Adherence to antiretroviral therapy in resource-limited settings: everything matters

Adherence is one of the major determinants of successful outcomes to antiretroviral therapy (ART) among HIV-infected individuals. In several studies, poor adherence, especially at levels < 95%, has been found to be associated with development of viral resistance, virological failure, progression of disease, and death [1]. Adherence is particularly important in resource-limited settings. This is mainly because of poor access to care and monitoring and limited availability of alternative regimens. Despite numerous studies of outcomes with ART in resource-limited settings, surprisingly few have addressed adherence. Those that have report mostly encouraging results, with adherence rates similar to those observed in developed countries, although notable declines among individuals on more long-term therapy are beginning to be observed [2]. Interesting findings concerning barriers to adherence have also been reported in these studies. More commonly, financial constraints and disruptions in the supply of medications are reported as the major barriers to adherence in resource-limited settings [3–5], contrasting significantly with the barriers typically reported in developed countries, such as side effects, stigma, or inconvenience. The findings should raise considerable concern given the uncertainty of long-term donor commitments in resource-limited settings and eventual probability of patients having to pay at least some if not all costs of therapy. The importance of financial stability and a sustainable drug supply are again highlighted in this issue of AIDS by Oyugi et al. [6]. The study, undertaken in Uganda, evaluated adherence, immununological, and virological outcomes in 97 ART-naive HIV-infected Ugandans initiating self-pay ART with two frequently used first-line generic regimens, Triommune and Maxivir. Although the cohort was relatively small, significant strengths of the study included the intent-to-treat analysis, minimal patient loss to follow-up, low mortality, and the use of superior measures of adherence. As expected, overall adherence rates were initially excellent, with mean levels ranging from 82 to 95% across all measures during follow-up. Adherence correlated well with clinical outcomes with 90–100% of those with 96% adherence achieving maximal viral suppression. Other less encouraging findings, however, concerned the issue of treatment interruptions. In over half of the participants with continuous EMM data, therapy had been interrupted for > 48 h. Almost half of the interruptions were reported to result from the costs of therapy (40% of participants with treatment interruptions cited financial difficulty obtaining medications compared to only 15% of those without interruptions). Interruptions in therapy, in turn, were found to be highly predictive of the development of resistance (13% of individuals with treatment interruptions and 0% of individuals with continuous treatment). Although the effect of interruptions on other clinical outcomes were not assessed in this study, several other studies have demonstrated significant associations between interruptions to therapy and poorer outcomes, giving us significant cause to be concerned [7]. Disruptions in drug supply also contributed significantly to treatment interruptions in this study. Disruptions in ART supply have been frequently observed in resource-limited settings and may occur for a number of reasons. These include weak drug procurement and distribution practices, delayed shipments, manufacturer stock outs, and inaccurate quantification estimates. Interruptions in drug supplies have been shown to be a significant cause of unplanned treatment interruptions in several other studies in resource-limited settings [8], as well as a significant cause of drug resistance and antiretroviral treatment failure [9]. Regardless of the causes in the relative short span of this study, adherence declined significantly over time and this was associated with treatment failure. Clearly, the ability to remain on therapy is important to ensure optimal clinical outcomes in HIV. As illustrated in this and other studies, treatment interruptions caused by external factors such as drug costs and availability may constitute more of a risk to success of therapy in resource-limited settings than an individual's own level of adherence. Addressing the issue of ‘non-adherence’ therefore may require somewhat different approaches to those practiced in the resource-rich areas. As stated by the authors, interventions to ensure a steady and reliable access to ART may be more important than behavioral interventions to ensure optimal adherence to therapy and successful clinical outcomes in resource-limited settings. Increasing efforts should, therefore, be tailored towards them.