Introduction Intermittent preventive treatment in pregnancy with sulphadoxine pyrimethamine (IPTp) is a key malaria prevention strategy in sub-Saharan African countries. We conducted an anthropological study as part of a project aiming to evaluate a community-based approach to the delivery of IPTp (C-IPTp) through community health workers (CHWs) in four countries (the Democratic Republic of Congo, Madagascar, Mozambique and Nigeria), to understand the social context in order to identify key factors that could influence C-IPTp acceptability. Methods A total of 796 in-depth interviews and 265 focus group discussions were undertaken between 2018 and 2021 in the four countries with pregnant women, women of reproductive age, traditional and facility-based healthcare providers, community leaders, and relatives of pregnant women. These were combined with direct observations (388) including both community-based and facility-based IPTp delivery. Grounded theory guided the overall study design and data collection, and data were analysed following a combination of content and thematic analysis. Results A series of key factors were found to influence acceptability, delivery and uptake of C-IPTp in project countries. Cross-cutting findings include the alignment of the strategy with existing social norms surrounding pregnancy and maternal health-seeking practices, the active involvement of influential and trusted actors in implementation activities, existing and sustained trust in CHWs, the influence of husbands and other relatives in pregnant women’s care-seeking decision-making, the working conditions of CHWs, pregnant women’s perceptions of SP for IPTp and persistent barriers to facility-based antenatal care access. Conclusions The findings provide evidence on the reported acceptability of C-IPTp among a wide range of actors, as well as the barriers and facilitators for delivery and uptake of the intervention. Overall, C-IPTp was accepted by the targeted communities, supporting the public health value of community-based interventions, although the barriers identified should be examined if large-scale implementation of the intervention is considered.
The study design followed an inductive approach, based on grounded theory,21 whereby no theoretical framework was used to guide the research or inform its design. Instead, emphasis was placed on generating theory through the interpretation of the data collected, thus grounding theory in data. Research questions were reformulated throughout the research process, including data collection and analysis, in a continuous and iterative manner. The study design was longitudinal, with a total of four annual data collection phases starting before project implementation in 2018 and through the last full year of implementation in 2021. A total of four data collection phases were conducted throughout the 4 years, one per year, each one generating sets of data that informed the data collection approach to follow during the subsequent phases. Although the overall approach that guided the research was inductive, the study design was founded on the notion that acceptability is ‘a multifaceted construct’.22 Furthermore, feasibility is understood in more structural systemic terms, in reference to the extent to which an intervention can be carried out,23 thus, ‘delivered’ and ‘taken-up’. Within the context of this study, it has been considered that acceptability, as a multifaceted construct, also addresses certain aspects of feasibility, which are captured in the study of the delivery and uptake of C-IPTp. In addition, the findings from the exploratory phase of data collection (ie, the first phase before implementation) led to the introduction of two conceptual frameworks that further guided the design to better address the study objectives, namely the barriers and opportunities analysis24 and the socioecological model.25 Data collected during the exploratory phase were analysed separately by the in-country research teams and then pooled by the research team at the coordinating institution. The pooled analysis consisted of identifying cross-cutting themes that then informed the choice of conceptual frameworks and their subsequent adaptation to the study aims and context. The decision to introduce the socioecological model was informed by the emphasis placed across study sites on the importance of the social context wherein pregnant women are embedded in influencing their health-seeking patterns, as well as the realisation of the complexity of the decision-making processes that underpin these patterns. The decision to introduce the barriers and opportunities analysis was informed by the multiple thematic dimensions found to be relevant to the anticipated acceptability of C-IPTp, as it was foreseen prior to its implementation. This required the introduction of a framework that allowed us to simultaneously address multiple themes and that was adaptable to the research interests delineated in the exploratory phase and to themes emerging throughout subsequent phases (ie, sensitive to the iterative nature of the research process). The use of both frameworks facilitated the examination of C-IPTp delivery, uptake and acceptability from a multidimensional perspective, and for understanding pregnant women’s health-seeking pathway as embedded in a complex and layered social system. The study was carried out in the three areas per each of the four countries where the TIPTOP project was being implemented: Kenge, Bulungu and Kunda health zones (HZ) in the DRC; Mananjary, Vohipeno and Toliary II districts in Madagascar; Nhamatanda, Meconta and Murrupula districts in Mozambique; and Ohaukwu, Akure South and Bosso local government areas (LGA) in Nigeria (see figure 1). In all countries, malaria is one of the leading causes of maternal and child morbidity and mortality, while the CHW models vary across countries. Map of study sites. C-IPTp was implemented similarly across the four TIPTOP project countries, with CHWs trained to identify pregnant women in the community, screen them for IPTp eligibility, provide IPTp to eligible pregnant women, and refer all women to ANC. The latter ‘referral to ANC’ was a key component to ensure all pregnant women received comprehensive care. Based on national Ministry of Health (MoH) recommendations and guidelines on CHW roles, notable adaptations were made to the strategy, such as the administration of the first dose of IPTp-SP. In the DRC and Nigeria, CHWs were authorised to administer all doses of IPTp-SP, including the first one. Importantly, in these countries, CHW did not physically assess pregnant