Authors' objectives: Tuberculosis remains a United Kingdom health concern. It occurs predominantly in people who have lived in tuberculosis endemic countries or have links there. Adherence to anti-tuberculosis treatment can be challenging, especially for people who experience severe side effects or social marginalisation. Poor adherence can lead to treatment failure. Current adherence support interventions make little difference to outcome. We identified the need for a ‘manualised’ approach to (1) improve case-managers’ ability to detect people likely to non-adhere and (2) guide targeted adherence support. Synthesise knowledge on drivers and interventions to support anti-tuberculosis treatment adherence Apply the Perceptions and Practicalities framework to understand poor adherence Develop a manualised intervention to identify adherence-related risks, modifiable barriers and support mechanisms Pilot the intervention and assess feasibility of data collection Evaluate implementation through fidelity and reach, and assess impact on adherence Assess intervention delivery costs to guide a full trial plus economic evaluation Tuberculosis (TB) remains a health concern in the UK. Effective treatment involves taking daily medication for at least 6 months. Adherence to anti-tuberculosis treatment (ATT) can be difficult, especially for people having severe side effects or are socially marginalised. Poor adherence can lead to treatment failure, increased transmission risk and drug resistance. Current interventions to support adherence (e.g. directly observed therapy) offer limited benefit. We identified the need, therefore, for a ‘manualised’ approach to (1) improve case managers’ (CMs’) ability to detect people most likely to non-adhere and (2) guide the delivery of targeted adherence support. Synthesise knowledge on (a) determinants of ATT adherence, and (b) interventions to support adherence, with particular emphasis on social and cultural barriers. Apply the Perceptions and Practicalities conceptual framework of adherence to identify and address the personal, sociocultural, and health systems context, mechanisms and pathways of poor adherence among NHS TB patients. Develop a manualised intervention with multiple components that when delivered in the NHS can identify (a) patients most at risk of non-adherence, (b) salient modifiable barriers and (c) support mechanisms required to match appropriate interventions with specific, individual barriers. Pilot the intervention in people at risk of poor adherence to determine how the components work in combination and separately and assess feasibility of data collection for a future full trial. Evaluate the intervention’s implementation through (a) describing challenges and facilitators in delivery (fidelity, reach) and (b) assessing impact through evaluation of adherence. Use the pilot to (a) assess intervention delivery costs and (b) guide development of a full trial proposal and economic evaluation.

Authors' results and conclusions: We developed a tuberculosis needs assessment for tuberculosis services. This appeared better than standard care at identifying people requiring adherence support [e.g. at baseline 21/36 (58.3%) intervention vs. 4/43 (9.3%) control] and social support (over 24 weeks, on 29 vs. 6 occasions respectively). Cumulative dose-taking was high across the study population at 24 weeks [84% (95% confidence interval 78–91%) overall; 81% (68–93%) intervention; 88% (67–100%) control]. Dose-taking patterns were similar between arms. The videos and booklet produced short-term improvements in beliefs, necessity, concerns and practical barriers. Collecting health economic data using self-completed questionnaires was feasible; retrieving data from records more challenging. The intervention was acceptable to patients and staff though took longer than control to perform. The intervention, a National Health Service first, is feasible to use. Its place in care and method of evaluation could be assessed in a larger, definitive study. Formative work key findings Quantitative literature (n = 24 studies): we found high variability in adherence definitions, methods of assessment, and non-adherence (range 0.9–89%). Determinants commonly associated with non-adherence were homelessness, incarceration and alcohol or drug misuse. Health system and psychosocial factors were less commonly evaluated. Qualitative literature (n = 28 studies): health systems factors were most reported (86%), followed by personal (82%), structural (61%), social (57%) and treatment-related factors (50%). Patient (n = 18) and caregiver (n = 4) interviews identified numerous factors influencing individuals’ clinical trajectory, service engagement, and capacity to adhere, for example competing priorities, access to social capital, and risk perception. HCP interviews (n = 27) revealed that experienced CMs delivered a comprehensive risk assessment by checking for items not listed on the standard form. Clinic observations supported this and additionally found that adherence support was often assigned based on CM’s judgement of the patient’s social and clinical needs. From the formative work, we concluded that our approach should: acknowledge how ATT disrupts and is managed within people’s lives appreciate that circumstances and access to resources may change during treatment display sensitivity around social and cultural norms affecting experiences of ATT. The IDG agreed that (1) standardising and expanding the risk assessment was beneficial, and (2) repeated needs assessments were required to account for changes in circumstances. It was suggested that some components should be site-dependent (i.e. based on resource availability), and that an electronic risk assessment would increase continuity of care. Final components are listed above. Recruitment and demographics Between October 2020 and March 2022 [i.e. during the coronavirus disease (COVID) pandemic], we recruited adults with TB. From a total 155, 130 (84%) were eligible, contactable, and approached. Consent was given by 82/130 (64%): 26/37 (70%) and 13/29 (45%) at intervention sites 1 and 2, respectively; and 18/30 (60%) and 25/34 (74%) at control sites 1 and 2, respectively. More males and Asian/Asian British adults declined to participate. Three participants withdrew from the intervention arm. Fourteen subsequently withdrew and were retained in the analysis to the point of withdrawal. Withdrawal was highest months 1–2; individuals who did and did not withdraw were demographically similar. Data from 79 participants were analysed (equal male and female, 49% each); 1 registered as ‘other’. Most [67/79 (85%)] were born outside the UK. Around 50% (39/79) were Asian or Asian British and 58/79 (73%) had extrapulmonary disease. The study arms appeared balanced demographically. At baseline, only 11/79 (14%) had ≥ 1 ‘classic’ risk factor for non-adherence recorded (six intervention, five control). Participants’ perceived need for ATT (BMQ-S) increased in the intervention arm and decreased in the control arm between week 2 and month 3 of treatment [adjusted mean difference for clusters: 0.49, t = 13.01 (95% CI 0.11 to 0.97); p = 0.05]. Between months 3 and 6, there was no significant difference between arms. Concerns about treatment for TB (BMQ-S) decreased over time and was no different between arms. Similarly, no differences were detected in changes to perceived practical barriers (MPRAQ) of taking ATT (increased over time) or in Side Effect Experience Questionnaire (SEEQ) (reduced over time). Data completeness Across time points for all participants, response rates were 71–99% for EQ-5D-5L and 71–97% for modified CSRI completion; time spent delivering needs assessment was recorded for 25–86% of participants; and costs could be calculated for TB-related medications in 78–99%. Thirty-one patients participated (n = 15 intervention; n = 16 standard care): interviews and observations were conducted with 25 and 23 patients and 5 with intervention site CMs. Patients appeared forthcoming with information when answering the TNA. The booklet was often shared with family or coworkers. There was no clear evidence that our intervention improved the CM/patient relationship. Some patients reported the videos helped change their perceptions of TB and treatment, improving understanding and reducing concerns. Almost all said the booklet was helpful and allowed them to monitor side effects and organise their medications. In general, CMs felt equipped to deliver the intervention; initial uncertainties resolved over time and with practice. Most CMs delivered the main intervention components to most of their patients. At intervention sites, the TNA was completed for 97% of baseline consultations and 57–74% of 1- to 5-month consultations. Case Managers were less sure about the feasibility of the intervention, citing mainly the length of time needed to administer it. Adaptations were sometimes made by CMs, including asking patients to read through the TNA themselves, asking patients to read the booklet at home (instead of in clinic), or relying on their own judgement to assign adherence support (instead of following the TNA’s recommendations). The additional activities mandated by the study protocol were also sometimes a problem. Less experienced CMs found the intervention helpful, as it provided structure and a standardised way to inform their patients. More experienced CMs said it was less helpful, as they felt they already had an effective approach. One CM felt that repeating the TNA made the most difference to patients. The COVID pandemic caused considerable study disruption and may have altered the intervention’s effects by changing: perceptions of respiratory illness, adherence behaviour (as people were working from home), and services, due to its health system disruption. We could design, build and test an intervention in the NHS that combined published evidence with patient and healthcare provider insights and assessments within a conceptual framework of adherence. In the pilot study, the TNA appeared better than the standard assessment at detecting people needing adherence and social support, with more being offered these. However, overall adherence support uptake was similar in both arms. Dosette boxes were most often used as adherence support. Treatment adherence was > 80% in both arms. Adherence patterns did not differ between arms. The videos and patient booklet produced non-sustained improvements in beliefs, necessity, concerns and practical barriers. Collecting health economic data using self-completed questionnaires was feasible; retrieving data from records and CRFs more challenging. To get the most from this multicomponent intervention package may require changes to service delivery ± some service redesign. Limitations include: pilot study sample contained few people most likely to non-adhere; large number of assessments may have altered the intervention’s effects; use of medication monitors in both arms may have affected results; performing the study during COVID and lockdowns may have affected participants’ adherence to treatment.

