Interventions to increase linkage to care and adherence to treatment for hepatitis C among people who inject drugs: A systematic review and practical considerations from an expert panel consultation.

BACKGROUND: Following advances in treatment for hepatitis C (HCV), optimizing linkage to care and adherence to treatment of people who inject drugs became of pivotal importance. An ECDC/EMCDDA stakeholders survey in 2018 indicated that two components of the cascade of care, linkage to care and adherence to treatment, were priority areas for inclusion in the updated guidance, planned for publication in 2022. This systematic review was commissioned with the aim to evaluate the effectiveness of interventions on HCV linkage to care and adherence to treatment among people who inject drugs.
METHODS: The full study protocol is available at PROSPERO (2020 CRD42020191116). We searched five databases to identify studies published between 2011 and 2020. Studies were included if they had a comparative study design and reported on the primary outcomes for linkage to care (visits, treatment initiation) and/or adherence to treatment (treatment adherence, treatment completion, SVR12) among people who inject drugs/people receiving opioid substitution therapy. Following the risk of bias (EPHPP) and quality of evidence assessment (GRADE), evidence to decision tables were produced and shared for critical review with an expert panel convened by ECDC and EMCDDA. The expert panel provided further considerations on the benefit, acceptability, and transferability of interventions.
RESULTS: Fourteen studies (using DAA therapy as well as DAA plus interferon-based therapy) met eligibility criteria and were included in the analysis. Integrated care with case management, peer support, psychological interventions, contingency management, and cooperation between health care providers improved engagement in and adherence to HCV care in most studies. However, the quality of evidence was predominantly low to moderate.
CONCLUSIONS: Available evidence suggests that integrated, people-centered approaches may improve engagement throughout the continuum of HCV care among people who inject drugs. For progressing HCV elimination efforts, interventions should be implemented in colocation with harm reduction and counselling activities and in combination with additional services, including opioid substitution treatment, directly observed therapy, peer support and/or contingency management.

Background

Hepatitis C virus (HCV) infection is a major public health threat due to disease burden and risk of complications and death (WHO, 2021). Injecting drug use is the most likely mode of HCV transmission reported in the EU/EEA and accounted for 49% of acute and 61% of chronic infections in 2018 (ECDC, 2020a). Compared to other drug-associated blood-borne viral infections, HCV is the most prevalent one among people who inject drugs (PWID) across Europe (ECDC, 2020b). Prevalence of HCV antibody among PWID estimated from nationally representative samples ranged between 15% and 86% during 2018–2019. The prevalence of current infections measured by HCV-RNA (or antigen) tests ranged from 15% to 64% between 2013 and 2019 in six countries with available data (EMCDDA, 2021a). PWID are therefore considered as a priority population in prevention, testing, linkage to care and treatment, and prevention of re-infections to achieve HCV elimination (EMCDDA, 2016; Grebely et al., 2018; Day et al., 2018).

PWID are indicated as a priority population to be targeted for voluntary, integrated testing for blood-borne infections by ECDC guidance on HIV, hepatitis B and C testing (European Centre for Disease Prevention and Control and European Monitoring Centre for Drugs and Drug Addiction 2018). The guidance provides details on testing frequency, settings and testing technologies and highlights the need to facilitate linkage to care and access to treatment of those that tested positive. The introduction of highly effective, short-course direct-acting antiviral (DAA) treatment provides major advantages in regard to achieving a cure of patient populations previously defined as difficult to cure (Alimohammadi, Holeksa, Thiam, Truong, & Conway, 2018; Asselah, Boyer, Saadoun, Martinot-Peignoux, & Marcellin, 2016; Iversen et al., 2013, 2017; WHO, 2017a) or even excluded from treatment such as PWID. WHO recommends DAA treatment to all those diagnosed with chronic HCV infection, including PWID, irrespective of drug use patterns and independent of currently receiving drug dependence treatment or not (WHO, 2018). Offering HCV treatment to all PWID will also prevent onward transmission of HCV with a public health benefit (EASL, 2018).

Linkage to care and adherence to treatment are considered core components of the HCV care cascade and together with HCV testing are pivotal for HCV elimination. The HCV cascade of care defines the steps required to progress from diagnosis to cure and is generally structured around the steps of testing, linkage to care, treatment, viral suppression or the effectiveness of treatment, though the exact outcomes measured vary by study (WHO, 2017b). Linkage to care connects individuals following the diagnosis of HCV infection, often identified through screening programs, to a care provider to initiate treatment. While guidelines recommend that all people with active HCV should be engaged in care (EASL, 2018), the loss to follow-up after receiving a diagnosis is a further challenge, contributing to delayed or no treatment initiation (Jordan, Perlman, Reed, Smith, & Hagan, 2017; Meyer et al., 2015; WHO, 2017a). The goal of treatment within the HCV care continuum is to attain sustained virologic response (SVR) as an indicator for the effectiveness of treatment. Viral load is the most widely used efficacy endpoint in clinical studies of HCV and is measured 12–24 weeks after completion of treatment (WHO, 2017b).

A target of 90% of HCV-diagnosed patients linked to care and adequately monitored, and of 75% of patients eligible for treatment receiving treatment were set by WHO European Action Plan target for 2020. As of 2017, according to data reported by seven EU/EEA countries, the percentage of HCV cases linked to care and successfully treated remains low, ranging from 2.3% in Denmark to 55.3% in Romania (ECDC, 2020b). Further on, only one of the 12 EU/EEA reporting countries have achieved in 2017 having 75% of patients diagnosed with chronic HCV infection and eligible for treatment, receiving treatment (ECDC, 2020b).

It is critical to monitor how well patients proceed along the HCV care cascade, including appropriate linkage to HCV-specific care and treatment. Such measures across the cascade provide an important basis for comparison and highlight factors impacting progression along the cascade. For example, factors that may hinder successful engagement in HCV treatment for PWID populations can be located at the system level (e.g., national regulatory framework or organization of testing and treatment, stigma and discrimination in society, lack of epidemiological monitoring), provider level (e.g., low level of HCV knowledge in drug services, current drug use defined as exclusion criteria for HCV treatment, lack of time or funding, insufficient staff or information material) as well as client level (e.g., insufficient knowledge of HCV and treatment options, fear of stigma, access barriers including competing problems such as lack of housing) (EMCDDA, 2021b). Monitoring the HCV affected population across stages of a cascade of care provides a measure of program effectiveness and informs targets for improved clinical care and public health intervention.

In 2011, ECDC and EMCCDA published a joint guidance for prevention and control of infectious diseases among PWID, recommending testing for and treatment of infections among the seven key interventions that also included harm reduction interventions (provision of injecting equipment, drug dependence therapy), vaccination and health promotion (ECDC & EMCDDA, 2011). A stakeholders survey in 2018, conducted for the purpose of guidance update, identified linkage to care and adherence to treatment for HCV as priority areas for inclusion in the updated edition (ECDC & EMCDDA, 2018). As a result, ECDC commissioned a systematic literature review with an aim to evaluate the effectiveness of interventions on linkage to care and adherence to treatment to inform the first update of ECDC's and EMCDDA's guidance on the prevention and control of infectious diseases among people who inject drugs.

With the availability of simple, well-tolerated, and highly effective DAA therapies that are now the treatment of choice, our review adds to the previously published work limited to the interferon treatment era, such as Bajis et al. (2017), by providing a contemporary approach in line with WHO and EASL treatment guidelines. This article presents the outcomes on two core components of the HCV care cascade among PWID: interventions to optimize linkage to care following HCV diagnosis; and achieving a successful treatment outcome (adherence and/or completion). We also summarize considerations by an expert panel on relevance, benefit and acceptability of HCV treatment interventions for PWID, as well as on implementation and transferability of interventions to enhance linkage to care and adherence to treatment for PWID to European settings.

