Adoption of evidence-based global policies at the national level: Intermittent preventive treatment for malaria in pregnancy and first trimester treatment in Kenya, Malawi, Mali and the Gambia

In 2012, the World Health Organization (WHO) updated its policy on intermittent preventive treatment in pregnancy with sulphadoxine-pyrimethamine (IPTp-SP). A global recommendation to revise the WHO policy on the treatment of malaria in the first trimester is under review. We conducted a retrospective study of the national policy adoption process for revised IPTp-SP dosing in four sub-Saharan African countries. Alongside this retrospective study, we conducted a prospective policy adoption study of treatment of first trimester malaria with artemisinin combination therapies (ACTs). A document review informed development and interpretation of stakeholder interviews. An analytical framework was used to analyse data exploring stakeholder perceptions of the policies from 47 in-depth interviews with a purposively selected range of national level stakeholders. National policy adoption processes were categorized into four stages: (1) identify policy need; (2) review the evidence; (3) consult stakeholders and (4) endorse and draft policy. Actors at each stage were identified with the roles of evidence generation; technical advice; consultative and statutory endorsement. Adoption of the revised IPTp-SP policy was perceived to be based on strong evidence, support from WHO, consensus from stakeholders; and followed these stages. Poor tolerability of quinine was highlighted as a strong reason for a potential change in treatment policy. However, the evidence on safety of ACTs in the first trimester was considered weak. For some, trust in WHO was such that the anticipated announcement on the change in policy would allay these fears. For others, local evidence would first need to be generated to support a change in treatment policy. A national policy change from quinine to ACTs for the treatment of first trimester malaria will be less straightforward than experienced with increasing the IPTp dosing regimen despite following the same policy processes. Strong leadership will be needed for consultation and consensus building at national level.

The study was undertaken in Kenya, Mali, Malawi and The Gambia between February 2017 and February 2018. The epidemiology of malaria varies across the four countries. Kenya has four major malaria epidemiological zones: endemic, highland epidemic, semi-arid and seasonal and low risk, determined mainly by altitude, rainfall patterns and temperature; in Mali, malaria is endemic in the central and southern regions (where >90% of the population lives), epidemic in the north and highly seasonal; in Malawi, malaria is intense across the country with local variation in intensity and in The Gambia malaria transmission is heterogenous across the country (very low to low prevalence in western and central Gambia and moderate prevalence in eastern Gambia), and highly seasonal. The timelines for global and national policy adoption for each policy are provided in Figure 1, which includes the retrospective and prospective periods of the study in relation to IPTp-SP and ACTs, respectively. The policy on treatment of first trimester malaria with ACTs (WHO Malaria Policy Advisory Committee, 2016a) was under review by the WHO Malaria Chemotherapy TEG and was not adopted in any of the countries at the time of the study. The rationale for studying both of these policies simultaneously was that it provided the opportunity to use lessons learned from the retrospective IPTp-SP policy adoption to support and manage the prospective first trimester malaria treatment policy adoption (Walt et al., 2008). Timeline of WHO policies and recommendations for treatment and prevention of MiP in relation to study An analytical framework was developed (Figure 2) to explore what happened and why with adoption of the revised IPTp-SP policy, and what is likely to happen with adoption of the policy on first trimester treatment of MiP and why (Walt et al., 2008). Our focus was the adoption of WHO global policy to national policy. We defined adoption as the national decision-making process for the policy change in the country. The analytical framework provided a guide to our study tool development and analysis. Our framework was adapted from the policy triangle framework of Walt and Gilson (Walt and Gilson, 1994) including policy content, policy context, policy process, actors and power; and Tesfazghi et al. (2015) who described the functions of actors in the normative process of policy adoption for vector control in Nigeria. In our adapted framework, policy content is the technical content of the policy; process are the stages of the adoption process; context is not just the wider distal social, political and economic influences but more proximal factors within the realist concept of context as ‘the characteristics of the conditions in which the interventions are introduced’ (Pawson and Tilley 1997); and actors are individuals