INTRODUCTION Cuba has an HIV prevalence of 0.1% in the population aged 15 to 49 years, very low despite increased incidence in recent years. In 2001, domestically-produced generic antiretroviral therapy was introduced and there has been complete coverage since 2003. In 2006, 1986 people with HIV/AIDS were receiving ART; by 2009, that figure reached 5034. Adherence to antiretroviral therapy is fundamental: nonadherence leads to treatment failure, development of resistance, progression to AIDS, and death. OBJECTIVE Measure levels of treatment adherence and its predictive factors in persons with HIV/AIDS receiving antiretroviral therapy in 2006 in Cuba. METHODS A cross-sectional study was carried out in 2006 of Cuban HIV-positive individuals receiving antiretroviral therapy. A sample size of 876 was calculated using two-stage sampling (first by strata, and then by simple random sampling in each stratum). An anonymous structured questionnaire was administered to participants. Reporting of doses taken on each of the three days and in the week preceding the survey was recoded into five categories. Participants were considered highly adherent if they reported taking ≥95.0% of their medication as prescribed. Reasons for nonadherence were described and logistic regression modeling used to develop hypotheses on associations between high adherence and its predictive factors. RESULTS Interviews were obtained with 847 individuals, 70.6% of whom self reported high adherence. There were no significant differences between highly adherent and less adherent patients with regard to sex, place of residence, treatment setting, time of diagnosis, or length of treatment. Variables associated with high adherence were communication with the specialist physician, change in treatment, memory, self-efficacy, as well as commitment to and opinions about treatment. CONCLUSIONS In Cuba, where treatment is free of charge to patients, adherence is good. Treatment adherence might be improved by achieving a closer doctor-patient relationship; taking measures to motivate patients and promote self-efficacy and commitment to treatment; publicizing treatment outcomes; and providing assistance to patients to help them remember their medication schedule. Further studies are required to determine current adherence levels; and longitudinal research to determine adherence over time.