Methods

This systematic review was conducted and reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (Page et al., 2021). Inclusion criteria, methods of analysis, PICO questions and search strategy were specified a priori and are available at PROSPERO (registration no.: CRD42020191116). This review is part of a larger review commissioned by ECDC and conducted to identify interventions that can increase linkage to care and/or adherence to treatment for hepatitis B and C, HIV, and tuberculosis (ECDC, 2022a). Here we focus only on HCV for PWID given the considerably larger body of evidence identified (in contrast with no studies retrieved for HBV, only four for HIV and one for tuberculosis) and due to availability of DAA therapy that is efficacious and well tolerated by PWID (EASL, 2020) and would allow those linked to care an HCV cure. The research question that guided this review was formulated as: “What interventions are associated with improved linkage to care and adherence to treatment regimens for HCV among PWID?”

Eligibility criteria

The PICO framework in this article is adapted from the methodological approach used in the original review (ECDC, 2022a) and defines more specific selection criteria focused on HCV and referring to DAA treatment regimens or combination of interferon/DAA. Thus, studies were included if they met the following criteria:

Population: People who inject drugs (PWID) or at least 50% of the study sample were comprised of people who reported ever injection drug use or receiving opioid substitution therapy (OST) with chronic HCV infection over the age of 14 years

Intervention: Intervention(s) aimed at improving engagement of study population at any (or combination) of the following stages along the chronic HCV care cascade:

linkage to care, defined as the clinical assessment of HCV infection or liver disease following diagnosis of HCV infection

adherence to treatment, with regimens combining interferon/DAA or solely DAA

Comparator: Comparative group composed of participants allocated to receiving care as usual or routine care as defined by study authors. For non-randomised studies, comparative groups include historical comparisons or before and after intervention implementation, or control groups derived by convenience sampling

Outcomes: the primary outcomes were

For linkage to care:

proportion of study population with chronic HCV infection that came in contact with a care provider (“visit”) and/or,

proportion of study population with chronic HCV infection initiating HCV treatment (“treatment initiation” as defined by the study authors)

For adherence to treatment:

proportion of study population adherent to HCV treatment and/or completing the HCV treatment course

SVR12 or SVR24 as surrogate endpoint

Studies that were not presented as peer-reviewed scientific articles or conference abstracts, study protocols, review articles including systematic reviews and non-comparative studies were excluded. Public health interventions targeting health care providers as opposed to individuals were also excluded. Concerning geographical considerations, studies in EU/EEA/EFTA member states, candidate countries to the European Union and comparable countries as Australia, Canada, New Zealand, United Kingdom, United States were included.

Information sources

The following electronic databases of literature were searched for full-text and abstract entries published between 2011 and 8 July 2020: MEDLINE (Ovid MEDLINE(R) 1946 to 6 July 2020), Embase, Cochrane Central Register of Controlled Trials (CENTRAL), Cochrane Database of Systematic Reviews (CDSR), APA PsycINFO 1806 to June Week 5 2020. The search strategy followed PRESS 2015 Guideline Evidence-Based Checklist by McGowan et al. (2016). In addition, ClinicalTrials.gov was searched for registered ongoing or unpublished studies in August 2020 and updated in April 2021 but yielded no additional results. Reference lists of selected articles and relevant review articles retrieved during the initial search were handsearched and forward citation checks were performed to identify additional studies.

Search strategy

A comprehensive list of search terms related to linkage to care and/or adherence to treatment among PWID and/or people in OST was used to develop a search strategy for each electronic database. Search strings for each database were peer-reviewed by the ECDC librarian not involved in the project. Search strings were formulated using a combination of keywords and indexed subject headings (MeSH and EMTREE terms). Terms within each of the population and hepatitis C disease groups were combined using the ‘AND’ operator; and the terms within groups belonging to each of the two components of the HCV care cascade were combined using the ‘OR’ operator. Indexed subject heading were broadened using the ‘explode’ operator. The detailed list of search terms used to search all the electronic databases and full search strategies used for the database are included in the Supplementary material (Supplementary data 1).

Study selection

Data retrieved through the search strategy were imported into EndNote X9 bibliographic software (Clarivate Analytics, Philadelphia, US) and duplicates removed. Titles obtained were screened by two reviewers (L.F. and I.R.) using Covidence systematic review software (Veritas Health Innovation, Melbourne, Australia. Available at www.covidence.org) and irrelevant citations were removed. A further abstracts screening was performed by two reviewers according to pre-defined selection criteria (T.S. and L.F.). Assessment of full text of all abstracts that passed eligibility criteria was independently conducted by two reviewers. (T.S. and I.R.). Any disagreements were resolved by discussion between the two reviewers and a third team member until consensus was reached. Reasons for exclusion were reported. Full text screening was done in Endnote X9. The results are presented in a PRISMA flow diagram (Fig. 1).

Fig. 1

Study selection process.

* The original review (ECDC, 2022a) included 20 HCV studies. Due to advances in therapy (DAA being recommended by current treatment guidelines), six HCV studies reporting interventions in interferon only era were subsequently excluded. The PICO framework in this article is adapted from the methodological approach used in the original review (ECDC, 2022a) and defines more specific selection criteria focused on HCV and referring to DAA treatment regimens or combination of interferon/DAA.

Data extraction and risk of bias assessment

Studies that fulfilled inclusion criteria for primary outcome (quantitative data) were extracted by two reviewers independently (T.S. and I.R.) in a standardised extraction form (Microsoft Excel 2010 Redmond, WA, USA) which was piloted in advance. The following variables were extracted: first author, title, publication year, study design, study period, study location, population characteristics (population in% of ever- and/or recent-injecting drugs or % of participants on OST), setting, intervention description (incl. duration), comparator description, sample size (including intervention and control arms), outcome description, number of participants achieving the outcome of interest (and proportions, if applicable) in both intervention and control arms, funding source and conflict of interest. To maximize comparability between studies, data were extracted and analysed according to intention to treat when possible, even if individual studies reported results or conclusions based on per-protocol analyses.

As stated in the PROSPERO protocol (2020 CRD42020191116), the use of Cochrane Collaboration's Risk of Bias 2 tool (revised tool for Risk of Bias in randomized trials) and the ROB-INS-1 tool (Risk of Bias in non-randomized studies - of Interventions) were intended for the risk of bias assessment. Both tools were pre-piloted with two NRS and three RCTs. ROBINS-1 tool was not appropriate for the assessment of NRS included for evidence synthesis because some did not meet the key criteria of the Cochrane “Study Quality Guide”. Following data extraction, risk of bias was assessed by one reviewer (I.R.) for both randomized controlled trials (RCTs) and non-randomised studies (NRS) using the Effective Public Health Practice Project (EPHPP) quality assessment tool (EPHPP, 2009). This tool accounts for the different study designs of the included studies. For consistency reasons, both study types (NRS and RCTs) were assessed with the EPHPP-quality assessment tool taking into account the accompanying EPHPP dictionary. Each study was examined against six dimensions: selection bias, study design, confounders, blinding, data collection methods, withdrawals and dropouts. The quality of each study was graded as strong, moderate or weak according to the individual ratings attributed to each dimension.

Data analysis and quality assessment

Given conceptual heterogeneity across studies, and variations in study designs, populations, interventions, and outcomes, a meta-analysis was not conducted. Further, reporting and publication bias could not be assessed with e.g. Eggers test, as there were not enough studies available. The characteristics and findings of the included studies were summarised and structured using tables. Forest plots were generated to show the main features of the studies identified and comprehensively visualizing study size, effect, and confidence interval of the included studies.

