Lead/Presenter: D. Keith McInnes,
COIN - Bedford/Boston
All Authors: McInnes DK (Center for Healthcare Organization and Implementation Research, Bedford/Boston; Boston University School of Public Health), Byrne, T (Center for Healthcare Organization and Implementation Research, Bedford/Boston, and Boston University School of Social Work), Troszak, LK (Center for Innovation to Implementation, Palo Alto) Midboe, AM (Center for Innovation to Implementation, Palo Alto) Shwartz, M (Center for Healthcare Organization and Implementation Research, Bedford/Boston, and Boston University Questrom School of Business) Fincke, BG (Center for Healthcare Organization and Implementation Research, Bedford/Boston, and Boston University School of Public Health) Montgomery, AE (VA National Center on Homelessness Among Veterans, Birmingham VAMC, and University of Alabama at Birmingham School of Public Health) Gifford AL (Center for Healthcare Organization and Implementation Research, Bedford/Boston, and Boston University School of Medicine)
Objectives:
Approximately 44% of homeless adults are infected with hepatitis C virus (HCV). New, direct acting antivirals (DAAs), have high cure rates and few side effects. We explored whether access to HCV medications in homeless and marginally housed (HMH) populations was ameliorated with the widespread use of DAAs in the VA healthcare system.
Methods:
We used VA medical record data to select 305,979 veterans who tested positive for HCV between 2012 and 2016. We followed them prospectively for up to five years and assessed their housing status (housed, at-risk of homelessness, currently homeless, formerly homeless, and multiple homeless indicators) and receipt of HCV treatment in a series of six-month windows. We estimated the probability of HCV treatment initiation in both the interferon-era (before 1/1/2015) and DAA-era (on/after 1/1/2015), controlling for patient and site level characteristics.
Results:
In the interferon-era, the probability of HCV treatment initiation in a given six-month period among housed veterans (6.2%, 95 % CI: 5.3% - 7.1%) was significantly higher than veterans who were currently homeless (2.6%, 95% CI: 1.9% - 3.3%), formerly homeless (3.9%, 95% CI: 2.5% - 5.2%), or who had multiple indicators of homelessness (2.0%, 95% CI: 1.3% - 2.7%), but not significantly different from veterans who were at-risk of homelessness (3.8%, 95% CI: 1.9% - 5.8%). Housed veterans, and 3 of the 4 homelessness groups, had significantly higher probabilities of initiating HCV treatment in the DAA-era compared to the interferon-era (veterans at-risk of homelessness who showed no significant change); the largest change was for currently homeless veterans, increasing from 2.6% (95% CI: 1.9% - 3.3%) in the interferon-era to 6.3% (95% CI: 5.7% - 6.9%) in the DAA-era. Overall, housed veterans, compared to HMH veterans, maintained a significantly higher probability of treatment initiation in the DAA-era as they had in the interferon-era.
Implications:
DAAs have helped all Veterans, housed and non-housed, access HCV treatment; however housed veterans have greater access than HMH groups, regardless of the treatment era.
Impacts:
The arrival of DAAs make this an opportune time to close the gap in access to HCV treatment for Veterans who are homeless and marginally housed.