Prospective randomized two-Arm controlled study to determine the efficacy of a specific intervention to improve long-term adherence to highly active antiretroviral therapy.

Background: Nearly perfect compliance seems to be indispensable to obtain the maximum benefit from highly active antiretroviral therapy (HAART). Interventions to ensure a high level of adherence during a relatively long-term period of therapy are necessary.

Methods: This is a prospective, randomized, two-arm controlled study including patients starting their first-or second-line HAART who were randomized to receive psychoeducative intervention to implement adherence (experimental group [EG]) or a usual medical follow-up (control group [CG]). We aimed to study the efficacy of a psychoeducative intervention to ensure long-term adherence to HAART, its relation with the virologic efficacy of treatment, and to determine the variables related to long-term adherence. Visits were made at weeks 0, 4, 24, and 48 for data collection. Self-reported adherence was registered at each visit and its veracity was tested by randomized blood analyses performed without previous warning to 40% of patients. Appropriate adherence was defined as the consumption of >/=95% of medication prescribed. Statistical analyses were performed both by the as treated (AT) and the intention to treat missing = failure (ITT) methods.

Results: In all, 116 patients were included. At week 48, 94% of patients in the EG versus 69% controls achieved adherence >/=95% (p =.008); 89% of patients in the EG versus 66% controls had HIV-1 RNA levels 400 copies/ml (p =.026). Overall, 85% of patients with adherence >/=95% but only 45% of those with adherence 95% had viral load (VL) 400 copies/ml (p =. 008). In multivariate analysis, variables significantly related to adherence were having received a psychoeducative intervention (odds ratio [OR], 6.58; p =.04), poor effort to take medication (OR, 5.38; p =.03), and high self-perceived capacity to follow the regimen (OR, 13.76; p =.04). Self-reported adherence and drug plasma levels coincided in 93% of cases. However, differences in adherence did not reach statistical significance in the ITT analysis although a clear tendency toward benefit was observed in EG.

Conclusion: In all, 116 patients were included. At week 48, 94% of patients in the EG versus 69% controls achieved adherence >/=95% (p =.008); 89% of patients in the EG versus 66% controls had HIV-1 RNA levels 400 copies/ml (p =.026). Overall, 85% of patients with adherence >/=95% but only 45% of those with adherence 95% had viral load (VL) 400 copies/ml (p =. 008). In multivariate analysis, variables significantly related to adherence were having received a psychoeducative intervention (odds ratio [OR], 6.58; p =.04), poor effort to take medication (OR, 5.38; p =.03), and high self-perceived capacity to follow the regimen (OR, 13.76; p =.04). Self-reported adherence and drug plasma levels coincided in 93% of cases. However, differences in adherence did not reach statistical significance in the ITT analysis although a clear tendency toward benefit was observed in EG.