Association Between Patient Portal Access and Viral Suppression Among People Living with HIV in a Large Southeastern Clinical Cohort

BACKGROUND: Viral suppression (VS) among people living with HIV (PLWH), the goal of the HIV care continuum, leads to improved patient outcomes and decreased HIV transmission. Patient portals are online tools that enable patient interaction with healthcare systems and may increase patient engagement and improve health outcomes. We examined whether portal access was associated with VS among PLWH. METHODS: We conducted an observational cohort study among PLWH aged ≥18 years who had ≥1 HIV healthcare provider visit at the Vanderbilt Comprehensive Care Clinic (Nashville, Tennessee) from January 1, 2011–December 31, 2015. Patient portal access was defined as being registered for a portal account at any point in the year prior. VS was defined as having ≥1 viral load (VL) measured and the last VL ≤200 copies/ml within a given year. The adjusted relative risk (aRR) of VS was estimated with modified Poisson regression and robust standard errors for multiple outcomes per individual. Models were adjusted for all covariates in the Figure and for year since first kept appointment. Missing data were multiply imputed. RESULTS: The study population included 4,237 PLWH; median age was 43 years (IQR 33–50), 78% were male, 41% were black, and 60% reported male–male sexual contact (MSM). Of the 57% who had portal access during the study period, median age was 42 years (IQR 31–49), 86% were male, 30% were black, and 75% were MSM. In adjusted analysis, portal access was independently associated with improved VS (aRR = 1.19, 95% CI 1.16–1.21 vs. no portal access) (Figure). Increasing age and sexual contact (vs. injection drug use) remained associated with improved VS; black race (vs. white race), lower socioeconomic status, and higher baseline VL remained associated with poor VS after accounting for portal access (Figure). CONCLUSION: Portal access was independently associated with improved VS, although sociodemographic disparities in VS persisted. Additionally, there were sociodemographic disparities in patient portal access. There may be important unmeasured confounders such as health literacy and educational attainment. Additional prospective studies are needed to determine whether patient portal access leads to improved VS among PLWH. DISCLOSURES: P. Rebeiro, NIH: Grant Investigator, Research grant; G. Jackson, Vanderbilt Center for Effective Health Communication: Grant Investigator, Research grant; Agency for Healthcare Research and Quality: Grant Investigator, Research grant; American Medical Informatics Association: Board Member, Research support; A. Pettit, NIH/NIAID - K08AI104352: Grant Investigator, Research grant