ABSTRACT The primary goal of this F31 proposal is to characterize the social and behavioral determinants of non-retention, including the influence of caregiver support, and to determine trajectories of depression and adherence self- efficacy among youth living with HIV (YLH)in Kenya. There are an estimated 3.4 million adolescents and young adults aged 10-24 years living with HIV globally. Compared to other age-groups, adolescents are less likely to be linked to care, retained, and achieve viral suppression, with many disengaging in care leading to poor clinical outcomes and onward transmission of HIV at the community-level. There are distinct subgroups of YLH that likely have distinct retention determinants. Younger YLH (10-14) often have parental support in HIV management while older YLH (15-19) desire increased autonomy in health decision-making; young adults (20-24) may have long-term partners who influence retention. The proposal will leverage data from a large 24-site cluster randomized trial in Western Kenya (DiSC; NCT#05007717; MPIs: Drs. Grace John-Stewart and Pamela Kohler) among 1904 YLH and 746 caregivers. Aim 1 will explore age-specific demographic, social, behavioral, and clinical cofactors of non-retention and loss-to-follow-up (LTFU) among YLH during 12-month follow-up, including adherence self-efficacy, social support, substance use, depression, and anxiety using generalized linear mixed effects models. Cofactors of LTFU will be estimated using Cox proportional hazards regression. Aim 2 will determine trajectory patterns of depression and adherence self-efficacy and predictors of trajectory groups using group-based trajectory modeling. Aim 3a will explore the relationship between caregivers’ depressive and anxiety symptoms and adolescent HIV outcomes including non-retention, ART adherence, and viral non- suppression. These associations will be modeled using generalized linear mixed effects models. Aim 3b will examine how caregivers’ mental health impacts caregiver motivation and behaviors to support adolescents in staying adherent to ART and retained in care. Findings from this project will advance our understanding of non- retention among YLH and contribute to interventions to optimize retention among YLH. This proposal will provide the F31 candidate with rigorous predoctoral training, including 1) advanced epidemiologic statistical methods including longitudinal data analyses, 2) experience with developing and leading a mixed methods study, 3) content expertise in retention and mental health among youth living with HIV, and 4) understanding of caregivers’ roles in supporting YLH.