Evidence of health system resilience in primary health care for preventing under-five mortality in Rwanda and Bangladesh: Lessons from an implementation study during the Millennium Development Goal period and the early period of COVID-19.

Background: The coronavirus disease 2019 (COVID-19) pandemic led to disruptions of health service delivery in many countries; some were more resilient in either limiting or rapidly responding to the disruption than others. We used mixed methods implementation research to understand factors and strategies associated with resiliency in Rwanda and Bangladesh, focussing on how evidence-based interventions targeting amenable under-five mortality that had been used during the Millennium Development Goal (MDG) period (2000-15) were maintained during the early period of COVID-19. Methods: We triangulated data from three sources - a desk review of available documents, existing quantitative data on evidence-based intervention coverage, and key informant interviews - to perform a comparative analysis using multiple case studies methodology, comparing contextual factors (barriers or facilitators), implementation strategies (existing from 2000-15, new, or adapted), and implementation outcomes across the two countries. We also analysed which health system resiliency capabilities were present in the two countries. Results: Both countries experienced many of the same facilitators for resiliency of evidence-based interventions for children under five, as well as new, pandemic-specific barriers during the early COVID-19 period (March to December 2020) that required targeted implementation strategies in response. Common facilitators included leadership and governance and a culture of accountability, while common barriers included movement restrictions, workload, and staff shortages. We saw a continuity of implementation strategies that had been associated with success in care delivery during the MDG period, including data use for monitoring and decision-making, as well as building on community health worker programmes for community-based health care delivery. New or adapted strategies used in responding to new barriers included the expanded use of digital platforms. We found implementation outcomes and strong resilience capabilities, including awareness and adaptiveness, which were related to pre-existing facilitators and implementation strategies (continued and new). Conclusions: The strategies and contextual factors Rwanda and Bangladesh leveraged to build 'everyday resilience' before COVID-19, i.e. during the MDG period, likely supported the maintained delivery of the evidence-based interventions targeting under-five mortality during the early stages of the pandemic. Expanding our understanding of pre-existing factors and strategies that contributed to resilience before and during the pandemic is important to support other countries' efforts to incorporate 'everyday resilience' into their health systems.