Social demographic determinants of maternal mortality in Zambia: a comparative analysis of 2002 and 2007 demographic health surveys
Nauluta, Cynthia Mwasile
MetadataShow full item record
This study is aimed at investigating the influence of the socio-demographic determinants on maternal mortality in Zambia drawing from an analysis of 2002 and 2007 demographic health survey data. A logistic regression was applied on both the 2002 and 2007 Zambia Demographic Health Survey data. The binary logistic regression was used to estimate the woman dying while pregnant, during delivery or within two months after child birth. In this study the independent variable was explained by the odds ratios of the explanatory variables. The dependent variable wasdenoted by the number(1) one if a woman died of the pregnancy related death and (0) zero if otherwise. Findings from the study indicate higher risks of dying from maternal mortality amongst adolescent mother in rural areas and outside marital unions. This is particularly evident from 2002 ZDHS for adolescent mothers in Eastern and Northern provinces. Adolescent mothers from Eastern and Northern provinces were 1.6 and 1.4 respectively at higher risk of dying from pregnancy related causes compared to women aged 44-49 years from Lusaka, the reference province. The observed early maternal mortality among adolescents outside marital unions was not significantly different in 2007 ZDHS,although there were more provinces experiencing this pattern, with education and religion of the mother having some influence on overall deaths. Mothers with no education were at 1.5 higher risk of dying compared to mothers with higher levels of education, the reference group. It is therefore, evident from both 2002 and 2007 ZDHS that pre-marital fertility especially in rural areas of Zambia and amongst women with no education is having significant influence on maternal mortality. In order to overcome this challenge, there is need to target young adolescent mothers for comprehensive health care interventions supported by measures to keep adolescents longer in schools. Even though previous surveys have had challenges in investigating factors associated with maternal mortality due to inadequate numbers resulting from maternal deaths, this study shows that the ZDHS data has great potential in explaining the factors associated with maternal mortality when the logistic regression is applied on such survey data. However, the ideal framework for investigating determinants of maternal mortality as proxy for maternal mortality should be supported by including data on biomedical variables that are otherwise unavailable from the Zambia Demographic Health Survey data.