RESUMO
PIP: Maternal education has been accepted almost unanimously as a major socioeconomic factor in infant and postinfant mortality, but anomalies have been found which call the relationship into question. Considerable evidence points to the existence of other important elements in infant mortality besides maternal education. Survey results from Kenya, Lesotho, and other countries in Africa and elsewhere do not consistently demonstrate higher mortality among children of less educated or illiterate mothers. It is thus important to determine how maternal education influences infant mortality. 3 principal interpretations have been that educational level is an indicator of living standard and consequently determines access to goods and services that influence mortality; that formal education permits acquisition of knowledge and modern standards in the areas of hygiene, child care, health and related domains that affect the level of mortality; and that education measures the degree of breaking with tradition. The relation between maternal education, selected intermediate variables influenced by it, and mortality was analyzed using data from the Survey of Infant and Child Mortality in Yaounde (EMIJ). This survey was conducted on a sample of infants born in 1978 in maternity centers of Yaounde to women residing in the city. The infants were visited at home at 1, 4, 8, 12, 16, 20, and 24 months. Intermediate variables analyzed included the type of milk provided at 8 months, whether meat and eggs were given at 8 months, whether or not water given to infants at 4 months was boiled, what vaccinations the child received by 12 months, the number of prenatal visits, and the source of household water. Infant mortality was found to be lower when the number of prenatal visits increased, when boiled water or mineral water was given to the child, when there was running water in the house, and when the infant was vaccinated. The risk of death increased with bottle feeding. A relationship was also found between the educational level of the mother and the intermediate variables, but the strength of the relationship varied for different variables. The relationship was very weak in the areas of type of foods given and number of vaccinations. The strongest relationship was with the mode of feeding, the very variable that sometimes leads to a positive relationship between maternal education and infant mortality. The factor of ethnic group was found to disturb the normal relationship between maternal education and infant mortality, probably because each ethnic group adopts elements of western culture at its own distinctive pace. The data demonstrated that it is only after 10 years of successful schooling that the changes in habits and customs become significant. A few years of school do not automatically determine a decrease in infant mortality.^ieng