or institutions involved in or with an influence on policy decision-making. Our concept of context thus included perceptions of stakeholders on policy legitimacy, evidence and power. Analytical framework adapted from Walt (1994) and Tesfazghi et al. (2015) Driven by our aim to understand the policy processes and actors we extrapolated from the functions of actors suggested by Tesfazghi et al. (2015) to describe four stages of the policy adoption process: Stage 1: identify a [policy] need; Stage 2: review the evidence; Stage 3: consult stakeholders and Stage 4: endorse and draft policy. We recognize that these stages may not in practice follow one after the other in a linear fashion (Sabatier, 2007) and that they may occur more than once. However, defining these stages aided our exploration of how different institutions and individual actors interacted within the policy adoption process, what roles they took, and what factors influenced their decision-making within these roles. We assumed that perceived legitimacy, credibility and salience were important in the policy adoption of IPTp-SP in our study countries and were likely to be important for first trimester treatment of malaria with ACTs as was shown by D’Souza and Parkhurst recently in the global development process of two malaria control policies (D’Souza and Parkhurst, 2018). These policies were IPTi and intermittent preventive treatment of children under 5 years of age, which is now re-named as SMC. Legitimacy is the perception that evidence generation and use was unbiased; credibility that the evidence was of sufficient strength and quality and salience that the evidence was relevant to the needs of the decision-makers (Cash et al., 2003). We were particularly interested in the capacity that stakeholders had to act and direct or influence the decision-making process of policy adoption and examined this through a power lens (Sriram et al., 2018). A document review of published and unpublished national documents was first undertaken. Documents reviewed included: organograms and structures of ministries of health; national malaria policies, strategies and guidelines; national reproductive health policies, strategies and guidelines and President’s Malaria Initiative (PMI) operational plans (Table 1). Documents were primarily accessed at the national level through national level contacts among the authors and requests to stakeholders and were supplemented by online searches for peer-reviewed publications on malaria and MiP policies in PubMed. Information was extracted from the documents into an excel spreadsheet on the policy making context, the content of malaria and reproductive health policy documents, and the key stakeholders involved. This information was used in the selection of stakeholders, development of the theme guides and analysis of the interview data. Documents reviewed by category 1. Programme Nationale de Lutte Contre le Paludisme. Politique Nationale de Lutte Contre le Paludisme au Mali. Bamako (Undated document). 2. Programme Nationale de Lutte Contre le Paludisme. 2013. Plan National de Suivi/Evaluation 2013–2017. 1. National Malaria Control Programme. 2014. Kenya Malaria Strategy 2009–2018 (Revised 2014). Nairobi. 2. Ministry of Public Health and Sanitation and Ministry of Medical Services. 2010. National Guidelines for the Diagnosis, Treatment and Prevention of Malaria in Kenya. Nairobi. 3. Republic of Kenya National Malaria Control Programme: http://www.nmcp.or.ke/index.php/resource-centre/download-centre/case-managment 4. Ministry of Health Division of Malaria Control. 2005. Malaria communication strategy. 5. National Malaria Control Programme (NMCP), Kenya National Bureau of Statistics (KNBS) and ICF International. 2016. Kenya Malaria Indicator Survey 2015. Nairobi, Kenya and Rockville, Maryland, USA: NMCP, KNBS and ICF International. 1. National Malaria Control Programme. Malaria Strategic Plan 2011–2015 Towards Universal Access. Lilongwe. 2. National Malaria Control Programme. 2011. Revised Guide for the Management of Malaria. Lilongwe. 3. National Statistical Office (NSO) [Malawi] and ICF International. 2016. Malawi Demographic and Health Survey 2015–16: Key Indicators Report. Zomba, Malawi and Rockville, Maryland, USA: NSO and ICF International. 1. Programme Nationale de Lutte Contre le Paludisme. 2017. Plan Strategique de Lutte Contre le Paludisme 2013–2017. 2. Programme Nationale de Lutte Contre le Paludisme. Politique Nationale de Lutte Contre le Paludisme au Mali. Bamako (there are no dates for this document). 3. Programme Nationale de Lutte Contre le Paludisme. 2013. Plan National de Suivi/Evaluation 2013–2017. 4. Cellule de Planification et de Statistique du Secteur Sante Developpement Social et Promotion de la Famille. Canevas de synthese des rapports d’evaluation 2015 et de programmation 2017 du programme national de lute contre le paludisme. 1. National Malaria Control Programme. National Malaria Strategic Plan 2014–2020. 2. Ministry of Health and Social Welfare. 