Globally, tuberculosis (TB) is one of the main causes of death and the leading cause from a single infectious agent. In 2020, an estimated 9.9 million people developed TB and 1.5 million died. Millions of people remain undiagnosed with TB, hindering efforts to end TB. TB tests have inadequate accuracy or performance characteristics for implementation across all populations and settings. None of the tests meet the WHO-Target-Product Profile for TB screening and most need specialized laboratory staff and infrastructure, making them unsuitable for primary health care (PHC).

The overall objectives for this project are to:

1. To evaluate the performance of selected TB tests and test combinations in primary health care settings and KVPs.
2. To identify test combinations that increase the proportion of people diagnosed with microbiologically confirmed TB in comparison to a microbiological reference standard.
3. To demonstrate that test combinations are cost-efficient and feasible for scale-up to facilitate enhanced access to TB diagnosis in locations and for populations where they are not currently available.

Methods in brief:

Objectives 1 and 2 will use the same methods and will be assessed simultaneously. These objectives will use cross-sectional surveys. Activities will be conducted in Bangladesh, Brazil, Kenya, Cameroon, Malawi, Nigeria, and Viet Nam.

Each country will study selected priority populations at risk of TB, including adults attending PHC centers and district hospitals; people living with HIV (PLHIV); marginalized populations (internally displaced, refugees and pastoralists), and children. Activities within countries will use standardized protocols for evaluating diagnostic tests and combinations.

Methods for Objective 3 (Phase 1) will be a realist evaluation consisting of a market landscape analysis, evaluation of the costs and cost-efficiency of implementing TB tests and combinations, and a combination of key informant interviews, focus group discussions and/or participatory workshops, depending on appropriateness to stakeholder group and local context, to gather qualitative data on acceptability of TB tests and combinations.

1.1 BACKGROUND AND RATIONALE:

The location where people with presumptive TB are tested and the timeliness of diagnosis are crucial to increase access to TB diagnosis. Ideally, clinical samples should be tested locally close to the TB-affected person or community. This is because, approaches that rely on sputum transport or referral of patients often result in TB diagnostic delays and poor linkage to treatment. In contrast, local testing is associated with increased numbers of TB cases detected, confirmed, and patients linked to treatment. Major limitations for local TB testing are financial constraints to purchase testing equipment and poor availability of tests that can be used in locations with no laboratories or limited laboratory capacity. In this study, we will demonstrate that TB tests can be adapted to be used locally close to those affected by TB and that such local testing increases the number of participants diagnosed and treated.

Moreover, although there is sufficient evidence on the performance of most TB tests, and some evidence on their performance when tests are used sequentially (e.g. CXR followed by Xpert® Ultra) this evidence is incomplete for most other test sequences or "combinations" (e.g. C-reactive protein (CRP) test followed by lipoarabinomannan (LAM) test). There is also scant evidence for test performance once the population to be tested has been selected by a prior test, which increases the pre-test probability of being TB positive. For example, it is not known whether urine LAM tests perform better (or worse) among individuals with a positive CRP. Consequently, there is also limited evidence on how the introduction of a prior screening test affects the cost and cost-effectiveness of TB screening and diagnosis, including through reducing the number of people who undergo a more expensive confirmatory step following a positive screening result. Performance of tests vary with the pre-test probability of a positive test and perform better in populations with high TB prevalence. However, it is not known how the performance of test combinations vary in different populations.

The purpose of this study is therefore to close these knowledge gaps by ascertaining the performance of tests suitable for local use at the point of need when deployed alone or in combination and thereby inform follow on trials of the most promising tests and test combinations.

1.2. Hypothesis:

Our main hypotheses are that:

1. Combinations of TB tests at primary health care level will have higher sensitivity than any of the individual tests.

2. The use of combinations of TB tests will increase the pre-test probability of subsequent tests being positive and therefore increase the efficiency of testing by selecting populations increasingly likely to have TB or be TB-positive; and

3. It is cost-efficient, feasible, and acceptable to implement TB tests at primary health care level.

   1.3. Participating countries:

   Bangladesh, Brazil, Cameroon, Kenya, Malawi, Nigeria and Viet Nam.

   2. OBJECTIVES: 2.1. Primary Objective: Objective 1: to evaluate the performance of selected TB tests and test combinations in primary health care settings and key, vulnerable and underserved populations (KVPs).

