Subnational assessment of primary health care interventions during COVID-19: Evidence from implementation research in Rwanda

The COVID-19 pandemic brought about significant disruptions to health care delivery worldwide, including in Rwanda. Countries experienced variable disruptions both at the national and the subnational level. In this descriptive explanatory analysis, we report results from mixed-methods implementation research on evidence from Rwanda to assess disruptions or subnational variability in the delivery of health care interventions during a health shock such as the COVID-19 pandemic. To estimate the coverage of primary health care interventions known to reduce under-5 mortality during the initial period of COVID-19 in Rwanda, we analyzed existing data from the health management information system. Using the available administrative health management information system data in Rwanda during 2019 and 2020, we calculated (i) monthly and (ii) cumulative disruption ratios at the district level for 3 interventions: facility-based deliveries, 4 or more antenatal care visits (ANC4+), and diarrheal cases treated at health facility and community levels in 2019 and in 2020. We conducted key informant interviews between February and April 2021 with 21 policymakers, donors, implementing partners, and direct health services providers to identify barriers and facilitators of subnational variability in the delivery of primary health care interventions focused on children under 5, as well as implementation strategies, across Rwanda’s 30 districts. We found minimal disruption across most districts in Rwanda in the first phase of the COVID-19 pandemic (March to December 2020) and, furthermore, we found minimal subnational variability in disruption across districts: for facility-based delivery across the study period, only 5 districts experienced more than 2 months of disruptions (<1), meaning fewer services delivered than expected; for ANC4+, the disruption ratio for 17 was >1, meaning more ANC4+ attendance than expected. For cumulative disruption for diarrheal diseases, 17 districts had disruption (<1)—fewer cases of diarrheal disease reported than expected—while 6 districts had no disruption, and 7 had higher than expected cases (>1). Implementation strategies such as community health worker interventions, community engagement and education, provision of transport, and use of command posts, appeared to be important in sustaining low disruption across most districts. Facilitating contextual factors included a culture of accountability and a strong preexisting community health system and structure. Rwanda’s experience offers potentially transferable knowledge for policymakers and decision-makers in other regions and countries to minimize disruptions at the subnational and national levels to essential health services during future health shocks.