District-Level Estimates of Home Deliveries in Ghana: Application of Small Area Analysis Using DHS and Census Data

Fiifi Amoako Johnson, University of Southampton
Sabu S. Padmadas, University of Southampton
James Brown, University of Southampton

National estimates show that home deliveries under the supervision of unskilled attendants are high in Ghana. Spatial inequalities across administrative regions, where health planning and monitoring is focal, are not known. This study aims to derive district-level estimates of home deliveries and assess spatial variations between districts using data from the 2000 Ghana census and the 2003 DHS. The Fay-Herriot model for small area analysis is used to produce estimates for the 110 districts of Ghana. Most of the districts in the Savannah zone, where maternal mortality is very high, have the highest proportion of home deliveries. Home deliveries range from 73% to 95% in districts of the Northern Region compared to 13% to 22% for districts in the Greater Accra Region. Using five diagnostic procedures, the reliability of the estimates were assessed. The estimates show a significant clustering effect in delivery care uptake across districts that warrants policy and program attention.