Logistic regression analysis of determinant of neonatal mortality in Ghana

Abstract

According to Black (2003), neonatal mortality accounts for almost 40 per cent of under-five child mortality globally. In Ghana, recent studies argued that mortality among infants and children has been declining. An understanding of the factors related to neonatal mortality is important to guide the development of focused and evidence-based health interventions to prevent neonatal deaths. This study aimed to identify the determinants of neonatal mortality in Ghana, from the Multiple Indicator Survey (MICS, 2006). Logistic regression, using a stepwise approach, was performed to estimate odds ratio for each of the independent variables associated with neonatal deaths, using community, socio-economic status and proximate determinants. This research utilized a conceptual framework that identifies the multi-level and diverse determinants of health inequalities. The study showed that, at the community level, the odds of neonatal death was significantly higher for infants considering the geographical settlement of the mother compared to the lowest with maternal education. A progressive reduction in the odds was found as the percentage of deliveries assisted by trained delivery attendants increased. At the socioeconomic level maternal religion indicated a high ratio-change on neonatal death. At the proximate variable level, maternal age at birth was outstanding predictor of neonatal death followed by knowledge of child weight at birth, size at birth and place of delivery respectively. As a recommendation, public health interventions directed at reducing neonatal death should address community, household and individual level factors; which were found to significantly influence neonatal mortality in Ghana.