<![CDATA[Introduction: Tuberculosis (TB) treatment requires sustained medication use, but adherence can be influenced by cognitive, psychosocial, and healthcare-related factors. Evidence on structured self-talk for supporting TB medication adherence remains limited. This study examined factors associated with adherence and whether assignment to the structured self-talk intervention was associated with short-term improvement in self-reported adherence. Methods: A sequential quantitative design was used. Phase I included 117 patients with pulmonary TB in a cross-sectional study assessing TB knowledge, access to health facilities, self-efficacy, and family support. Phase II used a quasi-experimental pretest–posttest control-group design with 68 patients. Participants were allocated according to clinic attendance schedules. The structured self-talk intervention involved twice-weekly practice for 4 weeks. Phase I data were analysed using Spearman's rank correlation and binary logistic regression. Between-group changes in the study-specific modified 11-item medication adherence scores in Phase II were compared using the Mann–Whitney U test; ANCOVA was used for the adjusted analysis. Results: In Phase I, TB knowledge, access to health facilities, self-efficacy, and family support were positively associated with medication adherence in bivariate analyses (all p < 0.001). After adjustment, only self-efficacy (OR = 2.34; 95% CI: 1.41–3.89) and family support (OR = 3.97; 95% CI: 2.16–7.31) remained significantly associated with good adherence. In Phase II, the primary non-parametric between-group analysis showed a Hodges–Lehmann location shift of 3.00 points (95% CI: 2.00–4.00; p < 0.001). ANCOVA yielded an adjusted between-group coefficient of 3.08 points for the post-intervention study-specific modified 11-item medication adherence score (95% CI: 1.85–4.31; p < 0.001). Conclusion: Cognitive, psychosocial, and contextual factors were associated with TB medication adherence. Assignment to the structured self-talk intervention was associated with greater short-term improvement in self-reported adherence. However, interpretation should consider the non-randomised design, self-reported outcome measurement, and short follow-up period. Randomised studies with longer follow-up and objective adherence measures are warranted.]]>
