Geographic barriers to establishing a successful hospital referral system in rural Madagascar

Background The provision of emergency and hospital care has become an integral part of the global vision for universal health coverage. It is recommended that at least 80% of a country’s population should be within two hours of a facility with essential surgery and anesthetic services. In order to strengthen health systems to achieve this goal, there needs to be an understanding of the time necessary for populations to reach a hospital. The goal of this study was to develop methods that accurately estimate referral and pre-hospital time for rural health districts in LMICs. We used these estimates to assess how the local geography can limit the impact of a strengthened referral program in a rural district of Madagascar.

Methods We developed a database containing 1) travel speed in a subset of routes by foot and motorized vehicles in Ifanadiana District; 2) a full mapping of all roads, footpaths and households in the district; and 3) remotely sensed data on terrain, land cover and climatic characteristics. We used this information to calibrate estimates of referral and pre-hospital time based on shortest route algorithms and statistical models of local travel speed. We compared these estimates with those from other commonly used methods in geographic accessibility modeling. Finally, we studied the impact of referral time on the evolution of the number of referrals completed by each health center in the district in 2014-2020 via generalized linear mixed models, using model estimates to predict the impact on referral numbers of strategies aimed at reducing referral time for underserved populations.

Results About 10% of the population lived less than two hours from the hospital, and more than half lived over four hours away, with variable access depending on climatic conditions. Only the four health centers (out of 21) located near the paved road had referral times to the hospital within one hour, which contributed over 75% of all 8,464 hospital referrals. Referral time remained the main barrier limiting the number of referrals despite health system strengthening efforts. The addition of two new referral centers is estimated to triple the population living within two hours from a center with higher acute care capacity and nearly double the number of referrals expected.

Conclusion This study demonstrates how adapting geographic accessibility modeling methods to local scales can occur through improving the precision of travel time estimates and pairing them with data on health facility data. Such information can substantially improve the design of a local health system to overcome existing barriers to care and achieve universal health coverage.