2009. Malaria Control in The Gambia Strategic Plan 2008–2015. Banjul. 3. The Gambia Bureau of Statistics (GBOS) and ICF International. 2014. The Gambia Demographic and Health Survey 2013. Banjul, The Gambia and Rockville, Maryland, USA: GBOS and ICF International. 4. National Malaria Control Programme. 2015. The Gambia National Monitoring and Evaluation Plan for Malaria 2014–2020. 1. Ministry of Health. 2007. National Reproductive Health Policy: Enhancing Reproductive Health Status for all Kenyans. Nairobi. 2. Division of Reproductive Health and National AIDS and STI Control Programme, Kenya. National Curriculum on Sexuality and Sexual Health Training for Health Service Providers: Facilitator’s Manual. June 2011. 1. Ministry of Health. 2006. National Reproductive Health Strategy 2006–2010. 2. Ministry of Health. 2012. Road Map for Accelerating the reduction of Maternal and Neonatal Mortality and Morbidity in Malawi. Lilongwe. 3. Ministry of Health. 2009. National Sexual and Reproductive Health and Rights (SRHR) Policy. Lilongwe. 1. Department of State for Health. National Reproductive Health Policy 2007–2014. 2. Sundby J. 2014. A rollercoaster of policy shifts: global trends and reproductive health policy in The Gambia. Global Public Health 9: 894–909. 3. Ministry of Women’s Affairs. The Gambia National Gender Policy 2010–2020. Banjul. Stakeholders were purposively selected to include individuals and institutions with an interest in either the revised IPTp-SP policy and/or the first trimester treatment of MiP policy, who were affected by these policies or who, because of their position had or could have an active or passive influence on the decision-making and implementation processes (Varvasovszky and Brugha, 2000). The document review formed the starting point for the identification of key individuals and institutions and was supplemented following discussion with the study lead in each country. The final list for each country was formulated at a stakeholder selection session, where lists were compared and adjustments made. The final list broadly included stakeholders from: Ministry of Health (MoH) or other national institutions; National Malaria Control Programmes (NMCPs), National Reproductive Health Programmes (NRHPs); United Nations (UN) or bilateral institutions; Non-governmental Organisations (NGOs); international donors and academia. An iterative process was applied whereby additional stakeholders identified as relevant during the interviews were approached and invited to take part in the study. In-depth interviews were conducted in: Nairobi, Kenya; Bamako, Mali; Blantyre and Lilongwe, Malawi and Banjul, The Gambia. Stakeholders were given information about the study and their written consent to be interviewed obtained. Interviews were then conducted by a trained social scientist in each country partner research institution using a theme guide. The theme guide related specifically to revised IPTp-SP policy (World Health Organization, 2012) and the first trimester treatment of MiP recommendation (WHO Malaria Policy Advisory Committee, 2016b), and was standard across the four countries and included: formal policy making structures; perceptions of policy legitimacy; prospective or retrospective support or opposition from competing interest groups; perceived fit within the health system of the intervention; anticipated future benefits and costs. However, an iterative process was applied to the interviews, such that additional themes were included and further interrogated as they arose during the interviews. These new themes were then included in subsequent interviews. Interviews were conducted in English except for Bamako where they were conducted in French. Data collected during stakeholder interviews was digitally recorded and supplemented by handwritten field notes. Data were transcribed, and translated (Bamako), and imported into NVivo (QSR International) Version 11 for data management and analysis. A coding framework was constructed where the primary coding nodes were based on the elements of our analytical framework. This deductive coding was supplemented by inductive coding based on additional themes that emerged from the data using content analysis (Bernard, 2006). Coding was conducted by one author but analysed and interpreted by three authors to enhance objectivity. For each stage of policy adoption the process and actors with a specific role were identified. Study participants were assigned an anonymous code during data analysis which related to their category of stakeholder (see below), and these were used to label quotes to ensure anonymity. This approach was followed for both the retrospective and prospective studies. The Standards of Reporting Qualitative Research (SRQR) which provides clear standards for reporting through a list of 21 essential items was used to guide the reporting of findings (O’Brien et al., 2014). The SRQR does not specifically include an analytical framework within its reporting guidance on qualitative approach, but does include guiding theory more broadly.

N/A

• Dima Health