Characteristics were compared across studies and used to group studies based on similarity of interventions. Studies which were clearly or predominantly attributable to a single type of intervention were assigned to a specific category. Studies combining multiple interventions, covering multiple needs of the PWID (packages of diverse and/or multi-disciplinary interventions) were grouped into “Integrated services and case management”.

The Risk Ratio (RR) and corresponding 95% CI was calculated for each study outcome by the project team, using the initial number of eligible participants included and the number achieving the outcome of interest in each arm. The statistical significance of an outcome was inferred from the 95% CI of RR. Analyses were conducted using MS Excel 365 (Microsoft, Redmond, WA, USA). The quality of evidence was assessed using Grading of Recommendations Assessment, Development, and Evaluation (GRADE) system (Guyatt et al., 2008). GRADE rates the quality of evidence for each outcome as high, moderate, low or very low. The strength of evidence expresses the level of confidence that the existing evidence correctly assesses the effect of the intervention.

Expert panel consultation

The effective interventions associated with linkage to care and adherence to treatment were identified through this review was critically reviewed by a panel of 21 experts covering a range of disciplines and affiliated with public health, clinical, community and harm-reduction organisations widely located across Europe. The expert panel members were appointed by ECDC and EMCDDA after assessment of declarations of conflict of interest. An Evidence to Decision (EtD) framework was developed based on a combination of GRADE (Burchett, Blanchard, Kneale, & Thomas, 2018) and SIGN 50 (Moberg et al., 2018) to support the expert panel to examine the evidence gathered through this review and, based on this, advise on recommendations. The expert panel met online in March 2021 when the discussions were guided by a structured agenda following the evidence to decision framework. They provided considerations for public health practice aimed to enhance the link between evidence and real-world settings considering the diversity of European healthcare systems and legal frameworks, and PWID patient needs. A summary report of discussions and proposed changes after the expert meeting was compiled by two project team members (I.H. and T.S.) (ECDC, 2022c). The expert panel suggestions and recommendations regarding the benefits, acceptability and transferability of interventions to enhance linkage to care and adherence to treatment are integrated in the Discussion section.

Results

Study selection

The numbers of studies identified, screened and selected and the reasons for exclusion are summarised in Fig. 1.

The literature search on interventions that can improve linkage to care and adherence to treatment for HCV, HBV, HIV and TB among PWID/ people on OST identified in total 11 075 records including one added through backward/forward citation check. After excluding duplicates, systematic reviews and meta-analysis, 7 318 remained for title and abstract screening. Of these, 251 were eligible for full text screening. Following the full-text review, 226 studies were excluded. A total of 25 studies for HCV, HBV, HIV and TB were identified of which 20 studies were identified for HCV. Of the 20 studies targeting linkage to HCV care or treatment adherence, six interventions evaluated interferon (IFN) regimens and 11 direct-acting antivirals (DAA) regimens. Three studies evaluated both treatment regimens. The availability of highly effective, interferon-free, short-course DAA treatment has transformed the clinical and public health management of HCV infection which is why this research focuses on DAA studies. A total of 14 studies were therefore identified for inclusion in the present review and analysis.

Study characteristics of included studies

The characteristics and results of included studies for each of the two outcomes are summarised in Tables 1 and 2. Forest plots for each of the two outcomes stratified by intervention type are shown in Fig. 2.

Table 1

Included studies of HCV linkage to care.

Author (year)	Study design	Location	Setting	Population (Ever IDU%; recent IDU %; OST %)	Intervention description	Comparator description	Outcome description	Intervention n/N (%)	Control n/N (%)	RR (95% CI)
Contingency Management
Norton et al. (2019)	NRS, convenience sampling	US	NSP center providing mental health counselling, referrals and other services	NSP clients (NA; 44%; NA)	Financial incentives for HCV visits, return of medication blister packs and successful early clinical outcome	Enhanced usual care: expedited appointment at center, round-trip transit fare card, reminders from care coordinator	Visit (conducting baseline HCV evaluation within 3 months); Treatment initiation (receiving prescriptions for at least one DAA within 1 year of baseline visit)	Visit: 14/19 (74%)TI: 9/12 (75%)	Visit: 6/20 (30%)TI: 4/4 (100%)	Visit: RR 2.46 [1.19, 5.05]TI: RR 0.75 [0.54, 1.04]
Ward, Falade-Nwulia, et al. (2019)	RCT	US	Clinic for HIV care	Patients from an outpatient clinic (NA; 46%; 28%)	Contingent cash incentives	Enhanced usual care: treatment according to standard protocol involving clinical visits and calls delivered by a nurse-led multidisciplinary team; all participants received $10 to $30 per visit	Treatment initiation (within 8 weeks)	41/54 (76%)	24/36 (67%)	RR 1.14 [0.86, 1.50]
Telemedicine
Cooper et al. (2017)	NRS	CA	Hospital and regional HCV program	Patients from hospital and regional HCV program (TM: 70.1%; NA; NA / Non-TM: 54.9%; NA; NA)	Majority of clinic visits conducted utilizing the Ontario TeleHealth Network video and audio system. Patient and remote site TM nurse are linked by audio and video to hospital site at which the HCV clinician, nurse and allied health care providers are located	Usual care: all assessments conducted at hospital outpatient clinic based within a tertiary care center	Treatment initiation	27/43 (62.8%)	319/608 (52.5%)	RR 1.07 [0.83, 1.39]
Peers
Broad et al. (2020)	RCT	CA	Three HCV program sites	Participants recruited from peer out-reach worker's personal networks or strangers/acquaintances found in non-health care settings (100%; 65.8%; NA)	Point of care HCV testing and linkage to care via outreach workers (research assistants with lived experience of HCV hired, trained and supported to deliver HCV education and POC HCV antibody testing)	Usual care, testing as usual (no POC testing)	Visit (within 6 months)	6/195 (3%)	5/185 (3%)	RR 1.17 [0.36, 3.77]
Ward, Falade-Nwulia, et al. (2019)	RCT	US	Clinic for HIV care	Patients from an outpatient clinic (NA; 46%; 28%)	Peer mentors (successfully treated for HIV and HCV); face-to-face meetings with mentees, contacted mentees before, during and after treatments	Enhanced usual care: treatment according to standard protocol involving clinical visits and calls delivered by a nurse-led multidisciplinary team; all participants received $10 to $30 per visit	Treatment completion (after 12 weeks DAA treatment); SVR (12 weeks after treatment)	45/54 (83%)	24/36 (67%)	RR 1.25 [0.96, 1.62]
Integrated services & case management
Radley et al. (2020)	RCT, cluster	UK	55 randomly assigned community pharmacies	Patients receiving OST in pharmacies for at least 3 months (NA; NA; 100%)	Modified directly observed therapy (mDOT) in pharmacy with DAA alongside supervised OST	Enhanced usual care within pharmacist pathway: HCV infection discussed with patients, testing offered if HCV status unknown, referral to treatment center offered if HCV positive	Treatment initiation	112/1 365 (8%)	61/1 353 (5%)	RR 1.82 [1.34, 2.46]
Ward, Falade-Nwulia, et al. (2019)	RCT	AU, NZ	13 primary care sites (INT) and specialist based local hospitals (CON)	Patients attending a primary care study site (INT: 100%; 49%; 77% / CON: 95%; 49%; 69%	Hospital employed community hepatitis nurses provided HCV education, assessments, linkage to care and support whilst patients were in treatment at primary care site; general practitioners provided OST	Usual care in local hospital specialist clinic; OST provided in primary care sites	Treatment initiation	43/57 (75%)	18/53 (34%)	RR 2.48 [1.54, 3.95]
Starbird et al. (2020)	RCT	US	Outpatient clinic for HIV and HCV care	Patients from clinic providing HIV primary care and HCV specialty care (NA; 24%; 52%)	Nurse case manager initiated HCV referral and assisted to schedule an appointment, discussion of barriers, reminders; HCV education (coaching participants, identify their strengths)	Usual outpatient care plus HCV fact sheet	Visit (at hepatitis practice within 60 days after enrolment); Treatment initiation (within 180 days)	Visit: 16/34 (47%)TI: 4/34 (12%)	Visit: 8/32 (25%) TI: 8/32 (25%)	Visit: RR 1.88 [0.94, 3.78]TI: RR 0.47 [0.16, 1.41]
Ho et al. (2015)	RCT	US	Three Veteran Affairs (VA) HCV medical centres	Patients attending VA with substance use and/or psychiatric risk factors for antiviral treatment (NA; 47%; NA)	Integrated care protocol with case management, incl. brief psychological interventions and contingent cash incentives provided in collaboration with clinic physicians, nurses, and other mental health providers (IFN+DAA)	Usual care within the HCV clinic or referral to standard mental health and substance use clinics for further treatment (IFN+DAA)	Treatment initiation	58/182 (31.9%)	34/181 (18.8%)	RR 1.70 [1.17, 2.46]
Messina et al. (2020)	NRS, prosp. cohort, pre/ post	IT	Facility for Substance Use Disorder (SUD)	Clients of an outpatient service for substance use disorders (100%; NA; NA)	Cooperation between SUD and Infectious Disease Unit incl. periodic pros. audits conducted by the infectious disease consultants to improve HCV knowledge; including screening, testing and starting treatment	Usual care in pre-intervention period without cooperation and HCV education	Treatment initiation	45/55 (82%)	3/16 (19%)	RR 4.63 [1.56, 12.19]

