Article Text

Download PDFPDF

Grandmother, aren’t you going to sing for us?’ Current childcare practices and caregivers’ perceptions of and receptivity to early childhood development activities in rural Burkina Faso
  1. Jennifer Hollowell1,
  2. Mari Dumbaugh2,
  3. Mireille Belem3,
  4. Sylvain Kousse3,
  5. Tessa Swigart1,
  6. Chantal Korsaga3,
  7. Pokiandi Solange Lankoande3,
  8. Kokovi Hogban Lawson3,
  9. Zelee Hill4
  1. 1Development Media International CIC, London, UK
  2. 2Insight Impact Consulting, Chicago, Illinois, USA
  3. 3Develpment Media International, Ouagadougou, Burkina Faso
  4. 4Department of Epidemiology and Public Health, Institute for Global Health, University College London, London, UK
  1. Correspondence to Dr Jennifer Hollowell; jennifer.hollowell{at}


Introduction Effective stimulation and responsive caregiving during the first 2 years is crucial for children’s development. By age 3–4 years, over 40% of children in sub-Saharan Africa fail to meet basic cognitive or socioemotional milestones, but there are limited data on parenting and childcare practices. This study, conducted to inform the design of a mass media intervention, explored practices, perceptions, motivators and obstacles to childhood development-related practices among parents and caregivers of children aged 0–2 years in rural Burkina Faso.

Methods We performed two rounds of six focus groups with 41 informants in two villages, using an adapted version of the Trials of Improved Practices methodology. These first explored beliefs and practices, then introduced participants to the principles and benefits of early childhood development (ECD) and provided illustrative examples of three practices (interactive ways of talking, playing and praising) to try with their children. One week later, further discussions explored participants’ experiences and reactions. Data were analysed inductively using thematic content analysis.

Results Existing activities with young children were predominantly instructive with limited responsive interaction and stimulation. Participants were receptive to the practices introduced, noted positive changes in their children when they adopted these practices and found engagement with children personally rewarding.

Conclusion Interactive, stimulating activities with young children did not appear to be widespread in the study area, but caregivers were receptive to information about the importance of early stimulation for children’s development. ECD messages should be tailored to the local sociocultural context and consider time limitations.

  • early childhood development
  • mothers
  • fathers, grandmothers
  • parenting
  • Burkina Faso
  • qualitative research

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Key questions

What is already known?

  • Lack of stimulation and responsive caregiving during a child’s early years can have long-term detrimental effects on health, educational achievement and economic prosperity throughout the life course.

  • Little is known about parenting and child care beliefs and practices in sub-Saharan Africa, nor the motivation and obstacles to conducting early childhood development (ECD) practices.

What are the new findings?

  • Caregivers currently engage in some ECD-related activities, but these are often instructive in nature, with limited responsive interaction or stimulation.

  • Parents and caregivers responded positively to introducing more interactive and stimulating activities into their child’s routine, illustrated through a radio format, saw positive changes in their children when they adopted these behaviours and found engagement personally rewarding.

What do the new findings imply?

  • Caregivers are motivated to help their children succeed and are receptive to messages about the need for interactive stimulation.

  • Interventions need to be carefully tailored to the local sociocultural context and promote activities that fit the daily routines of the intended recipients.


Effective interaction, stimulation and responsive caregiving during a child’s first years play a key role in brain development affecting cognition, motor skills and overall neurodevelopment. At this time, suboptimal parental interactions and poor nutrition can have long-term detrimental effects on health, educational achievement and economic prosperity.1 2

Around 250 million (43%) children worldwide are at risk of not reaching their developmental potential.3 Recent research suggests that by the age of 3–4 years, over 40% of children in sub-Saharan Africa (SSA) fail to meet basic cognitive or socioemotional milestones.4 Despite this there are limited data on parenting and childcare beliefs and practices in SSA. Much of what we know is derived from the early childhood development (ECD) questions added to the Unicef Multiple Indicator Cluster Surveys (MICS) in 20095–7 and some qualitative data from ethnographic studies.8–12 MICS data for countries in West and Central Africa indicate relatively low levels of adult, especially fathers’, support for early learning and limited access to books and playthings compared with other low/middle-income countries (LMICs).13 The limited available data on household childcare practices in Burkina Faso indicate similarly low levels of stimulation in the early years5 14 15 and limited access to play and other early learning resources at home.16 For example, 19% of mothers and 6% of fathers in Burkina Faso named, counted and drew with their child in the past 3 days compared with 44% and 17% respectively across all 39 LMICs included in MICS third round (2005–2010).14

