Cause of, and factors contributing to, stillbirth in sub-Saharan Africa

• Main Author: Aminu, M.
• Other Authors: van den Broek, N. ; Bar-Zeev, S.
• Published: University of Liverpool 2017
• Subjects: 616.9
• Online Access: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.733919

Background: Every year, an estimated 2.6 million stillbirths occur worldwide, with up to 98% occurring in low- and middle-income countries (LMIC). Most stillbirths are preventable. To develop strategies and take effective actions to end preventable stillbirths, a good understanding of the cause of death and its contributing factors is necessary. There is, however, a paucity of data from most LMIC settings. This study aimed to determine the cause of stillbirth in LMIC using three methods of assessment, and to assess quality of care delivered to mothers who had stillbirth. Methods: The study involved 1,563 stillbirths which occurred in 12 selected secondary and tertiary hospitals in Kenya, Malawi, Sierra Leone and Zimbabwe. The cause of death was determined by: (1) consensus of healthcare providers (HCPs) through stillbirth review; (2) expert review of cases and; (3) computer algorithms. Cause of death was classified using the classification according to Relevant Condition at Death (ReCoDe) and the International Classification of Diseases for Perinatal Mortality (ICD-PM). Quality of antenatal and intrapartum care and health system factors were reviewed using a set of criteria. Results: A total of 1,329 cases were reviewed, of which 1,267 (95.3%) stillbirths met the inclusion criteria. By country, the stillbirth rate ranged from 20.3 (Malawi) to 118.1 (Sierra Leone) per 1,000 births. The distribution of the major causes of stillbirth differed by method of assessment: asphyxia (18.5% – 37.4%), placental disorders (8.4% – 15.1%), hypertensive disorders in the mother (5.1% – 13.6%), infection (4.3% – 9.0%), cord problems (3.3% – 6.5%), and ruptured uterus due to obstructed labour (2.6% – 6.1%). Information was insufficient to assign cause of stillbirth in 17.9% - 26.0% of cases. Significant agreement was observed between cause of stillbirth assigned by the expert panel and by HCP (k=0.69; p < 0.0005) but there was a weaker agreement between expert panel and when using computer algorithms (k=0.34; p < 0.0005). Using ReCoDe, intrapartum events (mainly intrapartum asphyxia) contributed to most of the deaths, followed by maternal diseases (mainly hypertensive disorders and infection), placental and fetal conditions. With application of ICD-PM, 42.0% were antepartum, 50.7% were intrapartum and 7.3% could not be categorised. The major categories accounting for the death were: intrapartum hypoxia and fetal growth restriction. Major contributing maternal conditions in ICD-PM were: M1 (placental, cord and membranes) and M3 (other complications of labour and delivery). Poor quality of care during antenatal care was identified in 97.8% of cases, and only 30.7% of cases of Caesarean section were conducted within one hour of decision. For 414 (37.9%) stillbirths, the outcome could have been different with better care. Conclusion: Stillbirth rate was high, with high variations between countries. HCPs should be encouraged to conduct reviews and act upon findings to improve quality of care. Data requirements of computer algorithms need to be balanced between ability to find a cause and the availability of information. The new ICD-PM could work in LMIC, but there is the need for more guidance on how to handle cases of stillbirths whose time of death cannot be determined.