Cascade of Care for Hepatitis C Virus Infection Among Young Adults Who Inject Drugs in a Rural County in New Mexico

Abstract

Objective:

Treatment for hepatitis C virus (HCV) infection is highly effective; however, people who inject drugs (PWID), the population most affected by HCV, may encounter barriers to treatment. We examined the cascade of care for HCV infection among young adult PWID in northern New Mexico, to help identify gaps and opportunities for HCV treatment intervention.

Methods:

Young adults (aged 18-29 y) who self-reported injection drug use in the past 90 days were tested for HCV antibodies (anti-HCV) and HCV RNA. We asked participants with detectable RNA to participate in an HCV education session, prior to a referral to a local health care provider for treatment follow-up, and to return for follow-up HCV testing quarterly for 1 year. We measured the cascade of care milestones ranging from the start of screening to achievement of sustained virologic response (SVR).

Results:

Among 238 participants, the median age was 26 years and 133 (55.9%) were men. Most (90.3%) identified as Hispanic. Of 109 RNA-positive participants included in the cascade of care assessment, 84 (77.1%) received their results, 82 (75.2%) participated in the HCV education session, 61 (56.0%) were linked to care through a medical appointment, 27 (24.8%) attended the HCV treatment appointment, 13 (11.9%) attended their follow-up appointment, 6 (5.5%) initiated treatment, 3 (2.8%) completed treatment, and 1 (0.9%) achieved SVR.

Conclusions:

We observed a steeply declining level of engagement at each milestone step of the cascade of care after detection of HCV infection, resulting in a suboptimal level of HCV treatment and cure. Programs that can streamline testing and expand access to treatment from trusted health care providers are needed to improve the engagement of PWID in HCV treatment.

Providing treatment to people who inject drugs (PWID) with current hepatitis C virus (HCV) infection is an important public health priority, because this population has the highest prevalence and incidence of HCV infection.^1,2 In 2019, the Centers for Disease Control and Prevention (CDC) reported that 67% of all new cases of HCV infection that included information on injection risk were attributable to injection drug use. ³ Treatment for HCV changed dramatically in 2014 with the introduction of direct-acting antiviral (DAA) agents, which reduced pill burden (ie, the number of pills required to be taken), treatment duration, and the number of side effects.^4,5 Despite studies showing that PWID with HCV can be engaged in their health care and successfully treated for HCV,^6,7 HCV treatment and cure rates among PWID remain low.^8,9

The HCV cascade of care model includes a sequence of measurable indicators (milestones) that patients encounter on the path to being cured, including screening, diagnosis, linkage to care, treatment initiation, and sustained virologic response (SVR). The model allows for the documentation of health disparities among key populations by identifying losses at each milestone and can be a tool to inform and assess the performance of public health and clinical programs to treat HCV infection; cascades of care for HCV infection (and other conditions) are now well-recognized tools.^10-13 Studies have examined HCV cascade of care outcomes in populations vulnerable to reduced health care access and services, including PWID, veterans, and people experiencing homelessness.^8,14-16 Studies in rural populations have included HCV cascade of care outcomes among patients attending federally qualified health centers in rural Idaho ¹⁷ and assessment of health care contact and treatment uptake (but not SVR) in a long-standing cohort of PWID in rural Kentucky. ¹⁸ However, to our knowledge, no study has examined the full HCV cascade of care among PWID in rural areas in the United States. Thus, we examined the HCV cascade of care (from screening to SVR) in a cohort of young adult PWID recruited in a rural New Mexico county with a high prevalence of HCV infection. ¹⁹ Although efforts have been made to reduce the barriers to HCV treatment in New Mexico and access to treatment is increasing overall, ²⁰ it is not clear that these efforts have reached populations at high risk of HCV, such as PWID in rural areas; an evaluation of cascades of care could allow an understanding of whether these efforts are working.

Methods

Population and Procedures

We conducted our prospective study, named ¡Vále!, from September 2016 through May 2019; we aimed to recruit young adults (aged 18-29 y) who reported injection drug use in the past 90 days. The sampling and data collection methods are described elsewhere. ¹⁹ In brief, recruitment took place at The Mountain Center in Rio Arriba County in northern New Mexico. ¹⁹ The Mountain Center is a community-based organization that hosts a harm reduction program that includes a syringe services program (SSP). Other services include colocated (ie, services located at the same site) prevention, recovery, and drug treatment programs that provide counseling and buprenorphine from a health care provider. The Mountain Center also provides information and resources to clients on HCV infection treatment from local health care providers. Spanish-speaking staff are available if needed.