   2.2. Secondary Objectives: Objective 2: to identify test combinations that increase the proportion of people diagnosed with microbiologically confirmed TB in comparison to a microbiological reference standard.

   Objective 3: to demonstrate that test combinations are cost-efficient and feasible for scale-up to facilitate enhanced access to TB diagnosis in locations and for populations where they are not currently available.

   3. STUDY ENDPOINTS:

   In this four-year project, the Start4All team will work towards three joint portfolio-level outcomes, which focus on:

   • Accelerating the introduction and adoption of existing TB diagnostic tools and novel TB test combination packages at primary health care level and outreach approaches to increase access to timely detection and linkage to care in LMICs;

   • Developing the conditions for sustainable and equitable access to TB diagnostics tools and TB test combination packages integrated within primary healthcare; and
   • Strengthening global alliances and national partnerships to enable scale-up.

   During Phase 1 of the study, our primary and secondary study endpoints will be:

   3.1. Primary study endpoint: 3.1.1 Generation of robust estimates of the diagnostic accuracy and performance of TB tests and TB test combinations in primary healthcare settings and KVPs.

   3.2. Secondary study endpoints:

   3.2.1 Estimates of the predicted performance of TB test combinations and how they perform in specific populations.

   3.2.2 Selection of the optimal TB test combinations for implementation and scale-up in primary healthcare settings and key vulnerable populations based on their feasibility, acceptability, and modelled accuracy and cost-efficiency.

   3.3. Case definitions:

   Participants will be classified as:

   • Microbiologically confirmed TB case: Participants with a positive microbiological reference standard (MRS) test of MGIT culture (Bangladesh, Cameroon, Kenya, Malawi, Nigeria and Viet Nam) or solid culture (Brazil only). Where culture results are not available, a positive Xpert® Ultra and/or smear microscopy will be used.
   • Clinically diagnosed TB case: Participants in whom a healthcare professional has decided to initiate TB treatment based on clinical symptoms and/or signs and/or human-interpreted CXR consistent with TB and in the absence of a positive MRS
   • Unlikely TB case: Participants who are neither a microbiologically confirmed TB case nor a clinically diagnosed TB case.

   Sub-studies including children will use the 'Updated Clinical Case Definition for Classification of Intrathoracic Tuberculosis in Children, as described below.

   - Confirmed tuberculosis: Bacteriological confirmation obtained (Mycobacterium tuberculosis confirmed by culture or Xpert® Ultra assay from at least one clinical specimen including sputum, stool, gastric aspirate (GA) or nasopharyngeal aspirate (NPA))

   • Unconfirmed tuberculosis: Bacteriological confirmation NOT obtained AND at least 2 of the following:

     • Symptoms/signs suggestive of TB

     • Chest X-rays consistent with TB

     • Close TB exposure or immunologic evidence of M. tuberculosis infection

     • Positive response to TB treatment (requires documented positive clinical response on TB treatment-no time duration specified) AND no spontaneous improvement of symptoms in the absence of anti-TB treatment

   • Unlikely tuberculosis: Bacteriological confirmation NOT obtained AND criteria for "Unconfirmed TB" NOT met (including spontaneous improvement of symptoms in the absence of anti-TB treatment).

   Criteria for the clinical case definitions have been adapted from Graham S et al. definition to include the latest WHO recommendations such as the use of Xpert® Ultra assay with non-respiratory specimen for bacterial confirmation.

   - Clinical signs/symptoms suggestive of TB include: (a) Persistent cough: persistent (>2 weeks), unremitting cough and WHO criteria for presumptive TB in adults. In addition, criteria for children include (b) Weight loss/failure to thrive: (b1) Unexplained weight loss: >5% reduction in weight compared with the highest weight recorded in last 3 months OR (b2) Failure to thrive defined as (i) Clear deviation from a previous growth trajectory, and/or (ii) Documented crossing of percentile lines in the preceding 3 months, and/or (iii) Weight-for-age z score of ≤-2 in the absence of information on previous/recent growth trajectory, and/or (iv) Weight-for-height z score of ≤-2 in the absence of information on previous/recent growth trajectory AND (b3) Not responding to nutritional rehabilitation (or antiretroviral therapy if HIV infected). (c) Persistent unexplained fever: Persistent (>1 week) and unexplained fever (>38°C) reported by a guardian or objectively recorded at least once. (d) Persistent, unexplained lethargy or reduced playfulness: persistent, unexplained lethargy or decrease in playfulness/activity reported by the parent/caregiver. (e) Infants 0-60 d (or neonate): additional signs and symptoms suggestive of TB include: (e1) neonatal pneumonia or (e2) unexplained hepatosplenomegaly or (e3) sepsis-like illness.

