Background: With African health-care systems facing exploding demand for HIV care, reliable methods for assessing adherence and its influencing factors are needed to guide effective public-health measures. This study evaluated individual patient characteristics determining antiretroviral treatment (ART) adherence and the predictive values of different measures of adherence on virological treatment failure in a cohort of patients in a routine-care setting in Cameroon.

Methods: Longitudinal study over 6-months following ART introduction, using patients questionnaires and hospital and pharmacy records.

Results: At the end of the 6 months study period, 219 of 312 patients (70%) returned to the pharmacy to refill their medication, 17% (51) were lost to follow-up, 9% (28) were dead and 4% (14) were transferred to other care centres. Virological treatment failure at 6 months was experienced by 26 patients, representing 13% of patients with available viral load value. Pharmacy refill irregularity was the most powerful predictor (odds ratio 12.4; P < 0.001) of virological treatment failure, compared with CD4 cell count increase at 6 months (odds ratio 7.8; P = 0.002) or self-reported adherence at one month (odds ratio 1.1; P = 0.85). Low intensity of ART side-effects after one month was strongly associated with survival (odds ratio 0.11; P = 0.001). Patients starting ART with CD4 cell count <100 cells/mm3 had a greater risk of dying during the follow-up period (odds ratio 2.69; P = 0.02). Compared with asymptomatic CDC stage A patients, CDC stage B (odds ratio 5.72) and CDC stage C patients (odds ratio 16.9) had higher risk of becoming lost to follow-up (P < 0.001). In the multivariate analyses, pharmacy non-adherence was less frequent in women (adjusted odds ratio 0.56; P = 0.05) but more frequent in patients with high monthly income (odds ratio 3.24; P = 0.04).

Conclusion: Pharmacy-refill adherence might be considered as an alternative to CD4 count monitoring for identification of patients at risk of virological failure, especially in resources-scarce countries. The study confirmed the difficulty in demonstrating clear associations of individual patient factors and treatment outcomes. The substantial loss to follow-up and deaths occurring within 6 months after initiating ART emphasise the need to understand the best timing of ART initiation and further elucidate and educate on the underlying reasons for delaying initiation of ART in resource-limited countries.