Maternal education and community context in maternal and child health behavior in Nigeria.

Abstract

Maternal education has been shown to be among the most powerful determinants of child mortality in developing countries. The mechanisms that link maternal education to improved child survival remain largely unclear, however. Using the 1990 Nigeria Demographic and Health Survey, I study the relationship of maternal education to three important intervening behaviors: maternity care, preventive care for children in the form of immunization and curative care for acute childhood illness. The objectives of the analysis are twofold. The first is to establish the relationship of maternal education to health care behavior in a culturally diverse African setting. The second is to test sociological theories concerning the role of social context in the apparent health benefits of maternal education. Using community level data and appropriate multilevel methodology, the analysis examines the differential effect of maternal education on health seeking behavior under varying conditions of health service availability and relevant dimensions of women's status. In the first stage of analysis, I employ descriptive statistics and conventional linear modeling techniques to document the relationship of maternal education to health care behavior. While maternal education is generally associated with higher levels of utilization of modern sector services as expected, the magnitude and pattern of education effects vary considerably according to the type of outcome, the range of alternative sector treatments considered, and ethnic and regional context. The influence of education is particularly sensitive to ethnic context. In the second and third stages of the analysis, multilevel logit regression analysis is used to examine effect of maternal education under varying conditions of health service availability and women's status dimensions. Proximity to health facilities diminishes education differentials in the use of modern sector health services for children, but enhances education differentials in the use of modern sector maternity services. Aggregate patterns of women's work and work characteristics are employed as indicators of women's status. The level of women's work involvement at the community level is found to substitute for the effect of maternal education on utilization rates.