   - Chest X-rays will be considered consistent with TB if reviewers agree on the presence and location (right/left) of ≥1 lesion among the following: alveolar opacity, bronchial compression, excavation, ghon focus, gibbus, miliary, nodular infiltrates, paratracheal nodes, peri-hilar nodes, pleural effusion, tracheal compression

   4. STUDY DESIGN: 4.1. Study type:

   The study has a duration of 48 months and is divided into 2 data collection stages:

   1. Phase 1, with a duration of 18 months (this protocol)
   2. Phase 2, with a duration of 30 months

   Objectives 1 and 2 (Phase 1) will use the same methods and will be assessed simultaneously. These objectives will use cross-sectional surveys.

   Methods for Objective 3 (Phase 1) will be a realist evaluation consisting of a market landscape analysis, evaluation of the costs and cost-efficiency of implementing TB tests and combinations, and a combination of key informant interviews, focus group discussions and/or participatory workshops, depending on appropriateness to stakeholder group and local context, to gather qualitative data on acceptability of TB tests and combinations.

   4.2. Methodology:

   The inception period, starting in September 2022, comprises the development of documents for (and securing of) ethical approval, procurement, installation of and training related to equipment, development of standard operating procedures and training of staff, development of the administration and scientific groups underpinning the activities, developing data management routines and establishing databases for data recording and curation. This preparatory period does not require ethical approval.

   Objectives 1 and 2: Evaluating TB tests and test combinations Phase 1 will comprise multiple cross-sectional surveys (described in cohort/group section). This study design was selected because (together with diagnostic clinical trials) surveys are considered the best design for the evaluation of diagnostics. We will evaluate the performance of CRP, computer assisted detection (CAD) for CXR, pooling of multiple sputum samples and testing the pool with a single Xpert® Ultra cartridge, and urine LAM. Although the performance of these tests has been described among individuals with presumptive TB, there are now updated prototypes (e.g. LAM tests) or software (e.g. CAD CXR) for some platforms, there is limited information on their performance among KVPs (e.g. nomads/IDPs and children), and negligible information on their performance in primary health care settings. All tests will be compared to the MRS of sputum MGIT (Bangladesh, Cameroon. Kenya, Malawi. Nigeria and Viet Nam) or solid (Brazil only) culture or a composite diagnostic standard of one or more TB diagnostic tests. We will describe the sensitivity, specificity, positive and negative predicted value of the test. Research will be conducted in 7 countries, with participating population groups varying by country.

   Start4All will seek to recruit participants from defined groups including: adults attending primary healthcare facilities and adults attending district hospitals (which have adequate laboratories); and KVPs including nomadic populations, people living in informal settlements, people living rurally, refugees, internally displaced populations (IDPs), and children. For each group, the sample size has been estimated according to the expected prevalence of TB (except for children, where the main outcome is feasibility) and anticipated diagnostic accuracy of TB tests and TB test combinations (please see Table 3 in Section 4.3 "Sample size" below).

   All participants will be identified as having presumptive TB at the time of contact by the study team at a healthcare facility or during ACF activities. Individuals categorized as having presumptive TB based on either screening positive by WHO-recommended four-symptom screening (W4SS) criteria for TB (cough, fever, weight loss or night sweats) (healthcare facilities and ACF) or an abnormal CXR suggestive of TB (ACF only), will be invited to participate. Other people who are considered by WHO deffinitions to have a higher risk of TB based on either being a household contact of someone with active TB disease or being a PLHIV are also eligible to be enrolled in Start4All.

