The Tuberculosis Sentinel Research Network (TB-SRN) of the International epidemiology Databases to Evaluate AIDS (IeDEA): protocol for a prospective cohort study in Africa, Southeast Asia and Latin America


Before the onset of the COVID-19 pandemic, tuberculosis (TB) was the leading infectious cause of death globally by a single pathogen.1 The COVID-19 pandemic has disrupted TB and HIV services, with attendant challenges for optimal diagnosis, control and care management.1–4 In the years since the onset of the COVID-19 pandemic, global estimates of TB disease, drug-resistant TB and TB deaths have increased for the first time in many years.1–3 In 2021, an estimated 10.6 million people developed TB disease and 1.6 million people died from TB.1 Further—despite increasing global access to antiretroviral treatment (ART) and to TB preventive therapy (TPT) among people living with HIV (PLHIV)—TB remains the leading cause of morbidity and mortality among PLHIV.5–7 In this context, data are urgently needed to inform global strategies to address the dual TB and HIV epidemics.

Critical evidence gaps exist with regard to drivers of unfavourable TB treatment outcomes, such as mortality, TB recurrence and post-treatment sequelae/complications, including among PLHIV.5 8 9 Addressing these gaps is particularly critical in light of recent acceleration towards shorter TB treatment regimens for both drug-susceptible and drug-resistant TB, and given the prospect of possible individualised approaches to treat both TB and post-TB lung disease (PTLD).10 Key areas of ongoing research gaps relate to TB-HIV coinfection, treatment and associated complications; consequences of drug-resistant TB; pulmonary complications and post-treatment outcomes; the impacts of psychosocial and life course factors on TB outcomes; and mental health outcomes of TB. A global prospective cohort of individuals with TB and with TB-HIV coinfection enables harmonised data collection and procedures to inform questions in these key areas.

The International epidemiology Databases to Evaluate AIDS (IeDEA) global research consortium—established by the US National Institutes of Health (NIH) in 2006—collects and analyses observational data in a clinical cohort of over 2.2 million PLHIV and people at risk for HIV, in 44 countries.11 Data are organised by seven geographical regions and coordinated by regional data centres.11 IeDEA provides diverse global data from HIV treatment programmes. The TB Sentinel Research Network (TB-SRN) is a global platform for coordinated observational TB research within the IeDEA consortium, which receives funding from multiple institutes and centres within the US NIH. The TB-SRN working group of IeDEA developed an observational cohort study protocol. To facilitate possible pooled global analyses, protocol development was based in part on the framework and original protocol of the Regional Prospective Observational Research in TB (RePORT) International Consortium.12–14 The TB-SRN uses a common set of standards and definitions for prospective observational TB research. These were developed in alignment with the study concepts and measurement time points defined by RePORT International.12–14 The TB-SRN will facilitate the use of pooled data to study pulmonary TB treatment and post-treatment outcomes among people with and without HIV at TB-SRN sites in six of IeDEA’s global regions. The resulting findings and study infrastructure may be used to inform policy and practice regarding TB treatment, and create a platform for additional regional and multiregional TB research within IeDEA.

Methods and analysis

Objectives of the IeDEA TB-SRN

With its focus on HIV and associated coinfections and comorbidities, the global IeDEA research consortium is ideally positioned to study TB outcomes among people with and without HIV. To accomplish this, the TB-SRN will study outcomes of people diagnosed with pulmonary TB through a network of 16 sentinel sites (table 1) located in 11 low-income and middle-income countries in 6 IeDEA regions: Asia-Pacific, CCASAnet (Caribbean, Central and South America), Central Africa, East Africa, Southern Africa and West Africa (figure 1).15

Table 1

Planned initial study sites in the tuberculosis sentinel research network of the International epidemiology databases to evaluate AIDS (IeDEA), by IeDEA region and target sample size

Figure 1
Figure 1

Country locations of planned initial study sites in the Tuberculosis Sentinel Research Network of the International epidemiology Databases to Evaluate AIDS. Map created using CCASAnet, the Caribbean, Central and South America network for HIV epidemiology.

