Variation in the strength of the relationship of income with adult mortality and morbidity in selected countries

Jeroen van Ginneken, Netherlands Interdisciplinary Demographic Institute (NIDI)
George Groenewold, Netherlands Interdisciplinary Demographic Institute (NIDI)

In nearly all studies conducted in developed countries negative relationships were found between income and adult mortality and morbidity. There is much less information on this topic in developing countries and we hypothesize similar relationships. Another, more refined hypothesis is that the strength of the income - mortality/morbidity relationship varies in developing countries and that it varies according to income inequality (Gini index): the higher the income inequality in a country, the higher is the gradient in the income -mortality/morbidity relationship. The first objective is to summarize the available evidence on the income – mortality relationship in developing countries. The second objective is to test the more refined hypotheses, but this cannot be done for mortality due to lack of data. It can, however, be done for self-reported morbidity due to data collected in countries that participated in the World Health Survey conducted in the early 2000s. We selected nine countries in three regions (South Asia, Sub-Saharan Africa and Latin America) with roughly similar mortality levels, but varying in income inequality. We added three countries in Western Europe as a comparison group. The Self-Rated General Health (SRGH) measure and the Self-Rated Disability (SRD) index (from the WHO Disability Assessment Schedule) were used as the outcome variables and we determined how income was related to these two variables in the twelve countries (and also after controlling for other variables) in the age group 18-59 years old (men and women). We found that there was a positive association of income with SRGH and SRD in all countries. Tentative results indicate considerable variation in the strength of income with SRGH and SRD, but it is not yet clear if this variation is in accordance with the income inequalities observed. There is little information on the impact of income on health in developing countries and the relevance of our project is that it helps to fill this gap.