   Sampling and sample processing: For all participants in countries where CXR is incorporated into NTP guidelines for ACF activities, CXR will be taken in accordance with routine policy and individuals will then be assessed for eligibility into Start4All. For all other settings, including within healthcare facilities, CXR will be taken following informed consent along with all other study procedures. In total, participants will be asked to provide two spot sputum samples taken at least 30 minutes apart, a urine sample and a whole blood sample (preferentially using a finger prick with capillary sampling or, where unavailable venous blood sampling). The first sputum will be tested using the GeneXpert® Ultra cartridges locally within the PHC or district hospital, for sites without existing GeneXpert® testing platforms, Start4All will provide the GeneXpert® model GX-IV29,30. Where there is sufficient volume remaining, the remaining sample from the first sputum will be pooled with up to 3 other sputum samples from study participants for pooled GeneXpert® Ultra testing. The second sputum will be kept in a cold chain and transported to the reference laboratory for culture. Depending on the study country's standard of care, sputum smear microscopy may be performed on one or both sputum specimens. For children, sputum collection will follow the routine standard of care in each local health setting and will be collected using either expectorated sputum, induced sputum , gastric aspirate (GA) or nasopharyngeal aspirate (NPA). Thus testing and pooling of sputum may be reduced in this group and so stool will also be collected as a recognized alternative respiratory specimen for TB in children. If the local guidelines or approach changes during the study for any country, sputum will be collected using the new approach to be consistent with updated local routine practices. We will monitor the time to results and compare the time to results of tests processed using the routine country-level diagnostic processing pathways

   Locally, urine samples will be tested using the LAM lateral flow test and whole blood will be tested using the POC quantitative CRP, a semi-quantitative rapid test, and also for HIV testing for people with unknown HIV status. HIV testing will be performed following national guidelines (see Table 2). Standard of care in countries with high HIV-associated TB burden (Brazil, Cameroon, Kenya, Malawi, Nigeria) is to perform HIV testing on everyone attending health centers for TB diagnosis (and vice versa). In this scenario, clinical samples for TB and HIV diagnosis are taken simultaneously. In countries with low HIV-associated TB burden such as Bangladesh and Viet Nam, HIV testing will only be offered to people with confirmed TB. Additionally in Cameroon and Viet Nam where children are being recruited, a TB infection test will also be performed, in line with standard of care and NTP guidelines.

   CXR will be obtained using the facilities equipment, where available or study equipment using ultra-portable digital x-rays, depending on the setting, and classified using CAD software. All sites will use Qure.ai software and CXR images will be algorithmically classified by the proprietary software as presumptive TB (scored 0-1), no radiological sign of TB, other abnormalities, or invalid.

   The data from this study will be used to assess whether TB test combinations increase the proportion of microbiologically confirmed TB cases. We will use a MRS of sputum culture as previously described, and in the unlikely instance a culture result is not available we will use a Xpert® Ultra result and/or smear result. As we aim to identify the best combination of tests and how they perform in specific populations we will analyze all tests singly and model their performance in combination with other tests including how the use of prior tests modifies this performance. The analysis aims to maximize the number of individuals screened-in for confirmatory tests; evaluate whether combinations of screening and confirmatory tests improve performance; and identify the test combinations resulting in both the highest yield of TB cases (comparing against all people screened and also against only those with confirmed TB) and appropriate discriminatory power to triage people without TB. We will compare TB test combinations considering the number of tests, costs and cost-efficiency of tests, and their feasibility for implementation and scale-up in Phase 2.

   The modelled test combinations will be compared with standard of care diagnostic pathways based on the test combinations that are routinely available in each setting. Where standards of care have not yet been developed, as may be the case, for example, for ACF interventions in certain settings, modelled TB test combinations will be compared to recommended WHO screening algorithms.

   Objective 3: demonstrating cost-efficiency and feasibility of test combinations for implementation and scale-up.

   ● Realist evaluation: We will conduct a realist evaluation (RE) to maximize the impact of TB tests combinations, evaluating "what might work, for whom, under what circumstances and how" at micro/meso/macro-levels32. The RE will incorporate three complementary activities to identify feasible, acceptable, and scalable TB test combinations in suitable countries, healthcare levels, and populations: a market shaping and commercialization landscape analysis; baseline data collection and modelling of health system costs of delivering the TB test combinations; and in-person/virtual key informant interviews (KIIs), focus group discussions (FGDs), and/or participatory workshops, depending on stakeholder group and local context, to collect qualitative data on acceptability of tests and combinations.

