Estimation of survival functions for events based on a continuous outcome: a distributional approach

Abstract

The limitations resulting from the dichtomisation of continuous outcomes have been extensively described. But the need to present results based on binary outcomes in particular in health science remains. Alternatives based on the distribution of the continuous outcome have been proposed. Here we explore the possibilities of using a distributional approach in the context of time-to-event analysis when the event is defined by values of a continuous outcome above or below a threshold. For this we propose in a first step a distributional version of the Kaplan-Meier estimate of the survival function based on repeated truncation of the normal distribution. The method is evaluated with a simulation study and illustrated with a case study.