Introduction
Seventy-one million people live with hepatitis C virus (HCV) worldwide [1]. Globally, HCV disproportionately affects incarcerated individuals. People living in detention (PLD) are 9–13 times more likely to be HCV infected than the general population [Reference Larney2]. Of the estimated 10.2 million PLD worldwide on any given day in 2013 [Reference Walmsley3], Dolan et al. estimated that more than 1.5 million (15.1%) were living with HCV [Reference Dolan4].
Since 2014, highly effective, well-tolerated direct-acting antivirals (DAAs) with shorter treatment duration have allowed for the completion of treatment within the average length of imprisonment, which is 8.5 months globally [Reference Aebi and Tiago5, Reference Ermis and Senocak Tasci6]. As the disease can have a long asymptomatic phase, early detection and treatment can prevent spread, progression, and reduce long-term healthcare costs.
PLD are from the poorest and most marginalized sections of the population [Reference Karstedt7]. They often serve multiple, short-term sentences [Reference Aebi and Tiago5] due to crimes of poverty – violations committed primarily out of economic necessity – and drug-related offences – which are a leading causes of imprisonment globally, particularly in countries with punitive drug policies [8]. PLD spend most of their lives outside of prison, so treating them while they are inside reduces onward transmission risk in the community. Therefore, prison-related interventions such as the micro-elimination of HCV in places of detention by universal screening and treatment of the infection will not only deliver benefits to the individual but are likely to create a community dividend, that is, benefit for general population health [Reference Stürup-Toft, O‘Moore and Plugge9]. Such interventions that focus on equivalent health outcomes for PLD and the general population – rather than equivalent healthcare – are more likely to successfully contribute to equity in prison health [Reference Charles and Draper10]. They may also impact on community-level health disparities by providing healthcare to PLD that might not have been accessible prior or following their incarceration, thus helping to reduce health disparities when these individuals return to their communities.
The potential positive effect of prison-related HCV interventions on prison populations and the wider community has been highlighted numerous times but no one has yet synthesized the evidence that examines whether and in what ways HCV diagnosis and treatment of PLD benefit general population health. Therefore, a scoping review was conducted with the aim of mapping out sources, types, and characteristics of evidence on the existence of a community dividend and to identify key outcomes that make up the community dividend.
Inclusion criteria
Participants
Population of interest was defined as people living in detention or in secure psychiatric units for any length of time of all ages – including young offenders – and of all genders and sexualities. Recently released PLD or those on parole who were tested in prison, started treatment there or awaiting to start treatment in the community were also included. We considered sources reporting on both whole general prison populations and prison population sub-groups (e.g., injecting drug users living in detention, HIV-infected incarcerated individuals, etc.).
We excluded people living in police custody, at immigration removal centres, prisoners of war, and individuals recently released from detention or on parole if they were not offered an HCV test or/and treatment while in detention.
Concept
We included evidence to support or refute a community dividend caused by testing and/or treating HCV in detention. Community dividend was defined as the benefit of a prison-related intervention for general population health [Reference O’Moore11]. We excluded evidence with no clear link to a community dividend caused by testing and/or treating HCV in detention. Evidence was excluded when the HCV intervention resulting in the presence or absence of a community dividend was other than testing and/or treatment such as programmes directed at awareness raising, behaviour change, provision of syringes for IDUs, opioid substitution therapy, and so forth.
Context
The context was HCV infection. The markers and manifestations that made evidence eligible for inclusion were the following: HCV infection, HCV-induced liver disease, extrahepatic manifestations associated with HCV, mental and physical illnesses that are the consequences of HCV infection, health statuses, and levels of wellbeing that are associated with the presence and absence of HCV infection.
Markers that were not associated with HCV infection were excluded unless they could not be separated from HCV-associated outcomes.
Types of sources
We included multiple evidence sources and study types to allow for a broad conceptualization of the community dividend.
We included:
-
• primary and secondary research, health economic evaluations directed at testing and/or treating HCV in prison population where at least testing took place while already in custody,
-
• modelling studies whose outcome was a community dividend (e.g., transmission prevention, disease progression prevention, opportunity cost, etc.),
-
• and all other literature published or unpublished that discussed the community dividend resulting from testing and/or treating HCV in detained populations.
We excluded secondary research unless they contained novel information beyond the primary sources they reviewed; thus, excluded systematic reviews without meta-analysis.
Methods
The Joanna Briggs Institute (JBI) scoping review methodology guidance [Reference Peters12, Reference Peters13] was followed. The literature was searched from January 1991 to June 2022 because it was assumed that research activity directed at secure settings might have become more prominent after the publishing of the United Nations’ Basic Principles for the Treatment of Prisoners in 1991 [14]. No limitation was set on the country of origin and language.
