‘HepCheck Dublin’: an intensified hepatitis C screening programme in a homeless population demonstrates the need for alternative models of care

Hepatitis C virus (HCV) is one of the main causes of chronic liver disease worldwide [1]. The long-term impact of HCV infection is highly variable, from minimal changes to extensive fibrosis and cirrhosis with or without hepatocellular carcinoma (HCC) [2, 3]. Acute infection is asymptomatic in 60–70% of cases, meaning that many do not become aware that they are HCV positive until decades after initial infection, after progression of the disease and emergence of sequelae [4]. Effective diagnosis and treatment is heavily reliant on screening of at-risk individuals [5]. The number of chronically infected persons worldwide is estimated to be approximately 71 million [6]. The implementation of extended criteria for screening of HCV is currently the subject of major debate among different stakeholders [7].

A 2012 systematic review and meta-analysis of HCV in homeless populations found prevalence ranging from 3.9 to 36.2% [8]. Homelessness is associated with an increased prevalence of risk factors for HCV such as injecting drug use (IDU) [9, 10]. The link between homelessness and poor health is well established [11–13]. Homeless individuals generally delay going to a health service and so when they eventually do present, it is generally late on in their condition and can often have become so severe that they require hospitalisation. Preventing the transmission of blood borne viruses (BBV) such as HCV is therefore a key responsibility of community health services and their partners in homeless services [8, 14–17]. High levels of alcohol use, as well as diabetes and metabolic syndrome suggest probable higher rates of fatty liver disease among this population [9, 18].

In Ireland, where our study took place, it is estimated that 20,000–50,000 people are chronically infected with HCV [19]. Of the six HCV genotypes, Genotype 1 and 3 are most common in Ireland. National data on HCV infection in Ireland amongst drug users is restricted to certain treatment sites and prisons [20]. Keegan et al. (2017) in their study reported that prevalence of HCV infection among drug users was 63.6% with no significant gender difference and with prevalence lower in younger adults. General prevalence estimates within the drug user populations range from 54% [21] to 84% [22]. In terms of homeless adults in Dublin, in 2005 the rate of HCV infection was estimated to be 36% [23]. A 2015 study of people who are homeless/at risk of homelessness receiving free primary healthcare in Dublin reported a HCV prevalence of 23% [16]. Information on the extent of chronic liver disease amongst homeless individuals in Ireland is however limited. There is substandard uptake in HCV assessment and treatment among people who inject drugs (PWID) in Ireland [24].

The Irish Health Service Executive (HSE) funded Safetynet Primary Care service is a network of general practitioners and nurses providing primary care to homeless and other marginalised groups. This multifaceted service includes the offering of treatment in their own environment to homeless people who are addicted to heroin, thus removing many of the barriers that result in their exclusion from treatment. It provides a novel alternative to the centralised clinic approach, which was the only option for homeless people prior to the project’s establishment. In Dublin, homeless heroin dependant patients can access methadone maintenance treatment through Safetynet services (as well as HSE addiction centres). Our study sought to evaluate an intensified screening programme for HCV in homeless individuals attending Primary Care services in Dublin, Ireland, and in doing so investigate and establish the characterisation of HCV burden among this cohort.

HepCheck Dublin was undertaken as part of a larger European wide initiative called HepCare Europe. HepCare Europe is an EU-supported service innovation project and feasibility study at four European sites (Dublin, London, Seville and Bucharest) to develop, implement and evaluate interventions to enhance identification and treatment of HCV among vulnerable populations [25]. The project focuses on providing an ‘integrated care’ model for HCV treatment based on collaboration between primary and specialist care practitioners to allow for more efficient use of limited specialist resources. The HepCheck Dublin project looks specifically at screening individuals using rapid HCV testing and tracking results and pathways to care outcomes.

This study used a cohort design. The target population were homeless individuals accessing the Safetynet Primary Care services in Dublin. Individuals using homeless services and hostels (Merchant’s Quay Ireland (MQI), Capuchin Day Centre, Ana Liffey Drug Project (ALDP), Salvation Army, De Paul Ireland, Crosscare, the Dublin Simon Community and the Safetynet mobile health unit) were informed of screening for HCV in Safetynet clinics and encouraged to attend. Individuals who reported that they had already been diagnosed as having HCV were advised to attend a safetynet clinic for assessment and referral for treatment if appropriate. Screening took place in 11 Safetynet affiliated GP practices, as well as Safetynet in-reach locations such as coffee shops, needle exchanges, hostels, drop in centres, and the Safetynet mobile health unit.

Inclusion criteria were as follows: 1) attended Safetynet services during the study period, 2) age > 18, and 3) ability to provide informed consent. Exclusion criterion was defined as inability to provide informed consent. Subjects were given an information sheet on the study and provided written consent. Subjects were recruited over a 19-month period, from December 2014 to July 2016. As conventional probability methods, such as simple or stratified random sampling, are often not appropriate for populations of injecting drug users [26], targeted sampling, a non-probability sampling framework, was used to identify subjects on whom data was collected [27]. Subjects received no financial incentive for participation in the study.

