Multi-country clustering-based forecasting of healthy life expectancy.

Area of Science:

• Demography
• Biostatistics
• Public Health

Background:

• Healthy life expectancy (HLE) is crucial for healthcare planning and Long Term Care insurance.
• Forecasting HLE is essential for addressing the needs of aging populations.
• Understanding HLE patterns across countries is vital for global health strategies.

Purpose of the Study:

• To develop a methodology for simultaneous HLE forecasting across groups of countries.
• To identify and analyze similarities in HLE patterns among different nations.
• To improve the accuracy of HLE predictions for demographic and economic planning.

Main Methods:

• Functional data clustering applied to multivariate time series of HLE at birth (1990-2019).
• Utilized data from the Global Burden of Disease Study.
• Employed a multivariate random walk with drift for simultaneous forecasting within identified clusters.

Main Results:

• Identified three distinct clusters for HLE at birth for both genders.
• Successfully performed simultaneous HLE forecasting for countries within each cluster.
• Demonstrated statistical significance of parameters in the multivariate processes.

Conclusions:

• The proposed methodology effectively groups countries with similar HLE trajectories.
• Simultaneous forecasting within clusters provides a robust approach for HLE prediction.
• Findings offer valuable insights for demographic analysis and health policy development.