Daniel D. Reidpath, Ph.D.
It is hard to conceive of a situation in which an evidence base for national or regional health policy could be developed without measuring the health of the populations concerned. Knowing the health of a population can inform resource allocation, the prioritization of research and interventions, and the identification of the determinants. Which factors should be included in the measurement of population health, however is a matter of continuing debate. Current international measures of population health, such as the Disability Adjusted Life Year, tend to exclude from consideration the contexts in which the health condition occurs. Thus paraplegia of a fixed clinical manifestation will contribute identically to the measure of population health in one setting as it does in another without regard to social, cultural or environmental heterogeneity. The health impact of paraplegia on a person in rural Tibet will be equal to the health impact of paraplegia on a person living in Copenhagen in Denmark.
Using data from a recent study contrasting paraplegia in a developed country (Australia) and a developing country (Cameroon), the importance of context on the measurement of the burden of disease is illustrated. Stigma as a determinant of the impact of a health condition is the focus of the comparison; however, the general importance of context is highlighted. In general, paraplegia in Australia is less stigmatized than paraplegia in Cameroon, and it is suggested that there are cultural reasons for this, which interact with the level of economic development. An international research agenda in the measurement of health is proposed that takes account of stigma and other contextual factors on the impact of health conditions. It is argued that without this kind of approach, the effect of morbidity on population health will always be underestimated in stigmatized and contextually bound conditions, and because of the interaction with development will underestimate the effect among the poor and most marginalised.