   • Market shaping and commercialization landscape analysis: this will be a desk-based scoping review. First, it will evaluate existing evidence on the costs and cost-effectiveness of TB tests and test combinations and, using the health system and costs data collected below, will develop a value proposition and health technology assessment economic model, which will be crucial for future assessment of cost and cost-effectiveness. Second, it will assess the market landscape and market access to identify market gaps and challenges for current and new TB tests used in Phase 1 including developing the value proposition by considering the societal, economic, health system, and user pathways to impact of the potential TB test combinations. Third, it will review the data required to encourage adoption of new TB test combinations and the related market shaping including modalities, regulatory approvals, existing clinical guidelines, revenue, and commercial pathway/strategy/providers for manufacture, distribution, and global adoption. Fourth, it will shortlist potential participants from TB diagnostic and related policy-making bodies, manufacturers and producers, commercial research and development companies, and multilateral funding agencies, to invite to virtual KIIs of the Realist Evaluation.
   • Economic evaluation of TB test combinations: The combinations of tests are likely to influence the cost of a test cascade to both the provider and user. This is due to several mechanisms: (a) screening tests select participants that should undergo subsequent tests, increasing the proportion of follow-on tests that are test-positive, and thus reducing the cost per case detected; (b) some tests will be used more efficiently, again, potentially reducing the cost per TB case detected. This includes pooling, which in district hospitals decreases costs per test by 12%-46%. Evaluating whether the additional cost associated with prior screening tests is outweighed by the cost savings achieved through reducing the number of people who undergo a confirmatory test is central to the economic analysis.33 It is expected that the proportion of participants with positive tests will be low in primary health care, and that costs per patient tested using pooling would also be lower; and (c) some of the screening tests used have a lower price per patient than conventional tests. This includes CAD digital X-rays, which have a low cost per patient after capital costs. Taking capital costs into account, the price per test will be heavily influenced by the number of people screened, and is likely to vary from one setting to the next. In Phase 1, we will document the cost of each test, when used as (a) a single test and (b) in combination with multiple TB tests. Please see section on analyses for more details. We will also evaluate the cost of implementing novel TB tests and TB test combinations in primary health care from a health system perspective, comparing them to the cost of TB tests provided as part of the current standard-of-care. Although not collected during Phase 1, it is anticipated that costs to people accessing the tests, including out-of-pocket costs and lost time and income, will be collected during Phase 2 to inform cost-effectiveness analyses from both a health system and societal perspective. Consolidating the Market shaping and commercialization landscape analysis, unit and volume health system costs and charges will be collated from existing data sources and compared across countries and TB test groups for direct costs of TB diagnosis. We will document costs for implementing the new platforms and capital costs (e.g. training, security, technical support, procurement) and the recurrent costs of running the tests at primary health care (maintenance, supplies, technical support, etc.). The primary outcome will be the expected cost per TB case detected for different test TB combinations. Differences in costs and number of TB cases detected will be given by the incremental cost-effectiveness ratio (ICER) of potential screening and test combinations relative to routine TB tests as part of standard-of-care and will provide the basis for comparison across the range of theoretical combinations. In order to enable a detailed analysis and potential equity comparisons, results will be disaggregated by country, level and key population.

   Qualitative data on acceptability will be collected from key stakeholders using complementary techniques of KII and/or FGD and/or participatory workshops where appropriate as decided at local level. For example, previous experience in some study sites has suggested that FGDs may be the optimal method for gathering qualitative data on acceptability of diagnostics or interventions from people with TB whereas KIIs may be more productive for healthcare workers or National TB Programme staff, where there may be issues related to professional or sociocultural hierarchies that hamper data collection in FGD format.

   ● Key informant interviews (KIIs): KIIs will be conducted in-person with purposively selected and invited in-country key stakeholders in the field of TB and TB policy including Ministry of Health, WHO, and National TB Programme and implementing staff. These in-person KIIs will be supplemented by virtual KIIs with international leaders in the field of TB diagnostic tool design, development, procurement, and delivery as identified by the market shaping and commercialization landscape analysis above. To purposively select participants for the in-person KIIs, the Start4All core leadership team and partner country co-investigators will create a long-list of in-country stakeholders in each partner country. This long list will be iteratively reduced during team meetings to a short-list of potential participants. Virtual KII participants will be identified during the market shaping and commercialization analysis and refined by the core leadership team. Potential participants in the in-person KIIs will be ordered alphabetically by surname in each partner country and consecutively invited to participate until the sample size is reached. Potential participants in the virtual KIIs will also be ordered alphabetically by surname and consecutively invited to participate until the sample size is reached. Expanding our related research, we will develop topic guides to ask open questions with additional probes related to stakeholders opinions on the barriers and facilitators to access, delivery, and scale-up of TB test combinations at the individual, community, health system, policy, commercial, and political levels. Each KII, whether in-person or virtual, will be conducted by two Start4All investigators trained in qualitative methods and interview techniques with one member leading and the other taking field notes. These investigators will also lead data analysis and review information power (data saturation) in each KII and across KIIs.

