Incidence, Predictors and Reasons of Attrition of Patients on Antiretroviral Therapy for HIV in Eight Large-Cohort Sites in Conakry

Introduction: The purpose of this study was to estimate the incidence of attrition (death and lost to follow-up) among patients living with HIV on ART and to identify key predictors of this attrition. It also described the reasons why some patients are lost to follow-up. Methods: This was a historical cohort study of patients living with HIV put on ART between January 1, 2015 and December 31, 2020 in 8 large cohort sites in Conakry. An additional cross-sectional survey in the form of an investigation was conducted to describe the final status of patients reported lost to follow-up by the sites, as well as to describe the reasons for their loss to follow-up. Kaplan Meier techniques were used to estimate cumulative incidence, and the multivariate Cox proportional model was used to identify predictors of attrition. Results: The cumulative incidence of attrition was 19.50 over a median follow-up time of 2.5 years, for an overall attrition rate of 7.79 years per 100 person-years. Factors significantly associated with attrition were: Age 15 - 24 years [aHR = 2.212; 95% CI (1.321 - 3.704)], age >35 years [aHR = 1.723; 95% CI (1.041 - 2.852)], viral load >100,000 copies/ml [aHR = 2.056; 95% CI (1.668 - 2.534)], patients not on the 3-month or 6-month appointment system [aHR = 3.031; 95% CI (2.603 - 3.531)]. Conclusion: This study showed that the incidence of attrition increases with increasing follow-up time. Investigation of lost to follow-up reduced the estimated number of patients considered lost to follow-up and increased the number of deaths that were previously underreported. A prospective mixed study including many more variables would allow a better understanding of the attrition phenomenon among people living with HIV on ART in Guinea.