Mortality in rural Ethiopia: Levels, trends, differentials

Scattered health information from small towns, villages, and hospital admissions in large cities in Ethiopia suggest that infectious and parasitic diseases are still predominant. However, careful studies documenting the levels, trends and differentials in child and adult mortality are lacking for the rural areas. To fill this gap, we estimated child and adult mortality rates by sex for each region using data from the 1969-71 and 1981 rural demographic surveys. Estimates of mortality trends are made and regional life tables by sex are constructed. Child mortality correlates are also identified. Childhood mortality estimates are derived using a generalization of the Brass procedure. The orphanhood method is emphasized over others in the estimation of adult mortality. We generalized the orphanhood procedure and were able to express the multipliers not only as functions of the mean age of childbearing, as suggested by Brass, but also as functions of the variance and higher order moments of the fertility function. We also demonstrated that the validity of the independence assumption between the mortality of parents and that of their children is invalid in Ethiopia. Based on the 1981 data we estimated an IMR of 150 and a q(5) of 232 for both sexes combined, expectations of life at birth (e$\sb{\rm o}$) of 47.83 for males and 48.71 for females for the 12 rural regions considered. Because it appears that surviving children were more completely reported in the 1981 than in the 1970 surveys, we argue that even the nonsignificant child mortality decline implied by unadjusted figures is exaggerated. The range in e$\sb{\rm o}$ across regions is more than 14 years. The large variation in regional mortality levels suggests sharp social, economic, and cultural differences which might be operating through literacy status, health status, religion, and ethnicity. Literacy status and health status of parents are identified as key figures influencing child mortality. Unfilled needs for basic health inputs, such as adequate nutrition, clean water, access to health care facilities, and mass education have been the major obstacles to an accelerated mortality decline. The large social, economic, and environmental differences we detected in mortality highlight the importance of a more focused demographic and health survey in the country