Going beyond life expectancy in assessments of health systems' performance: life expectancy adjusted by perceived health status.

Abstract

International comparisons of health systems data have been used to guide health policy. Health systems performance is generally evaluated on how different factors contribute to mortality and longevity. Fewer studies scrutinize the factors that determine morbidity in different countries, partly because indicators that assess morbidity on a country level are not as widely available as mortality and longevity data. We introduce a new health status indicator able to combine mortality and morbidity in a single composite measure for each country and gender at a point in time (LEAPHS), yielding the average number of years that men (women) can expect to live in "good" (or better) health. Using the Sullivan method we combine the mortality risk, calculated for specific age and gender groups, with perceived health status for the same age and gender groups, and we estimate how medical care and various socio-economic, environmental and structural, lifestyle, and technological factors affect LEAPHS and life expectancy at birth for a large panel of thirty OECD countries. We find that some variables (alcohol consumption, urbanization) have a significant effect on both LEAPHS and life expectancy, while one variable (the number of hospitals) has a significant effect for both genders on life expectancy only. However, the effects of many other variables (health expenditure per capita, health expenditure per capita squared, GDP growth, and technology) were only significant predictors for LEAPHS. This leads us to conclude that LEAPHS is able to capture the impact of some health determinants not captured by life expectancy at birth. While we believe this new measure may be useful for health economists and statisticians doing cross-country analyses, further comparisons with other measures may be useful.