Hepatitis C virus (HCV) is a major health burden for the US population in general, and one that is difficult to treat among homeless persons. Studies among Veterans indicate HCV may be 4 times more prevalent among homeless compared to housed Veterans. The homeless are at elevated risk for HCV due to high rates of injection drug use and the associated needle sharing.
1. Describe how homeless HCV+ Veterans are distributed among VA medical centers.
2. Evaluate quality of care for homeless and non-homeless Veterans as measured by progress along the HCV Care Continuum and the relationship of quality to patient, facility, and housing characteristics.
3. Develop an intervention to improve HCV care for homeless Veterans, and submit a proposal to pilot test the intervention.
Overview: This is a mixed-methods study. It begins with retrospective cohort analysis of Veterans in VHA care in FY2009-2016. Each stage of the HCV Care Continuum will be evaluated with regression modeling. Later portions of the study involve site visits and qualitative data collection and analysis, leading to the development of an intervention.
For Aim 1 we will calculate the proportion of homeless Veterans who are HCV positive at each of the 128 VAMCs. We will assess variation in facility level HCV prevalence in the homeless with estimation of the mean, variance, standard error, skew, kurtosis, median and median absolute deviation. Inferentially, we will also assess the variance among VA sites via an unadjusted random-intercept logistic mixed model to assess unadjusted variance in HCV prevalence in homeless patients that is attributable to VA stations (j = 128).
For Aim 2 we estimate a separate regression model for each stage of the HCV Care Continuum. (1. tested vs. not tested, then 2. linked to care vs. not linked to care, then 3. retained in care vs. not retained in care, then 4. treated vs. not treated, and 5. achieving sustained viral response (SVR) vs. not achieving SVR) Each model will be estimated as a mixed-effects logistic regression model predicting binary outcomes for each stage of the HCV care continuum. In preparation for Aim 3 we will conduct risk adjustment in order to profile VAMCs from highest to lowest performing for each step in the HCV Care Continuum. This will be used to select VAMCs for site visits.
For Aim 3 we will use qualitative methods. We will visit 6 sites (3 high and 3 low performing, on the HCV Care Continuum). At each site we will conduct ethnographic observation of care processes, as well as conducting in-depth semi-structured interviews with key informants, including clinicians, staff, and patients. Data will be analyzed to identify practices, barriers, and facilitators associated with clinic performance on the HCV Care Continuum. These findings will inform development of an intervention.
We have some preliminary findings regarding prevalence of hepatitis C (HCV). We constructed a measure of homelessness to examine differences in HCV prevalence across three categories of homeless Veterans and across Veterans Affairs Medical Center (VAMC) facilities nationwide. We used VA administrative data to classify a cohort of 730,424 Veterans as at-risk of, currently or formerly homeless and examined variation in HCV prevalence using descriptive measures and mixed-effects logistic regression models. HCV prevalence was highest among formerly homeless Veterans, nearly 14%, followed by currently homeless Veterans, approximately 9%, and Veterans at-risk of homelessness approximately 6%.
None to date as of June 28, 2018
External Links for this Project
Grant Number: I01HX001767-01A2
- McInnes DK, Troszak LK, Fincke BG, Shwartz M, Midboe AM, Gifford AL, Dunlap S, Byrne T. Is the Availability of Direct-Acting Antivirals Associated with Increased Access to Hepatitis C Treatment for Homeless and Unstably Housed Veterans?. Journal of general internal medicine. 2022 Apr 1; 37(5):1038-1044. [view]