There are three specific objectives of the TB-SRN. First, the TB-SRN will collect and analyse clinical and treatment data among people treated for pulmonary TB with or without HIV coinfection, to improve our understanding of the prognosis of TB disease and its health-related outcomes, including quality of life and survival. Second, the TB-SRN will assess the individual-level effects of HIV and ART on TB symptomatology, diagnosis, treatment response and survival. As part of this aim, investigators will also explore the effect of site-level TB and HIV management and integration of TB and HIV services on pulmonary TB treatment and longer-term outcomes. Third, the TB-SRN will describe PTLD and associations with HIV infection, diabetes, chronic lung disease, mental health and tobacco, alcohol and substance use, including measuring physiological, structural, and functional impairment, health-related quality of life, and survival.

Study design

The TB-SRN is a prospective, observational study, with consecutive enrolment of PLHIV and HIV-negative individuals, ages 15 and above, with clinically diagnosed or microbiologically confirmed pulmonary TB disease. Microbiological confirmation of pulmonary TB is defined on the basis of either positive molecular diagnostic test (eg, GeneXpert), acid-fast bacilli smear and/or TB culture from sputum or other respiratory specimen. Microbiological confirmation of pulmonary TB may also be based on positive urine lipoarabinomannan assay in the presence of clinical signs, symptoms and/or radiographic findings of pulmonary TB. Clinically diagnosed pulmonary TB is defined by clinical diagnosis by medical providers through standard of care, in the absence of confirmatory testing, prompting initiation of TB treatment. Specific eligibility criteria that must be met as part of clinical diagnosis include having either (1) any signs or symptoms of active TB (eg, persistent cough, haemoptysis, fever, unintended weight loss, fatigue or lethargy, night sweats, pleuritic chest pain) together with chest X-ray findings consistent with pulmonary TB or (2) the presence of respiratory signs and symptoms (including chronic cough, haemoptysis or pleuritic chest pain) regardless of chest X-ray findings. Individuals who consent to participate will provide clinical, laboratory and radiographic data at study visits at specified time points from TB treatment initiation through 12 months after the end of treatment (table 2).

Table 2

Study procedures in the Tuberculosis (TB) Sentinel Research Network of the International epidemiology Databases to Evaluate AIDS

Most participants, if completing a standard duration of 6 months of TB treatment for drug-susceptible TB, will consequently be followed for a total of 18 months from TB treatment initiation. Time on study will be longer, however, if treatment duration is longer as determined by providers under standard of care. (Treatment duration may be longer than 6 months, eg, for some regimens for drug-resistant TB, or if treatment is interrupted, or if pulmonary disease coincides with infection at an extrapulmonary site warranting longer treatment duration.) In addition, some regions have longer follow-up periods after end of TB treatment, according to region-specific objectives. Some regions will also have data collection beyond this multiregional protocol, such as for laboratory biomarkers, pharmacokinetic data or biological specimens. Data will be aggregated for concept-driven analyses.

Sample selection

TB-SRN sites

Sixteen sites included in the TB-SRN are located in 11 countries: Brazil, Burkina Faso, Cambodia, Côte d’Ivoire, the Democratic Republic of the Congo, Haiti, Kenya, South Africa, Thailand, Uganda and Zambia. These sites represent a range of contexts for the dual TB-HIV epidemics, including differing prevalence of TB and HIV, care models and resources. Analyses in this study will include comparisons by site-level variables or by region.

Eligibility and exclusion criteria

The IeDEA TB-SRN will enrol participants ages 15 and older with clinically diagnosed or microbiologically confirmed active pulmonary TB who are initiating TB treatment at IeDEA TB-SRN sites. Participants must have either documentation of recent HIV testing or of HIV infection or willingness to be tested for HIV, as routinely indicated under TB treatment guidelines.16 Informed consent will be required for all participants, including parental/caregiver consent and minor assent for individuals younger than age 18 (or the legal age of majority). There will be no restrictions based on sex, gender identity, HIV status, pregnancy, ethnicity or nationality. Participants may be coenrolled in other research with the exception of clinical trials of novel TB treatment regimens.