The data were analysed using SPSS v24. Descriptive statistics (frequencies, percentages, median and interquartile range (IQR)) were used to summarise subjects’ characteristics. In analysing Phase Three data we used unadjusted negative binomial regression (NBR) models to examine the factors associated with the number of specialist care visits. The following explanatory variables were considered; age, gender, alcohol and drug use, key worker involvement and accommodation status. This model has been shown to have better fit than Poisson regression models by accounting for statistical overdispersion and excess zeros (individuals with no specialist care visits) and yields a more conservative result. The parameter estimates from the NBR models were converted to incidence rate ratios (IRRs) by exponentiating the regression coefficients, and 95% confidence intervals were estimated. Incidence rate ratio above 1.0 implies higher number of specialist care visit and IRR below 1.0 implies lower number of specialist care visits.

This is the first study in Ireland to specifically target and characterise homeless individuals and their prevalence of HCV and experiences of the HCV care pathway. The study presents a unique Irish profile of HCV burden among homeless patients accessing primary care services in Dublin, Ireland, and illustrates the complexities around provision of an intensified screening programme for HCV for homeless populations. The strengths of the study include the selection of a particularly hard to reach group of homeless individuals in Dublin, who are a HCV risk population who need to be studied and intervened with in order to better understand how to improve their experiences of HCV assessment and treatment.

The study highlights the complexities around provision of HCV cascade of care to homeless individuals living in Dublin. Not having stable accommodation is a major barrier to screening and treatment retention. Studies elsewhere in Europe have indicated how structural factors impede HCV screening and treatment access, with key barriers centring on social stigma, housing, criminalisation, healthcare systems, and gender [28].

As highlighted previously, hospital-based appointments are the only model of care available in Ireland. This study demonstrates that standard referral methods are inadequate in engaging this population. The current referral process exacerbates barriers to treatment engagement. Due to the transient nature of the cohort there are difficulties in contacting and sending reminders for appointments. To compound this further, not all participants had access to a keyworker to facilitate the referral process.

The opportunistic community-based screening uncovered concerning numbers of HCV Ab positive individuals in those accessing SafetyNet services, with 37% of individuals testing positive for HCV antibodies. Of note is that over half were ‘new positives’ and just under half were ‘known positives’, which underscores the need to enhance patient engagement with the HCV cascade of care, and to completion. Studies in Dublin have pointed to the issues around other disease morbidities in this population where symptoms of alcohol or drug withdrawals may mask HCV symptoms [24]. This highlights that symptom-based testing for HCV is futile, further emphasising the necessity for broad risk-based and network-based strategies in effectively identifying cases [5, 29].

The study supports Crowley et al. (2017) by emphasising the need to consistently target homeless individuals, particularly those who are actively using or have a history of injecting drug use, for HCV screening and treatment. Marginalised injecting drug users, especially those who are without stable accommodation are a hard to reach group and are less likely to access health care due to fear of discrimination, stigma, low health literacy, and health professional misunderstanding of their needs [20, 30–32]. Homeless patient characteristics in our study are equally complex in terms of physical and mental health problems such as continued drug/alcohol use, being on drug treatment, and experiencing forgetfulness, depression and anxiety, respiratory and dental conditions, all of which confound the HCV treatment care pathway. Reasons for non-attendance of HCV care such as active drug use, being incarcerated, fear of HCV treatment side effects and forgetfulness are similar to those reported by Crowley et al. (2017) in their study of Irish patient characteristics and barriers to Hepatitis C treatment in patients on opioid substitution treatment (OST) attending a community-based clinic where Fibroscans were performed.

Scaling up of HCV prevention, screening and treatment to address HCV transmission among the homeless, particularly those injecting drugs within existing community-based infrastructures and primary care are warranted [20, 31–33]. Of note is that among Phase 2 participants, 42% saw a GP once a week. This suggests a potential for high engagement within community-based primary care settings. A recent study by Barocas et al. (2017) describes excellent responses to community-based HCV treatment within a primary care program in a cohort of homeless and marginally housed adults, demonstrating that despite barriers and competing priorities, outcomes similar to those of clinical trials and other cohorts are possible for this traditionally difficult to treat group [34]. Similarly, Grebely et al. (2017) describes an effective implementation of point-of-care HCV RNA testing in people attending drug health and homelessness services in Australia, showing potential for improved screening, linkage to care and treatment retention [35].

We recognise that whilst community-based screening intervention can determine the prevalence of HCV in homeless populations, referrals to / attendance at secondary care remains a challenge for this cohort. Psychosocial factors are at the core of why patients do not attend secondary care for HCV management and in this study, the interplay between addiction, mental health and homelessness were especially problematic. With this in mind, community-based health services such as Safetynet are ideally placed to support HCV screening and facilitate vulnerable individuals in accessing the treatment and care pathway. This study demonstrates that the current hospital-based model of care is inadequate in addressing the specific needs of this population, reflected in only 2 of the 199 individuals testing positive accessing treatment and cure. This emphasises the need for a community-based treatment approach.

The homeless population remain a high risk and high prevalence population for HCV infection [8–10]. Health literacy, screening uptake and treatment engagement are challenging given the complexities of this patient cohort. Findings from this study represent a first step in Ireland to understanding the needs of homeless individuals and are intended to inform HepCare Europe in their development of a community-based model of care in order to engage with homeless individuals who are infected with or affected by HCV.