Epidemiology of perinatal mortality in rural Burkina Faso: A community-based prospective cohort study

Abstract

Background: Perinatal mortality is one of the major public health problems in Sub-Saharan Africa. It is estimated that over 6 millions infant deaths occur each year during the perinatal period either as stillbirths or early neonatal deaths. However, the accurate estimates on this burden are rare, especially in Africa where over 40% of all perinatal deaths take place. The lack of reliable data on PNMR in developing countries could be one of the reasons that make it invisible and therefore getting little attention from the funding agencies. We took the opportunity of the PROMISE-EBF trial, a randomized community-based study that aimed at assessing the effect of the promotion of exclusive breastfeeding by peer-counsellors on EBF rates and child morbidity at 12 weeks of age, to describe the magnitude of PNMR in Banfora health district, a rural area, South of Burkina Faso. Study objectives: To measure the PNMR in the EBF cohort in Banfora health district To identify potential risk factors for perinatal death in this cohort. Methods: We performed a secondary analysis on the datasets of the EBF study which was a cluster-randomized trial in 24 villages of Banfora with an intervention package consisting of one antenatal and 6 postnatal individual counselling sessions on EBF. Data of the two arms were considered as those of a single cohort and the PNMR, the stillbirth and the early neonatal mortality rates were estimated. In a multivariable logistic regression using baseline characteristics of the study participants as exposures and the perinatal death as outcome, we calculated crude and adjusted OR for perinatal death, stillbirth and early neonatal death. Covariates with an OR statistically significant (p<0.05) were considered as risk factors for PNMR. Results: 900 pregnant women were sampled for data collection in the EBF trial. Five women were excluded later (wrong inclusions) and 20 women got multiple births (20 pairs of twins), and were excluded from further follow-up. 875 women with a single birth were followed up to day 7 postpartum and included in the final analysis. The PNMR, the stillbirth and the early neonatal mortality rates, were 73.1‰ [95% CI: 55.8-90.4], 56‰ [95% CI:40.7-71.2], and 18.1‰ [95% CI:9-27.2], respectively. In the crude analysis, the young age of the mother (<20 years), the parity (nulliparous women), the season of birth (dry season), and the intervention appeared as the main risk factors for PNMR. In a multivariable logistic regression adjusting for all variables that were found to be important in the occurrence of perinatal deaths, we found that the young age of the mother (OR=2.93 95% CI:1.54-5.57), a birth during the dry season (OR=1.85 95% CI: 1.19-2.87), and the intervention (OR=2.16 95% CI:1.20-3.89) were factors that increased significantly the risk of perinatal death. The intention of the mother to not EBF the future baby had a marginal effect on PNMR (OR=1.55 95% CI:0.97-2.49) but a statistically significant effect on the risk of stillbirth (OR=1.90 95% CI:1.04-3.47). Conclusion: Our study showed the burden of perinatal deaths in a rural area in Burkina with the highest PNMR ever reported in this country. The risk factors identified in this study have been reported in previous studies except the intention of the mother to EBF that need further investigations.