On estimating the distribution function for quality of life in cancer clinical trials

Abstract

SUMMARY The analysis of quality-of-life data collected at periodic intervals in a cancer clinical trial must account for those patients' data that are ‘missing’ because of death or administrative censoring. We examine estimates of the cumulative distribution function of an interpolated area under a quality-of-life curve as summaries of the quality-of-life experience of a treatment group. A proposed estimator, although biased, offers an improvement over a Kaplan-Meier estimator that assumes incorrectly that the censoring is noninformative.