   ● Focus group discussions (FGDs) and/or participatory workshops: FGDs and/or participatory workshops will include (but are not limited to) people with TB disease, TB survivors, civil society leaders, and community representatives. We will aim to balance participation by age, gender and socioeconomic status. The proposed topic guide will explore perceptions of participants concerning potential barriers/challenges and facilitators/enablers to implementation of existing and new TB tests and test combinations across individual, health system, and community levels, but with a focus on the individual level. Thematic Framework analysis will be applied to the data with the potential incorporation of frameworks including Klaic et al's Conceptual Framework for Implementability of Healthcare Interventions, Sekhon's Framework for Acceptability of Healthcare Interventions, and Barker et al's Framework for Going to Full Scale, initially with open and closed first order categories, which will then be grouped into second and third order themes and recorded in an overarching codebook.

   If time, logistics, or budgets preclude conducting KIIs and/or FGDs, we might choose to hold a single-day in-country participatory workshop. We would invite all potential qualitative research participants and organize break-out group sessions with diverse groups throughout the day, achieving similar results while saving time and costs.

   4.3. Provisional study schedule:

   The programme has three stages: the inception period will involve protocol development, Phase 1 (activities described in this protocol) comprising a series of cross-sectional studies and a realist evaluation to inform Phase 2; and Phase 2 comprising trials or a multi-country trial of selected TB test combinations.

   5. STUDY ENROLMENT

   5.1. Study population for the cross-sectional surveys ● Recruitment for objectives 1 and 2 will take place in health facilities and during outreach activities outside health facilities through ACF. Recruitment in health facilities will take place in primary health centers and in district hospitals.

   Primary health care (PHC): Primary health care clinics are facilities providing diagnostic and treatment services to ambulatory patients. These centers have basic utility services (electricity, water, sewage), but have limited laboratory facilities and usually refer patients with serious conditions to the hospitals. Recruitment in these centers will be conducted in the general clinics and in HIV care and treatment clinics (where these exist).

   District hospitals: District hospitals are considered the first referral level for care and are responsible for a district of a defined geographical area. These centers typically have laboratory infrastructure that can support more advanced diagnostic services. Recruitment will be conducted in the outpatient TB clinics or general outpatient departments, and/or in HIV care and treatment clinics (depending on clinic flow in respective countries) of district hospitals. Patients spontaneously attending these clinics and those referred from surrounding health centers are most commonly received in the morning and are asked to go to the waiting area located in ventilated open areas of the hospital. The clinics typically receive >10 patients per day and about half are new patients and the rest are patients who come to replenish medicines and follow up appointments.

   In both PHC and district hospitals PCF and ICF will be utilized. During the triage process, individuals will be asked what the purpose of their visit to the health facility is. If the main reason for attending the health facility is because they are experiencing symptoms of TB, they will be classified under PCF enrolment if they meet the inclusion criteria. If the main reason for attending the health facility is not associated with TB (e.g. general or other clinics) but they still meet the inclusion criteria, they will be classified under ICF enrolment.

   Active Case Finding (ACF): ACF includes all methods for the identification and treatment of those with TB who have not presented to the healthcare system of their own accord. ACF normally implies screening that is implemented outside of health facilities. Partners will conduct ACF for refugees/IDPs (Nigeria), nomads (Nigeria), populations living in informal settlements (Bangladesh), populations living rurally (Cameroon), and urban vulnerable populations (Viet Nam). Testing for participants will include the same tests as for other stationary health facilities but mycobacterial culture may not be available given the remote location of some participants (especially nomads, refugees/IDPs), who may be long distances away from existing culture facilities. Depending on the local setting, Xpert® Ultra test results or sputum smear could still be used as the MRS provided there is an appropriate samples referral system. In the unlikely case where there is no sputum based diagnostic comparison, a separate statistical analysis will include the additional yield of single and additional tests, without specifying sensitivity and specificity. Participants will be identified from screening camps typically conducted at community gathering points and/or primary healthcare centers. The sites for the screening camps will be identified together with the local leaders in each setting. At the screening camp, attendees will be offered a free chest X-ray screen using an ultra-portable digital X-ray system and verbally screened for symptoms of TB by W4SS.

