Duke 2008 | ASHE | Equity and Efficiency in Health and Healthcare

The national health policy of most countries reflect concerted attempts at improving the average level of population health while reducing health disparities between subgroups, which might conflict with each other as resources are usually limited. Wagstaff recently proposed the generalized health achievement index as a summary measure for combining information on both level of health and income-related inequality in health. The index has been increasingly used in assessing health achievement across subgroups, particularly for the purpose of international comparison. The health achievement index is a weighted average of individual health with weights determined by a function of the individual's fractional rank in terms of income and hence balances efficiency and equity in assessing the performance of regional and national health systems.

The first paper in this session uses three waves (1992-93, 1998-99 and 2005-06) of India's National Family Health Survey (NFHS) to assess temporal and state-level (spatial) trends, and rural-urban differences in average levels and inequities in infant and child mortality in India over a decade and a half. Both concentration index and health achievement index are computed for the states participating in the surveys. The paper finds significant national improvements in infant and child health achievements, albeit with gaps between rural and urban areas. The inter-state and rural-urban disparities in health achievement demonstrate the need for national-level policy discussions.

The second paper derives an asymptotic variance estimator of health achievement index with the preference weighting parameter is set at two, which results in a widely used linear weighting scheme in terms of the fractional income rank. A simulation exercise is used to validate the estimator. Compared to the bootstrapping method, the analytical variance estimator has an advantage of computational efficiency, particularly when there are a large number of observations in the dataset. The implications of health achievement index and the use of the asymptotic variance estimator are illustrated in an empirical application using data from the U.S. Current Population Survey, 2001-2006.

The third paper examines the methodological issues in measuring socioeconomic inequalities among European elder populations. The paper utilizes two waves of the Survey of Health, Ageing and Retirement in Europe to assess inequalities in functional limitation by comparing the level of health inequalities using three different methodological approaches: odds ratios, concentration indices and overall health achievement indices. The paper finds evidence that health inequalities persist in later life and that the three indicators suggest somewhat different rankings of the subgroups. It is concluded that the strengths and limitations of a chosen approach should be recognized by practitioners when assessing socioeconomic inequalities in health.