Titolo del seminario: Reconsideration of the Kaplan-Meier Estimator: Censoring and Time-varying Covariates
Presentatore: Professor Rebecca Betensky, Chair of the Department of Biostatistics, School of Global Public Health at New York University
Quando: 11 Giugno 2021, 15.00-16.00
Dove: Zoom, https://ki-se.zoom.us/j/65377392771
In this talk I will present two extensions of the Kaplan-Meier estimator that address a nuance of censoring and incorporation of time-varying covariates. Clinical studies with time to event endpoints typically report the median follow-up (i.e., censoring) time for the subjects in the trial, alongside the median time to event. The reason for this is to provide information about the opportunity for subjects in the study to experience the event of interest. In most clinical studies, the censoring time is a composite measure, defined as the minimum of the time to drop-out from the study and time to the administrative end of study. The time to drop-out component may or may not be observed, while the time to the end of study is observed for each subject. I consider whether this decomposition of the censoring time into a time that is itself potentially censored and a time that is fully observed offers any improvement of the estimation of the censoring distribution.
Extensions of the Kaplan-Meier estimator have been developed to illustrate the relationship between a time-varying covariate of interest and survival, however they are limited to displaying survival for patients who always have a certain value of a time-varying covariate. I present extensions of these estimators that provide crude and covariate-adjusted estimates of the survival function for patients defined by covariate paths.