We asked study participants to undergo HCV screening, answer a baseline survey, and attend 3 follow-up visits at 3-month intervals during a 1-year period. First, after study participants underwent informed consent and enrollment, study personnel screened all participants for HCV antibodies (anti-HCV) and HCV RNA. Study personnel conducted anti-HCV screening with a rapid point-of-care anti-HCV test (OraSure Technologies), for which study personnel obtained capillary blood samples by fingersticks. To test for HCV RNA, study personnel trained in phlebotomy obtained blood samples through venipuncture and then transferred the samples to a laboratory. The test procedure used transcription-mediated amplification (discriminatory HCV transcription-mediated amplification assay component of the Procleix HIV-1/HCV assay; Gen-Probe Inc) to determine current HCV infection status.

Study personnel retested participants who had negative anti-HCV or HCV RNA test results at baseline or who had an insufficient quantity of blood available to test for RNA at the 3-, 6-, 9-, or 12-month follow-up visits.

Next, after participants underwent testing, but before study personnel disclosed anti-HCV test results, study personnel interviewed participants with a structured questionnaire (including questions on demographic characteristics, HCV and HIV treatment, substance use, and sexual practices and questions on access to medical, mental, and substance use disorder health care). Study personnel also provided infection prevention counseling and referrals for prevention services, drug treatment, the hepatitis A vaccine and hepatitis B vaccine, and behavioral health care to participants.

If HCV RNA test results (provided 1-2 weeks later) were positive, indicating current HCV infection, study personnel offered participants a one-on-one HCV education session with study staff and a referral for further HCV assessment and care. All participants received a $15 Visa merchandise card for completing the baseline visit and another $15 card for returning for their HCV RNA test results. Study personnel also offered HCV RNA–positive participants small incentives ranging from $10 to $25 merchandise cards to attend their HCV treatment visits at their chosen clinic.

Study personnel obtained releases of information from HCV RNA–positive participants to contact the clinic and to obtain the HCV-related portion of the medical records from those visits and any documentation of treatment and testing for SVR. SVR was defined as <15 IU/mL or undetectable HCV viral load at 12 weeks after the start of treatment. The assessment of SVR from clinical records was ongoing and conducted through 2020.

The University of New Mexico Health Sciences Center Institutional Review Board reviewed and approved the study. All participants provided written informed consent to participate. We received a Federal Certificate of Confidentiality for the study to enhance privacy protections due to the sensitive nature of the data collected. Study personnel informed all participants that treatment for HCV was not contingent on study participation and provided information about HCV treatment to all people screened and enrolled at The Mountain Center.

Statistical Analyses

We determined HCV infection status from the baseline anti-HCV test and from HCV RNA test results at baseline and at 3-month, 6-month, and 9-month visits. For the analysis, we included the total number of participants who were screened for anti-HCV and HCV RNA (at any visit). For assessment of cascade of care milestones, we determined the number and percentage of those who (1) were HCV RNA positive; (2) returned to receive their HCV RNA test results; (3) attended an education and referral session; (4) were linked to care (ie, study staff made an appointment for participants with a health care provider of their choice); (5) attended the initial assessment appointment; (6) obtained a confirmative, quantitative HCV RNA laboratory test; (7) attended a follow-up appointment; (8) initiated HCV treatment; (9) completed the prescribed HCV treatment; and (10) achieved SVR. We examined the HCV cascade of care with both a fixed denominator (all HCV RNA–positive participants), with losses shown across each milestone as a proportion of the HCV-positive population, and a running denominator, with losses shown at each consecutive step in the cascade of care. We obtained data on sociodemographic characteristics, injection practices, substances injected, health care utilization, and utilization of harm reduction services from survey responses. We reported variables using means, medians, and measures of dispersion. We conducted all analyses using SAS version 9.4 (SAS Institute Inc).

Results

Of 480 people screened, we enrolled 263 participants in the ¡Vále! study. Of those enrolled, 238 participants had anti-HCV and/or HCV RNA screening tests at any visit and, thus, were included in this analysis.

The median (interquartile range [IQR]) age was 26 (22.0-28.0) years. Of 238 people considered for our analysis, 133 (55.9%) were men; among 236 participants who provided data on race and ethnicity, 213 (90.3%) were Hispanic (Table). The median (IQR) age at first injection drug use was 19 (17.0-23.0) years. Among 232 participants who provided information on health care access, 135 (58.2%) had seen a health care provider in the past year and 26 (11.2%) had seen a health care provider in the past 30 days. Of 230 participants who provided information on drug use counseling, 159 (69.1%) reported ever receiving some kind of treatment or counseling for drug or alcohol use, and 137 (59.1%) of 232 participants reported having been previously tested for HCV. Of 220 participants who provided information on SSP attendance, 169 (76.8%) reported using an SSP in the past 3 months.

