Abstract:
Kaplan-Meier estimate or proportional hazards regression is commonly used directly to estimate the effect of treatment on survival time in randomized clinical studies. However, in an observational (or non-randomized) study, the treated and untreated groups cannot be compared directly because they may systematically differ at baseline characteristics, which would lead to biased estimation of the treatment effect. Researchers have developed various methods for adjusting biased estimates for confounding covariates such as matching, stratification, inverse probability treatment weighting, and proportional hazards regression adjustment using the propensity score. However, very few studies have compared the performance of these methods. In this paper, we conducted an intensive case study to compare the performance of various bias correction methods for non-randomized study and applied these methods to the Right Heart Catheterization study to compare the survival time of the RHC treated group and control group.