Abbreviations: AT Austria; AUT Australia; CA Canada; CON control; DAA direct-acting antivirals; DE Germany; ES Spain; FR France; HIV human immunodeficiency virus; IFN interferon; INT intervention; IT Italy; mDOT modified directly observed therapy; NA not available/applicable; NRS non-randomised study; NSP needle and syringe programmes; NZ New Zealand; OST opioid substitution treatment; POC Point of care; PWID persons who inject drugs; RCT randomised controlled trial; RR risk ratio; SVR sustained virologic response; TA treatment adherence; TC treatment completion; TI treatment initiation; UK United Kingdom; US United States; VA Veteran Affairs.

Table 2

Included studies of HCV adherence to treatment.

Author (year)	Study design	Location	Setting	Population (Ever IDU %; recent IDU %; OST %)	Intervention description	Comparator description	Outcome description	Intervention n/N (%)	Control n/N (%)	RR (95% CI)
Directly observed therapy
Akiyama et al. (2019)	RCT	US	OST-Centres	PWID from 3 OST programs (75%; NA; 100%)	Directly observed therapy linked to OST methadone visits, for non-pick up days doses packed in electronic blister packs for self-administration at home (IFN+DAA)	Usual care via self-administered medication at home (1-month supply was given at OST visit)	Treatment adherence (using daily timeframe); SVR (12 weeks after treatment)	TA: NA (86%) SVR12: NA (94%)	TA: NA (75%) SVR12: NA (87%)	TA: RR 1.15 [0.95, 1.39] SVR12: RR 1.08 [0.95, 1.22]
Coffin et al. (2019)	RCT, pilot	US	Community based research center	Patients recruited from HCV incidence cohort study (100%; 45%; NA)	Modified directly observed therapy (mDOT) incl. motivational interviewing, cash compensation and counselling for injection risk reduction and medication adherence	Unobserved treatment/dosing with Wisepill dispenser	Treatment adherence (mean weekly visit completion); SVR (12 weeks after treatment)	TA: NA (96.3%); SVR12: 18/20 (89.5%)	TA: NA (96.6%); SVR12: 10/11 (90%)	TA: RR 1.00 [0.87, 1.15]; SVR12: RR 0.99 [0.78, 1.27]
Schmidbauer et al. (2020)	NRS, convenience sampling	AT	Pharmacies and drug treatment facility (INT) and outpatient clinic of a tertiary care center (CON)	PWID and patients receiving OST (NA; 67.6%; 58.6%)	Directly observed DAA treatment together with OST at pharmacy or drug treatment facility for PWID at high risk for non-adherence (Mon-Sat, self-administered doses for Sunday)	Usual care via self-administered therapy for non-IDUs and PWID with excellent compliance (received prescriptions every month and were only seen for routine laboratory tests at the outpatient clinic)	SVR (12 weeks after treatment)	70/74 (94.6%)	69/71 (97.2%)	RR 0.97 [0.91, 1.04]
Contingency Management
Ward, Falade-Nwulia, et al. (2019)	RCT	US	Clinic for HIV care	Patients from an outpatient clinic (NA; 46%; 28%)	Contingency management: contingent cash incentives	Enhanced usual care: treatment according to standard protocol involving clinical visits and calls delivered by a nurse-led multidisciplinary team; all participants received $10 to $30 per visit	Treatment completion (after 12 weeks); SVR (12 weeks after treatment)	TC: 42/54 (78%)	TC: 23/36 (64%)	TC: RR 1.13 [0.84, 1.52]
Norton et al. (2019)	NRS, convenience sampling	US	NSP center providing mental health counselling, referrals and other services	NSP clients (NA; 44%; NA)	Contingency management: participants received financial incentives for HCV visits, return of medication blister packs and successful early clinical outcome	Enhanced Usual care: expedited appointment at center, round-trip transit fare card, reminders from care coordinator	SVR (12 weeks after treatment)	SVR12: 9/9 (100%)	SVR12: 3/4 (75%)	SVR12: RR 1.33 [0.76, 2.35]
Telemedicine
Cooper et al. (2017)	NRS	CA	Hospital and regional HCV program	Patients from hospital and regional HCV program (TM: 70.1%; NA; NA / Non-TM: 54.9%; NA; NA)	Majority of clinic visits conducted utilizing the Ontario TeleHealth Network video and audio system. Patient and remote site TM nurse are linked by audio and video to hospital site at which the HCV clinician, nurse and allied health care providers are located (IFN+DAA regimen)	Usual care: all assessments conducted at hospital outpatient clinic based within a tertiary care center	SVR (12 weeks after treatment)	18/19 (94.7%)	236/249 (94.8%)	RR 1.00 [0.90, 1.12]
Peers and group treatment
Ward, Falade-Nwulia, et al. (2019)	RCT	US	Clinic for HIV care	Patients from an outpatient clinic (NA; 46%; 28%)	Peer mentors (successfully treated for HIV and HCV); face-to-face meetings with mentees, contacted mentees before, during and after treatments	Enhanced usual care: treatment according to standard protocol involving clinical visits and calls delivered by a nurse-led multidisciplinary team; all participants received $10 to $30 per visit	Treatment completion (after 12 weeks); SVR (12 weeks after treatment)	TC: 42/54 (78%)SVR12: 41/54 (76%)	TC: 23/36 (64%) SVR12: 22/36 (61%)	TC: RR 1.22 [0.92, 1.62] SVR12: RR 1.24 [0.92, 1.68]
Akiyama et al. (2019)	RCT	US	Three OST-Centres	PWID from 3 OST programs (75%; NA; 100%)	Group treatment (weekly meetings with other patients and treatment team), self-administered individual treatment	Usual care via self-administered medication at home (1-month supply was given at OST visit)	Treatment adherence (using daily timeframe); SVR (12 weeks after treatment)	TA: NA (90%) SVR12: NA (87%)	TA: NA (91%) SVR12: NA (87%)	TA: RR 0.99 [0.93, 1.15] SVR12: RR 1.00 [0.93, 1.15]
Opioid substitution treatment
Grebely et al. (2016)	RCT, post hoc analysis, multic.	US, FR, DE, IT, ES, UK	Multicentre trial at sites in the US and Europe	Patients receiving OST (OST: NA; NA; 100% / Non-OST: NA)	Receiving OST	Usual care and not receiving OST	Treatment completion; Treatment adherence (≥80 of treatment doses); SVR (12 weeks after treatment)	TA: 65/70 (93%) TC: 68/70 (97%) SVR12: 66/70 (94%)	TA: 1 737/1 882 (92%) TC: 1 846/1 882 (98%) SVR12: 1 822/1 882 (97%)	TA: RR 1.01 [0.94, 1.08] TC: RR 0.99 [0.95, 1.03] SVR12: RR 0.97 [0.92, 1.03]
Christensen et al. (2018)	NRS, prosp. Cohort	DE	254 medical centres also providing OST	Patients from HCV-registry docu-mented by medical centres (NA; NA; 100%)	Receiving OST	Non-OST patients comprised patients with former/current drug use (non-OST/DU)	Treatment completion (at least at one follow-up documentation after 12–24 weeks after treatment completion); SVR (12 or 24 weeks after treatment)	TC: 528/739 (71%) SVR12/24: 450/528 (85%)	TC: 1 126/1 500 (75%) SVR12/24: 969/1 126 (86%)	TC: RR 0.95 [0.90, 1.00] SVR12/24: RR 0.99 [0.95, 1.03]
Integrated services & case management
Radley et al. (2020)	RCT, cluster	UK	55 randomly assigned com-munity pharmacies	Patients receiving OST in pharmacies for at least 3 months (NA; NA; 100%)	Modified directly observed therapy (mDOT) in pharmacy with DAA alongside supervised OST	Enhanced usual care within pharmacist pathway: HCV infection discussed with patients, testing offered if HCV status unknown, referral to treatment center offered if HCV positive	Treatment completion (8- or 12-week); SVR (12 weeks after treatment)	TC: 108/1 365 (8%) SVR12: 98/1 365 (7%)	TC: 58/1 353 (4%) SVR12: 43/1 353 (3%)	TC: RR 1.85 [1.35, 2.52] SVR12: RR 2.26 [1.59, 3.21]
Wade et al. (2019)	RCT	AU, NZ	13 primary care sites (INT) and specialist based local hospitals (CON)	Patients attending a primary care study site (INT: 100%; 49%; 77% / CON: 95%; 49%; 69%	Community hepatitis nurses provided HCV education, assessments, linkage to care and support whilst patients were in treatment at primary care site; general practitioners provided OST	Usual care in local hospital specialist clinic; OST provided in PC	SVR (12 weeks after treatment)	28/57 (49%)	16/53 (30%)	RR 2.22 [1.48, 3.33]
Ho et al. (2015)	RCT	US	Three Veteran Affairs HCV medical centres	Patients attending VA with substance use and/or psychiatric risk factors for antiviral treatment (NA; 47%; NA)	Integrated care protocol with case management, incl. brief psychological interventions and contingent cash incentives provided in collaboration with clinic physicians, nurses, and other mental health providers (IFN+DAA)	Usual care within the HCV clinic or referral to standard mental health and substance use clinics for further treatment (IFN+DAA)	Treatment adherence (≥80% of planned treatment); SVR (12 or 24 weeks after treatment)	TA: 95/182 (52%) SVR12/24: 29/182 (15.9%)	TA: 80/181 (44%) SVR12/24: 14/181 (7.7%)	TA: RR 1.18 [0.95, 1.46] SVR12/24: RR 2.06 [1.13, 3.77]

