Stillbirth in the African Great Lakes region: A pooled analysis of Demographic and Health Surveys.

BACKGROUND

The aim of this study was to estimate the rate and predisposing factors associated with stillbirth in the African Great Lakes region (Burundi, Congo Democratic Republic, Kenya, Rwanda, Tanzania and Uganda).

METHODS AND FINDINGS

Cross-sectional data from the most recent Demographic and Health Surveys (DHS) of countries in the African Great Lakes region were used in this study. DHS from Congo Democratic Republic was not included in the analyses because data was not collected for stillbirth in the country survey. A pooled sample of 57046 pregnancies of 7+ months' duration and 1002 stillbirths were included in the final analysis. The analyses were restricted to stillbirths reported in the 5 years preceding the surveys. Stillbirth was defined as foetal death in the third trimester (≥ 28 weeks' gestation). Multilevel logistic regression analyses that adjusted for cluster and survey weights were used to determine the factors associated with stillbirth in the Africa Great Lakes region. Health service variables and maternal medical condition variables were not included in the analysis because DHS do not collect data on these variables for pregnancies that did not result in a live birth. Burundi had the highest stillbirth rate per 1000 births [23% (95% CI: 20, 25)] within the region. Factors associated with stillbirth across the region were: no schooling [1.85 (95%Cl: 1.44, 2.38)] and primary education [1.64 (1.32, 2.05)], advanced maternal age [2.39 (95% CI: 1.59, 3.59)], smoking [1.99 (95% CI: 1.19, 3.32)] and drinking water from unimproved sources [1.18 (95% CI: 1.01, 1.37)].

CONCLUSION

To achieve Every Newborn Action Plan (ENAP) stillbirth target of 12 per 1000 births or less by 2030, policy interventions to prevent stillbirth should focus on promoting community-based socio-educational programmes which encourages a healthy lifestyle especially among uneducated women in the advanced age spectrum.