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

Area of Science:

• Biostatistics
• Epidemiology
• Health Economics

Background:

• Health expectancy is a key metric for population health assessment.
• Existing methods for decomposing health expectancy have limitations in handling complex longitudinal data structures.
• Accurate decomposition is crucial for understanding health disparities and planning interventions.

Purpose of the Study:

• To develop and present a novel attribution method for decomposing cohort health expectancy.
• To provide explicit formulas for stepwise decomposition applicable to longitudinal health data.
• To create user-friendly tools (R package, Shiny app) for implementing the method.

Main Methods:

• Development of a new attribution method tailored for longitudinal health data.
• Incorporation of techniques to handle interval censoring, semicompeting risks, and time-dependent covariates.
• Derivation of explicit mathematical formulas for stepwise decomposition of health expectancy.

Main Results:

• A new, robust method for health expectancy decomposition by age and cause has been established.
• The method successfully accounts for complex data features common in longitudinal health studies.
• Explicit formulas enable precise, stepwise analysis of health expectancy components.

Conclusions:

• The proposed method offers a significant advancement in the analysis of health expectancy.
• The accompanying R package and Shiny app facilitate wider adoption and application in public health research.
• This work provides a powerful tool for detailed health burden analysis and policy development.