TB Research

Baseline determinants of adherence for drug-sensitive TB treatment in a South African prospective cohort: a focus on HIV infection and anti-retroviral therapy, clinical care access, and TB stigma.

Adrian Steulet, Piotr Hippner, Noriah Maraba, Lauren Jennings, Israel Rabothata, Rachel Mukora, Nokhanyo Xaba, Lihle Mchunu, et al. (12 authors)

BMC infectious diseases · 2026-01

Abstract

BACKGROUND: Inadequate adherence to tuberculosis (TB) treatment increases the risk of treatment failure and recurrence. Identifying factors contributing to poor adherence could refine targeting strategies and optimize resource distribution. We examined specific individual-level factors for TB treatment adherence among adults with drug-sensitive TB, including HIV status, antiretroviral therapy, time to access clinical care, and perceived stigma.

METHODS: Data are from the &#x201c;TB Mate&#x201d; cluster-randomized trial, which evaluated a TB treatment adherence intervention across 18 public health clinics in South Africa (PACTR201902681157721). Treatment adherence was measured using smart pillboxes, with pillbox opening recorded as a proxy for the dose taken. Adults in the control group, utilizing the pillbox in silent mode, were included in this analysis. We used logistic regression to model poor adherence (<&#x2009;80% pillbox engagement), and negative binomial regression to model adherence as a count of pillbox engagement. Directed acyclic graphs informed confounder selection in the models.

RESULTS: Among 1,213 participants (nine clinics) in the control group, 61.2% (742) were male, the median age was 36 years, 63% (769) were living with HIV, with 66% (507/769) on antiretroviral therapy. The median time to access clinical care was 127&#xa0;min and 95% (1151/1213) reported no perceived stigmatization upon starting TB treatment. Overall 51% (614) exhibited adherence below 80%, with a geometric mean pillbox engagement of 59.6%. Living with HIV was associated with poor adherence to TB treatment, with an adjusted odds ratio of 1.68 (95% confidence interval [CI] 1.27&#x2013;2.22) for <&#x2009;80% adherence and an adjusted rate ratio of 0.90 (0.83&#x2013;0.97) for pillbox engagement, compared to being HIV-negative. Antiretroviral therapy, time to clinical care access, and perceived stigma showed no association with either measure of adherence.

CONCLUSIONS: The low adherence underscores the necessity for TB treatment support interventions, particularly among those living with HIV.

SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12879-025-12304-4.