MATERNAL DETERMINANTS OF CHILD SURVIVAL IN A RURAL AFRICAN COMMUNITY

Background. The aim was to determine maternal factors related to child survival in the rural area of Bwamanda, Northern Zaire. Methods. A pr ospective study of 30-months mortality was carried out in a cohort of 776 children aged 0-3 months, obtained by random cluster sampling. Inc lusion criteria were exclusive breastfeeding, no severe prematurity an d absence of severe protein-energy malnutrition, diarrhoea or acute re spiratory infection. Mortality was recorded by regular home visits and inspection of hospital and funeral registers. Maternal factors that r emain stable during follow-up were studied, Results. Factors associate d with excess mortality in bivariate and multiple logistic regression analysis were: (i) mother has parity >5 (relative risk [RR] = 1.5-4.2) ; (ii) distance from the health centre >5 km (RR = 0.9-2.9); (iii) inv aliding maternal diseases (RR = 1.2-9.0). Maternal school education (c onditional odds ratio [OR] = 1.0-5.0) was significant in the multiple regression. In contrast to the other risk factors, mother-child separa tion or problems with breastfeeding were rare and did not significantl y increase mortality. Conclusions. Chronic stress situations created b y maternal invalidity, high parity and distance from health care facil ities, increase child mortality. Acute stress in the mother-child dyad seemed to be efficiently compensated for. In subsistence economy area s, maternal school education can be a disadvantage.