An individual will be excluded if they meet any of the following criteria: have received >7 days of TB treatment within the prior 30 days, excluding TPT; have imminent plans to follow-up for TB care or relocate/return to a site distant from the enrolment site, which would interfere with the participant’s ability to complete all study visits; have substantial cognitive impairment that may interfere with the ability to give reliable informed consent; are currently imprisoned.

Enrolment began in September 2022 and is projected to continue through October 2024. Data collection is ongoing with a projected end date of April 2026.

Participant considerations

The inclusion of participants with and without HIV will facilitate analyses in which HIV status is evaluated as a factor potentially contributing to TB treatment or post-treatment outcomes. The proportion of participants with HIV coinfection is anticipated to vary across sites.

The inclusion of participants ages 15 and above will allow for dedicated analyses of participants in the age group of 15–24 years of youth with TB. While youth have specific needs that must be addressed in quality health services, TB programmes globally have not adopted youth-centred care models.17–20 Further, adolescents and youth have been neglected in TB research, either by failure to include individuals younger than age 18, or by not examining research data within stratified adolescent or young adult age groups.19 20 Guidance from the WHO now advises that youth and their specific needs should be included in global TB research and care efforts.21–23 This study will assess clinical characteristics, TB outcomes,and post-treatment outcomes in a subcohort of youth with TB across 5-year age strata (ie, older adolescents aged 15–19 and young adults aged 20–24 years).

TB during pregnancy can cause poor outcomes for both the mother and for the developing fetus (or infant, after delivery), including maternal complications, miscarriage, preterm birth, low birth weight or perinatal death.24–27 Existing data regarding clinical features and outcomes of TB in pregnancy are very limited. Pregnant and postpartum individuals with TB (who have been pregnant within the last 12 months) will be included in TB-SRN. Data collection will include specific variables related to pregnancy, receipt of TB and HIV medications during pregnancy, and maternal and infant outcomes. These will be collected over the course of the study period, including for individuals who become pregnant or give birth during the study.

Patient and public involvement

Key research questions of this study were informed by previous participatory research and advocacy from individuals with TB; in particular, calling for research in PTLD and other post-treatment outcomes,28 and for inclusion of adolescent minors in research.23 Individuals with TB were not involved in the study’s design. Draft case report forms (CRFs) were revised through iterative rounds of review and preliminary piloting at the study sites. In particular, clinical programmes at the sites were involved in revisions at this stage to ensure the feasibility of CRFs and to minimise burdens on individuals with TB participating in the study. Clinical programmes at the sites were also consulted related to study planning. This included preparations for referral for immediate and urgent health needs, such as for symptoms of depression and suicidal thinking (assessed by Patient Health Questionnaire (PHQ-9) and suicide risk assessment). Findings from this research will be shared with national TB programmes and HIV treatment programmes at the study sites, and disseminated to individuals with TB.

Assessments and data collection

Data (eg, chest X-rays, laboratory results, health-related outcome measures; table 2) will be collected according to a common schedule and methodology across sites, so that they can be harmonised and aggregated for analysis.

Participants who consent to the study will be followed during TB treatment and for 12 months after the end of their primary treatment course (ie, the treatment course initiated at study enrolment). For most participants, this will be approximately 18 months after provisional enrolment/treatment start if they have drug-susceptible TB and receive a 6-month TB treatment regimen, but it may be longer if they have drug-resistant TB or require a longer treatment regimen for other reasons (eg, if there is associated extrapulmonary TB disease). At the time of this study, TB programmes at the study sites are primarily using treatment regimens of 6 months’ duration as routine standard for drug-susceptible pulmonary TB.

