Urban-rural mortality differentials in adults in Burkina Faso : measures and effects of migration

Evidence regarding differences in mortality between urban and rural areas of Sub-Saharan Africa has been largely based on the experience of children. Urban-rural mortality differentials in adults are often obtained by inferring from child mortality rates and model life tables. However, the rapid urbanization in Sub-Saharan Africa is associated with some health challenges specific to adults, such as HIV/AIDS, tuberculosis, occupational hazards, violence, road traffic accidents, and a growing burden of non-communicable diseases. For this reason, urban-rural mortality differentials in adults should be measured, and not modelled. Measuring these differentials, and making sense of the resulting estimates also requires to take into account the issue of migration. Urban and rural areas are historically interlinked through important migration flows. In this context, the main objectives of this dissertation are to contribute to the measurement of urban-rural mortality differentials in adults and to the analysis of the impact of migration on that differential in Sub-Saharan Africa. The analysis draws mainly on the case of Burkina Faso, and combines various sources of data, (including sample surveys, Health and Demographic Surveillance Systems, and censuses) to circumvent the scarcity of data on adult survival. This dissertation shows that mortality indicators derived from censuses in rural areas can be severely underestimated, in particular in women. The measurement of urban-rural mortality differentials based on census data can be affected by underreporting of deaths, age misreporting and misreporting of the place of residence of deceased persons due to the mobility of dying individuals. Apart from censuses, other sources of data on adult survival based on the survival of close relatives (parents and siblings) are also plagued by underreporting of deaths, and most importantly missing information on the place of residence of individuals exposed to the risk of dying. Despite these limitations, a systematic analysis based on all available data indicates that in Burkina Faso, adults living in urban areas still benefit from a health advantage compared to their rural counterparts. Regarding migration, positive selection of migrants in rural areas plays a role in shaping the urban health advantage. However, the health status of migrants tends to deteriorate with their duration of residence in cities. The dissertation suggests that innovative strategies are needed to improve the quality of data collected on adult survival in Africa.(POLS - Sciences politiques et sociales) -- UCL, 201