Engagement in the Hepatitis C Care Cascade Among Homeless Veterans, 2015

Abstract

The Veterans Health Administration (VHA) is the largest provider of hepatitis C virus (HCV) care nationally and provides health care to >200 000 homeless veterans each year. We used the VHA’s Corporate Data Warehouse and HCV Clinical Case Registry to evaluate engagement in the HCV care cascade among homeless and nonhomeless veterans in VHA care in 2015. We estimated that, among 242 740 homeless veterans in care and 5 424 712 nonhomeless veterans in care, 144 964 (13.4%) and 188 156 (3.5%), respectively, had chronic HCV infection. Compared with nonhomeless veterans, homeless veterans were more likely to be diagnosed with chronic HCV infection and linked to HCV care but less likely to have received antiviral therapy despite comparable sustained virologic response rates. Homelessness should not necessarily preclude HCV treatment eligibility with available all-oral antiviral regimens.

An estimated 2.7 to 3.9 million people in the United States are chronically infected with hepatitis C virus (HCV),¹ including approximately 230 000 veterans.² The Veterans Health Administration (VHA), the largest national provider of HCV care, provides health care to >200 000 homeless veterans each year.³ Although homelessness among veterans has declined since 2009,⁴ it is a major obstacle to accessing health care for many veterans. In 2012, the US Department of Veterans Affairs (VA) Office of Inspector General reported homeless incidence rates of 3.2% to 4.0% among veterans.⁵

Small studies based on geographically limited data reported rates of HCV antibody positivity among homeless veterans ranging from 6.6% to 44.0% from 1993 to 2000.^6–8 These HCV antibody positivity rates are higher than the estimated rate for the general US population (1.3%).¹ Although chronic HCV infection is presumed to be highly prevalent among homeless veterans,⁹ large comprehensive analyses are lacking.

Afdhal and colleagues initially described the HCV care cascade in 2013 as a framework to describe and address chronic HCV infection on a public health level. The care cascade includes estimates of prevalence (including those undiagnosed but living with HCV), rates of diagnosis of chronic HCV by laboratory testing, linkage to care for people with HCV, receipt of HCV antiviral therapy, and rates of sustained virologic response (SVR) among those treated for HCV.¹⁰ Maier and colleagues used data from the VHA’s Corporate Data Warehouse and HCV Clinical Case Registry to describe the HCV care cascade among veterans in VHA care in 2013, before the widespread use of all-oral regimens of potent direct-acting agents (DAAs), which act to directly inhibit proteins (eg, the NS5A protein) and enzymes (eg, the NS5B RNA polymerase or the NS3/4A protease) that are essential for HCV viral replication. Using the data from the Corporate Data Warehouse, they estimated that 233 898 of 5 596 178 (4.2%) veterans in VHA care in 2013 had chronic HCV infection. Among those, 181 168 (77.5%) received a diagnosis of chronic HCV by laboratory testing, 160 794 (68.7%) were linked to HCV care, 39 388 (16.8%) had received HCV antiviral therapy, and 15 983 (6.8%) had achieved SVR.²

Homelessness was a major barrier to HCV care and treatment in the peginterferon era.^11–14 DAAs have shorter treatment durations, fewer side effects, and higher SVR rates when compared with the older peginterferon-based regimens.¹⁵ Rates of engagement in the HCV care cascade among homeless veterans have notably not been described since the introduction of DAAs. The objective of this study was to describe engagement in the HCV care cascade among homeless and nonhomeless veterans in this new era of HCV treatment.

Methods

Details of the methodology used by the VHA to estimate each step in the HCV care cascade are published elsewhere.² In brief, step 1 was estimating the total number of patients who ever had chronic HCV infection in the population receiving VHA health care in 2015, which was the sum of (1) those who received a diagnosis of chronic HCV infection by laboratory testing, (2) estimated additional cases from the projected prevalence among HCV antibody-positive patients who had not received RNA testing, and (3) estimated additional cases from the projected prevalence among the untested population. Step 2 was estimating the number of patients in VHA health care in 2015 who received a diagnosis of chronic HCV infection—defined as ever having a detectable HCV RNA or genotype—by laboratory testing. Step 3 was estimating the number of patients in VHA health care in 2015 who were linked to HCV care—defined by a health care provider entering HCV on a patient’s electronic medical record Problem List, which lists the patient’s medical conditions, and manual confirmation in VHA’s HCV Clinical Case Registry. Step 4 was estimating the number of patients in VHA health care in 2015 who had ever received HCV antiviral therapy in VHA on or before December 31, 2015. Step 5 was estimating the number of patients in VHA health care in 2015 with SVR, defined by undetectable HCV RNA on a test ≥12 weeks after the end of HCV antiviral therapy. For patients whose SVR status could not be determined (eg, still receiving antiviral therapy as of December 31, 2015), we applied the SVR rate for those whose status was determined.