Abbreviations: AT Austria; AUT Australia; CA Canada; CON control; DAA direct-acting antivirals; DE Germany; ES Spain; FR France; HIV human immunodeficiency virus; IFN interferon; INT intervention; IT Italy; mDOT modified directly observed therapy; NA not available/applicable; NRS non-randomised study; NSP needle and syringe programmes; NZ New Zealand; OST opioid substitution treatment; PWID persons who inject drugs; RCT randomised controlled trial; RR risk ratio; SVR sustained virologic response; TA treatment adherence; TC treatment completion; TI treatment initiation; UK United Kingdom; US United States; VA Veteran Affairs.

Fig. 2

Forest plot of included studies by outcome and stratified by intervention type. A. HCV linkage to care; B. HCV adherence to treatment

Overall EPHPP quality scores: Strong:Starbird et al. (2020). Moderate:Akiyama et al. (2019); Broad et al. (2020); Grebely et al. (2016); Ho et al. (2015); Ward, Falade-Nwulia, et al. (2019). Weak:Christensen et al. (2018); Coffin et al. (2019); Cooper et al. (2017); Messina et al. (2020); Norton et al. (2019); Radley et al. (2020); Schmidbauer et al., 2020; Wade et al. (2019).

Of the 14 studies included in the review, seven (50%) investigated the effect of interventions at both - linkage to care and adherence to treatment – stages of the HCV care cascade. Nine (64%) of the 14 studies were RCTs and five were NRS. All included studies were conducted in high income countries and published in English. The studies originated from the United States (6); Australia/New Zealand (1); United Kingdom (1); Canada (2); Germany (1); Austria (1); Italy (1) and include one multicentre trial at sites in the US and Europe. In total, 21 new or ongoing HCV studies were retrieved from the clinical trials register but contained no data and were therefore excluded from review and analysis.

Risk of bias assessment in individual studies

Risks of bias assessment using the EPHPP tool was performed and is available in the Supplementary material (Supplementary data 2). The risk of bias varied between studies particularly for non-randomised studies. According to the individual ratings attributed to each dimension, only one RCT with low risk of bias was identified. Half of the studies assessed were classified as having high risk of bias, including three RCTs and five NRS. All five included NRS were graded ‘weak’ with overall high risk of bias as they had more than two ‘weak’ individual ratings, particularly in the domains confounders, selection bias and data collection methods. One third of the studies assessed (only RCTs) had one dimension at high risk of bias (either selection bias or blinding) resulting in a moderate overall rating. For GRADE tables, including justifications for downgrading quality of evidence, see Supplementary material (Supplementary data 3).

Linkage to HCV care

Three studies evaluated interventions to enhance linkage to care and six studies investigated the effect of interventions on both linkage to care and adherence to treatment (Table 1 and Fig. 2A) (Broad et al., 2020; Starbird et al., 2020; Cooper et al., 2017; Ho et al., 2015; Messina et al., 2020; Norton et al., 2019; Radley et al., 2020; Wade et al., 2019; Ward, Falade-Nwulia, et al., 2019). Six of the nine studies were RCTs (of which one was a cluster RCT). The sample size ranged from 16 to 2 718 participants. Contingency management, telemedicine, peers, and integrated services & case management were evaluated in the studies investigating linkage to care and compared with usual care.

Contingency management

Two studies, one small-scale NRS (Norton et al., 2019) and one RCT (Ward, Falade-Nwulia, et al., 2019), assessed the effects of cash incentives on linkage to care and subsequent adherence to treatment. Based on the findings of the NRS by Norton et al. (2019), participants allocated to interventional contingent cash incentives were significantly more likely to visit a Needle and Syringes Programme (NSP) center for conducting baseline HCV evaluation within 3 months (74% vs. 30% in control group, RR 2.46, 95% CI 1.19–5.05, very low quality of evidence). In contrast, contingency management was found to have little or no impact on treatment initiation (RR 0.75, 95% CI 0.54–1.04, very low quality of evidence). In the RCT, participants received cash incentives to reinforce HCV treatment initiation. The initiation rate was higher in persons randomised to contingency management compared to usual care (RR 1.14, 95% CI 0.86–1.50, high quality of evidence), however not statistically significant (Ward, Falade-Nwulia, et al., 2019).