Participants will be requested to provide data during visits at key time points: at baseline (at initiation of TB treatment), month 1 (optional; performed at some but not all sites), month 2, end of TB treatment, 6 months post-treatment, 12 months post-treatment and at the time of suspected or apparent treatment failure, TB recurrence or study withdrawal if it occurs during treatment or the 12-month post-treatment follow-up phase. The baseline visit will include detailed clinical history, including course of TB symptoms and diagnosis, previous TB, HIV diagnosis and treatment (as applicable), history regarding recent or current pregnancy, and history regarding non-communicable comorbidities and their treatment (eg, diabetes mellitus, hypertension, pulmonary or cardiovascular disease, cancer, immune suppression, mental health diagnoses). HIV and TB clinical data will be extracted from medical records and TB registers, while current symptoms, pregnancy history and history of other conditions will be collected via patient interview. Sociodemographic information will be collected including specific information relevant to youth ages 15–24 at enrolment (eg, orphan status, caregiver characteristics, school attendance). Data collected at multiple visits during the study will include TB symptoms; ascertainment of immune reconstitution inflammatory syndrome, TB treatment failure or TB recurrence; clinical, radiographic and microbiological data related to TB; assessments for symptoms of depression (by the PHQ-9)29 and substance use (by the Alcohol, Smoking, and Substance Involvement Screening Test; ASSIST)30 31; and pulmonary investigations including repeated pulse oximetry, spirometry, functional test (1 min sit-to-stand test), and respiratory symptoms and health-related quality of life (by the Saint George’s Respiratory Questionnaire, SGRQ).32 33 These validated questionnaires have been used previously in the respective regions, with adaptations as appropriate (eg, PHQ-9, ASSIST and SGRQ).

Data will be collected on paper or electronic CRFs according to local capacity and regulatory requirements. A copy of the paper CRFs is provided (online supplemental material 1). Paper forms will be subsequently entered into the electronic system by trained personnel. All sites will use the secure, web-based REDCap data collection platform and/or the REDCap Mobile App for data collection.34 35 Common data management processes and procedures will be developed in collaboration with the Harmonist team at Vanderbilt University Medical Center, which provides informatics resources for IeDEA.36 37 For sites using film-based chest X-rays, films will be scanned and digitised using standard procedures defined by the NIH TB Portals platform.38

Supplemental material

Sites will work with the IeDEA regional data centres to enter, prepare and clean data using either the Vanderbilt REDCap server or a regional REDCap server to adhere to country regulations on research data storage. All sites will use the same REDCap project template to ensure variables and study events use the same names and code lists to facilitate subsequent data merging. Research staff will administer questionnaires using relevant local translations (in Bemba, French, Haitian Creole, Khmer, Lingala, Nyanja, Portuguese, Runyankole, Swahili and Thai) and adaptations as appropriate, implemented in paper CRFs and in the REDCap data collection platform. Regional data centres will conduct quality control or assurance activities for their sites based on guidance developed by the TB-SRN study team and the Harmonist team.36 37

Outcome measures

Multiple outcomes will be assessed in the TB-SRN (table 3). These include TB treatment outcomes, TB recurrence, mortality, other pulmonary, health-related quality of life and mental health outcomes.

Table 3

Select study outcomes in the Tuberculosis (TB) Sentinel Research Network of the International epidemiology Databases to Evaluate AIDS

TB treatment outcomes for both drug-resistant and drug-susceptible TB will follow the current definitions set by the WHO and will be in alignment with RePORT International outcomes.12 39 These include both clinical and biological criteria for TB treatment outcomes. TB recurrence (inclusive of TB relapse and new TB infection) will be defined as any new TB diagnosis after the end of TB treatment for the primary course. Mortality will include deaths from any cause, and will be assessed during the treatment and post-treatment periods. Information about cause of death will be recorded if available.

Post-treatment outcomes to be ascertained are informed by emerging research surrounding post-TB sequelae.28 40 41 PTLD will be defined by new, recurrent or persistent respiratory symptoms or signs that occur post-treatment; hypoxaemia (oxygen saturation <90%); pulmonary function impairment or chest X-ray abnormalities.40 Spirometry definitions for pulmonary function impairment include the following: forced expiratory volume in 1 S (FEV1)/forced vital capacity (FVC) ratio <lower limit of normal (LLN), FEV1<LLN and/or FVC<LLN. Global Lung Function Initiative (GLI) standard reference equations will be used to calculate LLN (fifth centiles) for each participant; these will be compared with observed values.42 43 Functional status will be assessed with 1 min sit-to-stand testing. Further measures of well-being after TB treatment will include symptoms of depression (by PHQ-9)29 and health-related quality of life (by SGRQ).32 33 The SGRQ has previously been validated32 and applied in studies of individuals treated for pulmonary TB.44–49 Both the PHQ-9 and SGRQ are among the assessments recommended by some experts for the evaluation of post-TB sequelae.28

