Methods: We conducted a retrospective cohort study to determine HCV incidence among ED adult patients. The study ED rolled out an ED-based HCV screening program since November 2015. A secondary data analysis was performed from a seroprevalence study on all adult patients who visited the study ED during December 10, 2015 and January 21, 2016. Patients who had at least two HCV antibody testings from two separate visits at the study hospital from 2003 to 2016 were included for this secondary data analysis. Patients who had reactive HCV antibody result at the first time point were excluded. Follow-up time (PYs) was calculated for each patient by the interval of between two HCV antibody tests. Time of HCV seroconversion was defined as the midpoint between the negative and positive HCV antibody test. Incidence rate ratio (IRR) and corresponding 95% CI was used to present the relative incidence between groups by mid-p exact test.
Results: A total of 302 ED patients were identified. The majority of them were female (60%), African American (79%), aged 35 years and older (60%). 68% of patients were born after 1965 (68%) and 25% born between 1945 and 1965 (birth cohort). 56% of patients had commercial insurance payor and 36% had Medicaid or Medicare. 34 (11%) patients had HIV infection and 7 (2%) were injection drug users (IDU). Overall, 6 (2%) had HCV seroconversion during 971.1 PYs, resulting in an HCV incidence of 6.2 per 1,000 PYs (95% CI: 2.5, 12.9 per 1,000 PYs). The incidence was significantly different by race [white: 30.9/1,000 PYs, African American: 2.9/1,000 PYs; RR: 12.3 (2.2, 95.8)] and IDU [IDU: 192.3/1,000 PYs, non-IDU: 4.2/1,000 PYs; RR: 46.2 (5.9, 260.3)] but not by birth cohort or HIV status.
Conclusion: The HCV incidence in urban ED patients was over 60 times higher than the general US population and even higher in some high risk groups, indicating ED is a critical venue for identifying high risk individuals for HCV prevention and detecting HCV-infected Americans for treatment.
D. Thomas, None
Y. H. Hsieh, None