Telemedicine

One study showed that a telemedicine program staffed by a multidisciplinary team of HCV competent healthcare professionals can lead to similar results for DAA treatment initiation in remote populations with limited access to healthcare facilities (Cooper at al., 2017). The study showed that participants in the intervention group were more likely to initiate treatment compared to the control, however not significantly (RR 1.07, 95% CI 0.83–1.39, low quality of evidence).

Peer interventional approach

Two RCTs evaluated the effect of a peer interventional approach on linkage to care. Broad et al. (2020) assessed the proportion of participants who had a visit with the HCV nurse within 6 months after point-of-care testing and contact with trained outreach workers with lived HCV experience (=peers). Facilitated linkage to care via outreach workers after testing did not increase the proportion of participants visiting the HCV nurse within 6 months (RR 1.17, 95% CI 0.36–3.77, low quality of evidence). Findings of another RCT by Ward, Falade-Nwulia, et al. (2019) suggest that engagement of peer mentors improves treatment initiation in HIV/HCV co-infected patients from an outpatient clinic (83% vs. 67% in control group, RR 1.25, 95% CI 0.96–1.62, high quality of evidence), however, the difference was not statistically significant.

Integrated services and case management

Integrated services and case management was evaluated in three studies. In a large cluster RCT (2 718 participants), treatment initiation was significantly improved for participants receiving OST through pharmacy-led DOT in comparison with usual care provided by a multidisciplinary service from community treatment centres in Scotland (RR 1.82, 95% CI 1.34–2.46, moderate quality of evidence) (Radley et al., 2020). Wade et al. (2019) assessed DAA treatment initiation in primary care (PC) settings with community hepatitis nurses providing HCV education, clinical assessments, linkage to care and support and general practitioners providing OST. Findings of the RCT show significantly increased initiation rates in the interventional PC group compared to participants receiving DAA in local hospital-based specialist care (75% vs. 34% in control group, RR 2.48, 95% CI 1.54–3.95, moderate quality of evidence). Integrated HCV care, drug use and psychiatric services delivered by a multidisciplinary team with case management services, increased the proportion initiating treatment (32% vs 19% in control group, RR 1.70, 95% CI 1.17–2.46, moderate quality of evidence) (Ho et al., 2015).

Adherence to HCV treatment

Five studies evaluated interventions to enhance DAA-based treatment adherence and six studies investigated the effect of interventions on both linkage to care and adherence to treatment (Table 2 and Fig. 2B) (Akiyama et al., 2019; Schmidbauer et al., 2020; Christensen et al., 2018; Coffin et al., 2019; Cooper et al., 2017; Grebely et al., 2016; Ho et al., 2015; Norton et al., 2019; Radley et al., 2020; Wade et al., 2019; Ward, Falade-Nwulia, et al., 2019). Seven of the 11 studies were RCTs (one cluster RCT, one pilot RCT and one post-hoc analysis). The study sample size ranged from 13 to 2 718 participants. Directly observed therapy (DOT), contingency management, telemedicine, peers & group treatment, opioid substitution treatment (OST), and integrated services & case management were evaluated in the studies investigating adherence to treatment in comparison with usual care.

Directly observed therapy

Akiyama et al. (2019) assessed the effects of DOT on treatment adherence in PWID receiving OST. The RCT was conducted in the transition period from interferon to DAA treatment, meaning that not all participants received the same therapy. Results show slight, however not significant improvements on treatment adherence in the DOT intervention group (86% vs. 75% in control group, RR 1.15, 95% CI 0.95–1.39, high quality of evidence). In a pilot RCT conducted in a community-based research center, the intervention included modified directly observed therapy (mDOT) with motivational interviewing and counselling for injection risk reduction and medication adherence (Coffin et al., 2019). Both study arms were compensated with cash and received transit cards to assist with transportation to the study site. Results show that mDOT with DAA did not increase the mean number of weekly visits (RR 1.00, 95% CI 0.87–1.15, low quality of evidence) and had no effect on SVR12 (RR 0.99, 95% CI 0.78–1.27, low quality of evidence) when compared to self-administered therapy (SAT) using Wisepill dispenser. The NRS by Schmidbauer et al. (2020) evaluated changes in SVR12 following DOT together with OST at pharmacy or drug treatment facility among PWID at high risk for non-adherence (intervention) and SAT for non-IDUs and PWID with known excellent compliance as defined by study authors (control). The proportion of participants achieving SVR12 in both groups was similarly high (95% vs. 97% in control, RR 0.97, 95% CI 0.91–1.04, very low quality of evidence).

The NRS and RCT described above evaluated the effect of contingency management intervention on treatment completion and SVR12 (Norton et al., 2019; Ward, Falade-Nwulia, et al., 2019). Ward, Falade-Nwulia, et al. (2019) found little effect of cash incentives on treatment completions (RR 1.13, 95% CI 0.84–1.52, high quality of evidence). Similarly, cash incentives did not substantially increase the proportion of patients achieving SVR12 (RR 1.12, 95% CI 0.82–1.54, high quality of evidence). Regarding SVR12, results in Norton et al. (2019) favoured the intervention (100% vs. 75% in control group, RR 1.33, 95% CI 0.76–2.35, very low quality of evidence), however, this study was not powered to show a (significant) difference in adherence rates between study arms (9 intervention, 4 control).

The NRS assessing the effect of telemedicine intervention on achieving SVR12 did not show differences between the groups (RR 1.00, 95% CI 0.90–1.12, very low quality of evidence) (Cooper at al., 2017).

Peer interventional approach and group treatment

Ward, Falade-Nwulia, et al. (2019) also evaluated the effect of engagement of peer mentors on the proportion of participants completing treatment and achieving SVR12. To increase treatment completions, engagement of peer mentors during and after DAA treatments was superior to the control, however not significantly (78% vs. 64% in control group, RR 1.22, 95% CI 0.92–1.62, high quality of evidence). Overall, 76% of study participants in the intervention group achieved SVR12 compared to 61% in the control (RR 1.24, 95% CI 0.92–1.68, high quality of evidence). Group treatment including weekly meetings with other patients/peers and treatment team, psychosocial support from peers and providers, HCV education, and SAT with DAA and IFN shows no significant effect, neither on adherence (RR 0.99, 95% CI 0.93–1.15, high quality of evidence) nor on achieving SVR12 (RR 1.00, 95% CI 0.93–1.15, high quality of evidence) (Akiyama et al., 2019).

Opioid substitution treatment

One RCT and one NRS evaluated if the provision of OST is effective in increasing adherence to HCV treatment (Christensen et al., 2018; Grebely et al., 2016). The NRS by Christensen et al. (2018) assessed DAA treatment completion and SVR12/24 in patients from a HCV-registry documented by OST centres vs. non-OST centres. Findings indicate that the provision of OST does not increase treatment completion (RR 0.95, 95% CI 0.90–1.00, low quality of evidence) or the proportion of patients achieving SVR12/24 (RR 0.99, 95% CI 0.95–1.03, low quality of evidence). A multicentre RCT at sites in the United States and Europe excluded participants with “clinically significant drug use within 12 months of screening” and found no effect of DAA treatment adherence (RR 1.01, 95% CI 0.94–1.08, moderate quality of evidence), treatment completion (RR 0.99, 95% CI 0.95–1.03, moderate quality of evidence) and SVR12 (RR 0.97, 95% CI 0.92–1.03, moderate quality of evidence) in participants receiving OST vs. not receiving OST (Grebely et al., 2016).

