RESUMO
Objectives: The objective of this study was to estimate the retention rate of patients in an ART program and identify the predictors of attrition. Methods: This was a historical cohort study of HIV patients who started ART between September 2007 and April 2020, and were followed up on for at least 6 months in nine large-volume sites. Kaplan Meier techniques were used to estimate cumulative retention and attrition probabilities. Cox proportional hazards models were used to identify predictors of attrition. Results: The cumulative probability of retention at 12 and 24 months was 76.2% and 70.2%, respectively. The attrition rate after a median follow-up time of 3.1 years was 35.2%, or an incidence of 11.4 per 100 person-years. Having initiated ART between 2012 and 2015; unmarried status; having initiated ART with CD4 count <100 cells/µL; and having initiated ART at an advanced clinical stage were factors significantly associated with attrition. Conclusion: The retention rate in our study is much lower than the proposed national target (90%). Studies to understand the reasons for loss to follow-up are needed.