Development of Context-Appropriate Recruitment and Retention Approaches for Pregnant Women in the SAFIRE Clinical Trial Assessing Safety and Efficacy of Artemisinin-Combination Therapies: A Formative Research Protocol

Abstract

Pregnant women, particularly those in the first trimester, have been historically excluded from clinical trials, resulting in substantial evidence gaps for the prevention and treatment of malaria in early pregnancy. Despite increasing calls from global health and regulatory bodies to include pregnant women in research, ethical, socio-cultural, and contextual challenges continue to impede their meaningful participation. Formative research is essential to understanding these challenges and informing trial procedures that are ethically robust and contextually appropriate. This study aims to explore stakeholder perceptions, beliefs, and decision-making processes related to early pregnancy and participation in clinical trials, to inform the development of culturally sensitive recruitment, retention, and community engagement strategies for the multi-country SAFIRE clinical trial on the safety and efficacy of artemisinin-combination therapies. Qualitative methods include in-depth interviews and focus group discussions with purposively sampled women of childbearing age, pregnant women, health providers and managers, community leaders, and other key health decision-makers. Data will be collected across five malaria-endemic countries in sub-Saharan Africa. A reflexive thematic analysis approach will be used, combining inductive and deductive coding within a collaboratively developed coding framework. Analysis will be informed by a Theory of Change to examine how identified barriers and facilitators may influence pathways to successful recruitment and retention. This protocol demonstrates how formative research can be systematically integrated into clinical trials to support the ethical inclusion of pregnant women.

Introduction

During pregnancy, women in malaria-endemic regions become more susceptible to malaria infections, often with significant consequences for both mother and baby (Desai et al., 2007). Findings from a meta-analysis attributed 20% of all stillbirths in sub-Saharan Africa to infections with Plasmodium falciparum (Moore et al., 2017). The World Health Organization (WHO) recommends chemoprevention with intermittent preventive treatment (IPTp) with sulfadoxine-pyrimethamine (SP) beginning in the second trimester to prevent malaria in pregnancy, but at present no safe drug options for chemoprevention in the first trimester of pregnancy have been identified. Until 2022, the artemisinin-based combination therapy (ACT) artemether-lumefantrine (AL) had been recommended for treatment of uncomplicated malaria only in the second and third trimester, before the WHO expanded its guidance to recommend AL as the drug of choice in the first trimester (World Health Organization, 2023). However, there was insufficient safety data to extend the recommendation to other ACTs. Whilst this treatment guideline update was welcomed, it highlights the dearth of evidence on the safety and efficacy of newer antimalarials for prevention and treatment options in early pregnancy. With the range of effective ACTs expanding, there are calls for pregnant women to be prioritised in the research agenda to ensure access to safe, well-tolerated, and effective drugs. A recent comment by El Gaaloul et al. called for new approaches within the malaria research community and, as drug development for new antimalarials gathers pace, stressed that the needs of pregnant women should be prioritised so they too can benefit from well tolerated, efficacious antimalarial drugs throughout pregnancy and beyond (El Gaaloul et al., 2022).

Pregnant women have been routinely and systematically excluded from clinical trials due to a combination of regulatory, ethical, and physiological factors (Blehar et al., 2013; Malhame et al., 2020). Though well-intentioned, this exclusion has led to some significant consequences for pregnant and lactating women that arise from the paucity of pregnancy-specific drug data including the delay or withholding of treatments due to concerns of harm to the foetus or the administration of inappropriate treatment from therapies that are neither safe nor effective (Malhame et al., 2020). Specifically, these policies have contributed to around 80% of pregnant women being prescribed medications with little or no available pregnancy safety data (Ayad & Costantine, 2015). Few clinical trials have included pregnant women in the first trimester, and none involving antimalarial treatments (Dellicour et al., 2017; Saito et al., 2023). Trials on the safety and efficacy of ACTs, specifically, have excluded women in the first trimester due to the heightened risk of embryotoxicity reported in animal studies, particularly between 6 and 12 weeks of gestation (Gomes et al., 2016). Currently there is a paradigm shift within maternal health research. Key regulatory bodies, including the WHO and the Task Force on Research Specific to Pregnant Women and Lactating Women (PRGLAC) and the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) ¹ are calling for a research agenda that includes pregnant and lactating women. To facilitate the inclusion of pregnant women in clinical trials, researchers need to consider procedures that are contextually appropriate and acceptable to pregnant women and their wider community, from enrolment through delivery to post-partum follow up.

