An epidemiological analysis of the determinants of childhood malnutrition and mortality in southwest Uganda
4320 children aged, 0-59 months, were measured (Weight, Height and Mid Upper Arm Circumference) in 31 villages in the district of Mbarara (Southwest Uganda) in March and April 1988. Socioeconomic and health characteristics of the families of the children were also collected. The socioeconomic variables were analysed through Multiple Correspondence Analysis which produced 7 socioeconomic classes. These were reduced to 3 socioeconomic groups (SEGs) because of similarities between some of the classes. SEG I was the most advantaged group with a higher level of education, a high prevalence of Government workers and professionals who were able to hire labour. SEG II was mainly composed of cattle keepers and producers of export crops with only primary education, not hiring labour and not working on other people's land. SEG III was the most disadvantaged group with minimal or no education and mainly comprised of subsistence farmers frequently working on other people's land. SEG I was the best off in every socioeconomic, health and nutrition indicator, while SEG III was the worst off. After 12 months, a follow-up survey was carried out in order to assess the number of children who had died. Mortality rates were inversely proportional to anthropometric indices. There was no interaction between mortality associated with different anthropometric cut-off points and SEGs. Malnutrition produced similar mortality across the SEGs. The most sensitive predictor for mortality was Mid Upper Arm Circumference and the weakest was Weight for Height. Malnourished children had a significantly higher risk of death from fever, measles, diarrhoea, acute respiratory infections and malnutrition while there was no evidence of a higher risk of death from other causes. Ownership of cattle, length of stay in the village, birth order, father's education and kind of lighting fuel were significantly associated with child mortality. Nutritional status was influenced by socioeconomic variables directly related to the poverty conditions of the family, to morbidity (especially diarrhoea), and to hygienic conditions in the house. Poverty was the main determinant of mortality and malnutrition, but even in the better off sector of the community, malnutrition per se put children at a higher risk of death.