Estimation of Overall Survival in an Illness-Death Model with Application to the Vertical Transmission of HIV-1

• Activity Number: 213
• Type: Contributed
• Date/Time: Monday, August 3, 2009 : 2:00 PM to 3:50 PM
• Sponsor: ENAR
• Author(s): Halina Frydman and Michael Szarek*+
• Companies: New York University and ImClone Systems
• Keywords: HIV ; illness-death model ; interval censoring ; self-consistency ; nonparametric maximum likelihood estimation

Abstract

We present a new approach to the analysis of trial data where infants born to HIV-1-infected women were randomly assigned to formula or breast feeding and followed for vertical HIV-1 infection. In this and other trials the focus is on estimation of the probability of the occurrence of the nonfatal event (NFE; e.g., HIV-1 infection) and NFE-free survival. When the NFE is assessed intermittently, the observed NFE times are interval censored, and when the NFE is not observed, NFE status is not known from the time of the last assessment until the end of follow-up. We use an illness-death model accounting for interval censoring and unknown NFE status to estimate the probabilities referred to above, the overall survival distribution and the cumulative intensity of death after the occurrence of the NFE. These estimates fully characterize how NFE occurrence modifies survival.