Background: Access to qualitative and equitable healthcare is a major challenge in Mauritania. In order to support the country’s efforts, a health sector strengthening programme was set up with participatory action research at its core. Reinforcing a health system requires a customised and comprehensive approach to face the complexity inherent to health systems. Yet, limited knowledge is available on how policies could enhance the performance of the system and how multi-stakeholder efforts could give rise to changes in health policy. We aimed to analyse the ongoing participatory action research and, more specifically, see in how far action research as an embedded research approach could contribute to strengthening health systems. Methods: We adopted a single-case study design, based on two subunits of analysis, i.e., two selected districts. Qualitative data were collected by analysing country and programme documents, conducting 12 semi-structured interviews and performing participatory observations. Interviewees were selected based on their current position and participation in the programme. The data analysis was designed to address the objectives of the study, but evolved according to emerging insights and through triangulation and identification of emergent and/or recurrent themes along the process. Results: An evaluation of the progress made in the two districts indicates that continuous capacity-building and empowerment efforts through a participative approach have been key elements to enhance dialogue between, and ownership of, the actors at the local health system level. However, the strong hierarchical structure of the Mauritanian health system and its low level of decentralisation constituted substantial barriers to innovation. Other constraints were sociocultural and organisational in nature. Poor work ethics due to a weak environmental support system played an important role. While aiming for an alignment between the flexible iterative approach of action research and the prevailing national linear planning process is quite challenging, effects on policy formulation and implementation were not observed. An adequate time frame, the engagement of proactive leaders, maintenance of a sustained dialogue and a pragmatic, flexible approach could further facilitate the process of change. Conclusion: Our study showcases that the action research approach used in Mauritania can usher local and national actors towards change within the health system strengthening programme when certain conditions are met. An inclusive, participatory approach generates dynamics of engagement that can facilitate ownership and strengthen capacity. Continuous evaluation is needed to measure how these processes can further develop and presume a possible effect at policy level.
We adopted a single-case study design and defined the ‘case’ as the health system strengthening approach of the AI-PASS programme in Mauritania. We chose two sites, namely the learning districts (“Moughataas d’Apprentissage”) of Dar Naim and Bababé. These districts were selected as AI-PASS sites on the basis of ITM’s experience and presence in Mauritania. There was a deliberate choice for one rural and one urban area in order to support the Ministry of Health in the implementation and adaptation of its national strategies in rural and urban zones. Both districts indeed present the deficits of the weak governance and low performance of the health system. Further details are described in Additional file 2. The study population of our case study included representatives of the communities and civil society of the two districts and the local, regional and central actors in the Ministry of Health of Mauritania. The case study had the following components: (1) a description of the development of the AI-PASS programme; (2) a description and assessment of the LHS analysis that was carried out in each district; (3) a description and assessment of the capacity-building and empowerment activities conducted in these two districts as a result of the situation analysis; and (4) identification and assessment of the effects of the action research approach. The study period was from mid-2016 to March 2019. Mauritania comprises a territory of more than 1 million km2, with about 4.3 million inhabitants and a low population density. For a decade now, internal migration has increased towards the capital, Nouakchott, where one-third of the population lives. Mauritania has substandard healthcare and problems of geographic inaccessibility. The maternal mortality ratio was estimated at 582/100,000 live births and infant mortality was estimated at 72/1000 live births in 2013 [15]. Healthcare is costly; the out-of-pocket expenditure (out of total health expenditure) is relatively high, at 43% [16]. This considerable cost of healthcare affects the vulnerable populations most, leading to catastrophic health expenditure and increased impoverishment. Currently, 42% of the population (59.4% rural, 20.8% urban) lives below the poverty level and 25.9% (40.8% rural, 7.7% urban) lives below the extreme poverty level [17]. Three methods were used to collect qualitative data. We carried out a document review to collect relevant national policy and programme documents, field visit reports, meeting summaries and workshop results. We conducted semi-structured interviews with key informants. To identify key informants, we used purposive and opportunistic sampling techniques. The selection was based on each informant’s current position and participation in programme activities (Table 1). After 12 interviews, data saturation was achieved. Written informed consent was obtained from all participants during the semi-structured interviews. Interviews were recorded when participants consented. Interviews lasted between 30 and 55 min, and they were conducted in French. An interview guide was elaborated (Additional file 1) and adapted iteratively during the process. Characteristics of interviewees MoH Ministry of Health, AI-PASS Institutional support for health sector strengthening We engaged in participatory observations of meetings, workshops and training sessions, in which local actors and representative authorities (at the local, regional and central levels of the Ministry of Health) participated. All participants were informed orally about the study. We collected data for each step of the case study (Table (Table22). Sources of data for analysing the health system strengthening approach of the AI-PASS programme in Mauritania (mid 2016 – March 2019) AI-PASS Institutional support for health sector strengthening The first recorded interviews were transcribed verbatim by the first author (KA). The remaining interviews were transcribed by an independent translator. All were checked for accuracy by the first author. The interviews were then entered into NVivo 12 software for data management and analysis. We used a thematic coding approach to analyse the primary data. Data from the document review, capturing the capacity-building activities and identified changes, were entered into a NVivo 12 project for analysis. Coding and thematic analyses were carried out by the first author (KA) and checked for accuracy by the last author (BC). An initial coding tree was elaborated deductively, based on our hypothesis and the objectives of the study. The coding tree evolved during the analysis. When we categorised the common elements in the interview transcripts and documents (meeting minutes and visit reports), diverse topics and patterns emerged. Recurrent themes included pathways of change, challenges, identified barriers and recommendations. Reflections of workshops and observations of meetings and trainings were systematically collected in a separate Excel file. By analysing the notes taken by the first author and by discussing experiences with the team members of the programme, these data were used for triangulation. We applied for and received ethical approval from the Institutional Review Board of ITM (Ref N° 1280/19). We received study approval from the Ministry of Health of Mauritania (Ref N° 003/2019).