Bartels, Christian and Dumortier, Thomas (2021) Inverse probability of censoring weighting for visual predictive checks of time-to-event models with time-varying covariates. Pharmaceutical Statistics. ISSN 1539-16041539-1612

Abstract

When constructing models to summarize clinical data to be used for simulations, it is good prac-tice to evaluate the model for its capacity to reproduce the data. This can be done by means of Visual Predictive Checks (VPC), which consist of (1) several reproductions of the original study by simulation from the model under evaluation, (2) calculating estimates of interest for each simulated study and (3) comparing the distribution of those estimates with the estimate from the original study. This procedure is a generic method that is straightforward to apply, in general. Here we consider the application of the method to time to event data and consider the case when a time-varying covariate is not known or cannot be approximated after event time. In this case, simulations cannot be conducted beyond the end of the follow-up time (event or censoring time) in the original study. Thus, the simulations must be censored at the end of the follow-up time. Since this censoring is not random, the standard KM estimate from the simulated study will be biased. We propose to use inverse probability of censoring weighting (IPoC) method to cor-rect the KM estimator. For analyzing the Cantos study, the IPoC weighting as described here proved valuable and enabled qualification of PKPD models for simulations. Here, we use a gen-erated data set, which allows illustration of the different situations and checking of the implemen-tation against the known truth.

Item Type:	Article
Keywords:	Visual predictive check, model diagnostics, time to event models, inverse probability weighting, time varying covariates
Date Deposited:	25 Jun 2021 00:45
Last Modified:	25 Jun 2021 00:45
URI:	https://oak.novartis.com/id/eprint/43151