Nonparametric estimation of conditional transition probabilities in a non-Markov illness-death model

One important goal in multi-state modeling is the estimation of transition probabilities. In longitudinal medical studies these quantities are particularly of interest since they allow for long-term predictions of the process. In recent years signi ficant contributions have been made regarding this topic. However, most of the approaches assume independent censoring and do not account for the influence of covariates. The goal of the paper is to introduce feasible estimation methods for the transition probabilities in an illness-death model conditionally on current or past covariate measures. All approaches are evaluated through a simulation study, leading to a comparison of two di erent estimators. The proposed methods are illustrated using real a colon cancer data set.This research was nanced by FEDER Funds through Programa Operacional Factores de Competitividade COMPETE and by Portuguese Funds through FCT - Funda ção para a Cência e a Tecnologia, within Projects Est-C/MAT/UI0013/2011 and PTDC/MAT/104879/2008. We also acknowledge nancial support from the project Grants MTM2008-03129 and MTM2011-23204 (FEDER support included) of the Spanish Ministerio de Ciencia e Innovaci on and 10PXIB300068PR of the Xunta de Galicia. Partial support from a grant from the US National Security Agency (H98230-11-1-0168) is greatly appreciated