RESUMEN
Antiretroviral therapy success is highly dependent on the ability of the patient to fully adhere to the prescribed treatment regimen. We present the results of a cross-sectional study that evaluates the predictive value of a self-administered questionnaire of adherence to antiretroviral (ARV) therapy. Study participants were interviewed using a 36-item Patient Medication Adherence Questionnaire (PMAQ) designed to assess knowledge about ARV therapy, motivation to adhere to treatment, and behavioral skills. Plasma HIV-1 RNA levels were correlated with the results obtained from the PMAQ. Of the 182 study participants, 82 (45%) were receiving their initial ARV regimen. Of the remaining patients, 39 (21%) and 61 (34%) were on a second or additional ARV regimen, respectively. An undetectable viral load was documented in 47/62 (76%) patients on their first regimen who reported missing medication on less than 4 days in the last 3 months. The PMAQ had a higher predictive value of plasma viral suppression for patients in the initial regimen than for patients in salvage therapy. The overall predictive value of the PMAQ to identify adherence was 74%, and 59% for nonadherence, with an overall efficacy of 64%. Of the 74 patients (45%) who did not understand the concept of antiretroviral therapy, 80% were failing or had previously failed the ARV treatment. Of 35 patients with doubts about their HIV status or skeptical of the benefits of ARV therapy, 29 (84%) were nonadherent. Despite the positive predictive value of PMAQ in identifying adherence, self-reported adherence is not a sufficiently precise predictor of treatment success to substitute for viral load monitoring. On the other hand, the use of such an instrument to identify factors associated with nonadherence provides an excellent opportunity to apply early intervention designed to specifically address factors that might be contributing to the lack of adherence prior to regimen failure.