Study Justification

To optimise the recruitment and retention of women in early pregnancy, researchers need to understand the unique contexts within which pregnant women make decisions about joining, and remaining in, clinical trials. Trial processes and informed consent procedures must be acceptable to women of childbearing age (WOCBA), pregnant women, and their wider community. Pregnancy practices and beliefs differ widely by country and context. As such, to optimise recruitment and retention of pregnant women in clinical trials it is important that researchers understand how different components of the trial will influence the willingness and ability of women to participate and use this knowledge to adapt the recruitment strategy where possible. At present, there is a dearth of evidence on the perceptions and experiences of pregnant women in clinical trials from sub-Saharan Africa. A study in Ghana exploring the willingness of pregnant women to participate in an antimalarial drug trial found that awareness of the disease, trust in healthcare providers, and acceptance of the study drug for prevention positively influenced participation (Osarfo et al., 2021). Other factors, such as number and quantity of required blood samples, suspicion about the real reason for the blood samples, and testing of placental tissue post-delivery influenced women’s willingness to participate (Compaoré et al., 2018; Osarfo et al., 2021). Communities with previous research experience may harbour mistrust of researchers or might conflate research with service provision – with expectations of diagnostics and treatments (Martínez Pérez et al., 2018). Community engagement is a pillar of successful clinical trials, helping to establish sustainable relationships between the researchers and the trial community (Nyika et al., 2010). Initially, communication with community leaders acts as a gateway into the community and creates a trusted two-way communication line where the researchers can share important information concerning the aims and specific details of the trial with key community actors, leading to enhanced transparency over the duration of the trial and beyond (Mtove et al., 2018). Building a positive relationship with communities at the trial sites can foster increased awareness among the research team of contextual sensitivities as well as a deeper understanding of important cultural practices, beliefs, and social networks, which can be used to inform the trial procedures and enhance the acceptability of the trial within the community. This is particularly important when seeking to recruit and retain pregnant women as they may experience reduced decision-making autonomy and specific cultural restrictions around pregnancy, particularly in the first trimester, and birth.

Conceptual Framework

In recognition of the challenges associated with the inclusion of pregnant women in clinical trials and the gaps in evidence on how to support their recruitment and retention, Medicines for Malaria Venture (MMV) undertook a study consisting of a literature search and interviews with key informants to identify potential strategies for optimising recruitment and retention of pregnant women in clinical trials. The evidence extracted from the literature review identified a wide range of factors that influence pregnant women’s decisions to participate in clinical trials, although there was scant evidence from sub-Saharan Africa (unpublished data). These factors were collapsed into five broad categories including personal factors, trial characteristics/activities, interpersonal factors, healthcare provider/researcher factors, and contextual factors, and used to develop a logic model to theorise the pathway from activities to achieving the desired recruitment and retention outcomes. This Theory of change was adapted for the SAFIRE trial (SAFIRE, PACTR202505891635396) (Figure 1). The pathway outlines a series of activities that would lead to outputs, including that the trial procedures fit within the community context, that recruitment approaches meet local and international expectations, and that community engagement includes the relevant members needed to disseminate information and monitor community feedback. These outputs would then achieve the following proximal outcomes 1) women feel willing and able to join the trial and participate in until completion, 2) the community accepts the trial and supports women to participate, and 3) trial staff provide correct support and advice to women, which would result in the recruitment and retention of women in the trial as per the protocol with minimal loss to follow up. Undertaking formative research was identified as the first activity on the pathway to optimal recruitment and retention of pregnant women in the logic model.

Figure 1.

Theory of change describing the pathway towards optimal recruitment and retention of pregnant women in the SAFIRE clinical trial

Information from the formative research conducted in the SAFIRE trial sites will be used to develop culturally sensitive and appropriate ways to recruit WOCBA and pregnant women, including how to identify women in early pregnancy (first trimester). It will capture main contextual factors to consider in the trial procedures that may influence both the willingness and ability of these women to join the trial and remain until completion. Finally, the formative research will also help identify key community members and inform the development of the community engagement strategy that will be used throughout the clinical trial to optimise the recruitment and retention of WOCBA and pregnant women.

Study Design

Study Objectives

The overall aim of the study is to support the development of culturally appropriate recruitment and retention strategies for the planned SAFIRE trial.

Specific objectives, to:

1. Assess the context, socio-cultural practices and beliefs around early pregnancy as well as collecting experience with and/or perception of previous clinical trials to inform the development of culturally acceptable and understandable communication materials about the SAFIRE trial, including the information sheets for obtaining consent.

2. Identify factors and individuals that may influence WOCBA and pregnant women’s decision-making around participation and retention in the clinical trial to be addressed as part of project procedures and communications.

3. Explore approaches for facilitating the inclusion of WOCBA and pregnant women (in early pregnancy) in the clinical trial, based on the findings from objective 2.

4. Identify key stakeholders (including household decision-makers, community leaders and opinion shapers, health sector actors), from community to national levels, to be included and involved in the community engagement strategy prior to trial implementation.

Methodology

This formative research study will be conducted at the sites of the SAFIRE trial, which aims to assess safety and efficacy of antimalarials for treatment in the first trimester of pregnancy. Prior to commencing trial enrolment procedures, the formative research will gather crucial contextual information from a range of key stakeholder groups to inform the trial procedures and to optimise the recruitment and retention of pregnant women. Preliminary findings will be synthesised into site-specific briefing notes to inform immediate adaptations in recruitment and community engagement strategies. This research will be carried out prior to the initiation of the clinical trial across sites in Burkina Faso, Kenya and Mali, and in two additional countries DRC and Uganda who will initiate the trial once further funding is secured. Timely feedback of key findings to the clinical trial teams in each country will support the development of the community engagement strategy and enable context-appropriate approaches to trial recruitment and retention of women in the first trimester of pregnancy.

Specifically, this research will help to (1) identify key community stakeholders who can support the recruitment and follow up of WOCBA and pregnant women in the clinical trial at each site, (2) understand the factors that influence WOCBA and pregnant women’s decision-making about participating in the SAFIRE trial, and (3) explore what approaches should be used for the community engagement strategy to facilitate women’s recruitment and retention at each trial site. Data will be collected through in-depth interviews (IDIs) and focus group discussions (FGDs) depending on suitability for each topic and the feasibility of gathering respondents together. The perceptions of a range of respondents will be explored, including WOCBA, pregnant women, community members, health managers and providers, including trial staff, to understand what factors could act as barriers or facilitators to women joining and completing the SAFIRE trial. In addition, we will seek respondents’ ideas on community engagement approaches that may enhance their inclusion in the trial.

Sampling

Respondents will be purposively sampled from proposed trial sites in each country to obtain cases that are “information-rich,” meaning inclusion of participants with specific knowledge or experiences relevant to the study (Palinkas et al., 2015). Respondents from the health sector will include facility and/or district/county and sub-county level managers, healthcare providers and community healthcare providers. We will aim to include both male and female respondents with a range of ages and levels of experience. In addition, we may also include clinical trial staff members where possible. Community members will include a range of leaders and opinion shapers, family members, WOCBA (15-49 years), including pregnant women. The inclusion of respondents is based on their ability to share perceptions on the socio-cultural context, as well as potential facilitators and barriers that could impact the recruitment and retention of women in first trimester pregnancy in the SAFIRE trial. Additionally, respondents will be asked to share their perspective on how to facilitate community engagement throughout the trial and provide recommendations on who might influence WOCBA and pregnant women’s decision-making on participating in the trial. The sample should include respondents with a mix of ages, ethnicities, education levels and socio-economic statuses. In addition, for pregnant women we will aim to include a range of parities and stages of gestation.

We will aim for a sample size of approximately 12 health providers/managers, 12-16 community leaders and health decision makers and between 6-10 WOCBA and pregnant women, approximately 30-38 respondents per country. In Burkina Faso and DRC, FGDs are considered an appropriate technique for discussing these topics among WOCBA and pregnant women so they will employ a mix of FGDs and IDIs. Each discussion will include approximately 6-8 participants to facilitate meaningful discussion while maintaining manageability. Sampling will remain flexible and iterative, allowing for the inclusion of additional participants or stakeholder groups should emerging findings indicate gaps in perspective or conceptual understanding.

Settings

In each country, IDIs and FGDs will be conducted in-person by local, trained social scientists. Due to the sensitive nature of the topic, the interviews and discussions with WOCBA and pregnant women, will be conducted by female social scientists. To maintain confidentiality and enhance privacy, respondents will be given the choice about where the interview will take place, including at their home, workplace, or other specified community location. In Burkina Faso and DRC, care will be taken to organise the FGDs in a private setting, either outside under a tree or in a community building.

Data Collection

Topic guides have been developed to capture relevant information from the different respondent groups, including information about 1) the socio-cultural context, 2) beliefs and practices in pregnancy – including when, how and to whom women first disclose their pregnancies, 3) previous experience with research in the community – including clinical trials, 4) important influencers of women’s decision-making, 5) specific components of the trial that could influence women’s ability and/or willingness to join and remain in a trial, 6) what factors specifically would help enable women to join and remain in a trial, and 7) appropriate community engagement activities – including who should be involved and the format of the activities (supplemental material – topic guides). Prior to commencing data collection, these guides will be piloted in the community to verify the contextual suitability, and clarity of the topics and questions for each respondent group. Interviews and focus groups will take around 60 minutes and will be audio-recorded to facilitate transcription and analysis. Field notes will be written by the researchers after the IDI or FGD is completed, and regular debriefing session will be held among the research teams to discuss emerging findings, adjust probing strategies were needed, and discuss saturation.

Data Analysis

IDIs and FGDs will be transcribed verbatim, translated into English where necessary, and coded using NVivo software. When possible, data coding will be carried out by the local research team to ensure meanings are accurately interpreted and important contextual nuances are captured. An initial coding framework (Figure 2), based around the study objectives, has been developed collaboratively and iteratively refined through regular cross-country analytic meetings to facilitate the timely use of key findings to support the development of community engagement strategies and finalise the recruitment approaches. A reflexive thematic analysis approach will be used, employing both inductive (using the coding framework developed a priori) and deductive coding, with emerging themes added to the framework. Discrepancies in coding or interpretation will be addressed through discussion, reflexive dialogue, and, where necessary, re-examination of the data until consensus is reached. Data will be coded by individual country teams but using a shared coding framework to enable both country level and cross-country analyses. Further thematic analyses will be conducted using the Theory of Change (Figure 1), together with data collected as part of the SAFIRE trial, to interrogate the data and determine if and how barriers and facilitators to recruitment and retention identified in this study influenced the pathway from activities to outcomes.

Figure 2.

Coding framework used to inform community engagement and recruitment strategies

Ethical Aspects

This study is sponsored by MMV and has been granted favourable ethical opinion by the Liverpool School of Tropical Medicine Research Ethics Committee. Ethical approval from the local ethical review board has been obtained in each of the five countries. Written consent will be obtained from all participants. As part of the informed consent process, respondents will be provided with information about the study and given time to consider participation. Importantly, WOCBA and pregnant women will be given up to 14 days to consider their participation in the study, ensuring sufficient time to discuss with family members and obtain permission if required. Those who agree to participate will be informed of the approximate length of the interview or discussion and reminded of their right to withdraw their consent at any time.

This study raises several ethical considerations, including discussions of pregnancy risk, medication safety, and past experiences of illness or adverse outcomes. These risks will be mitigated through careful development of study materials, use of trained interviewers, and provision of clear information about voluntary participation and the right to withdraw. Confidentiality is essential and will be strictly maintained, with data to be handled in accordance with applicable data protection regulations. Community sensitisation activities, including engagement with relevant community leaders (e.g. religious, administrative, civil society organisations etc.) will be undertaken prior to initiating data collection to ensure culturally appropriate study conduct and help build trust with participants and the wider community.

Validity and Rigor

Qualitative rigor will be ensured through strategies addressing credibility, dependability, reflexivity, and transparency. To enhance credibility, our sampling strategy includes a diverse set of stakeholders enabling us to capture a range of perspectives across the trial sites. Dependability is supported by systematic documentation of analytic procedures and use of a shared coding framework. Validity is strengthened through standardised training materials for field work across the countries and use of a single set of topic guides, adapted for cultural appropriateness where needed.

Reflexivity is central to the study design and analysis. As researchers exploring perceptions of, and experiences with, participation in clinical trials we need to critically examine our positionality, assumptions, and potential influences on data collection and interpretation (Dodgson, 2019). Mitigation strategies such as community sensitisation prior to field work where we clearly state our objectives and neutrality on the subject matter will be utilised. In addition, IDIs and FGDs will be conducted by local researchers familiar with the cultural context to minimise the differences between researcher and community. Attention will also be given to negative or deviant cases to ensure that minority perspectives are not overlooked in the analytic process. Transparency will be promoted through clear reporting of methods and analytic decisions, enabling readers to assess the trustworthiness and transferability of the findings.

