Estimating the scale of chronic hepatitis C virus infection in the EU/EEA: a focus on migrants from anti-HCV endemic countries

Chronic infection with the hepatitis C virus (HCV) is a global public health challenge and a leading cause of liver disease-related morbidity and mortality. The epidemiology remains poorly understood, however, and global, national and risk group-specific anti-HCV and viraemic prevalence estimates vary considerably. Recent studies suggest that between 105 million and 185 million people are anti-HCV positive worldwide and that global anti-HCV prevalence in adults could be as high as 2% [1, 2]. The Global Burden of Disease study estimated that chronic HCV (CHC) infection causes almost half a million deaths annually and is the 25th leading cause of death worldwide [3].

Chronic HCV infection affects the liver and has a mostly asymptomatic onset, but can lead to cirrhosis and hepatocellular carcinoma (HCC) decades later [4]. The asymptomatic nature of infection and the lack of adequate screening programmes means that the majority of people infected with CHC are unaware of their infection and only around one third of all estimated CHC infections in Europe have been diagnosed [5‐7]. Effective antiviral treatment can prevent the development of cirrhosis and HCC and, with newer direct acting antivirals (DAAs) reporting cure rates in more than 90% of cases, [8] the elimination of HCV infection is now possible in Europe [9]. This will require continued primary prevention of new infections in parallel with expansion of secondary prevention through effective screening, linkage to care and treatment.

Primary prevention measures in Europe, including a safe blood supply, improved infection control practices and harm reduction programmes, have led to a significant reduction of HCV transmission in many countries and a mathematical modelling study shows incident cases are declining [10]. Incident data is not systematically collected and reported in most EU/EEEA countries hampering a good understanding of time trends although iatrogenic and nosocomial transmission is reported to be rare in most EU/EEA countries [11]. However, models predict that the peak in the mortality is yet to be reached [9, 10, 12]. An estimated 7.4 million people are anti-HCV positive in the European Union/European Economic area (EU/EEA), although prevalence varies from 0.9% in Western Europe to 3.3% in Eastern Europe [1, 13]. Deaths from viral hepatitis now exceed those from HIV and tuberculosis combined and latest published estimates show that 96,000 people die each year in EU/EEA countries from HBV and HCV-related liver disease [3]. Some populations are disproportionately affected and have a high prevalence of chronic infection. One such population is migrants born in high anti-HCV prevalence countries [10, 14‐16], although estimates of the number of cases and the most affected migrant populations in Europe are lacking.

This epidemiological study seeks to inform targeted screening, linkage to care and treatment in the EU/EEA by: providing estimates, across and within all 31 EU/EEA countries, of the number of CHC cases among migrants from countries where anti-HCV prevalence is ≥1%; providing an estimate of the relative contribution of migrants to the overall burden of disease; and comparing the reported in-country of birth prevalence with that found among migrants living in European countries. In a sister paper to this, we conduct a similar analysis for chronic hepatitis B among migrants from endemic countries.

The data retrieval and analysis process are described in detail below and in a schematic representation (Fig. 1). Demographic data on the size of and countries of birth of migrant populations were extracted from statistical databases. Country of birth-specific and EU/EEA country general population anti-HCV prevalence estimates were derived from a systematic literature search. To assess the reliability of using country of birth-derived prevalence as a proxy for the prevalence among migrants, a systematic literature search was conducted to identify prevalence estimates among migrants in Europe to compare with country of birth-derived prevalence.

This is the first study we know of that attempts to systematically estimate the overall number of CHC cases among migrants, as well as the relative contribution of cases among migrants to the overall burden of CHC in EU/EEA countries. Migrants from endemic countries account for one in 12 EU/EEA adult citizens and for one in seven of all CHC cases in the EU/EEA. As the contribution of migrants to the overall burden of CHC varies between EU/EEA countries, effective approaches to secondary prevention will, therefore, also differ. Screening programmes targeting migrant populations will be most effective in EU/EEA countries such as Austria, France, Germany, the Netherlands and the UK where migrants account for a large proportion of the disease burden (32–92%, see Fig. 3) and a small proportion of the total population (7–15%, see Fig. 2).

In contrast, in countries where HCV prevalence is high in the general population and the contribution of migrants is low, it may be more cost-effective to implement population-based screening [27, 28]. Examples of such countries include Bulgaria, Poland, Romania and Slovakia where less than 4% of the CHC burden is attributable to migrants from endemic countries. The differences in general population prevalence between EU/EEA countries, together with the contribution of migrants moving within the EU/EEA from high to low prevalence countries, suggests that there may also be value in allocating EU health funding to scale up and systematise screening and treatment efforts in EU/EEA countries with a high general population prevalence, to strengthen efforts to reduce cross-border health threats and to improve overall population health in the EU. A recent modelling study estimated that only around a third of all CHC cases across Europe have been diagnosed and that there are wide differences in both the proportion diagnosed and the proportion on treatment comparing EU/EEA countries [6]. There is however no data on the estimated proportion of migrants from endemic countries that are diagnosed. The data reported, as well as the strategies and interventions suggested, in this study can contribute to increasing the proportion of cases of CHC diagnosed and on treatment.

