2207. Barriers to HCV Treatment in a Safety-Net Hospital System

BACKGROUND: In 2016, we implemented a hepatitis C (HCV) screening program for baby boomers (BB) born between 1945 and 1965) using a best practice alert (BPA) in the electronic medical record and patient navigation (PN) in our safety-net hospital system. We now examine barriers to HCV treatment among those who received PN for linkage to care (LTC). METHODS: The BPA prompts providers to order a HCV antibody (Ab) for any unscreened BB who has an outpatient appointment. Those with HCV Ab+ with a confirmatory RNA receive telephone navigation, using a pre-defined script, if LTC did not occur within 2 months of RNA testing. After LTC, a person was considered as untreated if HCV treatment had not occurred within 1 year of initial visit. Insured patients received treatment through prior authorizations and uninsured through pharmaceutical patient assistance programs. We examined demographics, homelessness, insurance, fibrosis score, substance use, and psychiatric illness, as potential predictors to treatment initiation using univariate and multivariate logistic regression analysis. RESULTS: Among the 16,363 BBs screened from March 1, 2016 to December 31, 2017, 1,445 (8.8%) were HCV Ab+ and 1,038 (72%) had HCV RNA completed. Among the 724 (5%) with confirmed HCV infection, 139 (19%) received LTC without navigation, 299 (41%) received navigation, and 286 (40%) could not be contacted after three attempts. Among those who received navigation, 225 (75%) completed a follow-up visit of which 81 (36%) did not start treatment, 34 (15%) are awaiting treatment initiation, and 110 (49%) started treatment. Gender, race/ethnicity, psychiatric illness, and homelessness were not predictive of starting HCV treatment. In univariate analysis, current substance use vs. none/past use (OR 0.52 (0.29, 0.93)) was associated lower likelihood of starting treatment and advanced fibrosis (OR 2.25 (1.20, 4.21)) was associated with higher likelihood of starting treatment). Compared with uninsured patients, Medicaid patients were less likely to start treatment (AOR 0.15 (0.67, 0.34)) in a multivariate analysis. CONCLUSION: Insurance status was independent predictor of starting treatment among patients at our safety-net hospital. Medicaid remains a barrier to HCV treatment access in safety-net systems. DISCLOSURES: A. Singal, Gilead Sciences: Grant Investigator, Research support. M. K. Jain, Gilead Sciences: Grant Investigator, Grant recipient and Research support. Janssen: Investigator, Research support. GSK/ViiV: Investigator, Consulting fee and Research support. Merck: Investigator, Research support