Effective interventions exist to improve developmental outcomes by improving stimulation and responsive caregiving17 and the benefits of well-targeted early interventions can still be measured many years later.18 These interventions commonly employ home visits or group meetings by community health workers/paraprofessionals, and there are concerns about the ability of programmes to maintain quality when scaled up,19 with limited rigorous evaluation at scale.20 Determining who can effectively deliver ECD interventions, and whether interventions can be taken to scale with sufficient integrity and fidelity, are key research priorities.21 To date there has been little focus beyond face-to-face approaches focusing on the mother–child dyad. This is despite calls to engage the wider family22 23and to use mass media.24 Mass media is a promising approach for behaviour change, as demonstrated by a recent randomised controlled trial (RCT) of a radio campaign that found a media campaign could be cost-effective mechanisms to deliver a health-related behaviour change intervention,25–27 but such an approach has not been tested for ECD.25–27 Radio can be an effective way to reach a large rural audience in countries such as Burkina Faso, where around 60% of rural women listen regularly to the radio.27 The qualitative results reported here are from the initial stages of a mixed-methods research programme designed to explore the feasibility of promoting ECD-related parenting behaviours in a radio campaign and to support the development of an intervention that could potentially be tested in an RCT. The aim of this formative qualitative study was to gain insight into current practices, perceptions, motivators and obstacles to ECD-related practices and behaviours among caregivers of children aged 0–2 years in rural Burkina Faso.


Study design

A main objective of this qualitative research was to understand the receptivity of local populations to anticipated intervention messages, behaviours and approaches. Therefore, we performed focus group discussions (FGD) to see how participants perceived and discussed ECD principles and behaviours and their interactions within the context and norms of a group setting. The FGDs were based on an adapted Trials of Improved Practices (TIPS) methodology.28 The TIPS approach is a formative research technique, usually employed during the intervention development phase of programmes, which focuses on behaviour and what people do, rather than on knowledge and beliefs, and usually involves multiple visits to study participants.28 We adapted the TIPS methodology to include two rounds of FGDs, with the same participants each attending two focus groups (FG) held 1 week apart. During the first round, facilitators asked participants about caregiving norms for children aged 0–2 years in their local communities. They then briefly introduced the general principles of ECD and benefits of engaging in ECD activities for children aged 0–2 years and offered examples of three specific ECD activities: deliberate and interactive ways of talking with, playing with and praising children. We also played a radio sketch delivered in the local language (Mouré), developed by the local research team. The sketch used a dramatised short story format (as used in previous campaigns)29 30 and portrayed a mother engaged in ECD activities with her child (playing with, naming and counting the child’s toes) interspersed with the mother chatting to a neighbour about the activities.

At the end of the first round of FGDs all participants were asked to consciously try out one or more of the behaviours that had been presented and discussed in the FG—interactive ways of talking, playing with and/or praising their children—over the course of the next week and to give feedback on their experiences in a follow-up FGD 1 week later. To reduce social desirability bias in participant responses, facilitators repeatedly emphasised that participants were not being ‘tested’ on their ability to implement behaviours; rather, researchers were interested in participants’ perceptions of the practices.

Data were collected in May 2018 by a Burkinabe team of male and female researchers, each with between 6 and 10 years of qualitative research experience, and all with a minimum of a bachelor’s degree (SK, CK, PSL, KHL), headed by a masters-level Burkinabe qualitative researcher with 15 years of experience (MB). An international, PhD-level qualitative researcher fluent in French with 6 years of qualitative research experience in West (MD) supervised fieldwork, and provided the team with 6 days of study-specific training prior to data collection. This included all researchers leading or observing a pilot FG using the study FG guide. FGDs were conducted in Mouré by a facilitator (male for fathers, female for mothers/grandmothers) and a note taker. Facilitators explained their credentials and reasons for conducting the study to the participants before beginning FGDs. Researchers used a semistructured guide piloted with four FGs and refined before data collection (online supplementary file 1). The guide included statements about relevant ECD activities modified from training materials used in an ECD programme in Guatemala,31 32 which were read to participants for feedback (see FG guide, online supplementary file 1).

