Progress Towards Elimination of Hepatitis C Infection Among People Who Inject Drugs in Australia: The ETHOS Engage Study

Abstract

Background

Evaluating progress towards hepatitis C virus (HCV) elimination is critical. This study estimated prevalence of current HCV infection and HCV treatment uptake among people who inject drugs (PWID) in Australia.

Methods

The Enhancing Treatment of Hepatitis C in Opioid Substitution Settings Engage is an observational study of PWID attending drug treatment clinics and needle and syringe programs (NSPs). Participants completed a questionnaire including self-reported treatment history and underwent point-of-care HCV RNA testing (Xpert HCV Viral Load Fingerstick; Cepheid).

Results

Between May 2018 and September 2019, 1443 participants were enrolled (64% injected drugs in the last month, 74% receiving opioid agonist therapy [OAT]). HCV infection status was uninfected (28%), spontaneous clearance (16%), treatment-induced clearance (32%), and current infection (24%). Current HCV was more likely among people who were homeless (adjusted odds ratio, 1.47; 95% confidence interval, 1.00–2.16), incarcerated in the previous year (2.04; 1.38–3.02), and those injecting drugs daily or more (2.26; 1.43–2.42). Among those with previous chronic or current HCV, 66% (n = 520/788) reported HCV treatment. In adjusted analysis, HCV treatment was lower among females (.68; .48–.95), participants who were homeless (.59; .38–.96), and those injecting daily or more (.51; .31–.89). People aged ≥45 years (1.46; 1.06–2.01) and people receiving OAT (2.62; 1.52–4.51) were more likely to report HCV treatment.

Conclusions

Unrestricted direct-acting antiviral therapy access in Australia has yielded high treatment uptake among PWID attending drug treatment and NSPs, with a marked decline in HCV prevalence. To achieve elimination, PWID with greater marginalization may require additional support and tailored strategies to enhance treatment.

Among this population of people who inject drugs in Australia, 24% were currently infected with hepatitis C virus. Of those who were ever infected, 66% had received treatment. Enhancing treatment uptake in key subpopulations is required to achieve elimination.

The World Health Organization’s (WHO’s) goal to eliminate hepatitis C virus (HCV) infection as a public health threat aims to reduce HCV incidence and related mortality by substantially increasing diagnosis and treatment [1]. Globally, an estimated 71 million people are infected with HCV, including an estimated 6.1 million who have recently injected drugs and a large population having injected drugs in the past [1–3]. Mathematical modeling has demonstrated the importance of rapid treatment initiation to reduce population-level HCV infection and prevent onward transmission among people who inject drugs (PWID) [4, 5]. Despite favorable treatment outcomes among PWID [6], system, societal, provider, and individual barriers persist and hinder optimal HCV care [7].

Since March 2016, adults with HCV have access to government-reimbursed direct-acting antiviral (DAA) therapy with no drug, alcohol, or fibrosis stage restrictions [8]. This public health approach in the provision of unrestricted DAA therapy engendered one of the highest HCV treatment uptakes globally, with Australia named as one of few countries on track to achieve the WHO target of reducing new infections by 2030 [9, 10].

Although studies have explored DAA treatment among PWID, they are limited with respect to reimbursement restrictions, population size, single/homogenous settings, or insufficient virological data [6, 11–14]. This study evaluated progress towards HCV elimination among PWID in Australia among a large, national cohort of PWID recruited from drug treatment and needle and syringe programs (NSPs) during an unrestricted HCV treatment era. The primary aim of this study was to evaluate the proportion of people with current HCV infection and associated factors. A secondary aim was to evaluate the proportion of people who had received HCV treatment and associated factors.

METHODS

Data Sources

The Enhancing Treatment of Hepatitis C in Opioid Substitution Settings (ETHOS) Engage is an observational cohort study. Participants were enrolled between 28 May 2018 and 6 September 2019 from 25 sites, including opioid agonist therapy (OAT) clinics (n = 21) and NSPs (n = 4), in New South Wales (n = 17), Queensland (n = 4), South Australia (n = 2), and Western Australia (n = 2).

Inclusion criteria were informed consent, age 18 years or older, history of injecting drug use, and either injection drug use in the previous 6 months or current OAT. Pregnant women were excluded given that FibroScan (Echosens, Paris, France) was contraindicated at the time of study protocol approval. The study protocol was approved by the Human Research Ethics Committees at St Vincent’s Hospital, Sydney, and the Aboriginal Health and Medical Research Council (HREC Ref: HREC/17/SVH/113).

Procedures

ETHOS Engage was advertised in the weeks preceding recruitment with posters, information cards distributed with injecting equipment, and word of mouth. Recruitment spanned 2 to 5 days at each site. Peer-support workers were on-site encouraging participation.

