Regression analysis in an illness-death model with interval-censored data: A pseudo-value approach:

摘要: Pseudo-values provide a method to perform regression analysis for complex quantities with right-censored data. A further complication, interval-censored data, appears when events such as dementia are studied in an epidemiological cohort. We propose extension of the pseudo-value approach data based on semi-parametric estimator computed using penalised likelihood and splines. This takes interval-censoring competing risks into account illness-death model. apply three mean value parameters interest studies dementia: probability staying alive non-demented, restricted survival time without absolute risk dementia. Simulation conducted examine properties pseudo-values this estimator. The is applied French cohort PAQUID, which included more than 3,000 non-demented subjects, followed 25 years.