Supplemental material

Location and participant selection

FGDs were conducted in two villages, located 17 km apart, in the Plateau-Central Region of the country. The villages were chosen based on their proximity to Ouagadougou, already-existing contacts in the villages and adequate numbers of target populations residing in the villages. The villages were without electricity and running water and roads to essential services (health, education, commerce) were undeveloped. The main income-generating activity was agriculture.

Participants were selected purposively with assistance from village leaders, who were asked to identify a diverse group of caretakers who were 18 years or older, namely mothers, fathers and grandmothers of children aged 0–2 years, without any specification for religion or ethnicity. Grandmothers living in the same households as their grandchildren were included as primary caretakers based on evidence that they are key influencers in child raising.33 The few participants invited by community leaders with children aged over 2 years were not turned away from first-round FGDs. However, participants with much older children were not asked back for second-round FGDs. Participants were asked to practise the activities with their youngest child, but discussion suggested that in some cases participants additionally practised these activities with older children. FGDs lasted 40–150 min and took place in neutral locations with reasonable privacy. Participants were interviewed in separate FGs of six to seven people, stratified by participant type: mothers, fathers and grandmothers. We conducted a total of 12 FGDs with 41 unique participants (75 interactions total between the two rounds of data collection) (table 1). The number of FGDs was determined by logistical constraints and suggestions from methodological literature exploring qualitative data saturation.34 In the Results section, first-round FGDs are labelled as ‘A’ and second-round discussions, after participants tried ECD activities in their homes for 1 week, as ‘B’.

Table 1

Participant demographic characteristics

Data analysis

FGDs were audio recorded and translated and transcribed into French by researchers and external contractors fluent in Mouré. Researchers (MD, MB, SK, CK, PSL, KHL) first established ‘sensitizing concepts’35 from original research questions to guide the first round of inductive analysis. Next, they established a first set of descriptive and interpretive themes by analysing three FGDs. The principal analysts (MB, SK) then applied thematic content analysis to all FGDs, adjusting, combining and adding themes as inductive analysis progressed (online supplementary file 2). They did this by manually coding participant responses from the transcripts and grouping them thematically. MD independently analysed half of the FGDs for analytical triangulation using Atlas.ti software36; the analysis team discussed any differences in interpretation until consensus was reached. MD translated quotes for this article from French to English.

Supplemental material


We received informed written consent from all participants and verbal consent to audio record discussions before the start of each FGD. Participants received salt and sugar as compensation for their time. There was minimal loss to follow-up between research visits except for a group of four fathers who were called away to an urgent village matter.


Analysis revealed a number of descriptive and interpretive themes allowing researchers to evaluate the feasibility of introducing a scaled-up ECD informational intervention in this and similar contexts.

Young children’s social context

It was widely agreed that children’s primary caregiver for the first years of life is their mother with many participants citing breast feeding as the obvious facilitating factor. However, each member of the family plays their own role in childcare and rearing. Mothers tended to acknowledge yet minimise the role of fathers in direct care; fathers themselves similarly defined their ‘important’ role in raising young children in terms of provision of material needs such as food and medical care. Both women and men affirmed that though men’s time spent with children is limited, men do sometimes help with childcare ‘when they have time’ usually in the evenings so that women can ‘attend to their work.’ Many fathers desired to spend more time engaging with children.

The clear separation of childrearing tasks between mothers and fathers—caregiving versus material provision—related to local gendered divisions of household roles and labour. Mothers’ sociocultural duties as wives physically tied them to the house; fathers’ income-generating activities, however, often kept them away from the house, limiting their time and ability to perform direct childcare tasks.

