Risk factors for stillbirth and early neonatal death: a case-control study in tertiary hospitals in Addis Ababa, Ethiopia

BACKGROUND: Ethiopia is a Sub-Saharan country that has made significant improvements in maternal mortality and under-five mortality over the past 15 years. However, the nation continues to have one of the highest rates of perinatal mortality in the entire world with current estimates at 33 deaths per 1000 live births. METHODS: This case-control study was conducted between October 2016 and May 2017 at Tikur Anbessa Hospital and Gandhi Memorial Hospital. All women who had a stillbirth or early neonatal death (i.e. death within 7 days) during this period willing to participate were included as cases. A systematic random sample of women delivering at the hospital were approached for recruitment as controls to generate a 2:1 ratio of controls to cases. Data on risk factors were retrieved from medical records including delivery records, and treatment charts. Statistical differences in background and social characteristics of cases and controls were determined by t-test and chi-squared (or fisher’s exact test) for quantitative and categorical variables respectively. Binary logistic regression analysis was completed to determine any associations between risk factors and stillbirth/early neonatal death. RESULTS: During the study period, 366 women delivering at the hospitals were enrolled as cases and 711 women delivering at the hospitals were enrolled as controls. Records from both hospitals indicated that the estimated stillbirth and neonatal mortality rates were 30.7 per 1000. Neonatal causes (43.4%) were the most common, followed by antepartum (32.5%) and intrapartum (24.5%). Risk factors for stillbirths and early neonatal death were low maternal education (aOR 1.747, 95%CI 1.098–2.780), previous stillbirth (aOR 9.447, 95%CI 6.245–14.289), previous preterm birth (aOR 3.620, 95%CI 2.363–5.546), and previous child with congenital abnormality (aOR 2.190, 95% 1.228–3.905), and antepartum hemorrhage during pregnancy (aOR 3.273, 95% 1.523–7.031). CONCLUSION: Antepartum hemorrhaging is the only risk factor in our study amenable for direct intervention. Efforts should be maximized to improve patient education and antenatal and obstetric services. Moreover, the most significant cause of mortality was asphyxia-related causes. It is imperative that obstetric capacity in rehabilitation services are strengthened and for further studies to investigate the high burden of asphyxia at these tertiary hospitals to better tailor interventions.