Participants provided 100-µL finger-stick capillary whole-blood sample to test for HCV RNA using the point-of-care Xpert HCV Viral Load Fingerstick assay (Cepheid, Sunnyvale, CA; lower limit of quantification, 100 IU/mL; upper limit of quantification, 10⁸ log₁₀ IU/mL) [15] and self-completed a computer tablet–based questionnaire collecting demographic, behavioral risk, previous HCV testing, infection status, and treatment information. Fibrosis stage was assessed using transient elastography (FibroScan) and median stiffness (kPa) was discussed with appropriate clinical staff. Participation was compensated with a shopping voucher ($30 [Australian dollars]).

Hepatitis C virus RNA results were returned to clinics after quality assurance checks. Staff were encouraged to contact participants with current HCV infection to initiate treatment. Postcampaign treatment initiation and outcomes will be assessed in the 3 years proceeding campaign days.

Outcomes

The primary outcome was current HCV infection (HCV RNA detected with the Xpert HCV Viral Load Fingerstick assay). Previous work has demonstrated the high sensitivity (100%) and specificity (100%) of this assay in HCV RNA quantification [15, 16] and fingerstick testing acceptability among PWID [17].

The secondary outcome was self-reported history of HCV treatment among participants with either previous (self-reported history of HCV treatment) or current HCV infection (in participants who have been treatment eligible). Participants who were never infected (HCV RNA undetectable and self-reported as never having been diagnosed with HCV) and who had spontaneously cleared (HCV RNA undetectable, self-reported as having been diagnosed with HCV, and not having received HCV treatment) were also identified (Figure 1).

Figure 1.

ETHOS Engage participant flowchart, current HCV status (N = 1468). *Determined by self-report. Abbreviations: ETHOS, Enhancing Treatment of Hepatitis C in Opioid Substitution Settings; HCV, hepatitis C virus; HCV Ab-ve, HCV antibody negative.

Statistical Analysis

Logistic regression models were used to estimate the unadjusted and adjusted odds ratio (aOR) for (1) factors associated with current HCV infection among the total cohort and (2) factors associated with a history of HCV treatment among those with evidence of previous chronic or current HCV infection.

Demographic and behavioral factors hypothesized to be associated with current HCV infection and HCV treatment were determined a priori, comprising the following: (1) age at survey (stratified around median: <45, ≥45 years), (2) gender, (3) Indigenous ethnicity (Aboriginal and/or Torres Strait Islander), (4) homelessness, (5) OAT status (never, past, within the last month/current), (6) incarceration history (never, >1 year ago, within the last year), (7) recency and frequency of injection drug use (>1 year ago, within 1–12 months ago, within the last month less than daily, and daily or more), (8) main drug injected in the last month (none, heroin, other opioids, methamphetamine, other) and (9) hazardous alcohol consumption in the previous year, defined by the Alcohol Use Disorders Identification Test (AUDIT-C) [18].

All exposures were analyzed in unadjusted analyses and considered for adjusted models if no collinearity was observed. Collinearity was assessed using variance-covariance matrices, with variables removed from adjusted models if ≥0.5 correlation was identified.

Each outcome was assessed for the overall eligible population, and subsequently restricted to participants with recent (last month) injecting drug use. In analyses restricted to participants with recent injecting drug use, injecting-related variables were recategorized as recency and frequency of injecting (less than daily, daily or more) and main drug injected in the last month (heroin, other opioids, methamphetamine, other). In post hoc analysis, predictors of HCV treatment were stratified by gender. Analyses were conducted using Stata 14.0 (StataCorp, College Station, TX).

RESULTS

Sample Characteristics

Among 1468 participants in ETHOS Engage, 5 (<1%) had insufficient questionnaire data, 16 (1%) withdrew participation, and 4 (<1%) duplicate enrollments were identified across sites, resulting in 1443 participants (98%) eligible for analysis (Figure 1).

The median age was 43 (interquartile range [IQR], 37, 50) years, 65% (n = 932) were male, 74% (n = 1070) were receiving OAT, and methamphetamine was the most common main drug injected (31%, n = 449). Nearly two-thirds (64%) of participants injected drugs in the last month, and 30% injected drugs daily or more (Table 1). Characteristics stratified by recent injecting drug use, OAT status, and gender are presented in Supplementary Tables 1–3.

Table 1.