Three RCTs assessed integrated services and case management interventions to increase adherence to treatment. Results in Radley et al. (2020) show that using pharmacists to carry out DOT and deliver an integrated HCV care pathway for participants receiving OST significantly improves SVR12 (7% vs. 3% in control, RR 2.26, 95% CI 1.59–3.21, moderate quality of evidence). Providing DAA in primary care settings with community hepatitis nurses and general practitioners providing OST significantly increased the proportion of participants achieving SVR12 in the intervention group compared to the control (RR 2.22, 95% CI 1.48–3.33, moderate quality of evidence) (Wade et al., 2019). The integrated care approach that included mental health care services by Ho et al. (2015) increased DAA and IFN treatment adherence (RR 1.18, 95% CI 0.95–1.46, moderate quality of evidence) and significantly increased the proportion of participants achieving SVR12/24 (15.9% vs. 7.7% in control group, RR 2.06, 95% CI 1.13–3.77, moderate quality of evidence).

Discussion

We have undertaken a systematic review to identify interventions that can increase linkage to care and adherence to treatment of HCV among people who inject drugs (PWID) and people receiving opioid substitution therapy (OST). Our review adds to the previously published work by including studies that evaluated DAA treatment regimens with at least 50% of the study sample reporting ever injecting drug use or receiving OST and excluding studies without comparators. We also improve the rigor of our results by using GRADE to assess quality of evidence and by involving an expert panel to review the evidence retrieved through this analysis, to participate in a structured discussion and to comment on recommendations based on evidence retrieved. The 14 included studies represented a diverse group of interventions spanning various settings. Studies reporting on linkage to care interventions aimed to increase the likelihood of PWID visiting a provider/specialist after having tested positive for HCV for an initial evaluation in order to start treatment. Those interventions included contingency management, telemedicine approaches, peer involvement, and integrated services and case management interventions. Studies evaluating strategies to enhance adherence to treatment looked at impact on treatment adherence, treatment completion and achieving SVR12. Interventions included the interventions aimed to increase linkage to care and, in addition, group treatments, directly observed therapy (DOT) and provision of OST.

The effectiveness of contingency management interventions was not supported by the evidence retrieved. The expert panel indicated that incentives may facilitate linkage and engagement in HCV care, but will not necessarily ensure continuity of care. To improve the entire cascade of care for PWID, experts highlight that contingency management should only be considered in addition to other low-threshold services. Importantly, countries’ legal framework must be considered, and caution is advised to not create inequalities when financial incentives are only established in specific sites and given to specific subgroups of HCV-infected persons (i.e. PWID/people receiving OST).

Telemedicine (TM) services are a broadly defined field of intervention based on the use of remote interventions. TM could offer a novel approach in bringing health and health equity into all population groups and regions, in particular when referral to specialists is limited. Results of a Canadian study (Lepage et al., 2020) indicate the use of real-time telemedicine as successful in the engagement of rurally based patients to HCV treatment among highly vulnerable and marginalized patient groups (IDU history >72%). Similar success (97% overall SVR rate) can be seen among incarcerated populations following DAA for HCV (Jiménez Galán et al., 2019). A recently published review supports TM as a cost-effective and successful approach to improve patient's engagement and treatment rates with quality HCV services (Khoja, Ali, & Feroz, 2021). The expert panel considered TM to be more useful to increase adherence to treatment than linkage to care (ECDC, 2022c). However, relevant barriers exist for some client groups to benefit from TM interventions, in particular older PWID, and clients with severe psychiatric or other complex co-morbidities (EMCDDA, 2021c). Based on Covid-19 pandemic experiences, where access to conventional treatment services has been restricted, telemedicine gained additional relevance. During the first confinement period in most European countries in March 2020, TM by phone or video generally became an alternative for drug services’ operational work (EMCDDA, 2021c) and the use of remote interventions to manage HCV care in the current Covid-19 pandemic is considered as effective (Khoja et al., 2021). Further research is recommended focussing on the adequacy of TM for specific PWID subgroups as well as useful integration of TM in integrated care structures (e.g. primary care).

Peer involvement interventions showed a positive but not significant effect on linkage to care and adherence to treatment, based on the results retrieved. However, peer support is widely acknowledged in HCV elimination, in particular when addressing and engaging hard-to-reach populations such as PWID in the care cascade (WHO, 2018). An observational study in the UK conducted in the framework of HepCare Europe found high level referral to secondary care (>89%) as well as >50% referrals to treatment centres among a population with complex needs as a result of active peer involvement (Surey et al., 2019). Enhanced peer involvement to increase HCV testing and treatment uptake among marginalized groups also shows high cost-effectivity (Ward, Campbell, et al., 2019). The expert panel supported the importance of peer involvement to better approach lived realities and vulnerabilities and therefore reach underserved population at risk. Peers act as facilitators by promoting trusting environments and provide a unique access point (Henderson, Madden, & Kelsall, 2017), even to hard-to-reach sub-populations of PWID (e.g. migrant background, homeless, people with insufficient reading or writing skills). A recent collection of European models of practice aimed to identify best-practice evidence of interventions to improve community-based testing, linkage to care, adherence to treatment for infections among PWID as well as for prevention of infections, implemented in real-life settings. In fact, two third of the selected projects included peer involvement (ECDC, 2022b). To maintain engagement of PWID in treatment, various models of peer involvement in HCV diagnosis and treatment can be identified (Henderson et al., 2017), differing between marginal or central roles of peers (ECDC, 2021b; Surey et al., 2019). Highly trained peers can take a relevant role in the care cascade and in services by expanding pathways and navigating clients from testing to treatment completion. Surey et al. (2019) consider the role of peers in the project design as success factor for achieving the outcomes. Corresponding trainings for peers are seen as a pre-condition by the expert panel members. Trainings qualify peers to successfully act as peers and are, in addition, key to raise awareness and to reduce discrimination of peer work among health care professionals.

Directly observed therapy (DOT) as a delivery mode is rather unlikely to influence treatment initiation but may enhance treatment adherence, based on the evidence retrieved in this review. In a systematic review and meta-analysis, McDermott, Lockhart and Devine (2018) found that directly observed therapy may lead to significantly higher odds of attaining SVR when compared to treatment as usual. For the most vulnerable population usually seen in the streets and more often not engaged in OST but in low-threshold programs, based on expert panel input, DOT may often not be feasible or desired. However, for some PWID with chaotic lifestyle, unstable housing, who do not have adequate places to store medication, practical experience shows that low-threshold settings with their trusting atmosphere could facilitate HCV linkage to care by offering an enhanced form of DOT. Expert panel conclusions emphasize that DOT as a delivery mode should under no circumstances be a formal requirement to receive HCV treatment and should not be used as a controlling measure, as there is a need of different care pathways decided upon by clients.

Providing opioid substitution treatment (OST) to PWID is seen as an effective response to reduce the risk of the transmission and acquisition of bloodborne infections (EMCDDA, 2017). According to EMCDDA, all EU Member States provide OST but in some, access to OST for those in need still remains limited (EMCDDA, 2021). Presented results suggest that OST did not directly impact HCV treatment adherence, completion or SVR12. Data demonstrates that DAA HCV therapy is well tolerated and effective among both PWID receiving OST and those with illicit drug use during HCV therapy (Grebely et al., 2020; Schmidbauer et al., 2020). According to the expert panel, an enrolment in OST should therefore not be considered as a pre-requisite to HCV treatment access, a demand possibly resulting from the former exclusion of PWID in the interferon era. In contrary, OST treatment provides a fixed setting and a regular meeting point for former or recent PWID, which can be seen as beneficial elements for successfully managing the individual's HCV treatment regimen. Integration of OST and HCV treatment should be recommended as part of patient-centered care model, considering different groups of PWID diverse in size, consumption behavior and other features.

