Decomposing Differences in Cohort Health Expectancy by Cause and Age With Longitudinal Data.

We propose a new method to decompose cohort health expectancy by age and cause. We develop a new attribution method for longitudinal data. This method handles interval censoring, semicompeting risks, and time-dependent covariates. We derive explicit formulas for stepwise decomposition of cohort health expectancy. We provide an R package and Shiny app to support use of the proposed method.