Characteristics of Participants Enrolled in ETHOS Engage

Characteristics	Total, n (column %)
Total, N	1443
Age at survey
<45 years	791 (55)
≥45 years	652 (45)
Gender
Male	932 (65)
Female	508 (35)
Transgender	3 (<1)
Indigenous ethnicity
No	1106 (77)
Yes	337 (23)
Homeless
No	1286 (89)
Yes	157 (11)
OAT status
Never	205 (14)
Past	168 (12)
Current	1070 (74)
Incarceration history
Never	469 (32)
>1 year ago	715 (50)
Within last year	259 (18)
Recency of injecting
>12 months	215 (15)
Within 1–12 months	307 (21)
Within last month, <daily	494 (34)
Within last month, ≥daily	427 (30)
Main drug injected in last month
None	522 (36)
Heroin	312 (22)
Other opioids	132 (9)
Methamphetamine	449 (31)
Other	28 (2)
Excessive alcohol consumptiona
No	915 (64)
Yes	525 (36)

Abbreviations: ETHOS, Enhancing Treatment of Hepatitis C in Opioid Substitution Settings; OAT, opioid agonist therapy.

^aNot reported for transgender participants.

Factors Associated With Current Hepatitis C Virus Infection

Among all participants (n = 1443), 1388 (96%) had valid Xpert Viral Load Fingerstick point-of-care results. Invalid results (n = 55, 4%) included early withdrawal (n = 16, 1%) and operator/machine error (n = 39, 3%). Among those with valid results, 24% (n = 331) were currently infected with HCV (HCV RNA detectable). The prevalence of current HCV infection stratified by characteristics is shown in Figure 2 and Table 2.

Figure 2.

Current HCV prevalence among ETHOS Engage participants with known point-of-care HCV RNA results (n = 1388). *Main drug injected in the last month. Data for participants injecting other drugs (n = 24) are not shown. Abbreviations: ETHOS, Enhancing Treatment of Hepatitis C in Opioid Substitution Settings; F, female; HCV, hepatitis C virus; HCV Ab-ve, HCV antibody negative; HCV RNA+ve, HCV RNA positive; M, male; m, month(s); Meth, methamphetamine; OAT, opioid agonist therapy; RNA+ve, RNA positive.

Table 2.

Unadjusted and Adjusted Analysis of Factors Associated With Current Hepatitis C Virus (HCV) Infection in All ETHOS Engage Participants With Available Point-of-Care HCV RNA Results

Characteristics	Total Valid Point-of-Care Result, n (column %)	Current HCV Infection, n (row %)	OR (95% CI)	aOR (95% CI)
Total, N	1388	331 (24)	…	…
Age at enrollment
<45 years	760 (55)	190 (25)	Ref	Ref
≥45 years	628 (45)	141 (22)	0.87 (.68, 1.11)	0.92 (.71, 1.20)
Gender
Male	891 (64)	216 (24)	Ref	Ref
Female	494 (36)	113 (23)	0.93 (.71, 1.20)	1.03 (.78, 1.35)
Transgender	3 (<1)	2 (67)	6.25 (.56, 69.26)	omitted
Indigenous ethnicity
No	1064 (77)	253 (24)	Ref	Ref
Yes	324 (23)	78 (24)	1.02 (.76, 1.36)	0.93 (.69, 1.26)
Homeless
No	1241 (89)	282 (23)	Ref	Ref
Yes	147 (11)	49 (33)	1.70 (1.18, 2.45)	1.47 (1.00, 2.16)
OAT status
Never	199 (14)	44 (22)	Ref	Ref
Past	160 (12)	48 (30)	1.51 (.94, 2.43)	1.38 (.85, 2.25)
Current	1029 (74)	239 (23)	1.07 (.74, 1.53)	1.16 (.78, 1.71)
Incarceration history
Never	455 (33)	77 (17)	Ref	Ref
>1 year ago	685 (49)	179 (26)	1.74 (1.29, 2.34)	1.79 (1.30, 2.45)
Within last year	248 (18)	75 (30)	2.13 (1.48, 3.07)	2.03 (1.38, 3.01)
Recency of injecting
>12 months	209 (15)	31 (15)	Ref	Ref
Within 1–12 months	299 (22)	67 (22)	1.67 (1.04, 2.66)	1.54 (.95, 2.49)
Within last month, <daily	477 (34)	109 (23)	1.70 (1.10, 2.63)	1.54 (.99, 2.41)
Within last month, ≥daily	403 (29)	124 (31)	2.55 (1.65, 3.95)	2.29 (1.45, 3.62)
Main drug injected in last month
None	508 (37)	98 (19)	Ref	Omitteda
Heroin	296 (21)	75 (25)	1.42 (1.01, 1.99)
Other opioids	127 (9)	44 (35)	2.21 (1.44, 3.29)
Methamphetamine	431 (31)	108 (25)	1.40 (1.02, 1.90)
Other	26 (2)	6 (23)	1.25 (.50, 3.20)
Excessive alcohol consumptionb
No	880 (64)	197 (23)	Ref	Ref
Yes	505 (36)	132 (26)	1.25 (.48, 3.20)	1.20 (.92, 1.56)

Abbreviations: aOR, adjusted odds ratio; CI, confidence interval; ETHOS, Enhancing Treatment of Hepatitis C in Opioid Substitution Settings; OAT, opioid agonist therapy; OR, odds ratio; Ref, reference.