Participants widely acknowledged that caregivers change with a child’s age. For example, when a child ‘no longer has to be carried on a [mother’s] back’ (generally agreed to be around 1 year) they would spend more time away from their mother either with a grandmother, cowife, older siblings or on occasion taken by fathers to seek medical care or go to the shops. Grandmothers in particular played a central role in caregiving, watching children if mothers had to leave the house and showing first-time mothers how to care for young babies. Other participants mentioned that ‘grandparents’ in general took on caregiving duties suggesting that grandfathers are also sometimes implicated. Table 2 contains illustrative quotes related to caregivers and how these change over time.

Table 2

Young children’s social context

Existing ECD practices

During the first FGDs all participants described some talking with, playing with and praising their young children. Some participants recounted stimulating, interactive activities with children, which exhibited ECD principles of active interaction (subtheme A in table 3). However, our analysis and participants’ own reflections during the second round of FGDs confirmed that most caregiver activities with young children prior to the first TIPS sessions were predominantly instructive, and not necessarily involving responsive interaction and stimulation. For example, participants described giving advice or reprimanding children, rather than interacting/conversing with their child (subtheme B in table 3).

Table 3

Existing ECD practices

Receptivity to and perceptions of ECD activities

All participants expressed an overall positive perception of ECD principles and activities both before and after they were asked to try them at home, receptive to the fact that engaging in these activities would benefit their children. Participants often used observations and shared analogies from their own experiences to explain their perceptions of the process and effects of early learning. In this way, participants validated the positive benefits of ECD cited by facilitators by relating new ECD information shared in FGDs to their own lives (subthemes A and B in table 4).

Table 4

Receptivity to and perceptions of ECD activities

While many caregivers stated that children younger than 2 years ‘forget’ what they learn, caregivers also saw the value of repeating activities with children, so they would adopt good ‘habits’ and learn. After facilitators probed this seeming contradiction, participants consistently explained that ECD activities ‘awaken’ a young child’s spirit or intellect. Therefore, even if young children ‘forget’ the specifics of early interactions, these repetitive interactions create necessary foundations on which the child will continue to learn and grow (subtheme C in table 4).

Despite their general understanding of the value of ECD best practices, throughout discussions most participants continued to link the benefits of ECD activities to learning through instruction—specific motor or language skills (walking, learning particular words), sociocultural/moral values (how to be a good person) and/or tangible success at school. For example, while some participants acknowledged that praise can generally encourage and instil confidence in children, most participants perceived praising (and reprimanding) children as serving to demonstrate the difference between positive and negative behaviours, especially related to completing household tasks or learning a practical skill (subtheme D in table 4).

Motivations to engage in ECD activities

All participants communicated strong motivations to engage with their children through different ECD activities both during and after our study (subtheme A in table 5). Most frequently parents referenced love for their children, their desire to make their children happy, the positive reactions of children when they implemented activities and to help their children ‘succeed’ in life as motivations. ‘Success’ was usually related to intellect-enhancing competencies for school achievement and socialisation to become an upstanding community member and citizen. Though participants reported varying degrees and frequencies of interactions with their children between the FGD rounds, no participant reported not trying/being able to implement any ECD activities during the study.

Table 5

Motivations and obstacles to engaging in ECD activities

Many participants, notably many fathers, reported noticeable changes take place in both themselves and their children, including young, non-verbal children, after engaging in ECD activities over even a short period. But, for some this increased engagement was not discussed in child development terms but instead for the amusement of the child. For example, in first-round FGDs some fathers initially noted that giving children store-bought games makes them happy and occupies them while their mothers are trying to work. This view echoed perceptions of ECD activities as directing the child’s behaviour rather than responsive interaction. However, in second-round FGDs, participants, especially fathers, mentioned more intentional engaged play with children such as playing ball with their child. While describing engaged play, these interactions were still framed as amusing the child rather than as an opportunity to stimulate the child and help them learn.

Participants highlighted that external validation of ECD messages—in this case the research team’s visits to the villages—added to their understanding of the importance of ECD activities for their children thereby motivating them to continue engaging with children in future. Finally, while acknowledging that some people may see ECD activities as strange, such as talking to a child who cannot yet talk, those few opinions would not deter caregivers who were aware of the benefits of ECD. In fact, participants said they would serve as examples to others in the community to adopt ECD activities.

