG
Author Information
1. 复旦大学公共卫生学院流行病学教研室-公共卫生安全教育部重点实验室 上海 200032
- Publication Type:Journal Article
- Keywords:
propensity score matching(PSM);
observational studies;
confounding factors
- From:
Fudan University Journal of Medical Sciences
2025;52(6):917-922
- CountryChina
- Language:Chinese
-
Abstract:
Observational studies are important approaches for obtaining real-world evidence.However,due to the lack of randomized allocation,differences in baseline characteristics between groups often introduce confounding bias,which may distort the results.Traditional stratified analyses and multivariable regression models have limited ability to control confounding when multiple covariates are involved.Propensity score matching estimates the probability of receiving an intervention based on observed covariates,then matches individuals with similar propensity scores between treatment and control groups,thereby balancing covariate distributions and reducing confounding.In recent years,propensity score matching has been widely applied in various fields,including chronic disease management and drug effectiveness evaluation,public health policy and health service assessment,vaccine effectiveness studies focusing on population disparities,and evaluations of telemedicine interventions.This review summarizes common matching methods and application scenarios of propensity score matching in observational studies.