Database searches were carried out in EMBASE (Table 1), Scopus, ASSIA, UWE Library, CINAHL Plus, and MEDLINE using the keywords in the logic grid displayed in Table 2. The searches were conducted in titles and abstracts without limits except for the limitation of publication period. Search results were exported into Zotero, duplicates were removed manually, and the remaining items were screened by reading the titles and abstracts. Of the 63 records sought for retrieval, 21 could be included and are listed in Table 3: 19 published articles [Reference Palmer15–Reference Godin33] and two conference posters [Reference Wong34–Reference Manca35]. The reference lists of included literature were searched as per JBI protocol. While more information could not be retrieved on the content of the two conference posters, they were included for more comprehensive mapping of the geography and type of evidence available on the topic.
Table 1. Example library search
Table 2. Logic grid
Table 3. Reference list of included sources
PRISMA ScR flow diagram was completed (Figure 1). It depicts the flow of information through the different phases of the scoping review study selection and maps out the number of records identified, included, and excluded as well as the reasons for exclusion.
Figure 1. PRISMA Flow diagram
From: Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021;372:n71. doi: 10.1136/bmj.n71. For more information, visit: http://www.prisma-statement.org/
Relevant data that relate to the review question and objectives were extracted and charted (Table 4) using the following refined charting table headings: Study Number, Reference, Title, Year, Country, Study Type, Aim(s), Methodology, Population, Description of Intervention(s) and comparator (covers Test (T) and/or Treatment (Tx) and/or Linkage to Care (LtC)), Outcome Measures, Key Findings, Domain of Community Dividend. For each included source of evidence, findings were collated and reported by charting table headings, thus providing an overview of the characteristics of the sources of evidence, the distribution of studies by year and country of publication, the distribution of study designs, intervention types, target populations, and methodologies used. Using the completed charting table, a list of outcomes relating to the community dividend was collated and their distribution across the included sources illustrated.
Table 4. Charting table
As per JBI guidance, the formal assessment of the methodological quality of the included studies was not performed with the aim of providing a basis for inclusion or to facilitate the development of recommendations for practice. It was carried out to determine the trustworthiness and rigour of the included sources, to confirm the quality of the evidence base, and to help draw well-founded dependable conclusions. The Consolidated Health Economic Evaluation Reporting Standards (CHEERS) criteria list [Reference Husereau36], an appropriate tool for the quality assessment of both economic and non-economic modelling studies, was used.
Results
Most of the included studies were published after 2014 except for 5 [Reference Martin18, Reference Sutton, Edmunds and Gill21, Reference Sutton31–Reference Castelnuovo32, Reference Wong34] published between 1991 and 2013. The studies of the scoping review were sourced from a very limited number of high-income countries – United Kingdom [Reference Martin18, Reference Mohamed20–Reference Sutton, Edmunds and Gill21, Reference Martin26, Reference Stone28, Reference Sutton31–Reference Castelnuovo32, Reference Manca35], United States [Reference Assoumou17, Reference He29–Reference Liu30, Reference Wong34], Spain [Reference Marco, Dominguez-Hernandez and Casado16, Reference Nicolas Perez24–Reference Chhatwal25], Australia [Reference Palmer15, Reference Kwon23], Canada [Reference Godin33], Ireland [Reference Ward19], Switzerland [Reference Girardin27], and Taiwan [Reference Chen22].
The aim of the included papers was to evaluate the impact of an HCV intervention or interventions on either the individual living in detention, the prison community, the wider community, or the combination of these. Some evaluated only the economic [Reference Palmer15, Reference Martin18, Reference Sutton, Edmunds and Gill21–Reference Chen22, Reference Girardin27], some only the epidemiological impact [Reference Stone28, Reference Godin33]; others a combination of economic, clinical, and epidemiological impacts: economic and clinical [Reference Marco, Dominguez-Hernandez and Casado16–Reference Assoumou17, Reference Mohamed20, Reference Nicolas Perez24, Reference Liu30–Reference Castelnuovo32, Reference Wong34–Reference Manca35], economic, clinical, and epidemiological [Reference Ward19, Reference Kwon23, Reference Chhatwal25–Reference Martin26, Reference He29].
The interventions could be grouped into four distinctive types according to their coverage of the HCV care cascade: two sources evaluated the impact of HCV testing strategies only [Reference Sutton, Edmunds and Gill21, Reference Manca35], five HCV treatment strategies only [Reference Chen22, Reference Chhatwal25, Reference Stone28, Reference Liu30, Reference Wong34], and ten studies evaluated both testing and treatment [Reference Palmer15–Reference Marco, Dominguez-Hernandez and Casado16, Reference Mohamed20, Reference Kwon23–Reference Nicolas Perez24, Reference Martin26–Reference Girardin27, Reference He29, Reference Sutton31–Reference Castelnuovo32]. Four sources also discussed the effect of linkage to care (LtC) in the community in addition to testing and treatment [Reference Assoumou17–Reference Ward19, Reference Godin33] and demonstrated the positive impact on economic, clinical, and epidemiological markers.
19 of the 21 included studies [Reference Marco, Dominguez-Hernandez and Casado16–Reference Sutton, Edmunds and Gill21, Reference Kwon23–Reference Manca35] used prospective longitudinal modelling – economic or epidemiological – as their research method to simulate the long-term costs and/or individual and population benefits of different HCV cascade of care strategies and scenarios. