What is the feasibility and acceptability of stool sampling and testing for childhood tuberculosis diagnosis? A qualitative study in The Gambia Authors: Sheila Agyeiwaa Owusu, Victory F Edem, Esin Nkereuwem, Madikoi Danso, Martina Boakarie, Sarjo Wassa Koita, Sheriff Kandeh, Musa Jallow, Tijan Baldeh, Marie Gomez, Joanna Busza, Toyin O Togun Published: 2026 ABSTRACT Abstract AbstractObjectiveTo explore the feasibility and acceptability of stool sampling and testing for childhood tuberculosis (TB) diagnosis among adolescents, child caregivers and healthcare workers (HCWs) who participated in a diagnostic accuracy study and among National Leprosy and TB Control Programme (NLTP) officials in The Gambia.DesignThis was a qualitative cross-sectional study that used semi-structured interviews. Braun and Clarke’s thematic analysis framework guided our analysis.SettingThe study was conducted in The Gambia at the TB clinic of the Medical Research Council Unit The Gambia at the London School of Hygiene and Tropical Medicine.ParticipantsWe purposively sampled and interviewed 27 participants, including adolescents and caregivers of children who participated in and provided stool samples for the diagnostic accuracy study of the Childhood TB Research in West Africa (CHIRWA-2), HCWs involved in the CHIRWA-2 study, and NLTP officials responsible for policy and programme implementation in The Gambia.ResultsHCWs and NLTP officials described stool sampling and testing as feasible for TB diagnosis, given the ease of collection and the potential to integrate them into existing workflows. However, availability and sustainable supply of resources are required for successful implementation. Caregivers were willing to provide stool samples for TB diagnosis, although they were surprised by the use of stool for diagnosing TB. Adolescents, on the other hand, felt embarrassed about providing a stool sample. HCWs preferred stool samples over sputum samples because stool sampling was easier and less time-consuming, and further suggested sustained logistics support, adequate community sensitisation and regular incentives for HCWs as necessary for successful implementation.ConclusionParticipants perceived that using stool sampling and testing for TB diagnosis in children is feasible and acceptable in Gambia. Its integration into routine health service delivery requires careful consideration of the available resources.Keywords: Tuberculosis, PAEDIATRICS, QUALITATIVE RESEARCH STRENGTHS AND LIMITATIONS OF THIS STUDY The inclusion of multiple stakeholder groups, adolescents, caregivers, healthcare workers (HCWs) and policy officials provides a comprehensive understanding of feasibility and acceptability across the implementation chain. The inclusion of multiple stakeholder groups, adolescents, caregivers, healthcare workers (HCWs) and policy officials provides a comprehensive understanding of feasibility and acceptability across the implementation chain. The use of participants with direct experience with stool sampling and testing, through their participation in the parent study, ensures that the perspectives are grounded in actual experience rather than hypothetical scenarios. The use of participants with direct experience with stool sampling and testing, through their participation in the parent study, ensures that the perspectives are grounded in actual experience rather than hypothetical scenarios. Limitations include social desirability bias, particularly among HCWs. This was mitigated by emphasising confidentiality at the beginning of the interviews. Limitations include social desirability bias, particularly among HCWs. This was mitigated by emphasising confidentiality at the beginning of the interviews. This study contributes contextually relevant evidence to inform policy decisions in The Gambia and provides substantive insights applicable to other resource-limited settings, particularly regarding stool sampling and testing for tuberculosis. This study contributes contextually relevant evidence to inform policy decisions in The Gambia and provides substantive insights applicable to other resource-limited settings, particularly regarding stool sampling and testing for tuberculosis. Introduction Tuberculosis (TB) is preventable and treatable, yet globally, an estimated 174 000 children and young adolescents under the age of 15 years died from TB in 2024.1 A modelling study estimates that globally, approximately 96% of children who die from TB are not treated.2 Additionally, about 58% of TB cases among children under 5 years of age are missed due to underdiagnosis and underreporting.3 This case detection gap in children is due to non-specific clinical and radiological presentations, difficulty in obtaining good-quality sputum samples and the paucibacillary nature of TB in children.4,6 This is particularly pronounced in resource-limited settings where access to diagnostic tools and trained personnel is limited.7 Currently, the WHO recommends alternative samples such as induced sputum, nasopharyngeal aspirates and stool samples, using low-complexity automated nucleic acid amplification tests and concurrent testing of stool and respiratory samples in children for diagnosing TB.8 Stool sampling and testing for TB offers a promising non-invasive alternative for childhood TB diagnosis. Stool samples are relatively easy to obtain and do not require specialised training or complex processing techniques. However, they may require an additional hospital visit if the child is unable to produce stool on demand.9 A systematic review and meta-analysis by MacLean et al evaluating the diagnostic accuracy of stool testing with the Xpert Ultra assay for childhood pulmonary TB reported a pooled sensitivity of 67% (95% CI 52% to 79%) and a specificity of 99% (96% CI 98% to 99%), indicating its importance as a first-line, non-sputum-based test for microbiological diagnosis.10 However, the successful implementation of any diagnostic tool extends beyond technical performance. Even if a sample has a high yield, its implementation depends on other factors, such as acceptability, feasibility and sustainability, which could be context and health system-specific.11 12 Proctor et al defined acceptability as ‘the perception among stakeholders that a given treatment, service, practice, or innovation is agreeable, palatable, or satisfactory, and feasibility as the extent to which a new treatment or an innovation can be successfully used or carried out within a given agency or setting’.12 Understanding the perspectives and experiences of policymakers and end users, such as healthcare workers (HCWs) and patients, is critical to ensuring adoption and sustainability.13 14 Recent studies have reported varied views on the acceptability and feasibility of using stool samples to diagnose paediatric TB among patients and HCWs across Africa.15,18 It was preferred by caregivers and laboratory staff over respiratory samples in a study conducted in Ethiopia.17 However, challenges with stool collection have been reported in South Africa, particularly among children aged 7–12 years, due to embarrassment around producing stool.16 A study by Lima and colleagues among HCWs in southern Mozambique found that stool sampling and testing were advantageous and easy to perform, but patient acceptability was influenced by prior experience with other sampling techniques, clear information about the procedure, misconceptions and fear of handling stool.19 Although stool sampling and testing show promise in TB diagnostics, addressing barriers at different user levels is necessary for their successful implementation in high-burden settings. In the Gambia, a country in West Africa, the TB incidence rate is 138/100 000 population/year.1 The country has made substantial progress in TB control through its National Leprosy and TB Control Programme (NLTP), yet childhood TB diagnosis remains challenging, with the case notification rate reaching only 6% in 2023, falling short of the 10% target outlined in the 2023 National Strategic Plan for TB.20 In 2024, the country adopted the WHO’s recommendation to use stool samples for diagnosing childhood TB. Despite increasing global interest in stool sampling and testing for childhood TB, there remains a limited understanding of how end users experience and perceive these methods in real-world contexts. Existing studies have largely prioritised laboratory accuracy, workflow robustness and transport conditions, while paying less attention to the perspectives of adolescents, caregivers and frontline health workers, who ultimately determine whether stool collection is feasible in practice.15 17 21 22 This gap is highly consequential for policy and implementation: stool sampling and testing require caregivers to collect and transport samples and hinge on health workers’ willingness to recommend and integrate the method into routine care. Without insights into their concerns, preferences and practical challenges, policies promoting stool sampling and testing risk underutilisation or poor uptake, even when the method is technically sound. Understanding end-user perspectives is therefore essential to inform guidance, design supportive implementation strategies and ensure that the method is both acceptable and scalable within routine TB diagnostic pathways in high-burden settings. This study explored the feasibility and acceptability of stool sampling and testing for the diagnosis of childhood TB among caregivers, adolescents and HCWs who participated in a diagnostic accuracy study, as well as NLTP officials in The Gambia. Methods Study design and setting This study employed an exploratory qualitative design embedded within the CHIRWA-2 study, a multicountry diagnostic accuracy study which evaluated the added value of the Computer-Aided Detection for TB version 7 (CAD4TBv7, Delft Imaging Systems, Waterstraat, Netherlands) and stool testing using the Xpert Ultra assay (‘stoolXpert’) in The Gambia, Ghana and Benin between November 2022 and December 2024. The qualitative component was conducted in The Gambia at the Medical Research Council (MRC) Unit The Gambia at the London School of Hygiene and Tropical Medicine (MRCG at LSHTM). The Unit is located in the Greater Banjul Area, where about 70% of all annual TB cases in The Gambia are reported. It has urban, peri-urban and rural settings with a population of about 700 000.23 The childhood TB clinic operates as an outpatient clinic within the Unit’s clinical service department. A total of 163 out of 169 eligible CHIRWA-2 paediatric participants provided stool samples, from which qualitative participants were recruited following completion of the parent study. The qualitative study aimed to inform the ongoing national rollout of stool sampling and testing for TB diagnosis in children. The study design and reporting followed the Consolidated Criteria for Reporting Qualitative Research guidelines (COREQ).24 Stool sample collection and procedures After routine TB diagnostic procedures within the parent CHIRWA-2 study, caregivers of younger children were provided with containers and ziplock bags to collect stool samples, while adolescents received these materials directly. Participants either provided a stool sample at the study site or were instructed on how to collect it at home and transport it to the MRCG at LSHTM, in accordance with study guidelines. At the laboratory, stool samples were processed using the KNCV simple one-step (SOS) stool processing method, as described by Haas et al.25 Conceptual framework Wademan et al’s conceptual framework for assessing overall TB treatment acceptability was adapted for this study to frame the research questions and guide coding.26 There are three main domains within the framework: Usability, defined as how easy or difficult it was to provide stool samples, and the likes/dislikes of this method, Receptivity that explores the extent to which children and caregivers feel it is worth it to give stool samples compared with what they expected from TB diagnosis procedures and Integration which explores how well the stool sampling approach fit in with the routine health system and daily activities of all the different health providers and the TB programme managers. Participants’ sampling and data collection A purposive sampling approach was used to ensure variation in perspectives across key respondent groups involved in stool sampling and testing. This approach was chosen because the study aimed for depth and diversity of experiences, rather than representativeness. Participants were eligible if they: Had participated in the CHIRWA-2 stool sample sub-study and were able to provide stool samples (caregivers of children younger than 5 years, caregivers of children aged 5–10 years, adolescents aged 10–15 years). Had participated in the CHIRWA-2 stool sample sub-study and were able to provide stool samples (caregivers of children younger than 5 years, caregivers of children aged 5–10 years, adolescents aged 10–15 years). Were HCWs directly involved in the CHIRWA-2 study. Were HCWs directly involved in the CHIRWA-2 study. Were NLTP officials (programme manager, monitoring and evaluation officer, officer in charge of childhood TB, and three other programme officials) in The Gambia. Were NLTP officials (programme manager, monitoring and evaluation officer, officer in charge of childhood TB, and three other programme officials) in The Gambia. Caregivers and adolescents who met the inclusion criteria were selected and contacted by phone to assess their availability and willingness to participate in in-person interviews, with representation from those with both positive and negative StoolXpert results. All HCWs involved in the CHIRWA-2 study were contacted by phone and invited to participate in the qualitative study. Data were collected between June and August 2024. Semi-structured interview guides (online supplemental file 1) were tailored to each respondent category and informed by Sekhon’s theoretical framework of acceptability, Bowen et al’s feasibility assessment framework, and Waldeman et al’s conceptual framework for determining overall TB treatment acceptability.1326,28 The guides were pretested with four participants (one per group) to ensure clarity, relevance and cultural appropriateness. Interviews were conducted by three trained research assistants fluent in local languages (Wolof, Mandinka or Fula), supervised by an experienced social scientist and by the first author, a paediatrician and PhD student. Interviews were conducted face-to-face in private rooms at the MRCG at LSHTM and at participants’ convenience, with caregivers, adolescents and HCWs. Written informed consent (and assent for adolescents) was obtained prior to interviews. The interviews explored the acceptability of stool sampling and testing in TB based on perceptions (a detailed account of the experience compared with other methods of TB testing), their likes and dislikes, and feasibility questions regarding barriers, facilitators and recommendations or comments on incorporating stool as a sample for TB diagnosis within their current work environment. For caregivers and adolescents, we asked about perceptions of stool use, experiences providing a sample and preferences compared with other samples. For NLTP officials, the questions focused on the perceived usefulness of stool samples for TB diagnosis, as well as barriers and logistical challenges. For NLTP officials, a group interview was conducted at the NLTP office for convenience. The decision to conduct a group interview rather than individual interviews was based on the complementary nature of participants’ roles and the need for different people to contribute to answering questions to ensure a comprehensive response. Acceptability was specifically explored in terms of participants’ feelings, perceptions and willingness to engage with stool-based testing, while feasibility focused on practical, logistical and system-level factors influencing implementation. The interviews were conducted in English or the participants’ preferred local language, and were recorded with permission. Those conducted in the local languages were translated and transcribed verbatim into English. Researchers’ characteristics and reflexivity The research team included a paediatrician (PhD student) trained in qualitative research (SAO); a social scientist experienced in qualitative and mixed-method evaluation (JB); three TB researchers with mixed-methods background (VFE, EN, TOT); and five research assistants (two nurses (TM, SF), a medical doctor (MK) and two field technicians (SA, TA) who were involved in the project. While our dual roles across the qualitative and quantitative aspects of the research risked introducing bias, they enhanced contextual understanding. Patient and public partnership Patients were not involved in the design and conduct of the research; however, dissemination is planned across the Ministry of Health in The Gambia and the Vaccines and Immunity Unit at the Medical Research Unit in Gambia. Data analysis The transcripts were exported to NVivo (V.14) for coding and data management. Thematic analysis was conducted following Braun and Clarke’s six-step process.29 30 SAO conducted the initial analysis with inputs from JB and EN. First, the transcripts were read repeatedly to become familiar with the data. For steps two and three, main themes and subthemes were developed inductively and compared the conceptual framework and study objectives. The fourth and fifth steps encompassed theme refinement, including reviewing, defining and revising the thematic labels. Finally, the identified themes and corresponding codes were documented and presented in this paper. Selected quotes that illustrate and corroborate the identified themes are also presented. Quality control The study context, processes and participants were described in detail to allow transferability to similar settings. Bilingual researchers were employed to help verify translation quality, and an audit trail was maintained, with weekly peer debriefings during data collection, thereby improving dependability. A reflexive note was also kept to document our reflections and thought processes, reducing bias and enhancing the confirmability of the findings. Results 27 participants were interviewed: ten caregivers (CG; five in each age category), six adolescents (AD), five HCWs (two female nurses, one male doctor, one male field worker and two male laboratory technicians on the CHIRWA-2 study) and six NLTP officials (PO). One eligible HCW declined to participate in the study without providing a reason. Two main themes (acceptability and feasibility) and eight subthemes (four on acceptability, four on feasibility) were derived from our analysis of the study’s objectives (table 1). The themes, subthemes and associated quotes are described in the following sections. Table 1Themes and subthemes with links to the conceptual frameworkThemeEmergent subtheme from analysisComponent from Waldeman’s TB FrameworkApplicationAcceptabilityWillingness to provide stool samplesReceptivityAcceptance, perceived benefits and concerns about providing stool samplesDo caregivers and children consider providing stool acceptable? What are their feelings and perceptions about this process?Emotional responses and perceived burdenReceptivityResponses to the use of stool samples for TB diagnosisWhat is the perceived burden and emotions concerning the use of stool samples for TB diagnosis?Perceived value of stool-based testingReceptivityConfidence that stool samples will accurately diagnose TBDo caregivers and health workers believe that stool samples are effective for diagnosing TB?Preference in the TB sampling methodReceptivityPreference of TB sampling methods ie respiratory and stool samplesWhat is the preference of various stakeholders for TB sampling methods?FeasibilityAvailability of resourcesIntegrationSystem capacity, logistics and workflow integrationAre the necessary resources and system capabilities in place to support the collection and processing of stool?Integration into existing workflowsIntegrationSystem capacity, logistics and workflow integrationIs the stool sampling process compatible with the existing TB diagnostic workflows?Perceived advantages of stool samplesIntegrationSystem capacity, logistics and workflow integrationIs there a perceived need for stool-based TB diagnostics within the health system?Operational challenges: sample collection and transportation processesUsabilityEase of stool collection and handling proceduresWillingness to transport stools, challenges of stool transportIs the process of collecting and handling samples straightforward or burdensome for caregivers and adolescents?Are caregivers willing to transport stool samples?TB, tuberculosis. Themes and subthemes with links to the conceptual framework TB, tuberculosis. Acceptability of using stool samples for TB diagnosis in children within the Gambian health system Overall, stool sampling and testing for TB diagnosis in children were perceived as acceptable by most participants. Acceptability was reflected in participants’ willingness to provide stool samples, their emotional responses to the procedure and their perceived value of stool-based testing. Four subthemes were identified: (1) willingness to provide stool samples, (2) emotional responses and perceived burden, (3) perceived value of stool-based testing and (4) preference in the TB sampling method Willingness to provide stool samples Although some caregivers were surprised and not familiar with the use of stool for diagnosing TB, they were willing to provide the samples because they wanted the best for their children. I was surprised because I never thought that TB could be found in the stool, because even when I was young, our elders gave a container to put sputum and submit it. (CG 07) That’s the first time I heard about using a stool for TB diagnosis, I never knew. I only know of the sputum as a test for Tb diagnosis. I had no problem when they told me, and i just want my child to be heatlthy. (CG 04) On the other hand, as illustrated by the quote from participant AD 02, some adolescents found it challenging to provide a stool sample because they were not feeling the urge and were uncomfortable with the procedure. Some were embarrassed by the idea of providing stool samples for diagnosing a chest infection. I was not ready to poop, but my mother forced me … I was very shy and very nervous. I thought they would make a joke of me … When they told me [to provide a stool sample], I was afraid. Because I said, how can I do that for TB? (AD 02) Emotional responses and perceived burden While caregivers generally did not express discomfort, adolescents reported embarrassment, shyness and anxiety when asked to provide stool for diagnosing a chest condition. These emotional responses influenced their willingness and preference for different sample types. I didn’t find it difficult. The stool was collected from the diaper privately without anyone knowing. I had no problem transporting it in my bag because I was prioritising my child’s health. (CG03) Some adolescents felt uneasy about the procedure, especially in clinical settings where privacy was perceived to be limited. The need to produce stool on demand also contributed to psychological discomfort, with some describing the process as stressful or unnatural. I was very shy and very nervous. I thought they would make a joke of me … how can I do that for TB? (AD 02) HCWs were also confident that caregivers, once instructed, would be able to provide stool samples, as the process is easy. Comparing stool to sputum, it will be easier for the mother or caregiver, because with the sputum induction, the mothers are sometimes not happy seeing the way it is done, especially gastric lavage or induced sputum. (HCW 04) Perceived value of stool sampling and testing Participants had varied views on the diagnostic accuracy of stool samples for TB, but most were optimistic about its high accuracy. Most caregivers said that once the physicians requested it, they knew it would be accurate. Despite the delay in obtaining stool samples, some caregivers preferred stool samples over sputum samples because sputum induction was unpleasant. I feel a little bit anxious because I saw a pipe going inside, so I did not know what will come next, so it was a little bit psychological. I don’t know what I can call it, but later, thank God in my presence, it was done successfully. (CG 02) Preference in the TB sampling method Some caregivers had no preference and would do whatever the doctor requested. Most adolescents and caregivers found it easier to produce a sputum sample than a stool sample because they did not feel shy about providing the samples. I don’t have a choice I prefer both of them. Whatever the doctor says I will go with it. I don’t have a problem with it. (CG 12) Because for the sputum I just coughed and provided it and the stool I had to force myself to produce it. I would prefer sputum because it was easier. (AD 05) However, almost all the HCWs preferred stool samples to sputum, emphasising minimal patient contact during sampling, non-invasiveness and the challenges of obtaining sputum samples. However, they were concerned that caregivers with no formal education might not accept it. I prefer stool collection because it doesn’t require direct contact with patients, unlike sputum, which involves invasive procedures that can distress children and mothers. (HCW 03) I think this could cause some challenges. Because some mothers will be asking themselves, what is the connection between TB and stool? To an educated person, this may not cause problems. Hopefully, with time, I don’t think we should have issues. (PO 02) Feasibility of using stool samples for TB diagnosis in children within the Gambian health system HCWs and NLTP officials generally perceived stool-based testing as feasible within the Gambian health system. Feasibility was influenced by structural, logistical and health system factors, including resource availability, integration into workflows, perceived advantages and operational challenges. Availability of resources HCWs and NLTP officials considered the availability and sustainability of human resources and the sustained supply of essential logistics and equipment at various levels of healthcare as a determinant of the feasibility of successful implementation; however, they did not think it would be a challenge at the moment because these resources are available, but will have to be sustained. They also considered supply chain and logistics to be necessary for implementation. Yes, it’s feasible because stool collection is non-invasive and relatively simple. However, healthcare facilities need proper resources, like clean toilets and sanitisers, to motivate caregivers. (HCW 2) When it comes to the machines [for testing], we don’t have issues, if our presumptive registers were not being filled adequately, meaning we were missing cases in terms of recording, we would be under-budgeting for the procurement of cartridges. So, you may have issues with cartridge shortages if not well budgeted … (PO 4) Staff attrition and logistical issues were identified as systemic barriers to successful implementation. Although refresher training is conducted periodically, rapid staff turnover was identified as a cause of personnel factor in essential services. Therefore, it is crucial to maintain regular training to ensure the sustained implementation of these programmes. I think the other challenge we may anticipate is the staff turnover, because these trainings are conducted, if I’m not mistaken, once a year. Staff are leaving on a very regular basis. That turnover may affect the capacity of some facilities to do the test effectively because the reference lab may not have the resources to train everybody, even though that’s desirable. (PO 2) Recommendations, mainly from HCWs, included raising community awareness, training HCWs, adhering to standard operating procedures, incentivising HCWs, and securing political buy-in. Sensitisation will also be essential—mass campaigns on TV and radio over several months (three to six months) to inform people. (HCW 01) Integration into existing workflows HCWs stated that it is possible to integrate stool sample testing for TB into existing workflows due to the ease of sample processing, their experience with GeneXpert machines and its fast turnaround time, especially with the SOS stool-preparation method. The availability of GeneXpert platforms across regions will facilitate nationwide rollout. Stool processing is simpler and faster. It uses a direct one-step procedure where you take a spoonful of stool and mix it with the buffer in the same container. (HCW 2) In all seven regions, we have Gene Xpert in major health facilities and tertiary health facilities, like general hospitals, so laboratory technicians and scientists were trained recently on how to use stool to test. So, we hope that once it is fully rolled out, it will enhance the case finding. (PO 2) However, some HCWs believed that the country was not yet ready and would need pilot studies and proper caregiver education on the use of stool samples for TB diagnosis before it could be fully implemented. No, we’re not ready as a country [to implement stool testing]. This analysis needs to be completed first … . If it [stool sampling] is just introduced without explanation, there might be resistance. But once you explain, I believe it will be widely accepted. (HCW 01) Programme managers also pointed out that although there have been no pilot studies on the feasibility of stool sampling and testing for TB diagnosis in children in The Gambia, the practice can proceed to implementation, as it is already a WHO recommendation. They highlighted the NLTP officials’ commitment to using stool samples and anticipated that this would help reduce gaps in case detection. No pilot study [on the feasibility of stool samples yet]. In most cases, like you said, these are WHO recommendations, and once countries agree to do it and they have the capacity, they just go into it. There is no [need for] pilot study. (PO 1) Perceived advantages of stool samples Programme managers commented on the demand for and perceived value of using stool samples. They highlighted the potential impact of using stool samples to improve childhood TB diagnosis, given the perceived ease of sample collection. If they are sick or exposed and the clinician thinks that this might be TB, other biological specimens might be practically impossible, but the stool can easily be obtained and analysed, and TB identified. (PO 2) Also, HCWs preferred stool samples over sputum, emphasising the minimal patient contact during stool sampling, its non-invasiveness, and the challenges of obtaining sputum samples. Collecting quality sputum is often challenging—patients might not produce it or provide saliva instead, leading to low sensitivity. But stool samples can be collected from almost anyone without difficulty. So, for me, stool is the better option. (HCW 01) Operational challenges: sample collection and transportation Although stool collection itself was perceived as simple, participants highlighted several logistical and operational challenges that could affect implementation. Caregivers reported difficulties related to transportation, particularly when travelling long distances or using public transport. Concerns about odour, hygiene, and discomfort when carrying stool samples in public were also expressed. Transportation is one of the hectic things because I come all the way … especially when it is raining. (CG 02) If you don’t have a car that will take you directly from home to the hospital, instead, you are in the midst of people, it’s something that’s not clean. It’s better to provide it while in the hospital. (CG 07) Additionally, some caregivers suggested that sample collection should ideally take place in healthcare facilities to minimise these challenges. These findings indicate that while stool sampling is technically simple, practical barriers related to transport and infrastructure may affect its feasibility in routine settings. Discussion