A0366
Title: Concordance index for interval-censored survival data
Authors: Yi-Kang Tseng - National Central University (Taiwan) [presenting]
Tzu-Ling Wang - National Tsing Hua University (Taiwan)
Abstract: The concordance index (c-index) is crucial in medical research for assessing predictive models in distinguishing patient risk levels and estimating event probabilities. While well-studied for right-censored survival data, its estimation for interval-censored data remains underexplored. Parametric and semiparametric methods are developed to estimate the c-index for interval-censored survival data. The framework is extended to time-dependent c-index estimation with longitudinal covariates, and measurement errors are addressed using a joint modeling approach. Large-sample properties are examined, and performance is validated through simulations. The methodology is applied to Taiwan AIDS cohort data, enhancing risk assessment in clinical research.
