Multimodal Implementation Research on Primary Health Care Services in sub-Saharan Africa: Implementation Outcomes, Service Readiness and Costs

Abstract

Primary health Care (PHC) serves as the cornerstone for achieving universal health coverage (UHC), one of the health-related targets in the 2030 Sustainable Development Goals (SDGs). PHC facility managers play a key role in ensuring quality of care by overseeing daily operations, managing human resources, and effectively implementing national PHC guidelines. This dissertation reports findings on implementation outcomes, service readiness, and costs through multimodal implementation research on strategies to improve PHC services in sub-Saharan Africa (sSA). Two of the three research Aims (Aims 2 and 3) are embedded within a district-based implementation and dissemination program in central Mozambique's PHC settings, called the Integrated District Evidence-to-Action (IDEAs) program for neonatal mortality reduction, implemented from 2016 to 2020. IDEAs applied a system-level Audit and Feedback (A&F) strategy that included three core components: routine facility and district readiness assessments, district-level biannual health facility performance review meetings, and targeted facility support through supportive supervision to help health facilities implement their micro-interventions developed during the biannual meetings. There were nine cycles of IDEAs. For Aim 1, this study synthesized implementation strategies and outcomes of evidence-based management interventions for PHC managers in sSA through a systematic review. Nine case studies from six countries—Ethiopia (3 studies), South Africa (2), and one each from Botswana, Kenya, Tanzania, and Zambia—were identified. The interventions included evidence-based training programs, peer-to-peer learning, electronic systems for monitoring management practices, and supportive supervision. Common implementation strategies included training and education, coaching and mentoring, knowledge sharing through learning collaboratives, interactive and continuous learning, and A&F using routine data. Interventions were implemented using more than one strategy. Acceptability was a consistent positive implementation outcome reported, with management effectiveness improving in areas such as financial and resource management, organizational climate, and human resource management. For Aim 2, we assessed effectiveness of management training and IDEAs intervention on improving basic obstetric and neonatal service readiness in PHC facilities. We found that IDEAs intervention's effectiveness in enhancing service availability was highest when health facility managers in the intervention sites had received management training, with an average increase of 11.1 points out of 100 per year (95% CI: 0.7 to 21.5, p=0.037), after adjusting for potential confounders. Hence, capacitated PHC managers were better able to optimize a system-level A&F strategy to improve PHC services, bundling management training with A&F strategies could enhance effectiveness. For Aim 3, a mixed costing approach (gross and microcosting) was used to estimate the cost of implementing IDEAs. We found the total cost of the program across 12 districts over five years (2016-2020), discounted to 2020 US dollars was USD $2,197,971 with $495,323 (23.8%) allocated to capital costs and $1,702,648 (77%) to recurrent costs. The average cost of IDEAs activities annually per district was $36,693; A&F meetings made up $10,893 (29.7%) of costs, with per diem as the main cost driver; Capital cost were $8,255 (22.5%), with vehicle purchase as the main cost driver; Targeted support were divided into two parts, district focused and facility supervision. The performance review meetings occurred biannually, each lasting five days, resulting in a total of 10 days per year per district. The average hours spent per year attending A&F meetings for the seven key positions from health facilities, district and province was 2320 hours (80 hours per person annually per district), while five district staff conducting supervision spent a total of 240 hours per year (60 hours per person annually per district). We were not able to estimate staff hours for routine data collection since it was contracted to local agencies. This study provides new insights into the cost of implementing iterative system-level A&F strategies in low-income settings. As demonstrated in this dissertation, we applied a multimodal implementation research approach — using evidence synthesis to improve health system coordination and management, strengthen management capacity, and incorporate economic evaluation — helped identify pathways for the systematic integration of national PHC guidelines, offering insights for improving PHC services. The key takeaway is that while implementation research is valuable for improving healthcare systems, there is a significant knowledge gap regarding what works for PHC management in sub-Saharan Africa and the costs associated with implementing a system-level A&F strategy. This dissertation aims to spread knowledge about strengthening health systems to improve primary healthcare and achieve universal health coverage.