Veterans who had at least 1 outpatient visit in 2015 to any of the clinics designated by the VHA’s National Center on Homelessness Among Veterans and who received homeless services were included in the homeless cohort. The following clinics were included: Community Outreach to Homeless Veterans, US Department of Housing and Urban Development/VA Supportive Housing, Grant and Per Diem, Health Care for Homeless Veterans/Homeless Chronically Mentally Ill, Homeless Veteran Community Employment Services, and Veterans Justice Outreach. Overall, 5 667 452 patients had a VHA outpatient visit in 2015, including 242 740 homeless veterans. This study was performed as part of VHA health care operations and was considered exempt from institutional review board review, as determined by the VHA Office of Research Oversight.

Results

A total of 5 667 452 patients had a VHA outpatient visit in 2015, including 242 740 homeless veterans. Of these homeless veterans, 189 508 (78.1%) had ever had any VHA HCV testing (HCV antibody or RNA-based testing). The estimated total number of homeless veterans who had ever had chronic HCV infection and who constituted the starting population for the HCV care cascade was then the sum of (1) 29 063 veterans who had received a diagnosis of chronic HCV infection by RNA laboratory testing, (2) 543 estimated additional veterans from the projected prevalence of 82.6% (derived from the observed prevalence of positive RNA in antibody-positive veterans with RNA testing) applied to the 657 antibody-positive veterans who had not received RNA testing, and (3) 2843 estimated additional veterans from applying the projected prevalence in 3 birth cohorts to those untested (born before 1945: 157 of 3936, 4.0%; born 1945-1965: 2093 of 23 517, 8.9%; born after 1965: 593 of 25 770, 2.3%). The projected prevalence in birth cohorts was calculated from the prevalence of those in care in 2015 and first tested for HCV in 2015, multiplied by the quotient of the prevalence of those in care in 2015 and first tested in 2015, divided by the prevalence of those in care in 2015 first tested for HCV in 2014. Thus, the estimated total number of homeless veterans in VHA care in 2015 who ever had chronic HCV infection was 32 449 of 242 740, for a prevalence of 13.4%.

In similar calculations for the 5 424 712 nonhomeless veterans in VHA care in 2015, 3 227 572 (59.5%) had ever had any VHA HCV testing. The estimated total number of nonhomeless veterans who had ever had chronic HCV infection was then the sum of (1) 144 964 who received a diagnosis of chronic HCV infection by RNA laboratory testing, (2) 4371 estimated additional veterans from the projected prevalence of 73.3% applied to the 5963 antibody-positive veterans who had not received RNA testing, and (3) 38 821 estimated additional veterans from application of the projected prevalence in 3 birth cohorts to those untested (born before 1945: 580 of 929 958, 0.1%; born 1945-1965: 30 579 of 804 722, 3.8%; born after 1965: 2312 of 462 460, 0.5%). Thus, the estimated total number of nonhomeless veterans in VHA care in 2015 who had ever had chronic HCV infection was 188 156, for a prevalence of 3.5% (188 156 of 5 424 712).

Proceeding along the HCV care cascade, 29 063 of 32 449 (89.6%) homeless veterans who were estimated to have chronic HCV infection had been diagnosed with chronic HCV infection by laboratory testing, as compared with 144 964 of 188 156 (77.0%) nonhomeless veterans (Table). Linkage to HCV care was roughly equivalent between homeless veterans (25 786 of 29 063 [88.7%] of those diagnosed with HCV infection) and nonhomeless veterans (136 169 of 144 964 [93.9%] of those diagnosed with HCV infection). The rate of receipt of HCV antiviral therapy among homeless veterans (7421 of 25 786, or 28.8% of those linked to HCV care, including 4012 treated in 2015 with all-oral DAA regimens) was lower than the rate among nonhomeless veterans (58 321 of 136 169, or 42.8% of nonhomeless veterans linked to HCV care). Thus, the percentage of the total homeless population with chronic HCV infection who had ever received HCV antiviral therapy (7421 of 32 449, 22.9%) was lower than the percentage of the total nonhomeless population who had ever received HCV antiviral therapy (58 321 of 188 156, 31.0%). However, the cumulative SVR rates achieved among homeless veterans who had ever received HCV antiviral therapy (5041 of 7421, 67.9%) and nonhomeless veterans who had ever received HCV antiviral therapy (42 878 of 58 321, 73.5%) were comparable.

