Evaluating adherence to antiretroviral therapy using pharmacy refill records in a rural treatment site in South Africa

Abstract

Optimal adherence to combination antiretroviral therapy (cART) is critical to maintain virologic suppression, thereby ensuring the global success of HIV treatment. We evaluated adherence to cART using pharmacy refill records and determined the adherence threshold resulting in >90% virologic suppression in a community run treatment site in South Africa. Additionally, we analysed factors associated with adherence using univariable and multivariable logistic regression models. Logistic regression was also performed to determine the relationship between adherence and virologic suppression and the adherence threshold resulting in <10% virologic failure. The overall median (interquartile range) adherence was 95% (88.6-98.4%). Out of the study participants, 210/401 (52.4%) had optimal (≥95%) adherence while only 37/401 (9.2%) had poor (≤80%) adherence. The majority (90.5%) of patients with optimal adherence had virologic suppression. Having TB at registration into care was found to be negatively associated with adherence (adjusted odds ratio [AOR], 0.382; p≤.05). Compared to nonadherent individuals, optimally adherent participants were more likely to achieve virologic suppression (OR 2.92; 95% CI: 1.63-5.22). Only adherence rates above 95% were observed to lead to <10% virologic failure. cART adherence measured by pharmacy refill records could serve as a useful predictor of virologic failure; adherence rates >95% are needed to maintain optimal virologic suppression.