Kaplan-Meier Approach to Censored Data in Clinical Trials

Abstract

In clinical studies, patients survival time are subject to paucity of patients who report for treatment. For this reason, clinical data are mostly censored. Estimating mean, variance and standard deviation in censored data are different from those in uncensored data. Censored data sometime do not follow a specific distribution, hence using parametric method in estimating parameters in censored data calls for specific assumptions. KM method is a nonparametric approach that does not depend on any assumption. KM approach also known as product limit approach has been shown to provide more robust estimate of the mean and standard deviation for both left and right censored observations than other methods. In this paper, the method has been used to estimate mean survival time of patients in multiple censored data.