Retention and predictors of loss-to-follow-up among patients on antiretroviral therapy in the Test-and-Treat Era: evidence from a retrospective cohort study in Maputo, Mozambique

Retention in antiretroviral therapy (ART) is critical for HIV treatment success. Evidence on the impact of Test-and-Treat (TT) strategy in high-burden urban settings remains limited. We evaluated TT implementation effect on ART patients’ retention and identified predictors of loss-to-follow-up (LTFU) in Maputo, Mozambique. A retrospective cohort study (2013–2020) was conducted among ART patients aged 15–49 years using data from 12 health facilities (HFs). Patients were grouped into pre-Test-and-Treat (BTT, 2013–2016) and post-Test-and-Treat (ATT, 2017–2020) cohorts for comparative purposes. Retention probabilities were estimated using Kaplan–Meier methods, and predictors of loss to follow-up (LTFU) were assessed using Cox proportional hazards models. Population attributable fractions (PAFs) were calculated to quantify the population-level impact of Test-and-Treat implementation. Among 16,968 patients (9,036 BTT; 7,932 ATT), 7,908 experienced LTFU over 487,958 person-months. Loss to follow-up incidence decreased from 17.3 to 15.0 per 1,000 person-years in BTT versus ATT, a 13.3% reduction. Median retention time (MRT) increased from 39.8 to 49.0 months. Test-and-treat reduced LTFU hazard by 25% (adjusted PAF 23%). Higher LTFU risk was associated with male sex, WHO stage III/IV, HIV/TB co-infection, second-line ART, and non-membership in community ART groups (CAGs). Test-and-Treat substantially improved ART retention, particularly when combined with differentiated service delivery models (DSDMs). Targeted interventions for high-risk groups may further enhance retention. Future research should assess TT effects in specific subpopulations and account for undocumented deaths and self-transfers.