Table.

Demographic and behavioral characteristics and substance use among young adults, aged 18-29 years, who inject drugs in a rural New Mexico county, 2016 through 2019 ^a

Demographic characteristic or survey response	No./total b	% (95% CI)
Age, median (IQR), y	26.0 (22.0-28.0)	—
Sex
Female	105/238	44.1 (37.7-50.7)
Male	133/238	55.9 (49.6-62.2)
Hispanic ethnicity	213/236	90.3 (85.7-93.7)
Race and ethnicity
American Indian	18/234	7.7 (4.5-11.7)
Black	5/234	2.1 (0.7-4.8)
Other (Asian/Pacific Islander, mixed)	79/234	33.8 (27.2-39.6)
White	132/234	56.4 (48.9-61.9)
Education
High school diploma or GED	89/233	38.2 (31.2-43.9)
Less than high school diploma	87/233	37.3 (30.4-43.0)
Some college or higher	57/233	24.5 (18.7-29.9)
Housing
Other/shelter/street/outdoors	16/230	7.0 (3.9-10.7)
Own or rent apartment/room/house	99/230	43.0 (35.3-48.1)
Someone else’s apartment/room/house	115/230	50.0 (41.8-54.9)
Ever been on probation/parole	122/228	53.5 (46.8-60.1)
Health care utilization responses
Saw a health care provider in the past year	135/232	58.2 (51.6-64.6)
Has seen a health care provider in the past 30 d	26/232	11.2 (7.5-16.0)
Can get to medical appointments (car, walking, a ride, public transportation)	219/233	94.0 (90.1-96.7)
Ever previously tested for HCV	137/232	59.1 (52.4-65.4)
Positive result from most recent hepatitis C antibody test	16/30	53.3 (34.3-71.7)
Detectable virus in most recent HCV RNA/viral load test	15/49	30.6 (18.3-45.4)
Positive result from most recent hepatitis C test, type unknown	31/69	44.9 (32.9-57.4)
Ever been tested for HIV	113/232	48.7 (42.1-55.3)
Currently has health insurance	207/232	89.2 (84.5-92.9)
Ever received any kind of treatment or counseling for drug or alcohol use	159/230	69.1 (62.7-75.0)
Ever received any medication for drug dependence/addiction	109/193	56.5 (49.2-63.6)
Attended or participated in any form of treatment for drug use in the past 3 months	58/228	25.4 (19.9-31.6)
Wanted to get drug treatment in the past 3 months but did not go	91/209	43.5 (36.7-50.6)
Harm reduction services
Attended an SSP in the past 3 mo	169/220	76.8 (70.7-82.2)
Average times went to SSP in the past 3 mo, median (IQR) (n = 147)	2.0 (1.0-3.0)	—
Average number of syringes obtained in 1 visit to SSP, median (IQR) (n = 156)	100 (42.5-200.0)	—

Abbreviations: —, does not apply; GED, General Educational Development; HCV, hepatitis C virus; IQR, interquartile range; PWID, people who inject drugs; SSP, syringe services program.

^aParticipants were aged 18-29 years and had been recruited to the ¡Vále! study at The Mountain Center in Rio Arriba County in northern New Mexico.

^bAll values are number/total unless otherwise indicated.

At the baseline visit, 235 of 238 (98.7%) participants were tested for HCV (n = 233 anti-HCV; n = 211 HCV RNA, because 24 participants had an insufficient quantity of blood collected to perform the RNA test). Of the 233 participants who could be tested at baseline for anti-HCV, 147 (63.1%) had a reactive anti-HCV test result. For the HCV RNA test, 209 participants had results at baseline, and 99 (47.4%) were HCV RNA positive. In total, 9 participants had a positive HCV RNA test result and a negative anti-HCV test result, indicating acute HCV infection.

Among the 64 additional HCV RNA tests with results available conducted during follow-up visits, 10 (15.6%) had detectable RNA, and 6 (12%) of the 50 anti-HCV tests conducted during follow-up had new positive anti-HCV test results. Overall, 283 anti-HCV tests and 275 HCV RNA tests were conducted among 235 people during the study period.