In this study, we explored the perceptions of caregivers, adolescents, HCWs and NLTP officials regarding the feasibility and acceptability of using stool sampling and testing for TB diagnosis in children in The Gambia. HCWs reported that the SOS stool preparation method is simple and has a rapid turnaround time, consistent with findings from a survey in Ethiopia by Yenew et al, in which most laboratory staff also indicated that the SOS stool method was easy to perform.17 Our study additionally provides a nuanced description of their perceptions. The ease of integration into existing workflows and the availability of GeneXpert platforms nationwide can be leveraged to scale up nationwide. This further enhances the utility of stool sampling and testing within WHO algorithms, as it leverages infrastructure already deployed for sputum-based testing and requires no additional specialised equipment for specimen processing. Although this may confer a sustainability advantage in The Gambia, it must be balanced against the need for ongoing quality assurance, continuous capacity building and staff motivation to ensure the smooth integration and decentralisation of stool sampling and testing. Sustainability considerations emerged as an important cross-cutting issue influencing feasibility. Although existing infrastructure, such as GeneXpert platforms, supports the integration of stool-based testing, long-term sustainability will depend on consistent supply chains, ongoing training and health system financing. Without these, initial gains in implementation may not be maintained. These findings suggest that sustainability should be explicitly considered within feasibility assessments, particularly in resource-limited settings where programme continuity is heavily dependent on external and domestic support. The strong preference for stool collection over traditional gastric aspiration and sputum induction among caregivers and HCWs corroborates studies reporting that stool samples are an advantageous method, particularly for individuals unable to expectorate.14 16 19 Despite some scepticism about the role of stools in TB diagnosis and the occasional delay in children’s ability to produce stool samples, especially when they had to be given during clinic hours, caregivers preferred it to sputum due to the difficulty in obtaining sputum samples, particularly in young children. However, the adolescents’ embarrassment identified in our study represents an important age-related barrier to acceptability that warrants careful consideration. This is consistent with a study by Lima et al in South Africa, which reported that older children had difficulty providing stool samples due to embarrassment, attributed to increased self-awareness and their need for privacy and autonomy in this age group.16 The contrast between caregiver enthusiasm and adolescent reluctance highlights the importance of age-appropriate counselling and privacy provisions during sample collection. This underscores the complexity of acceptability as a multi-dimensional construct that varies across stakeholder groups and age categories. Finally, they emphasise the importance of engaging multiple stakeholders, such as HCWs, caregivers and adolescents, in the planning and implementation of new diagnostic approaches. The perceived value of stool sampling and testing, particularly regarding diagnostic accuracy, time to diagnosis and impact on the child’s health outcomes, motivated caregivers to provide them. This aligns with the broader acceptability framework proposed by Sekhon et al13 and with other studies conducted in low- and middle-income countries, which showed that caregivers consider the benefits of the sampling technique irrespective of its invasiveness, stigma or time required.17 19 Although induced sputum and gastric lavage have demonstrated higher diagnostic accuracy than stool samples, the acceptability and feasibility of the latter, especially in resource-limited settings, pose a critical trade-off.21 31 The ‘one-size does not fit all’ principle applies equally to the selection of diagnostic algorithms and specimen types. While induced sputum may offer superior diagnostic yield in settings with robust laboratory infrastructure and trained personnel, in resource-limited settings like The Gambia, combining stool sampling and testing, which is highly acceptable among caregivers, will increase bacteriological confirmation rates, as recommended by the WHO. Strengths and limitations To the best of our knowledge, this is the first qualitative study conducted in The Gambia to explore the feasibility and acceptability of stool sampling and testing for diagnosing childhood TB. As such, the findings provide useful guidance on introducing stool sampling for the diagnosis of childhood TB within the Gambian healthcare system and sociocultural context, with likely resonance in other settings. However, each country or local context will have its own combination of facilitators and barriers, not all of which will align with those of The Gambia. Eliciting multi-stakeholder perceptions provides a comprehensive understanding of the experiences and preferences of stool sampling and TB testing among providers and patients. Furthermore, this study considered the lived experiences of caregivers and adolescents, which differed considerably, underscoring the importance