Using GIS to Identify Health System and Infrastructure Changes to Improve Geographic Access to Routine and Emergency Obstetric Services in Mozambique

Session: Using Data Visualization to Improve Maternal and Newborn Health Outcomes

Presenter: Emily Keyes, FHI360
Download presentation slides

Background: Since 2007, Mozambique has seen substantial improvements in the institutional delivery rate, met need for emergency obstetric care (EmOC), population-based cesarean delivery rate, and the direct obstetric case fatality rate; though, the number of functioning EmOC facilities declined. Even with this progress, availability of and access to some critical life-saving interventions remains inequitable. We investigate geographic access to EmOC services in 2012 and, through spatial modeling, explore how resources, e.g. ambulances, referral infrastructure, and functioning EmOC facilities, could be better distributed to improve geographic equity in EmOC access.

Methodology: We use multiple spatial data sources including population, expected births, land cover, and road networks, as well as Mozambique facility-level data from 2012 on provision of EmOC services. In ArcGIS, we model multi-modal travel times to various service levels: 1) home to first level, 2) home to functioning comprehensive EmOC, and 3) transfer times between service levels. Finally, we propose improvements to the distribution of resources and model resulting changes in geographic access.

Results: We present national level results for 2012 that: (1) quantify access to EmOC, measured by the percentage of the population with timely access to functioning EmOC services; (2) identify geographic areas with poor access to EmOC and describe the referral readiness of lower level health facilities in these areas; (3) locate health facilities with critical referral infrastructure, and present a reconfigured system with referral resources distributed to maximize EmOC access; and, (4) for each proposed system improvement, summarize marginal changes in percent of the population with access to each service level.

Conclusion: Modeling geographic access to health services within a GIS is a powerful tool that can inform health planning at the national and sub-national levels. The methodology facilitates testing the impact of system improvements as well as planning with critical sectors beyond health such as roads and transportation.