Integrated care structures, where a variety of services are brought together, and nurse case management approaches that facilitate referral for HCV assessment and scheduling of specialist appointments for clients were identified as successful interventions to enhance the care continuum among PWID and people receiving OST. This is consistent with findings from the systematic review by Bajis et al. (2017) and Zhou et al. (2016), which identified the importance of integrated approaches to hepatitis care and treatment for PWID and other specific vulnerable populations. The role of case managers (peers or social workers) and the cooperation and integration of services (e.g. combining testing, linkage to care and treatment) are key elements of success for managing drug user health and considering their multiple needs (WHO, 2018). Expert panel discussion revealed that delivering a variety of targeted services is also an option to act timely in responding to new developments or small but highly vulnerable groups contributing to incidence, always depending on existing health care structures.

Cooperation between institutions, in particular between tertiary, pharmacy and primary care setting, was found to be successful in order to implement integrated care strategies for rapid linkage to care and DAA treatment programmes. As PWID groups are regularly using different services, e.g. needle and syringe programs, OST, etc., networks and the co-location and embedding of health services allowing for a ‘one-stop shop’ model of care may be a key reason for the success of these interventions and an important part to achieve integrated care (Delile et al., 2018; Socías et al., 2019). Experts emphasized that these institutions should be located in the same geographic area and should reduce barriers by treating individuals holistically and actively accompanying clients in the referral to other services. However, further studies are needed to assess which combined interventions are most effective to reach, link and treat hard-to-reach PWID groups.

Results suggest that HCV-infected patients can be effectively treated in primary care. This is consistent with the findings of the systematic review and meta-analysis by Radley et al., which demonstrates the feasibility of decentralizing care and providing local services with reach into communities of people infected with HCV (Radley et al., 2019). Broadening access to DAAs beyond hospital-based services must occur if their benefits are to be fully realized. The primary care setting provides many advantages for providing HCV treatment, including patient comfort and familiarity, and easier integration into patients’ existing treatment plans. Further, Wade et al. highlight the potential benefit of providing HCV care in community-based clinics, including OST clinics or using telehealth, instead of conventional tertiary services (Wade, Veronese, Hellard, & Doyle, 2016). Delivery of DAA treatment by general practitioners, addiction medicine specialists, nurses or other primary care providers must be assessed in the view of legal aspects, differences in health care systems, national frameworks, and characteristics of settings, as highlighted in the expert panel discussion. However, in countries where OST is well established in primary care, DAA treatment may work equally well in primary care-based settings (Delile et al., 2018). Additional research is needed to evaluate if referrals to specialist care could be limited to more complex patients with psychiatric comorbidities, decompensated cirrhosis, or treatment failure to DAA.

Our analysis showed that a multidisciplinary, people-centered approach improves HCV linkage to care and adherence to treatment in a variety of setting ranging from primary care and pharmacy settings, to community health clinics, to HIV clinics and other tertiary services, to OST programmes, to harm reduction centres. Although the number of services (e.g., linkage, treatment) and type of intervention implemented (e.g., nurse case management, peer mentors) differed, our findings highlight the importance of integrated approaches to hepatitis care and treatment for PWID populations. Understanding their needs and embedding HCV elimination strategies at points where they are already accessing services will be key to implement effective and acceptable integrated models of care.

This review has several limitations. First, the mentioned outcomes are intermediate endpoints related to linkage to care and treatment, not disease endpoints such as morbidity and mortality associated with HCV. For both research questions, different outcome measures with different endpoints and time intervals were used by the included studies. Second, the quality of studies was limited with one third of all studies being non-randomised studies. Most studies were small in sample size causing many RRs to be non-significant with high risk of bias. It is likely that the requirement for a comparator meant that many publications reporting on innovative interventions and programmes were not included. This may be acerbated by the exclusion of conference abstracts, omitting strategies and approaches not published as full articles. Third, all included studies were conducted in high-income countries, of which 9 were conducted in non-European settings. Further research is needed to address diversity of settings in the EU/EEA and other European countries. Fourth, as with most systematic reviews, the heterogeneity of the study populations included in the analysis presents a limitation of this study. Study populations differed in terms of former, recent, and current injection drug users or people receiving OST and may therefore not be directly comparable with one another (i.e. stable living conditions vs. risky behaviours etc.). Fifth, due to the heterogeneity of data extracted and the limited number of studies with similar interventions, it was not possible to conduct a meta-analysis. The heterogeneity of interventions implemented, and treatment settings documented in this review may reflect the variety of ways in which patients can access HCV treatment and that this is highly context-specific. Finally, the review excluded interventions focusing on screening and (point of care) testing and did not include costs or cost-effectiveness. Reviewing the cost-effectiveness of the interventions with the strongest evidence-base and incorporating provider perspectives would enable decision-makers to combine the impact evidence synthesis with implementation cost and provider acceptability of the strategy, to make an informed choice.

We identified several important research gaps and priorities for future research. First, most of the included studies for HCV were conducted outside of Europe. This particularly applies to studies that assessed outcomes for linkage to care, impairing the transferability of the results. In addition, well-powered RCTs or comparative studies to validate the effects of the interventions are lacking. Studies investigating the entire care cascade from testing to cure in one study are urgently needed, whereby the implementation of a variety or combination of different interventions should be pursued further. The expert panel emphasized the need to develop practice-based evidence focusing on implementation experiences, apart from peer-reviewed literature based scientific evidence. A technical report summarizing models of good practice with documented impact on linkage to care and adherence to treatment of PWID in European settings has been published in conjunction with the updated ECDC/EMCDDA guidance (ECDC, 2022b). Factors such as peer involvement, trust and conveniences to patients, community or low-threshold approaches covering a broad spectrum of patient's needs are rarely measured in RCTs and require more attention in further research agenda. Further, success factors are often limited to biomedical indicators and not considering (more qualitative) structural factors and enabling favourable environments, which often seems to be key pre-conditions to engage hard-to-reach population at risk. Experts highlight that the preparatory work for treatment done by drug services prior to the actual intervention can be seen as pre-conditions for its success. Prior participation in service provision to PWID (e.g. pharmacies already involved in OST, implementation of harm reduction facilities) will positively impact HCV treatment. In addition, motivational factors such as treatment willingness and preparedness of the clients play an important role.

Conclusion

This review found low to moderate quality evidence that integrated, people-centered approaches may improve engagement throughout the HCV treatment journey for PWID populations in a variety of settings. The diversity of integration approaches suggests that interventions on linkage to care and adherence to treatment - as crucial stages in the HCV care cascade - need to be adapted to the lived realities of PWID population groups and embedded at settings where PWID are already accessing services (i.e. harm reduction and counselling services). Strategies to improve engagement along the continuum of HCV care should consider contextual factors, such as available resources, national healthcare system characteristics and legal frameworks in order to be successful. Integrated approaches and the use of case managers may be particularly beneficial when combined with additional services, including opioid substitution treatment, directly observed therapy, peer support or contingency management. As such, there is an urgent need for higher quality studies, to inform integrated care strategies to optimize HCV care access and outcomes among PWID. Our review provides an evidence-base for policy makers, public health researchers and national and international program coordinators involved in the prevention and control of HCV among highly vulnerable groups in EU/EEA countries, and elsewhere.

Financial support

This research was commissioned by the European center for Disease Prevention and Control (ECDC) through service contract No. ECD.10793/2020, coordinated by Otilia Mårdh, and produced by Gesundheit Österreich GmbH (GOEG). The Austrian National Public Health Institute (GOEG) is funded by the Austrian federal government. The views expressed in this publication do not necessarily represent the position of the Austrian Government.

Declarations of Interest

The authors have no conflicts of interest to declare. All co-authors have seen and agree with the contents of this review and there is no financial interest to report.