women for gestational age and only gave IPTp1 if the woman reported having experienced foetal movement. In Madagascar and Mozambique, CHWs were not authorised to administer the first dose of IPTp-SP but were able to administer subsequent doses. When identifying pregnant women that had not received any dose of IPTp-SP, CHWs were required to refer them to the health facility to receive their first dose during their routine ANC visit. Although the C-IPTp strategy incorporated existing CHW cadres in all countries, an additional community-based role was introduced in Mozambique to support existing CHWs, given their relatively low ratio to pregnant women. This new category of volunteer health worker—called lay community counsellor (LCC)—was introduced to support the C-IPTp approach. LCCs were trained to perform CHW tasks related to C-IPTp, with the exception of administering SP and the addition of referring pregnant women to CHWs as well as to ANC. These tasks included the identification and mapping of pregnant women, conducting follow-up home visits, referring pregnant women to the health facility for continued ANC, and liaising between pregnant women and CHWs. The study involved a total of 3235 participants across the four countries (see table 1) over the 4-year duration of the project. Participants included pregnant women, women of reproductive age (15–45 years), ANC health providers, CHWs, traditional healers, health management representatives, traditional birth attendants (TBAs), community leaders and relatives of pregnant women. The identification of participants was based on non-probabilistic methods, following purposive sampling combined with snowballing. Complementary sampling strategies included opportunistic sampling and, to a lesser extent, convenience sampling. Although field entry strategies varied between countries, all research teams benefited from the support of local leaders, regional health authorities and healthcare workers during the sampling process. Informants meeting the aforementioned target group criteria were then invited to participate in research activities. Those willing to participate and provide written informed consent were enrolled. A local social scientist from each in-country partner institution coordinated research teams were composed of local guides, interviewers, facilitators, translators and transcribers. Research teams were trained to employ various sampling approaches and conduct data collection. Participants recruited per site and phase (IDIs/FGDs) *Although TIPTOP was piloted in three areas per country, in the early phases of the study research activities were planned to be conducted only in two areas per country. In order to strengthen the scientific value of the results, the third project area was included in the study design after the first phase of data collection. †No fieldwork was carried out in Nhamatanda in 2019 due to cyclone Idai. DRC, Democratic Republic of Congo; FGDs, focus group discussions; IDIs, in-depth interviews; NA, not available; TIPTOP, ransforming IPTp for Optimal Pregnancy. The data collection techniques employed were drawn from qualitative methodologies and ethnography.26 These consisted of in-depth interviews (IDIs), focus group discussions (FGD), informal conversations and direct observations. IDIs and FGDs were conducted using semi-structured question guides to capture participants’ knowledge of maternal health, malaria prevention in pregnancy and IPTp, as well as perceptions of C-IPTp and potential barriers and facilitators that could influence the acceptability of the intervention. Template guides in English were adapted to meet local context specificities and translated into the relevant language of local use by in-country research teams. Researchers performed live translations into other local languages, where appropriate. FGDs and IDIs were conducted in the language preferred by the participant, digitally recorded and transcribed verbatim. As needed, transcriptions were then translated into English in Nigeria, Portuguese in Mozambique and French in Madagascar and the DRC. Researchers conducted non-participant observations of the following activities: the delivery of ANC services, both at health facilities and in TBA’s homes; the administration of IPTp-SP by CHWs, whether delivered in pregnant women’s homes or in CHWs’ homes; communication activities for IPTp and maternal health promotion, carried out both in health facilities and communities; and the day-to-day activities of pregnant women. Observations were recorded in field notes, following observation guides to ensure notes were comprehensive and captured several key aspects: description of the physical space, people involved, performance of the activity, timing, dynamics between participants and the general mood. Informal conversations carried out spontaneously with key informants encountered during fieldwork activities were also captured in field-notes. In total, 265 FGDs, 796 IDIs and 388 observations were performed throughout the study phases and across sites (see table 2). Data collection activities performed per tool, phase and site *No FGDs were conducted in the Madagascar sites during the first/baseline phase of data collection. †No fieldwork was carried out in Nhamatanda in 2019 due to cyclone Idai. DRC, Democratic Republic of Congo; FGD, focus group discussions; IDIs, in-depth interviews; NA, not available. Given the iterative nature of the grounded theory approach, data collection and analysis progressed simultaneously, allowing the tools to be modified in light of themes emerging from ongoing analysis. The data collected was interpreted through a combination of content and thematic analysis.27 Content analysis involved categorising data based on the themes predetermined by the two conceptual frameworks introduced after the initial exploratory phase. Thematic analysis consisted of identifying emerging themes relevant to the study aims that subsequently contributed to further shaping the research process. Research teams coded the data manually. The social science team at the coordinating institution then compared the analyses from the different sites in order to elucidate commonalities and divergences. During this process, there was a constant interaction between research teams to resolve key interpretation issues. More information about the collaboration between research institutions and their involvement in the research process is presented in the authors reflexivity statement (see online supplemental appendix 1). bmjgh-2022-010079supp001.pdf Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
N/A