Discussion

Our methodology highlights the importance of embedding formative research into the lifecycle of clinical trials. Thematic analysis will generate key contextual insights that will underpin the development of culturally sensitive strategies to recruit WOCBA and pregnant women, including how to identify women in early pregnancy (first trimester).

Work Progress to Date

Community sensitisation activities with key community actors were conducted at the study sites to appraise them of the formative research study and objectives, and to allow them to ask questions. Across the five countries 194 IDIs and six FGDs (Burkina Faso and DRC only) have been conducted to date with health providers/managers, community leaders and health decision-makers, WOCBA and pregnant women. Some research teams faced challenges. For example, in Burkina Faso limited availability of community participants due to agricultural commitments, and CHWs due to ongoing health campaigns in the area. In addition, heavy rains affected transportation and access. Regular team debriefs during data collection facilitated the identification of emerging themes and key findings, which were synthesised into site-specific briefing notes. Data processing and coding is ongoing.

Early findings from this study have been used to inform the mobilisation, recruitment, and retention strategies in the SAFIRE trial, ensuring that strategies are ethically robust, culturally appropriate, and responsive to local realities. More broadly, the protocol offers a transferable model for integrating qualitative formative research into the life cycle of other clinical trials, particularly for studies seeking to include pregnant women. In doing so, it contributes to the advancement of qualitative methodology and good participatory practices as essential components of equitable and responsible global health research. This research further supports ongoing efforts to generate policy-relevant evidence for malaria treatment in early pregnancy. The findings will be disseminated in both single and cross-country articles published in peer reviewed journals. This will enable the publication of rich, contextual data unique to the individual countries as well as important high-level syntheses that examines the similarities and differences across contexts.

Strengths and Limitations

A key study strength is the range of perspectives included across both respondent groups and country contexts, allowing for a granular, nuanced understanding of contextual influences whilst also facilitating the identification of potential barriers and facilitators to recruitment and retention of women in first trimester pregnancy across varied settings. Several limitations should be noted. Respondents were purposively sampled based on their perceived relevance to recruitment and retention processes; however, our sample includes health managers and providers without previous research experience and those who may not be involved in the trial. In addition, a purposive sampling approach may under-represent more marginalised or less visible perspectives, including women with limited community engagement, lower autonomy – including unmarried pregnant women and adolescents, or prior negative experiences with research or healthcare services. Discussions about research participation, pregnancy, and decision-making may be influenced by social desirability bias, particularly in settings where researchers are perceived as authority figures or linked to health service provision. Power differentials between respondents and interviewers—especially when discussing sensitive topics such as pregnancy disclosure or medical risk—may shape their responses.

Conclusion

This qualitative formative research protocol outlines a rigorous, contextually grounded approach to understanding the socio-cultural, ethical, and relational factors that shape the recruitment and retention of women of childbearing age and pregnant women—particularly those in early pregnancy—in the SAFIRE trial. This approach will be relevant to other clinical trials recruiting women in early pregnancy. By engaging multiple stakeholder groups across diverse malaria-endemic settings, this study is designed to generate nuanced insights into how pregnancy, risk, decision-making, and research participation are understood and negotiated at individual, household, community, and health system levels.

Supplemental Material

Supplemental material - Development of Context-Appropriate Recruitment and Retention Approaches for Pregnant Women in the SAFIRE Clinical Trial Assessing Safety and Efficacy of Artemisinin-Combination Therapies: A Formative Research Protocol

Supplemental material for Development of Context-Appropriate Recruitment and Retention Approaches for Pregnant Women in the SAFIRE Clinical Trial Assessing Safety and Efficacy of Artemisinin-Combination Therapies: A Formative Research Protocol by Jenna Hoyt, Adélaïde Compaoré, Samba Diarra, Freddy Bikioli Bolombo, Susan Nayiga, Serge Henri Zango, Moussa Djimde, Stephanie Dellicour, Japhet Kabalu Tshiongo, Hypolite Muhindo-Mavoko, Abel Kakuru, Simon Kariuki, Hellen Barsosio, Kassoum Kayentao, Innocent Valea, Jenny Hill, Maud Majeres Lugand in International Journal of Qualitative Methods.