For our cascade of care assessment (Figure), we included 109 of 238 (45.8%) participants who had a positive HCV RNA test result at baseline or at any of the follow-up visits. Of 109 participants, 84 (77.1%) returned to receive these results, 82 (75.2%) took part in the HCV education session with research staff, and 61 (56.0%) were linked to care through a medical appointment. The most common reason participants were not linked to care was that they indicated on the structured questionnaire not to be linked, with some declining because they were not interested in pursuing treatment at that time and a few wanting to make arrangements on their own. Of 109 HCV RNA–positive participants, 27 (24.8%) attended the initial medical appointment, but only 19 (17.4%) obtained a confirmatory laboratory test for quantitative HCV RNA (required by health care providers at the time for HCV treatment). In addition, only 13 (11.9%) participants attended their follow-up appointment, 6 (5.5%) initiated treatment, 3 (2.8%) completed treatment, and 1 (0.9%) achieved SVR. When we examined results based on the running denominator, we found that 61 of the 82 (74.4%) participants who attended the education and referral session were linked to care, but just 27 (44.2%) of these 61 participants attended their initial assessment appointment.

Figure.

Hepatitis C virus (HCV) cascade of care for young adults (aged 18-29 y) who inject drugs in a rural New Mexico county (¡Vále! study), 2016 through 2019. Percentages above the bars represent participants with a positive HCV RNA test result (fixed denominator), and the percentages shown in the arrows represent participants advancing by milestone (running denominator). Abbreviation: SVR, sustained virologic response.

Discussion

The high prevalence of HCV infection among PWID in the United States warrants a concerted population approach to HCV treatment to reduce prevalence and ongoing transmission. ²¹ The young adult PWID participants residing in rural northern New Mexico, where HCV infection is widespread, ¹⁹ had a steeply declining level of engagement in cascade of care milestones after HCV infection detection, resulting in a suboptimal level of HCV treatment and cure. Young adult PWID are more likely than older PWID to acquire HCV infection, potentially because of reduced knowledge of HCV status and differences in seroadaptive behaviors that vary by age, which can increase high-risk injection exposures, ²² making young adult PWID a critical population to reach for treatment and elimination efforts.

Of 109 people who screened positive for HCV RNA and were included in the cascade of care analysis, we documented only 1 person who achieved SVR. We were encouraged that about 75% of those who tested HCV RNA positive attended the education and referral session. High attendance may be linked to the higher incentive ($25 gift card) offered for attending that session, which was more than the remuneration offered for other visits ($10-$15). However, engagement declined at each step of the cascade of care. Similar results have been shown in other studies of DAA treatment adherence among PWID. In an observational study of young adult PWID with acute HCV infection in San Francisco, 50.0% (30 of 60) of the participants accepted a referral to a health care provider for treatment assessment, but 5 (10.2%) of 49 participants with acute HCV infection initiated treatment and achieved SVR. ⁸ In a recent retrospective analysis of Florida Medicaid data and HCV cascade of care milestones among substance users, ²³ a subanalysis of the PWID population reported that 18% were tested and 8% had a positive HCV diagnosis; however, only 11% of the PWID group initiated HCV treatment, 96% of whom completed treatment. Slightly higher engagement outcomes were reported in an analysis of PWID from the 2018 Seattle-area National HIV Behavioral Surveillance survey, in which 26% of respondents with a diagnosed HCV infection said they had received treatment and 18% said they had been cured. ⁹ Engagement rates for all cascade of care milestones were lower among our PWID population and the PWID populations in the other studies compared with non-PWID populations. ²⁴

Interestingly, PWID in our study and other studies of HCV cascade of care with DAA as a standard of care demonstrated engagement in cascade of care milestones that were not notably different from milestones described in similar studies with interferon-based treatment as a standard of care. For example, among 254 PWID with positive anti-HCV test results recruited by Stephens et al ¹⁸ in rural Appalachian Kentucky from 2008 to 2010, when interferon-based treatment was the standard of care, 51.8% of those who were counseled to seek care contacted a health care provider and 7.8% received treatment. HCV treatment has become increasingly accessible in New Mexico: the number of patients treated for HCV infection increased 8-fold from 2014 to 2017, and this trend has continued, ²⁰ with no current policies restricting DAA access in New Mexico. ²⁵ Of note, despite the small financial incentives given to participants in our New Mexico ¡Vále! study to attend HCV follow-up appointments, the proportions of PWID who engaged in the cascade of care, received treatment, or achieved SVR were similar to or even lower than proportions shown in the Seattle-area or rural Appalachian PWID cohorts.^9,18

Several studies have assessed factors associated with seeking HCV treatment. In Stephens et al, ¹⁸ PWID who sought follow-up care were more likely than PWID who did not seek care to have health insurance, internet access, prior substance use treatment, probable anxiety disorder, and recent marijuana use. In addition, PWID with major depressive disorder and those with prior methadone use had lower odds of seeking care than those without these factors. In the National HIV Behavioral Surveillance sample of PWID in the Seattle area, older age was associated with having DAA treatment. Because the Seattle PWID cohort included a larger age range than our study, we suggest that age difference is a partial explanation for the higher proportion (26%) of PWID who reported treatment compared with our sample (6%). ⁹ Despite DAA treatment shown to be as effective in PWID as in people without a history of drug use, ²⁶ we observed a low uptake of treatment, indicating that more work needs to be done to advance HCV elimination efforts.