of including older children’s perspectives in health systems research. Further research is needed to explore the perspectives of children, who represent a crucial demographic in TB control efforts, using age-appropriate data collection methods, such as drawing or storytelling, to capture their unique experiences and perceptions. Furthermore, exploring caregivers’, children’s and adolescents’ perceptions of the acceptability of other non-sputum-based approaches could lead to the incorporation of more child-friendly diagnostic tools as new technologies emerge. This study has some limitations. Participants were drawn from those who had already participated in the diagnostic accuracy study, potentially introducing selection bias. To minimise this, participants were purposively selected for this qualitative study, with varying levels of engagement and diverse demographic characteristics, to capture a range of perspectives. Also, the participants’ direct involvement in the parent study provided a good description of their lived experiences. Research assistants were HCWs who were involved in the quantitative study. This dual role may have introduced bias in the interview process, as participants may have been less forthcoming or may have responded in a socially desirable way due to their relationship with the interviewer as a service provider. Additionally, research assistants’ clinical involvement may have influenced their approach to the interview or interpretation of participants’ responses. These were mitigated by training research assistants in reflexivity and neutral interviewing, using interview guides to streamline data collection, and by peer debriefing and reflexive journaling by research assistants. However, future studies should consider separating data collection and clinical roles to minimise potential role-related bias. Social desirability bias may have influenced responses, particularly among HCWs who might have felt pressure to provide positive feedback on the diagnostic method. To prevent this, informed consent emphasised confidentiality, reducing concerns about repercussions for honest feedback. Again, this study was conducted in a research context with support structures that may not be available in routine settings, potentially affecting perceptions of feasibility. This was addressed by including questions about routine care settings without research support, and, for the NLTP officials, we explored resource requirements, sustainability concerns and operational challenges that would need to be addressed for routine implementation. Of note, findings from the CHIRWA-2 study have not yet been published and therefore cannot be cited for diagnostic accuracy. Conclusion Stool sampling and testing for the diagnosis of childhood TB is feasible in The Gambia from the perspectives of HCWs and NLTP officials. Successful implementation will, however, require addressing healthcare system capacity and ensuring a consistent supply of logistics. Also, age-sensitive approaches to education and community engagement will be required, especially for adolescents, to address concerns about embarrassment. Supplementary material Acknowledgements We are grateful to all caregivers and health workers who generously shared their time and experience with us. To Mr Tunkara Abdoulie, who helped with translation during the interviews. Also, all the staff of the National Leprosy and TB Control Programme, Ministry of Health, The Gambia, for their participation. Footnotes Funding: This study was funded by a European and Developing Countries Clinical Trials Partnership (EDCTP) grant to the West African Networks of Excellence for TB, AIDS and Malaria (WANETAM) (ref: EDCTP-RegNet 2015-1049). The funder had no role in the study design, data acquisition, data analysis, data interpretation or manuscript writing. The content is solely the responsibility of the authors and does not represent the official views of the funders. Prepublication history and additional supplemental material for this paper are available online. To view these files, please visit the journal online (https://doi.org/10.1136/bmjopen-2025-114409). Provenance and peer review: Not commissioned; externally peer reviewed. Patient consent for publication: Not applicable. Ethics approval: This study involves human participants and was approved by The Gambia Government and the MRC Joint Ethics Committee before the commencement of all study procedures (ethics ref. 3112). Written informed consent was obtained from caregivers, HCWs and NLTP officials, and assent from all adolescents prior to each interview. The participants were assured of their right to withdraw at any time without any consequences. To ensure confidentiality, all personal details were removed from the transcripts, and audio recordings were stored on an encrypted device. Additionally, participants were kept anonymous and not linked to the quotes presented, which are identified solely by respondent type and interview number. Patient and public involvement: Patients and/or the public were involved in the design, or conduct, or reporting, or dissemination plans of this research. Refer to the Methods section for further details. Data availability statement Data are available upon reasonable request. References Supplementary Materials Data Availability Statement Data are available upon reasonable request.