Predicting adherence to combination antiretroviral therapy for HIV in Tanzania: A test of an extended theory of planned behaviour model

Objectives: Combination antiretroviral therapy (cART) for HIV is widely available in sub-Saharan Africa. Adherence is crucial to successful treatment. This study aimed to apply an extended theory of planned behaviour (TPB) model to predict objectively measured adherence to cART in Tanzania.

Design: Prospective observational study (n = 158) where patients completed questionnaires on demographics (Month 0), socio-cognitive variables including intentions (Month 1), and action planning and self-regulatory processes hypothesised to mediate the intention–behaviour relationship (Month 3), to predict adherence (Month 5).
Main outcome measures: Taking adherence was measured objectively using the Medication Events Monitoring System (MEMS) caps. Model tests were conducted using regression and bootstrap mediation analyses.
Results: Perceived behavioural control (PBC) was positively (β = .767, p < .001, R2 = 57.5%) associated with adherence intentions. Intentions only exercised an indirect effect on adherence (B = 1.29 [0.297–3.15]) through self-regulatory processes (B = 1.10 [0.131–2.87]). Self-regulatory processes (β = .234, p = .010, R2 = 14.7%) predicted better adherence.

Conclusion: This observational study using an objective behavioural measure, identified PBC as the main driver of adherence intentions. The effect of intentions on adherence was only indirect through self-regulatory processes, which were the main predictor of objectively assessed adherence.