Access, Service Readiness, and Use of Facility-Based Delivery Care: A Spatial Analysis of Health Facility Data and Population Data in Haiti

Session: Technology to Improve Coverage and Quality of Maternal Newborn Care

Presenter: Wenjuan Wang, ICF International
Download presentation slides

Background: In Haiti only 36% of births take place in health facilities, putting women at high risk of maternal mortality. One reason for the low use of facility-based delivery care could be the limited availability and poor quality of delivery care provided in health facilities. We applied a geo-spatial methodology to estimate the influence of the obstetric service environment on women’s use of facility delivery care.

Methods: Data for the study came from the 2012 Haiti Demographic and Health Survey (DHS) and the 2013 Haiti Service Provision Assessment (HSPA) survey, both of which collected GPS data of clusters or health facilities. We linked each DHS cluster to the SPA facilities within a specified distance (5 km for urban and 10 km for rural). The availability of obstetric care was measured by the number of facilities providing normal delivery care within the specified distance. Service readiness was measured with a range of indicators defined by the WHO. Random-effects logit regressions were used to model the variation in individual use of facility-based delivery care and cluster-level service availability and readiness with adjustments for individual-level factors. Separate analyses were conducted for urban and rural areas.

Results: About 42% of rural clusters and 24% of urban clusters were linked to no or only one facility offering delivery care within the specified distance. Urban clusters had access to better service readiness than rural clusters with a wide range of variation in both areas. Multilevel models indicate that in rural areas, both high levels of service availability and service readiness are associated with women’s greater likelihood of use of facility-based delivery care; in urban area, only service readiness was significant.

Conclusions: Improving service availability should be prioritized in rural areas. This may imply a need for changing the actual mandate of lower-level facilities to meet the local reality and health needs.