Clustering countries to evaluate health outcomes globally

Area of Science:

• Global Health
• Health Disparities
• Child Mortality

Background:

• Assessing global health disparities requires clustering countries by health outcomes.
• Inconsistent health outcome data across sources (World Bank, WHO, UNICEF) pose challenges.
• New Institute for Health Metrics and Evaluation (IHME) under-five mortality data offers improved country-level estimates and confidence intervals.

Purpose of the Study:

• To identify country clusters using under-five child mortality and WHO's Healthy Life Expectancy (HALE) data.
• To analyze global health outcome distributions and identify nonlinear patterns.
• To propose country clusters for studying societal factors influencing health outcomes.

Main Methods:

• Utilized IHME under-five child mortality data and WHO HALE data for 160 countries.
• Employed model-based clustering techniques to identify country groupings.
• Accounted for uncertainty within the data and clustering results.

Main Results:

• Identified four distinct country clusters in 2000 and six in 2003.
• Demonstrated nonlinear distributions of health outcomes globally.
• Indicated no single, uniform trajectory for health outcome progression across countries.

Conclusions:

• Country clustering reveals complex, nonlinear global health outcome distributions.
• The identified clusters provide a framework for further investigation into determinants of health.
• Future research should leverage these clusters to explore societal conditions impacting health over time.