Data management and harmonisation

TB-SRN data from multiple REDCap installations will be accessed via the secure REDCap Application Programming Interface (REDCap API) and automatically merged on a monthly basis to generate study enrolment and monitoring reports. These reports will allow tracking of study progress and ensure the distributed data collection remains aligned in variable formats and naming. Merged research datasets will be generated on demand for analyses associated with approved research concepts. Study data procedures include methods for ensuring the privacy and confidentiality of participant data, including using codes in place of names, implementing password-protected and encrypted data collection systems, training of site personnel on data management best practices, and applying data pseudonymisation where required for compliance with national data protection regulations. Analyses will be conducted by the designated regional data centre. Data-sharing agreements and management procedures will be overseen by the IeDEA Executive Committee.

The NIH, which funds both the IeDEA consortium and RePORT International, provides guidance for the coordination and linkage of these parallel streams of research. In addition, the Harmonist project, which supports IeDEA through development of data standards and software to support research operations, will coordinate with RePORT regions to streamline their existing data structures and identify points of data alignment with IeDEA to enable future cross-consortium data harmonisation and research.

Data analysis plan

While data may be used by the individual TB-SRN sites or regions, they are primarily being collected and harmonised for multiregional research, following IeDEA’s standard operating procedures governing research collaboration.50 The TB-SRN observational cohort study is designed to inform multiple analyses. Analyses of global TB-SRN data will be proposed through concept sheets, for detailed review and feedback from collaborators in the TB-SRN and other IeDEA working groups relevant to the study, with subsequent final review and approval by the IeDEA Executive Committee.50 Concepts will centre on major research questions in TB and HIV clinical epidemiology.51 These will include analyses of TB severity, TB treatment and post-treatment outcomes including PTLD, health-related quality of life and associated clinical, mental health and life course factors. Youth with TB (ages 15–24) will be assessed as a subset of this cohort, with attention to their clinical, psychosocial and lung health findings. The subset of pregnant and postpartum participants will also be described, to include specific variables and outcomes in this group.

Current, initial TB-SRN concepts delineate analyses in the following areas: baseline TB severity and associated factors; baseline depressive symptoms and substance use; chronic hypoxaemia and respiratory symptoms; and PTLD in youth.

Sample size considerations

The IeDEA TB-SRN will enrol 2600 participants across all study sites, including 300 participants in each of 5 IeDEA regions (Asia-Pacific, Central Africa, East Africa, Southern Africa and West Africa) and 1100 participants in CCASAnet. It is estimated that between 5% and 10% of treated TB cases will result in TB treatment failure or TB recurrence. Thus, if 2600 participants with active TB are enrolled, it is expected that between 130 and 260 episodes of treatment failure or recurrence will occur, with 200 being an approximate midpoint estimate. Furthermore, the majority of recurrent episodes are estimated to occur within 6 months of treatment completion, and thus >90% of all such episodes are expected to be detected during the follow-up period. Pulmonary function impairment may be anticipated in approximately 50%–60% of participants after completion of TB treatment.52 The overall cohort sample size of 2600 participants will enable precise estimates of key treatment and post-treatment outcomes. Given the anticipated HIV prevalence of 20%–30% across the global cohort, this sample size will also allow for multivariable analyses, including on HIV coinfection and treatment-related factors, in addition to sex, age and additional demographic or clinical factors.

As the TB-SRN is a descriptive study encompassing multiple planned outcomes and analyses, statistical considerations and power will vary by the research question proposed within the concept process of IeDEA. For analyses of youth with TB, this age group is estimated to make up 17% of new TB cases globally.53 In a cohort of 2600 individuals with TB, we anticipate including several hundred in this age group; representing a valuable contribution to the evidence base for this group.19 20 For pregnant/postpartum participants with TB, while low numbers are anticipated, relevant variables collected in this study for maternal and infant outcomes will be described. Given the very limited existing data on this population, these data will add to the existing literature.24 25 27 Aggregation with data from other cohorts may be considered for pooled analyses of priority questions for TB in these subgroups.19 20 51

This post was originally published on