The decreases that we observed at each level of the cascade of care suggest the presence of barriers, including system-, provider-, and patient-level barriers, which may result in loss of forward momentum.^27,28 The system-level and patient-level barriers for engaging in primary health care reported by PWID ²⁹ also exist for HCV treatment. With each milestone in the cascade of care, there are concomitant opportunities to improve engagement among PWID with HCV. For example, at a system level, HCV RNA testing for diagnostic purposes still requires a venous blood draw and a wait of 1-2 weeks for laboratory results. Insufficient blood quantity resulting from poor venous access limited our ability to inform all participants of their HCV infection status, which underscores the need for advanced diagnostics, including dried blood spots and point-of-service RNA tests, to close gaps in HCV testing. Although this stipulation does not apply to health insurance or Medicaid policy in New Mexico, health care providers in other states often postpone initiation of HCV treatment until patients have attended a set number of appointments or have achieved an abstinence requirement as a means of demonstrating adherence to medication. ²⁰ Patient-level barriers such as housing, transportation, information access, and other health and social priorities may affect HCV treatment engagement among PWID.^27,28 PWID in our study needed to attend multiple in-person visits before treatment initiation; we suggest that many of these visits could be combined to reduce this burden. For example, the education session could be included in the initial assessment visit, and treatment initiation could occur when HCV RNA test results are received.

More than 75% of the participants in our study reported accessing an SSP. SSPs serve as a bridge to HIV and HCV testing and other health services, including treatment for substance use.^30,31 Patients in a Florida cohort with substance use disorder who accessed SSPs were more than twice as likely as patients who did not access SSPs to be tested for HCV. ²³ Integrating HCV treatment into services provided at SSPs has the potential to improve engagement in HCV cascade of care milestones. Among PWID who accessed HCV treatment colocated at an SSP, Winetsky et al ³² reported that 42.2% (43 of 102) of patients completed treatment and achieved SVR. Although this outcome varied by mental health and substance use status during treatment, the proportions are still more than 20-fold higher than observed in our study sample. These outcomes suggest that when HCV treatment is provided in a harm reduction setting where reducing barriers to prevention and care for PWID are more normative and stigma is less of an issue than in other health care settings, meaningful population-level effects on SVR can be obtained.^33-36 In New Mexico, we suggest that cascade of care milestones in HCV treatment could be integrated with the New Mexico Department of Health Harm Reduction Program, which is widely accessed in many rural and medically underserved areas, areas that have been especially affected by the opioid epidemic. ³⁷

Limitations

We recognize some limitations. First, participants were sampled from 1 county in New Mexico, and the results might not be generalizable to other locales in the United States. Second, the sample size was modest, which also limited the generalizability of the findings. Third, while loss to follow-up is generally a limitation in prospective studies, in this study, losses were a central outcome for each step in the cascade of care and, thus, important to examine. However, we acknowledge that, because contact with some participants was challenging, we may not have ascertained some cascade of care outcomes, thereby underestimating the number of participants who initiated treatment, completed treatment, and achieved SVR.

Conclusions

With more than 75% of our study participants reporting attendance at an SSP in the past 3 months, we suggest that programmatic interventions that colocate HCV treatment services with SSPs, opioid treatment programs,^38-40 and other health services (eg, wound care clinics, pharmacies) that PWID access have great potential to close gaps in cascade of care milestones. ⁴¹ Simplifying HCV treatment with a patient-centered approach is feasible, safe, and effective, ⁴² and further work is needed to promote similar programs that reduce the multiple laboratory and clinical visits for PWID, especially in rural areas. Treatment in correctional facilities should be promoted to reduce the incidence of infection in the population. ⁴³ A test-and-treat model for HCV infection, in which people are tested, diagnosed, and then initiate HCV treatment at the point of service, can improve engagement in HCV treatment. ⁴⁴ As HCV elimination efforts proceed, the development and testing of new interventions directed at improving engagement in HCV care for PWID may also inform models to address and improve other health outcomes. Enacting programs that streamline testing and expanding access to treatment from trusted health care providers remain crucial to improving the engagement of PWID in HCV treatment.