Obstacles to engaging in ECD activities

The most widely mentioned obstacle to engaging in ECD activities was limited time. Daily responsibilities generally kept caregivers occupied during the day and tired them by night. Gendered divisions of labour necessarily put the responsibility of direct interactions of childrearing on women. Gendered strains on time are less discriminatory during the rainy season however when most able community members, including grandparents, are heavily engaged in agricultural activities and have very limited time. Some fathers noted that the positive or negative ‘understanding’ or ‘harmony’ between spouses could also affect both parents’ and children’s desire to play. Another potential obstacle was some participants’ belief in the innate intelligence or predestination of children: certain children are born smart or with a ‘good’ character and others not which could impact their ability to succeed in ECD activities and/or already determined ‘who they would become’ in future. Multiple participants reported that they simply did not know about the benefits of engaging with young children in intentional ways before FGDs, which suggests that educational campaigns could be effective in increasing the practice of ECD activities with young children. The only specific ECD activity some participants approached with caution was praising a child: too much praise, especially in front of other children, could have a negative effect. Illustrative quotes related to these obstacles are in subtheme B in table 5.


This study qualitatively investigated current ECD practices in Burkina Faso, focusing on praising, talking to and playing with children from birth through the first 2 years. The purpose was to gain insight into how to promote stimulating caregiver–child interactions in a mass media campaign. Findings are discussed below in light of other available evidence on parenting and childcare practices in SSA.

Existing research suggests that the instructive caregiver–child communication that was reported is also common in other settings. A qualitative study of carers of children aged 0–2 years in Malawi similarly saw that caregiver–child communications tended to be one way and that participants were unfamiliar with language surrounding the concept of responsiveness.37 Consistent with our findings, caregivers in Malawi lacked knowledge about the importance of talking to children from an early age and of responsive interactions, both verbal and non-verbal, with babies.37 Harkness and colleagues also found that in both West and East Africa parents typically do not have extended conversations with their young children and that, as we found in Burkina Faso, interactions tend to be directive and concerned with training children to carry out tasks.9 10

Gladstone’s study revealed that parent-initiated play in Malawi tended to be physical—although it could also include singing and telling stories—and parents saw these physical activities as a way of making their child stronger and healthier.37 This emphasis on physical play is consistent with studies from elsewhere in West Africa showing that parents tend to place importance on children achieving physical milestones in the early years.9 Our findings did not suggest such a strong emphasis on physical play, but it is possible that our discussion guide tended to emphasise play with toys and other objects over physical play. Our findings indicated that play was generally seen as a normal and important part of childhood. While some participants engaged in interactive, stimulating play activities it appeared more typical in our setting for busy parents to use playthings as a way of occupying their children and not as an opportunity to interact with or stimulate their children. Grandmothers seem to have more time away from household/income-generating tasks during the day than mothers or fathers, and mentioned imaginative games from ‘the past’ they would play with children. Grandmothers could, therefore, be a valuable resource for incorporating more interactive, low-cost play into children’s days.

We found that caregivers do already praise children, particularly for tasks well done. However, participants expressed some reservations about the benefits of praising children, especially in front of other children. In addition, fathers tended to be seen as having a disciplinary role. This could reflect local cultural values and norms that attach importance to socialisation, cooperation and acceptance of hierarchy.11 These concepts contrast with the Western cultural emphasis on individualism and self-actualisation, where a child’s cognitive achievements may be considered praiseworthy, suggesting that while our participants were open to the benefits of praising their children, ECD messages around praise should be tailored to local cultural values.38

Lack of time, particularly for fathers but also affecting other caregivers, especially during seasonal agricultural work, was a barrier to engaging in some ECD activities. This was similarly found to be a barrier in Malawi.37 Fathers in many West African cultures are not traditionally involved in routine care of children, particularly during infancy.39 Yet, similar to findings elsewhere, our results suggested that fathers considered that they had an important co-parenting role,40 were willing to engage more with their children41 and found it personally rewarding when they did so. Fathers may have more discretionary time than mothers at the end of the working day when they may ‘watch’ the child while the mother is engaged in domestic tasks. Since greater paternal involvement in stimulating activities is associated with improved child development in LMICs,42 a greater focus on fathers’ engagement in stimulating activities with their children might be effective in introducing more ECD activities into the child’s day and have a positive effect on overall development. However, although not mentioned by our participants, in parts of West Africa cultural taboos restrict fathers’ relationships with their children until they start walking or talking.39 Messages that both emphasise the importance of ECD activities as well as highlight the benefits of a father’s particular role in developing children’s potential might be empowering for them to engage more with their children and at younger ages. Offering specific suggestions for ECD activities could similarly give all caregivers ‘permission’ to engage in a wider range of stimulating interactions.