Table.

Engagement in the HCV care cascade among homeless and nonhomeless veterans with chronic HCV infection in VHA care, United States, 2015^a

Steps in the HCV Care Cascade	Homeless Veterans With HCV Infectionb			Nonhomeless Veterans With HCV Infection
	No.	%	Engaged in Prior Step of Care Cascade, %	No.	%	Engaged in Prior Step of Care Cascade, %
Estimated No. of veterans with HCV infectionc	32 449	100.0		188 156	100.0
Diagnosed with chronic HCVd	29 063	89.6		144 964	77.0
Linked to HCV caree	25 786	79.5	88.7	136 169	72.4	93.9
Received HCV antiviral therapyf	7421	22.9	28.8	58 321	31.0	42.8
Achieved SVRg	5041	15.5	67.9	42 878	22.8	73.5

Abbreviations: HCV, hepatitis C virus; SVR, sustained virologic response; VHA, Veterans Health Administration.

^aData sources: Veterans Health Administration Corporate Data Warehouse and HCV Clinical Case Registry, 2015.

^bHomeless veterans were identified per an outpatient visit to Veterans Administration homeless services in 2015.

^cThe total HCV population was estimated from the sum of those diagnosed with HCV, and projected prevalence applied to those who were untested for HCV.

^dDefined as ever having a detectable HCV RNA or genotype.

^eRequired HCV entry on the patient’s medical record Problem List and entry in the VHA’s HCV registry.

^fDefined as ever receiving HCV antivirals from VHA as of December 31, 2015.

^gDefined as undetectable HCV RNA on an HCV RNA test at least 12 weeks after the end of treatment, with the SVR rate among those evaluable for SVR applied to those without definitive SVR status.

Discussion

These data describe a large national cohort of homeless veterans with high rates of HCV testing, including high rates of RNA testing as appropriate for positive antibody results. Our findings can inform the care of homeless veterans with chronic HCV infection.

First, the laboratory-diagnosed prevalence of chronic HCV infection of 15.4% of homeless veterans, particularly in comparison with the prevalence of 4.5% of nonhomeless veterans, confirms the purported high prevalence of chronic HCV infection in the homeless veteran population. In previous studies, rates of HCV antibody positivity among homeless veterans ranged from 6.6% to 44.0%.^6–8 Rates of HCV antibody positivity observed in small general homeless populations (7.0% to 32.4%) were consistently higher than rates of HCV antibody positivity reported in nonhomeless populations.^16–20 Our analysis of a large national cohort with high rates of HCV testing that included RNA testing as appropriate provides a robust national estimate of chronic HCV infection prevalence among homeless veterans.

Our analysis revealed several notable findings: (1) VHA providers do a better job of testing for and diagnosing chronic HCV infection among homeless veterans than they do among nonhomeless veterans, (2) rates of linkage to HCV care were roughly equivalent between homeless and nonhomeless veterans, (3) receipt of HCV antiviral therapy among homeless veterans was substantially lower than among nonhomeless veterans, and (4) the cumulative SVR rates achieved by homeless and nonhomeless veterans who had ever received HCV antiviral therapy were comparable. For veterans deemed eligible for HCV antiviral therapy by VHA providers, homelessness did not appear to substantially influence treatment outcomes. Although homelessness was a critical barrier to HCV treatment in the previous peginterferon era,^11–14 our data suggest that this perceived barrier could be overcome in the era of all-oral DAAs. Additional research is needed to better describe the characteristics of homeless patients who were deemed eligible for therapy and any services provided that supported successful antiviral therapy.

Limitations

This study had several limitations. Homelessness is often a transient state, with people transitioning through degrees of housing stability over time. We used criteria from the VHA’s National Center on Homelessness Among Veterans to identify veterans using homeless services, which rely on accurate administrative data and coding. However, we did not have information on the timing of veteran homelessness in relation to their receipt of HCV care.

Conclusion

To our knowledge, these estimates are the largest, most comprehensive estimates of the prevalence of chronic HCV infection and rates of engagement in the HCV care cascade among homeless veterans. These data substantiate high prevalence rates of chronic HCV infection among homeless veterans and underscore the importance of continued HCV screening efforts. These data also demonstrate lower receipt of HCV antiviral therapy among homeless veterans compared with nonhomeless veterans despite comparable SVR rates in homeless veterans who received HCV antiviral therapy. Efforts are needed to identify appropriate interventions to ensure that more homeless veterans are candidates for HCV antiviral therapy. Homelessness should not necessarily preclude receipt of HCV antiviral therapy as we consider the direction of future HCV care and treatment eligibility criteria with all-oral DAA regimens.