Authors' recommendations: Scaling up intervention implementation and conducting a larger cluster-randomised trial to test its wider feasibility, efficacy and cost-effectiveness Consider evaluations of intervention single components and more flexible packages Consider consent at clinic- than individual-level Consider alternative study designs, for example, a stepped-wedge trial Consider measuring alternative outcomes to adherence, such as health status Consider monitoring long-term clinical outcomes such as relapse and post-TB disability.

Authors' methods: The study ran April 2018–September 2022. Formative work included scoping reviews of adherence literature; National Health Service patients, caregivers, and health worker interviews; and clinic observations. A multidisciplinary group, including people with lived experience of tuberculosis, healthcare professionals, and researchers, coproduced the intervention package. We performed a (1 : 1) pilot cluster-randomised trial (N = 79 participants evaluated), randomising four London tuberculosis clinics, in preparation for a definitive cluster-randomised trial. Participants in control clinics received standard care. The primary outcome was adherence, doses taken of a possible 168 measured using evriMED boxes and other sources. We recorded treatment outcomes and changes in participants’ needs, health-related beliefs and perceptions, costs, and health status. We conducted a mixed-methods process evaluation, using questionnaires, interviews, case-report forms, checklists and clinic observations. Study sample contained few people more likely to non-adhere. Assessments may have altered intervention’s effects. Use of medication monitor in both arms may have affected results. Research pathway Data from formative work (scoping reviews, in-depth interviews, and clinic observations) guided an Intervention Development Group (IDG) comprising key stakeholders. The resulting multicomponent intervention was trialled at four NHS TB service sites. A process evaluation explored possible mechanisms of effect and assessed the data collection feasibility and intervention delivery.

Project Status: Completed

URL for project: https://www.journalslibrary.nihr.ac.uk/programmes/hta/16/88/06

Year Published: 2026

URL for published report: https://www.journalslibrary.nihr.ac.uk/hta/GJML0604

URL for additional information: English

English language abstract: An English language summary is available

Publication Type: Full HTA

Country: England, United Kingdom

DOI: 10.3310/GJML0604

Organisation Name: NIHR Health Technology Assessment programme

Contact Address: NIHR Journals Library, National Institute for Health and Care Research, Evaluation, Trials and Studies Coordinating Centre, Alpha House, University of Southampton Science Park, Southampton SO16 7NS, UK

Contact Name: journals.library@nihr.ac.uk

Contact Email: journals.library@nihr.ac.uk