   5.2 Recruitment for objective 3

   Participant recruitment for objective 3 relates to the KIIs and FGDs and/or participatory workshops taking place as part of the Realist Evaluation. Participant recruitment is not relevant to the Market Shaping and Commercialization Landscape Analysis or economic evaluation components of the Realist Evaluation, which will be desk-based.

   We will recruit prospectively to and conduct the KII, FGDs, and/or participatory workshops as previously described until the team feels, through iterative review of themes and sub-themes collated, that information power (data saturation) has been reached. People with TB recruited to Start4All who are also eligible to participate in FGDs will concurrently be recruited at the project study sites, be that at healthcare facilities or other locations such as camps. Eligible participants for the virtual, international KIIs will be selected purposively from a long list identified in the Market Shaping component of the Realist Evaluation and NTPs.

1. Inclusion criteria for Individuals attending stationary health facilities:

   health-seeking persons presenting at participating study site health facilities who are:

   • Age 15 years and above
   • PLHIV irrespective of their CD4 count and antiretroviral therapy (ART) status
   • At risk of pulmonary and/or extra-pulmonary TB

   OR

   • People with a positive WHO-recommended four-symptom screen (W4SS) or parenchymal abnormalities on chest radiography

   • Written informed consent and agreement to follow-up after enrolment (e.g. not planning to relocate)

   • A parent or guardian present at the time of enrolment to provide signed informed consent and assent signed for those aged 15 years to age of majority

     • Exclusion

   • Current anti-TB treatment (counting from the third dose)

   • Any anti-TB treatment within 60 days prior to enrolment

   • Any TB preventive therapy within 6 months prior to enrolment

2. Inclusion criteria for refugees and IDPs

   • health-seeking persons resident in official refugee camps who have cross international borders from other countries; OR

   • internally displaced individuals (IDPs), defined as persons forced or obliged to flee or to leave their homes to avoid the effects of armed conflict, situations of generalized violence, violations of human rights or natural or human-made disasters, and who have not crossed an internationally recognized state border.

   • IDPs can be recruited from refugee camps or among individuals sharing accommodation with local residents.

   • At the time they are presenting at participating study site health facilities.

   • Clinical criteria will be the same as described for individuals attending health facilities

     • Exclusion

   • Same as described for individuals attending health facilities and

   • Individuals who intend to leave the local area in the short term and have no means of contact

3. Inclusion criteria for nomadic populations

   • Participants will be selected using active case finding through community screening camps targeting the nomadic population (nomadic are defined as a member of a group of people who have no fixed home and move according to the seasons from place to place in search of food, water, and grazing land and are usually herders with traditional hierarchies.

   • The dates and locations of screening days will coincide with community market days, after consultation with nomadic community leaders

   • Clinical criteria will be the same as described for individuals attending health facilities

     • Exclusion

   • Same as described for individuals attending health facilities and Individuals who intend to leave the local area in the short term and have no means of contact

4. Inclusion criteria for informal settlements, urban vulnerable groups and the rural poor

   • Participants will be selected using active case finding through community screening activities targeting urban vulnerable groups and the rural poor.

   • Screening will be conducted on dates and in locations agreed with community leaders, depending on the setting.

   • Clinical criteria will be the same as described for individuals attending health facilities.

     • Exclusion

   • Same as described for individuals attending health facilities and Individuals who intend to leave the local area in the short term and have no means of contact.

5. Inclusion criteria for children health-seeking children presenting at participating study site health facilities who:

   • Are between 12 months and 15 years of age.

   • Have presumptive TB based on at least one criterion among the following:

     • Persistent cough for more than 2 weeks

     • Persistent fever for more than 2 weeks

     • Recent failure to thrive (documented clear deviation from a previous growth trajectory in the last 3 months or Z score weight/age < 2)

     • Failure of broad-spectrum antibiotics for treatment of pneumonia to improve symptoms and signs

     • Suggestive CXR features OR

     • History of contact with a person with TB and any of the symptoms listed above with a shorter duration (< 2 weeks), if the child is HIV infected or presents with Severe Acute Malnutrition

     • A parent or guardian present at the time of enrolment to provide signed informed consent and assent signed by children > 7 years old.

       • > 5 days of anti-TB treatment in the last 3 months or individuals receiving TB preventive therapy
       • Children <12 months old

Individuals suspected of having TB.