Mortality in the era of HIV/AIDS in Tanzania
Understanding mortality in sub-Saharan Africa has been hampered by the lack of data and by the use of methods that do not consider changes in the burden of disease and environment in which new diseases are emerging. Demographers working in this region depend mostly on model life tables to make estimates and predictions. The basic assumption is that mortality in the population under study closely resembles the mortality shown by one of the model life tables. Recent studies on mortality in Africa have found stagnations or reversals of mortality in countries where HIV prevalence is high. We use DSS data from three communities and from two censuses in Tanzania to examine current levels and patterns of mortality and causes of death in the Tanzanian population, and whether these age patterns reflect a significant impact of the HIV/AIDS pandemic. We employ Preston-Bennett method to investigate changes in mortality from 1988 to 2002. We use Lee-Carter models to forecast impact of HIV/AIDS on mortality at different levels of HIV-prevalence. We compare our estimates with estimates from national populations with similar mortality conditions. We examine the extent to which the observed age patterns of mortality conform to the existing model life table systems. We find high levels of adult mortality in the most productive ages as a result of the HIV/AIDS pandemic, and that the pandemic significantly distorts the underlying age patterns of mortality. For populations experiencing the HIV/AIDS pandemic, the Coale-Demeny and the UN life table systems are no longer appropriate for modeling, estimating, and projecting African mortality. Despite few success stories in controlling the HIV/AIDS, the pandemic will continue to have a devastating impact on health and well-being of the population in sub-Saharan Africa. Although the HIV/AIDS pandemic clearly reduces life expectancy in affected populations, there is much more uncertainty about levels of mortality, particularly among adults in countries impacted by HIV/AIDS, than implied in the current available demographic estimates. Estimates of the impact of HIV/AIDS on mortality are sensitive to the type of data, the assumptions one makes about the quality of data and the estimation methods employed.
Mswia, Robert George, "Mortality in the era of HIV/AIDS in Tanzania" (2006). Dissertations available from ProQuest. AAI3246207.