^aMain drug injected in last month was not included in adjusted analyses due to collinearity with recency and frequency of injecting.

^bNot reported for transgender participants.

In adjusted analyses, factors associated with current HCV infection included homelessness (aOR, 1.47; 95% confidence interval [CI], 1.00–2.16), incarceration history (vs never, >1 year ago: aOR, 1.79; 95% CI, 1.30–2.45; within the last year: aOR, 2.03; 95% CI, 1.38–3.01), and at least daily injecting drug use (aOR, 2.29; 95% CI, 1.45–3.62) (Table 3). In adjusted analyses among PWID in the previous month, factors associated with current HCV infection were unchanged (Supplementary Table 4).

Table 3.

Unadjusted and Adjusted Analysis of Factors Associated With Self-reported Historical Heptatis C Virus (HCV) Treatment Among Total ETHOS Engage Participants Who Had Evidence of Previous or Current HCV Infection

Characteristics	Previous or Current HCV Infection, n (row %)a	Treated, n (row %)	OR (95% CI)	aOR (95% CI)
Total, N	788 (55)	520 (66)	…	…
Age at enrollment
<45 years	395 (50)	237 (60)	Ref	Ref
≥45 years	396 (61)	283 (71)	1.63 (1.22, 2.06)	1.47 (1.07, 2.02)
Gender
Male	543 (58)	372 (69)	Ref	Ref
Female	242 (48)	147 (61)	0.71 (.52, .98)	0.68 (.48, .96)
Transgender	3 (100)	1 (33)	0.23 (.02, 2.55)	Omitted
Indigenous ethnicity
No	613 (55)	410 (67)	Ref	Ref
Yes	175 (52)	110 (63)	0.84 (.59, 1.19)	0.87 (.60, 1.26)
Homeless
No	707 (55)	481 (68)	Ref	Ref
Yes	81 (52)	39 (48)	0.44 (.27, .69)	0.59 (.36, .96)
OAT status
Never	69 (34)	29 (42)	Ref	Ref
Past	90 (54)	53 (59)	1.97 (1.04, 3.73)	1.88 (.97, 3.63)
Current	629 (59)	438 (70)	3.16 (1.90, 5.25)	2.60 (1.51, 4.49)
Incarceration history
Never	196 (42)	130 (66)	Ref	Ref
>1 year ago	435 (61)	291 (67)	1.02 (.72, 1.47)	0.91 (.61, 1.34)
Within last year	157 (61)	99 (63)	0.87 (.56, 1.34)	0.85 (.52, 1.39)
Recency of injecting
>12 months	117 (54)	91 (78)	Ref	Ref
Within 1–12 months	159 (52)	105 (66)	0.56 (.32, .96)	0.65 (.37, 1.14)
Within last month, <daily	273 (55)	189 (69)	0.64 (.39, 1.07)	0.82 (.48, 1.39)
Within last month, ≥daily	239 (56)	135 (56)	0.37 (.22, .61)	0.51 (.30, .86)
Main drug injected in last month
None	276 (53)	196 (71)	Ref	Omittedb
Heroin	188 (60)	132 (70)	0.96 (.64, 1.44)
Other opioids	83 (63)	45 (54)	0.48 (.29, .80)
Methamphetamine	227 (51)	139 (61)	0.64 (.44, .94)
Other	14 (50)	8 (57)	0.54 (.18, 1.61)
Excessive alcohol consumptionc
No	495 (54)	337 (68)	Ref	Ref
Yes	290 (55)	182 (63)	0.80 (.59, 1.08)	0.85 (.62, 1.16)

Abbreviations: aOR, adjusted odds ratio; CI, confidence interval; ETHOS, Enhancing Treatment of Hepatitis C in Opioid Substitution Settings; OAT, opioid agonist therapy; OR, odds ratio; Ref, reference.

^aProportion of overall population (N = 1443).

^bMain drug injected in last month was not included in adjusted analyses due to collinearity with recency and frequency of injecting.

^cNot reported for transgender participants.

