Geospatial Prescriptions for Health Services

by Taurai Bwerinofa, MSc, formerly with MEASURE Evaluation Strategic Information for South Africa, and Sarie Podges, MEASURE Evaluation

Geography and health services may seem like an odd couple, not intuitively connected. Yet, by connecting the two, countries and locales may discover an opportunity to improve health outcomes for people, streamline performance and wise use of resources, and save big money.

Consider Mpumalanga Province in South Africa, where an investigation grounded in the province’s geography could mean a healthcare savings of US$1.6 million annually. Mpumalanga means “the place where the sun rises ” in several indigenous South African languages—fitting, as the province lies on the eastern South African borders with Swaziland and Mozambique.

In 2017, the province undertook a geospatial analysis, conducted by MEASURE Evaluation Strategic Information for South Africa (MEval-SIFSA), a project funded by the United States Agency for International Development (USAID) and the President’s Emergency Plan for AIDS Relief (PEPFAR). The purpose was to guide decisions about planning for health services for the population. For the most part, questions were simple: how many beds should be in a given hospital; how many ambulances were needed to serve a given area; and how many people and what amounts of money would be needed in each district to provide services?

The analysis is following two paths: (1) determine the most efficient way to distribute services with available resources, or (2) determine the 

Establish a referral chain from clinics to hospitals, sending clients first to clinics and, if necessary, onward to hospitals—to ease the burden on hospitalsoptimum distribution of resources throughout the province to ensure efficiency and equity in healthcare access. The province began with some ideas of improvements it wanted to consider:

• Consider how people move around the province and what disease trends exist in the province
• Ensure that healthcare is accessible, equitable, and efficient
• Review the status of hospitals, community health centers, and clinics in terms of staff, usable beds, coverage, etc.

© 2010 Andrew Bernish, Courtesy of Photoshare

The geographic information system (GIS) analysis used an ArcGIS network analysis tool. For example, it calculated drive time from ambulance stations to hospitals, incorporating data on typical speeds along specific road networks to determine how far an ambulance could go in 40 minutes. It mapped population centers around health facilities to determine the catchment population of each facility. And then, it came up with data to provide answers.

The key findings of the GIS analysis in Mpumalanga Province provided the province these insights:

• Health facilities were generally located in areas of high population density.
• Community health centers were located mostly in former homelands that had dense, but rural, populations, where people have lower incomes.
• Most district hospitals were situated in areas with large commercial farms, where population is less dense.
• Some areas had no hospitals within easy access, including some areas of dense population.
• Many households in the province could not get to health facilities by road.
• Some facilities were overused and therefore had the potential of being upgraded to a hospital to meet the needs of the community.
• Ambulances could get to an incident within 40 minutes as per the national core standard for rural areas, and thus were already optimally located.

The GIS analysis suggested, in general, that facilities should be reviewed so that the level of service offered could be upgraded, downgraded, or kept the same—a determination made according to standardized classification of facilities, service packages and national standards.

And then, GIS offered—Very. Specific. Advice:

• Immediately reconfigure some facilities that were either underused or are expected to render service packages at a level other than what is currently provided.
• Establish health posts in underserved areas and expand services to dense population areas with high growth rates.
• Establish contingency measures for ambulances during peak travel periods.
• Periodically review the classification of health facilities to ensure they align with population needs.
• Develop and strengthen health staff skills to use information and communication technologies, including GIS.

The recommendations made would potentially save the provincial department of health USD $1.6 million annually—depending on political will, government processes, and requisite approvals. Based on the findings of the GIS analysis, a 10-year strategic plan to transform service delivery in Mpumalanga Province is being developed.

Geography—more powerful than you might have guessed.

For more information, please visit www.measureevaluation.org and for more on GIS, please see https://www.measureevaluation.org/our-work/gis.