Implementation research protocol on the national community health policy in Guinea: A sequential mixed-methods study using a decision space approach.

The overall goal of this study is to explore the rollout of the community health policy in Guinea in the context of decentralization, and the role of decision space (the decision authority, capacities, and accountability of local officials) in explaining gaps between the policy's conceptualization and actual implementation. The implementation research study will employ a sequential explanatory mixed-methods design. The study will be conducted in 27 communes purposefully selected across the country and include communes where the national community health policy is fully, partially, and not yet being implemented. The quantitative component, based on a survey questionnaire and secondary data, will use ordinary least squares (OLS) multiple regression to compare maternal and child health (MCH) coverage indicators according to the level of policy implementation in the commune. An interrupted time series analysis will be conducted to assess changes in routine MCH service delivery indicators associated with implementation of the community health policy, comparing indicators from one year prior to implementation. OLS regression will be conducted to assess the association between decision space and MCH indicators; all analyses will be carried out in Stata. Findings from the quantitative study will be used to inform the key qualitative questions and areas to explore in greater depth, to develop the interview and focus group guides, and to generate an initial codebook. Qualitative data will be double coded in NVivo by two qualitative analysts, and results generated using thematic analysis. Findings from the quantitative and qualitative components will be integrated and triangulated for interpretation and reporting. Findings and recommendations of this study will inform revisions to the National Community Health Policy to improve its rollout and effectiveness.