Free healthcare for some, fee-paying for the rest: adaptive practices and ethical issues in rural communities in the district of Boulsa, Burkina Faso

In Burkina Faso, in July 2016, user fees were removed at all public healthcare facilities, but only for children under 60 months of age and for “mothers”, i.e. for reproductive care. This study was conducted in five rural communities in Boulsa District (Burkina Faso) (1) to understand the perceptions and practices of stakeholders regarding compliance with eligibility criteria for free care and (2) to explore the ethical tensions that may have resulted from this policy. Semi-directed individual interviews (n = 20) were conducted with healthcare personnel and mothers of young children. Interviews were recorded and transcribed, and a thematic content analysis was conducted. The study reveals the presence of practices to circumvent strict compliance with the eligibility criteria for free access. These include hiding the exact age of children over 60 months and using eligible persons for the benefit of others. These practices result from ethical and economic tensions experienced by the beneficiaries. They also raise dilemmas among healthcare providers, who have to enforce compliance with the eligibility criteria while realizing the households’ deprivation. Informal adjustments are introduced at the community level to reconcile the healthcare providers’ dissonance. Local reinvention mechanisms help in overcoming ethical tensions and in implementing the policy. A cross-sectional qualitative study was carried out in the Boulsa health district in Burkina Faso to understand the ethical issues related to compliance with the eligibility criteria for free healthcare. We conducted semi-structured individual interviews with the main beneficiaries and providers of free care: mothers of young children (n = 10) and members of the healthcare staff (n = 10). A COREQ checklist (consolidated criteria for reporting qualitative research) is available in Additional File 1. Interviews were first conducted with healthcare staff until saturation was reached. After reviewing personal notes and discussions between the interviewer and other team members, the interview guide for mothers was reworked. To balance the volume of information collected between the two categories of participants, ten more interviews were conducted with mothers; again, saturation had been reached. The collection took place between September and December 2018, while the implementation of the free service was in a routine phase (two years after its introduction). At the time of their enrolment, participants were unknown to the research team. This study was conducted in five rural communities in the Boulsa District: Niega, Bonam, Boala, Yarcé and Zambanga. The district was selected for convenience, as it is a rural district in a safe and accessible area where staff from the health district authorities were known and trusted by the research team, and where there had not been a pilot project of free healthcare before the introduction of the national policy in July 2016. The Boulsa District is located in the North-Central region, about 100 km from the capital, and has a population of approximately 210,000 served by 19 primary health centres (Ministère de la Santé, 2017). The five communities were purposively selected after consultation with the district’s authorities, based on the proximity to the nearest primary health centre with maternity services. In each primary health centre, the head nurse was approached and, after briefly outlining the purpose of the study, we sought his/her consent. We then approached a dedicated maternal health staff member (one midwife or auxiliary midwife per health centre) for recruitment. In each of the five communities, two households were purposefully identified using the assistance of the local community health worker. The eligibility criteria for the households were that they well established in the community (resided there for several years) and had at least one child under 5 years of age and one mother above 18 years old. In each household, the head was informed of the project and, upon approval, an adult mother of a child under five was approached to proceed with recruitment. All approached individuals agreed to participate and gave consent. For the health staff, interviews were conducted at the health centre in a consultation room, while the mothers were interviewed in their homes. All interviews were individual and semi-structured, with an interview guide specific to the type of participant that was developed for this study (Additional File 2). Interviews were conducted in a room or a remote location that guaranteed the confidentiality of the respondents. The guide was slightly enriched as the data were collected, particularly with respect to sub-questions used to reopen the discussion or to explore a theme in greater depth. Interviews with caregivers lasted approximately 30 min, while those with mothers lasted between 30 and 60 min. They were all conducted by a single female interviewer with vast experience in qualitative research (A Bila) and who speak the local language. Interviews were conducted in French or in Mooré, depending on the participant’s preference. They were recorded, transcribed verbatim and translated into French (in the case of those conducted in Mooré). The transcripts were validated by a second person who listened to the original audio recordings. Field notes were taken during the interviews and used during the transcripts’ validation to add non-verbal information to the material. Transcripts were not returned to participants. A thematic content analysis was carried out on all the material (Miles et al., 2014). We developed a common coding grid and independently coded the transcripts line by line (Additional File 3). We identified the dominant categories within the collected data and defined them as themes. Initial themes were discussed with team members and reformulated as necessary; some emerging themes that were accepted as significant were added. Double coding was performed to ensure that the themes were understood in the same way. The rare discrepancies (n < 10) were discussed and reconciled after clarification. QDA Miner software was used to facilitate the analyses. The participants did not provide feedback on the findings. This study has been approved by the Ethics Committee for Health Research in Burkina Faso (Deliberation #2018-6-075) and has received research authorization from the Ministry of Health of Burkina Faso (#2018-3032/MS/SG/DGESS/DSS). All participants provided written consent after the researcher explained the objectives of the study, read the consent form, and ensured that the information was understood. The researcher also presented the research institutions and their role in the project. The audio recordings and transcripts were stored on a computer with secure access. Participation was not remunerated.