Longitudinal Analysis of Adherence to First-Line Antiretroviral Therapy: Evidence of Treatment Sustainability from an Indian HIV Cohort.

INTRODUCTION

Given the chronic nature of HIV infection and the need for life-long antiretroviral therapy (ART), maintaining long-term optimal adherence is an important strategy for maximizing treatment success. In order to understand better the dynamic nature of adherence behaviors in India where complex cultural and logistic features prevail, we assessed the patterns, trajectories and time-dependent predictors of adherence levels in relation to virological failure among individuals initiating first-line ART in India.

METHODS

Between July 2010 and August 2013, eligible ART-naïve HIV-infected individuals newly initiating first-line ART within the national program at three sites in southern India were enrolled and monitored for two years. ART included zidovudine/stavudine/tenofovir plus lamivudine plus nevirapine/efavirenz. Patients were assessed using clinical, laboratory and adherence parameters. Every three months, medication adherence was measured using pill count, and a structured questionnaire on adherence barriers was administered. Optimal adherence was defined as mean adherence ≥95%. Statistical analysis was performed using a bivariate and a multivariate model of all identified covariates. Adherence trends and determinants were modeled as rate ratios using generalized estimating equation analysis in a Poisson distribution.

RESULTS

A total of 599 eligible ART-naïve patients participated in the study, and contributed a total of 921 person-years of observation time. Women constituted 43% and mean CD4 count prior to initiating ART was 192 cells/mm3. Overall mean adherence among all patients was 95.4%. The proportion of patients optimally adherent was 75.6%. Predictors of optimal adherence included older age (≥40 years), high school-level education and beyond, lower drug toxicity-related ART interruption, full disclosure, sense of satisfaction with one's own health and patient's perception of having good access to health-care services. Adherence was inversely proportional to virological failure (IRR 0.55, 95%CI 0.44-0.69 p<0.001). Drug toxicity and stigma-related barriers were significantly associated with virological failure, while forgetfulness was not associated with virological failure.

CONCLUSION

Our study highlights the overall high level of medication adherence among individuals initiating ART within the Indian national program. Primary factors contributing towards poor adherence and subsequent virological failure in the proportion of individuals with poor adherence included drug toxicity, perceived stigma and poor access to health care services. Strategies that may contribute towards improved adherence include minimizing drug interruptions for medical reasons, use of newer ART regimens with better safety profiles and increasing access with more link ART centers that decentralize ART dispensing systems to individuals.