Factors Associated With Hepatitis C Virus Treatment

Overall, 55% (n = 788) of participants had evidence of previous chronic (n = 457) or current (n = 331) HCV infection. Among these (n = 788, 55%) (Table 3, Supplementary Figure 1), 66% (n = 520) had self-reported ever initiating HCV treatment (Table 2). The majority (85%) had initiated treatment in the DAA era (2016–2018), and 31% (n = 162) reported receiving HCV treatment at a drug treatment clinic, 28% (n = 148) from a hospital-based specialist clinic, 19% (n = 100) from a general practitioner, 16% (n = 85) in prison, 3% (n = 14) within other community-based clinics, and 2% (n = 9) within an NSP.

Hepatitis C virus treatment was lower in females (vs males; 60% vs 69%), those who were homeless (48% vs 68%), those who had never received OAT (vs those currently receiving OAT; 42% vs 70%), and those with at least daily injecting drug use in the last month (vs ≥1 year ago; 56% vs 78%) (Table 3, Figure 3).

Figure 3.

Self-reported historical HCV treatment among ETHOS Engage participants with evidence of previous or current HCV infection (n = 788). *Main drug injected in the last month. Data for participants injecting other drugs (n = 24) are not shown. Abbreviations: ETHOS, Enhancing Treatment of Hepatitis C in Opioid Substitution Settings; F, female; HCV, hepatitis C virus; M, male; m, month(s); Meth, methamphetamine; OAT, opioid agonist therapy.

In adjusted analyses, HCV treatment was less likely among females (aOR, .68; 95% CI, .48–.96), people who were homeless (aOR, .59; 95% CI, .36–.96), and people with at least daily injecting drug use (vs no injecting in last year; aOR, .51; 95% CI, .30–.86). People aged 45 years or older (vs <45 years; aOR, 1.47; 95% CI, 1.07–2.02) and people receiving OAT (aOR, 2.60; 95% CI, 1.51–4.49) were more likely to receive HCV treatment (Table 3). In analyses restricted to people with recent injecting drug use (n = 921), the main drug injected in the last month was assessed in regression models. The factors associated with treatment were unchanged among this group (Supplementary Table 5).

To further investigate the association between gender and HCV treatment, stratified analyses were performed (Supplementary Tables 3 and 6). In adjusted analyses among males with evidence of HCV infection (ever) (n = 543/932, 58%), HCV treatment was less likely among those who were homeless (aOR, .49; 95% CI, .27–.80), and those with at least daily injecting drug use (aOR, .49; 95% CI, .25–.95). Hepatitis C virus treatment was greater among males who had ever received OAT either in the past (aOR, 2.51; 95% CI, 1.13–5.58) or currently (aOR, 2.86; 95% CI, 1.50–5.49) (Supplementary Table 6). Age was not associated with HCV treatment (aOR, 1.14; 95% CI, .77–1.70).

In adjusted analyses among females with evidence of HCV infection (ever) (n = 242/508, 48%), the only factor independently associated with HCV treatment was age, with females aged 45 years or older more likely to have received treatment compared with those younger than 45 years (aOR, 2.62; 95% CI, 1.47–4.66) (Supplementary Table 6).

DISCUSSION

In this national, well-characterized sample of PWID attending drug treatment clinics and NSPs in Australia, 24% were currently infected with HCV and 66% of individuals who had previous chronic or current HCV infection had ever received HCV treatment. Indicators of higher marginalization were negatively associated with HCV treatment and positively associated with current HCV infection. This study provides important insight into the impact of unrestricted DAA access and will inform policies and targeted strategies to further facilitate HCV elimination in Australia and globally.

Current HCV infection was higher (31%) in participants who reported at least daily injecting drug use. Given the potential for HCV treatment to prevent onward transmission of infection [5, 19], treatment scale-up among people with frequent injecting drug use combined with harm reduction (OAT and NSP) will be critical for HCV elimination, particularly in countries where the majority of new infections occur among PWID. Enhanced support within low-threshold and targeted primary health settings, including individualized, tailored adherence support and peer-to-peer education, has been positively associated with treatment uptake and adherence among people with frequent injecting drug use [20–24]. These strategies should be explored in the context of HCV treatment as prevention.

Treatment uptake was lower (56%) in those injecting daily or more, consistent with previous studies [14, 25]. Despite an increased association with serious injection-related injury and blood-borne virus infection, people who frequently inject drugs are more likely to experience barriers to healthcare access due to discrimination [26]. Such discrimination is associated with lower uptake of OAT and less access to a general practitioner [27, 28], both associated with enhanced HCV knowledge and engagement [23, 29–31]. Overcoming these barriers may be possible through partnerships with peer-based organizations in primary health and harm-reduction settings. Enhancing these partnerships may facilitate psychosocial support mechanisms, leading to improved healthcare-related communication between PWID and healthcare professionals, and greater treatment-related knowledge [24].

