Advantages and limitations of propensity score methods to analyze non-randomized clinical trials

Abstract

The propensity score (PS), defined as a patient’s probability of receiving a specific treatment conditional on the observed covariates, is a method that leads to control the bias associated with the non-randomly assigned treatment allocation in observational studies. PS methods have been increasingly used in the last 10 years. The quality of reporting PS in the intensive care medicine literature should be improved. This review aims to present the theory of PS, illustrated with an example and provides recommendations to the investigators in order to improve the reporting of PS analyses.