This week's Biostatistics Seminar will feature Dr. Jong H. Jeong, Professor and Interim Chair, department of Biostatistics, University of Pittsburgh, presenting, "A new summary measure for censored time-to-event data: Quantile Life Lost."
Time-to-event data may be analyzed based on (1) cumulative information up to a specific time point through the hazard function or the survival function, or on (2) residual information beyond that time point through the mean or quantile residual life function. In this talk, a new summary measure, the quantile life lost, is introduced, which has several advantages over the existing summary measures for censored time-to-event data. The distribution of life lost is characterized by the reverse hazard function. Nonparametric inference on the quantile life lost for one-sample and two-sample cases are presented, together with some extension to a regression setting. To avoid nonparametric estimation of the probability density function of the underlying time-to-event distribution under censoring, martingale representation of the estimating equation is used to estimate the variance of the quantile life lost estimator. Asymptotic distributions of the estimators and test statistics are derived and some simulation results are presented to demonstrate their finite sample behaviors. The proposed method is illustrated with a real data set from a breast cancer study.