Participants who were homeless were more likely to have current HCV infection and less likely to report HCV treatment. These results are unsurprising given the strong associations between unstable housing and injection drug use [32] and the multiple barriers faced by people who are homeless in accessing healthcare: high prevalence of psychiatric comorbidities, competing priorities regarding day-to-day shelter and food security, increased stigmatization, and lack of necessary identification for prescriptions [33, 34]. Interventions to address HCV in the context of these barriers are challenging. Previous work has indicated that housing complications were often cited as a common reason for missing appointments, despite being within models of care that integrate HCV therapy into specialized community medicine or within traditional low-threshold settings [35]. Higher marginalization, such as experiencing homelessness, is associated with loss to follow-up and disengagement with HCV-related services [33, 34, 36]; however, treatment uptake among people who are homeless may be enhanced through one-stop-shop models that utilize point-of-care testing and offer immediate, same-day treatment initiation [20, 36] and policy interventions to improve housing stability [32]. Innovative, holistic strategies to engage people who are homeless with harm-reduction and HCV care are required [33, 34].

Current OAT was associated with higher HCV treatment, consistent with published research [13, 14, 31, 33]. Opioid agonist therapy engagement is associated with increased awareness of HCV therapy and its effectiveness [30]. Furthermore, OAT is associated with reductions across multiple health outcomes, including injecting risk behavior [37], risk of human immunodeficiency virus (HIV) and HCV [38, 39], criminal activity [40], and all-cause and overdose mortality [41]. Ensuring high coverage and access to OAT is critical in achieving HCV elimination and improving health outcomes among PWID. Further, a significant proportion of PWID may not be opioid dependent, and efforts to increase HCV treatment among people who inject stimulants are warranted.

In line with previously published results, age was associated with HCV treatment [14]. Older PWID typically report fewer high-risk injection practices and increased uptake of health-related services, making this group generally easier to reach compared with younger PWID [42]. Surveillance suggests population-level aging of PWID in Australia; however, in some settings, there is a fast-growing population of younger PWID at risk of, or infected with, HCV and should therefore be considered a key population to engage in HCV care [43, 44]. While this study was insufficiently powered to analyze outcomes solely among younger PWID (<25 years; n = 57), these results suggest the importance of monitoring HCV and treatment initiation among this group.

The gender-specific differences in reported HCV treatment corroborate previous evidence, with females less likely to initiate treatment than males [45, 46]. Among women, the only independent factor associated with HCV treatment uptake was age, with older women more likely to have received treatment than younger women. Despite these results, gender was not associated with current HCV infection, related to the higher likelihood of spontaneous clearance among females [47]. Previous work has highlighted increased marginalization among women who inject drugs and the higher vulnerability in this population that contributes to disengagement with healthcare [45]. The intersectionality of gender, age, and other factors—such as ethnicity and receipt of OAT—is associated with treatment deferral [45, 46]. Gender-specific interventions that reduce vulnerabilities and marginalization among women who inject drugs are key. Further research is necessary to understand the complexity of treatment deferral among younger women.

Considering the criminalization of drug possession in Australia, the proportion of participants who had a history of incarceration (68%) was unsurprising. While not a factor associated with HCV treatment, incarceration was significantly associated with current HCV infection, highlighting prisons as key settings for HCV prevention and treatment. Although injecting frequency attenuates following incarceration, among those who continue to inject there is increased sharing of needles and syringes [48]. Increased coverage of harm-reduction and novel person-centered strategies may be needed to ensure prevention, timely diagnosis, and initiation onto HCV therapy, both in prison and after release [49, 50].

This study has limitations. Serological status was based on self-report and virology, potentially underestimating true HCV antibody prevalence; however, the inferred prevalence found here is similar to annual surveillance of PWID in Australia [43] and the utilization of the Xpert Viral Load Finger-stick assay for HCV RNA has allowed characterization of current HCV prevalence among the vast majority of participants (>95%), differentiating these results from previous studies [13]. Furthermore, participation in ETHOS Engage was voluntary, and recruitment was performed in healthcare settings, the majority of which operated primarily as, or in conjunction with, OAT. The annual Australian NSP survey indicates that nearly half of PWID last injected methamphetamine (48%) and daily or more (51%). While it is encouraging that this study was able to engage a large population of PWID who were mainly injecting methamphetamine (31%) and injecting drugs daily or more (30%), these results may be underrepresentative of the wider injecting population. This oversampling of PWID engaged in OAT has potentially introduced selection bias, possibly overestimating HCV treatment and underestimating current infection compared with a wider population of PWID. Finally, questionnaire data rely on recall and self-report. Although self-report is considered a reliable source of data collection among people who use drugs, some may not have provided accurate answers [51]. While recall bias could not be systematically minimized, we aimed to reduce social-desirability bias by providing self-administered surveys and ensuring anonymity.