Previous studies of hepatitis B/C screening implemented among migrant populations describe different models. These include: outreach offering awareness raising and/or on-site screening by public health teams in social, civic or cultural locations familiar to the target community [29‐31]; invitation-based screening where municipal, population or patient registries are used to send postal invitations to attend screening to people born in higher prevalence countries [32, 33]; opportunistic offering of screening to patients with country of birth-related risk factors who attend health care services for other health issues [34, 35]; and adding viral hepatitis screening to an existing screening programme, such as for tuberculosis, that targets people from high prevalence countries [36]. Each of these models differs in terms of logistical and resource requirements, uptake and case yield, but few studies have compared the characteristics and effectiveness of different models [37]. The characteristics of screening programmes that have demonstrated success in uptake and yield include: involvement of the community in planning and raising awareness; provision of screening in suitable and accessible locations for the target community; provision of language support, for example, translated materials and interpreters; planning and provision for people without health insurance coverage; cultural sensitivity about and efforts to reduce or eliminate stigma; and availability of follow-up and care in the community. Retention within a follow-up care and treatment pathway is crucial to ensure that the public health benefits of screening are realised [38]. A summary of migrant-specific screening programmes, an appraisal of factors contributing to success, and a range of other scientific and practical resources are available as part of the HEPscreen Toolkit, produced by the EU-funded HEPscreen project, which focused on screening for chronic viral hepatitis among migrants [39].

In four countries, (Cyprus, Lichtenstein, Luxembourg and the Netherlands) the upper range of the estimated relative contribution of migrants as a proportion of the total number of estimated cases was found to be over 100%. This reflects unmeasured correlation between the input parameters (prevalence in countries of birth and the size of the migrant population) for the Delta method as well as the strong skew in the distribution of cases in the general population in these countries. It is also possible that the prevalence in the general population in these EU/EEA countries is under-estimated, due to unrepresentative sampling or low participation of risk populations who are harder to reach, such as PWID and ethnic minorities as well as migrants. Under-representation of migrants and ethnic minorities is a widely recognised phenomenon in clinical trial and health survey research, [40, 41] although literature specifically focused on sero-prevalence sampling and uptake is very limited [23]. In addition, it is possible that estimates based on prevalence in countries of birth results in over-estimation of the prevalence among migrants. For example, it is probable that no longer living in a high prevalence country would reduce the risk of transmission of HCV, especially since much of the transmission in higher prevalence countries is nosocomial [42] and most EU/EEA countries have successfully controlled nosocomial transmission through health care infection control procedures. Over-estimation may also be due to the characteristics of migrants to the EU/EEA, who may be younger and healthier, and so less likely to have experienced hospitalisation or serious medical intervention and more likely to have benefited from improved primary prevention in the last two decades. Migrants to the EU may also be from higher socio-economic groups in their countries of birth and able to afford better, safer health care [43]. We sought to test out this theory of over-estimation and found that, for just three of the ten countries of birth for which we retrieved estimates for, [15, 29, 30, 44‐47] the prevalence in migrants was lower than the reported in-country prevalence.

Despite the systematic nature of data retrieval, there are some data limitations. Detailed demographic data was available from Eurostat for only 21 EU/EEA countries. There is also heterogeneity in the parameters and methods used by the different demographic databases and in the way that demographic data on migration is collected and reported by EU/EEA countries. From other literature, we know that countries such as Germany, France and the Netherlands require municipal registration upon arrival in a new area and collect country of birth data as part of registration. Other countries, like the UK, rely on population census data to systematically collect population size and origin [48]. However, there is very limited information in the literature on the methods used by each database or on the heterogeneity of demographic data collection methods across the EU/EEA. Despite these differences, the demographic data used in this study are derived from databases that in turn derive data from national statistical institutes. We believe it to be the best and most reputable available. These analyses do not include undocumented migrants, partly because the size and origin of this population is very dynamic, but mostly because of the imprecision and unreliability of the data [49]. The use of systematic reviews and meta-analyses for the country of birth prevalence input has reduced the reliance on less reliable single study estimates, and whilst we critically appraised the quality of the systematic reviews and meta-analyses, which all applied quality criteria to select studies, we did not directly assess the quality of the estimates included in these reviews. A further note of caution relates to the studies retrieved on prevalence in migrants, as just ten countries of birth were represented in the estimates and few studies had the specific aim of estimating the prevalence in migrants. The use of convenience sampling in many of these studies increases the chance of selection bias and, specifically, that people already diagnosed may not present for screening.

Advances in antiviral treatment open up the possibility of eliminating hepatitis C infection in Europe, but achieving this will require countries to scale up and better target screening, linkage to care and treatment. This study provides strategic, timely and detailed epidemiological intelligence for EU/EEA countries on the hepatitis C burden among migrant populations, a key population group affected by this infection in Europe. It also provides prevalence estimates for 224 countries and territories, which should serve as a useful resource for other countries and regions wishing to understand the relative contribution of migrants to the burden of hepatitis C. This intelligence, together with the learning from previous migrant-specific screening projects in the EU/EEA [39], can help to inform the design of screening programmes to reach migrant populations most affected by chronic hepatitis C infection. A targeted approach in higher risk populations makes more effective use of health care resources and contributes to reducing health inequalities. The World Health Organisation’s Global Strategy for Viral Hepatitis [50] and the European Action Plan [51] both share the ambitious goal of elimination of viral hepatitis by 2030. If this goal is to be realised, it will be essential to dramatically increase the proportion of people who are diagnosed, aware of their infection and on treatment. Future research can contribute by focussing on improving the evidence base on effective strategies to reach and retain migrant and other risk populations in screening and treatment and on cost-effectiveness across the treatment cascade.