>For example, Japan's average healthy life expectancy based on the WHO's
>Disability Adjusted Life Expectancy indicator is 71.9 years for men and 77.2
>years for women, while the country ranked third in assessment of health
>inequalities concerning mortality rates for children under 5.
That does not seem a valid measure of a "health system." Healthy life expectancy is likelyu to be affected by a host of environmental factors, ranging from life styles, to diet, to presence/absence of pollutants. Including these factors as a part of "health system" is quite problematic, because it dillutes the meaning of a "system" which implies a coherent organizational structure assembled in response to environmental factors.
A better approach would be to evaluate the contributions of institutional aspects of health care while controling for environmental factors. For example, country A might have lower life expectancy than country B, but A's health care system migh be superior to that of B, because health risk in A is much higher than in B. Stated differently, if B's health system were implemented in A, A's life expectancy would be even lower.
wojtek