Conclusions and Implications

In the context of HCV elimination, high treatment uptake across subpopulations was encouraging. These results highlight the successes of an unrestricted HCV treatment strategy in reaching marginalized populations of PWID and suggest progress towards achieving incidence-related HCV elimination targets. It is estimated that, among the 93 500 people who have recently injected drugs in Australia, an estimated 37 500 are infected with HCV [2, 52]. As such, it will be critical to enhance efforts to engage with the most marginalized PWID subpopulations, including people who are homeless or incarcerated, to maintain this progress. To engage those who remain untreated and those who may require follow-up and retreatment, interventions that reduce barriers to testing and treatment, including utilization of dried blood spot and point-of-care technology and provision of financial incentives to initiate treatment within drug clinics should be further explored [53–55]. Additionally, the importance of a peer workforce for the facilitation of HCV elimination should not be underestimated [22, 24].

Although largely indicative of a good news story on the path towards elimination of HCV among PWID, challenges remain. It is imperative that innovative strategies and holistic approaches to improve linkage to HCV-related care are adopted to further enhance engagement with people living with HCV who may delay treatment for competing priorities. There is an urgent need for increased efforts to address the gaps in care highlighted here to ensure that HCV elimination is equitable across all PWID in Australia and globally.

Supplementary Data

Supplementary materials are available at Clinical Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.

Notes

Author contributions. J. G., G. J. D., and P. M. conceived and designed the Enhancing Treatment of Hepatitis C in OpioidSubstitution Settings (ETHOS) Engage study, including construction of the participant questionnaire. H. V., J. G., G. J. D., and M. A. contributed to these research aims, analysis, and interpretation of results. H. V., under the supervision of J. G., wrote the first draft of the manuscript, and all coauthors contributed to the critical review and development of this final manuscript for publication.

Acknowledgments. The authors thank all participants who took part in the ETHOS Engage study. The authors first give special acknowledgement to the following peer workers and organizations who helped invaluably with participant recruitment: The NSW Users and AIDS Association (NUAA): Sara Adey, Rodd Hinton, Melanie Joyce; Youth Link Needle and Syringe Program, Cairns: Astrid Carthew; Hepatitis South Australia: Lisa Carter, Carol Holly; Harm Reduction Western Australia: Lyn Murphy. We also thank the contributions of members of the ETHOS Engage Study Group: Protocol Steering Committee: Jason Grebely (Kirby Institute, University of New South Wales [UNSW] Sydney), Gregory J. Dore (Kirby Institute, UNSW Sydney), David Silk (Kirby Institute, UNSW Sydney), Nicky Bath (LGBTI Health Programming and Development), Carla Treloar (Centre for Social Research in Health, UNSW Sydney), Andrew Milat (Centre for Epidemiology and Evidence, NSW Health), Adrian Dunlop (Hunter New England Local Health District), Janaki Amin (Macquarie University, Kirby Institute, UNSW Sydney), Jo Holden (Population Health Strategy and Performance, NSW Health), Charles Henderson (NUAA), Kyle Leadbeatter (Hepatitis NSW), Emma Day (Australasian Society for HIV, Viral Hepatitis and Sexual Health Medicine [ASHM]), Nikitah Habraken (ASHM), Louisa Degenhardt (National Drug and Alcohol Research Centre, UNSW Sydney), Clarke Scott (Centre for Addiction Medicine, Nepean Blue Mountains Local Health District), and Phillip Read (Kirketon Road Centre). Coordinating Centre (The Kirby Institute, UNSW Sydney): Jason Grebely (Co-Principal Investigator), Gregory J. Dore (Co-Principal Investigator), Maryam Alavi, David Silk (Clinical Project Coordinator), Heather Valerio (PhD candidate), Shane Tillakeratne (Data Manager), Pip Marks (Clinical Trials Manager), Indika Jayasinghe (Laboratory Coordinator), Hannah Reid, Valerie Gleeson, Jodi Van Dyk, Gerard Estivill Mercade, Alison D. Marshall, Stephanie Obeid, Samira Hosseini Hooshyar, Beth Catlett, Andrey Verich (campaign day implementation). Site Principal Investigators: Nadine Ezard (Rankin Court), David Reid (Illawarra Shoalhaven Local Health District), Carla Gorton (Cairns Sexual Health Service), Jeremy Hayllar (Metro North Hospital and Health Service, Brisbane), Thao Lam (Western Sydney Local Health District), Adrian Dunlop (Hunter New England Local Health District), Prasun Datta (Nepean Blue Mountains Local Health District), Alex Wade (Mid North Coast Local Health District), Sally Spruce (Mid North Coast Local Health District), Vicky Cock (Drug and Alcohol Services, South Australia), Mark Cornwell (Northern NSW Local Health District), Michael Christmas (Next Step, Perth), Craig Connelly (Next Step, Perth), Angela Cooper (Townsville Hospital and Health Services), Mark Montebello (Northern Sydney Local Health District). Site Coordinators: Robert Cherry (Rankin Court), Julie Dyer (Rankin Court), Shikha Arawal (Rankin Court), Nadine Horasak (Youth Link Needle and Syringe Program, Cairns), Rhondda Lewis (Cairns Sexual Health Service), Kathy Clark (Youth Link Needle and Syringe Program, Cairns), Daniel Morris (Youth Link Needle and Syringe Program, Cairns), Kathy Donohue (Youth Link Needle and Syringe Program, Cairns), Kathy Griffiths (Biala), Jason Dalla Lana (Biala), Sue Shin (Biala), Connie Graf (Illawarra Shoalhaven Local Health District), Adele Hampson (Illawarra Shoalhaven Local Health District), Carina Burns (Southwest Sydney Local Health District), Ravina Raidu (Southwest Sydney Local Health District), Kylie Stolzenhein (Southwest Sydney Local Health District), Wanda Brabender (Southwest Sydney Local Health District), Nargis Abram (Nepean Blue Mountains Local Health District), Rick Turner (Nepean Blue Mountains Local Health District), Stuart Larter (Nepean Blue Mountains Local Health District), Fiona Goodberg (Nepean Blue Mountains Local Health District), Jennifer Luksza (Western Sydney Local Health District), Michelle Hall (Hunter New England Local Health District), Susan Hazelwood (Hunter New England Local Health District), Krista Zohrab (Mid North Coast Local Health District), Belinda McClurg (Mid North Coast Local Health District), Kate Salisbury (Northern NSW Local Health District), Julie Markham (Queensland Injectors Health Network, Townsville), Jacky Talmet (Drug and Alcohol Services, South Australia), Sandy Dunn (Drug and Alcohol Services, South Australia), Fionnualh Smyth (Western Sydney Local Health District), Lisa Snell (Western Sydney Local Health District), Elizabeth Laing (Next Step, Perth) Martin Clark (Next Step, Perth), Justin Dorigo (Next Step, Perth), Brent Fergusson (Townsville Hospital and Health Service), Bonny Puszka (Northern Sydney Local Health District), Gai Duncan (Northern Sydney Local Health District), Fiona Baker (Northern Short Local Health District), and Jayde Walsh (Northern Sydney Local Health District).

