Determinants of stillbirths in Ghana: does quality of antenatal care matter?

BACKGROUND Each year, over two million babies die before they are born. Like maternal deaths, the great majority of these stillbirths occur in developing countries, with about a third of all cases worldwide in sub-Saharan Africa (SSA). Few studies have, however, examined the determinants of stillbirths in SSA. In addition, the emphases on promoting deliveries by skilled birth attendants and/or in health facilities to prevent maternal deaths, though important, may have undermined efforts to provide good quality antenatal care (ANC), which may have an additional role in preventing stillbirths. This study examines the factors associated with stillbirths in Ghana, focusing on the role of ANC quality. METHODS Data are from the Ghana Maternal Health Survey (N = 4,868)-a national survey of women of reproductive age. The main analysis includes women who had a pregnancy ending in a live birth or stillbirth in the five years preceding the survey and who received ANC at least once. ANC quality is measured by an index based on receipt (or otherwise) of nine antenatal services during the last pregnancy, including education about pregnancy complications; with receipt of at least of eight services classified as higher quality ANC. Stillbirths refer to babies born dead at seven or more months of pregnancy. Analytic techniques include multilevel logistic regression, with moderation and mediation analysis to examine conditional and intervening effects respectively. RESULTS Higher quality ANC decreases the odds of a stillbirth by almost half after accounting for other factors, including the type of delivery provider and facility. Educating pregnant women about pregnancy complications contributes significantly to this difference by ANC quality. The type of delivery facility and provider account for a small proportion (14 %) of the ANC quality effect on stillbirths and a larger proportion of the rural/urban difference (27 %) in stillbirths. Completing the recommended four antenatal visits decreases the odds of a stillbirth. Having a pregnancy complication, a multiple gestation, a past stillbirth, or a sister who died from pregnancy complications increases the odds of a stillbirth. CONCLUSIONS Good quality ANC can improve birth outcomes in two ways: directly through preventative measures, and indirectly through promoting deliveries in health facilities where complications can be better managed. Targeted programs and policies to increase ANC quality, including adequately educating women on pregnancy complications, will help improve birth outcomes in Ghana, and in SSA as a whole.