People in underserved groups have higher rates of tuberculosis (TB) and poorer treatment outcomes compared with people with no social risk factors.

This scoping review aimed to identify interventions that improve TB treatment adherence or completion rates.

Studies of any design focusing on interventions to improve adherence or completion of TB treatment in underserved populations in low incidence countries.

MEDLINE, Embase and Cochrane CENTRAL were searched (January 2015 to December 2023).

Piloted data extraction forms were used. Findings were tabulated and reported narratively. Formal risk of bias assessment or synthesis was not undertaken.

47 studies were identified. There was substantial heterogeneity in study design, population, intervention components, usual care and definition of completion rates. Most studies were in migrants or refugees, with fewer in populations with other risk factors (eg, homelessness, imprisonment or substance abuse). Based on controlled studies, there was limited evidence to suggest that shorter treatment regimens, video-observed therapy (compared with directly observed therapy), directly observed therapy (compared with self-administered treatment) and approaches that include tailored health or social support beyond TB treatment may lead to improved outcomes. This evidence is mostly observational and subject to confounding. There were no studies in Gypsy, Roma and Traveller populations, or individuals with mental health disorders and only one in sex workers. Barriers to treatment adherence included a lack of knowledge around TB, lack of general health or social support and side effects. Facilitators included health education, trusted relationships between patients and healthcare staff, social support and reduced treatment duration.

The evidence base is limited, and few controlled studies exist. Further high-quality research in well-defined underserved populations is needed to confirm the limited findings and inform policy and practice in TB management. Further qualitative research should include more people from underserved groups.