Disclaimer. Cepheid and Merck, Sharp, and Dohme Corporation were not involved in the study design, methodology, and writing of this manuscript. The opinions expressed in the paper are those of the authors and do not necessarily represent those of Cepheid or Merck Sharp, and Dohme Corporation. The views expressed in this manuscript do not necessarily represent the position of the Australian Government.

Financial support. The ETHOS Engage study is funded by a National Health and Medical Research Council (NHMRC) Partnership project grant (grant number 1103165), including funding from New South Wales Health and Cepheid. This study was also supported in part by a research grant from Investigator-Initiated Studies Program of Merck, Sharp and Dohme Corporation. The Kirby Institute is funded by the Australian Government Department of Health and Ageing. L. D. is supported by a US National Institute of Health (NIH), National Institute on Drug Abuse (NIDA) grant (R01DA1104470). G. J. D. is supported by an NHMRC Practitioner Fellowship. J. G. is supported by an NHMRC Investigator Grant. The contents of the published material are solely the responsibility of the individual authors and do not reflect the views of NHMRC.

Potential conflicts of interest. J. G. is a consultant/advisor and has received research grants from AbbVie, Cepheid, Gilead, Hologic, Indivior, and Merck, and personal fees from AbbVie, Cepheid, Gilead, and Merck, outside the submitted work. G. J. D. is a consultant/advisor and has received research grants from Merck, Gilead, Bristol Myers-Squibb, and AbbVie and reports travel support from Gilead, AbbVie, and Merck, outside the submitted work. C. T. has received speaker fees from AbbVie and Gilead, grants from Merck and Bristol Myers-Squibb, and an unrestricted education grant from Terumo. P. R. has received speaker fees from Gilead, AbbVie, and Merck, and research funding from Gilead. L. D. has received investigator-initiated untied educational grants for studies of opioid medications in Australia from Indivior, Mundipharma, and Seqirus. J. Hayllar has received travel, accommodation, speaker fees, and WOWS support from Janssen, Lundbeck, Servier, and Invivior. A. D. reports grants from Braeburn/Camurus. All other authors report no potential conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.