3309 Clinic-Level Factors and Retention in Care among People Living with HIV (PLWH) in a United States (US) Multi-Site Cohort, 2010-2016

OBJECTIVES/SPECIFIC AIMS: The National HIV/AIDS strategy aims to increase retention in care (RIC) to reduce HIV transmission and mortality. Previous studies have evaluated clinic-level interventions such as appointment reminders and peer navigation. However, few studies have investigated the association between multiple clinic-level factors and RIC among PLWH across the United States. We conducted a multi-site cohort study to identify clinic-level factors associated with RIC in the United States from 2010-2016. METHODS/STUDY POPULATION: We included PLWH with at least one HIV primary care visit from 2010-2016 at seven sites of the Center for AIDS Research (CFAR) Network of Integrated Clinical Systems (CNICS). Individual-level data collected as part of routine clinical care were abstracted from the electronic health record. Clinic-level data were gathered through a survey and included questions on site characteristics (i.e. clinic volume) as well as services available at the site during each year of the study: peer navigation, RIC posters/brochures, laboratory test timing, flexible scheduling, appointment reminder types, and stigma support services defined as intensive HIV education and advocacy regarding support to address stigma at outreach visits. RIC was defined as ≥2 encounters per year, ≥90 days apart, observed until death, administrative censoring (December 31, 2016), or loss to follow-up (no visit for >12 months with no future visits). Poisson panel regression with robust error variance, clustering by site and adjusting for calendar year, age (modeled with a cubic spline), sex, race/ethnicity, and HIV transmission risk factor, was used to estimate incident rate ratios (IRR) with 95% confidence intervals (CI) for RIC. Clustering by site has been shown to absorb for clustering that could occur at the individual level as well. RESULTS/ANTICIPATED RESULTS: Among 21,046 patients contributing 103,348 person-years, 67% of person-years were RIC. Text appointment reminders (IRR = 1.13; 95% CI: 1.03-1.24) and stigma support services (IRR=1.11; 95% CI:1.04-1.19) were significantly associated with RIC. RIC disparities in individual-level patient characteristics were observed even after accounting for clinic-level characteristics. Older patients were more likely to be RIC demonstrated through year comparisons due to the use of a spline; for age 50 years (IRR = 1.07, 95% CI: 1.06-1.08) and 60 years (IRR = 1.15, 95% CI: 1.13-1.17) compared to 45 years. Female PLWH were more likely to be RIC compared to males (IRR = 1.03, 95% CI: 1.02-1.05) and Hispanic PLWH were more likely to be RIC compared to white, non-Hispanic PLWH (IRR = 1.09, 95% CI: 1.05-1.13). Although commonly found to be associated with worse RIC, Black race and injection drug use were not associated with RIC in this population. DISCUSSION/SIGNIFICANCE OF IMPACT: In this multi-site US cohort study from 2010-2016, availability of text appointment reminders and stigma support services at a clinic were associated with approximately 10% higher probability of RIC than at clinics without those services. RIC disparities persisted with respect to individual-level characteristics of age, sex, and race/ethnicity even after accounting for these clinic-level factors. Prospective studies examining the impact of these clinic-level factors and individual-level uptake of these services on RIC are needed.