When introduced to more interactive ways of talking to and playing with their children as part of the TIPS component of the research, the participants’ initial responses suggested that they did not consider these activities as materially different from things they already did. However, after engaging in these activities at home for a week with more intention, the participants themselves noticed positive differences in their children’s behaviour and in interactions with their children. Finding ways to convey the concept of responsive interaction, particularly with preverbal children, may be challenging. Yet the immediate positive feedback that parents experienced with their children when they engaged in these activities is an important motivator. Additionally, barriers to the adoption of some behaviours that are important for children’s cognitive development can potentially be addressed by framing messages in the context of activities that the child will need to undertake in later life. For example, in rural Senegal where children need to learn to ‘run errands’, one early intervention project found that mothers could only be persuaded to name objects to their young babies by putting this in the context of errands.9 Our participants also recounted interactions with children related to everyday tasks such as fetching water and agricultural work.

Overall, our findings suggested that participants were receptive to information on child development and advice on activities to help their children to develop, and they responded positively to the ‘vignette’ format of the radio spots used to stimulate discussion.

Strengths and limitations

This is one of the few studies to provide insights into ECD practices of rural mothers, fathers and grandparents in a low-income SSA country. Our use of an adapted version of the TIPS methodology was a particular strength as it allowed for a more profound exploration of the kinds of improved practices that the population will be the most motivated and willing to adopt. Although our specific conclusions are limited to the contexts in which we worked, the congruence of our findings with those of similar studies in SSA suggests that our findings may be more widely applicable. We focused on a limited range of activities because available evidence indicated that activities such as reading, counting and storytelling with young children were not common among our population (and illiteracy is high).14 However, future research should incorporate a wider range of activities and positive disciplinary practices and could also benefit from discussions with other caregivers such as grandfathers and siblings. It is possible that findings were influenced by social desirability bias, that is, that participants responded to information and questions about ECD in a way they thought would be pleasing to researchers. However, as experienced researchers, our FGD facilitators took specific steps throughout data collection to limit desirability bias and the responses, enthusiasm and examples from participants appeared genuine.


Interactive, purposefully stimulating activities with young children do not appear to be widespread in rural families in Burkina Faso, but caregivers are receptive to information about the importance of early stimulation for children’s development. There do not appear to be major barriers for carers to incorporate ECD activities into their daily routines. ECD messages should be carefully tailored to the local sociocultural context and should promote activities that fit well into the realities of rural families’ lives. Fathers and grandmothers could potentially play a greater role in engaging in developmental activities with young children and expressed motivation to do so.

It appears to be possible to convey meaningful information about ECD practices using short, instructional statements combined with 1 min entertaining radio spots portraying ECD activities suggesting that further exploration of mass media as a mechanism to deliver ECD messages may be merited.


We thank Gretchen Domek for providing materials from the Niños Sanos: Early Childhood Health and Development Program in Guatemala, and Odette Ky-Zerbo for her guidance on practical aspect of using the TIPS methodology.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 41.
  42. 42.


  • Handling editor Stephanie M Topp

  • Contributors JH, TS and ZH conceived the project. JH and TS designed the methodology and draft focus group guides, and MD refined them. CK, SK, PSL and KHL carried out the focus groups and MD, MB and SK carried out the data analysis with input from the other researchers involved. Focus groups and analysis were supervised by MD. JH drafted the manuscript introduction and discussion and MD drafted the methods and results, with input on all sections from TS and ZH. All authors provided input to the interpretation of findings and approved the final version of the manuscript.

  • Funding The study was funded by Dubai Cares.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Ethics approval This study received ethical approval from the Ethical Committee for Health Research in the Burkina Faso Ministry of Health in September 2017 (reference number 2017-9-134).

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement Data sharing requests should be addressed to the corresponding author.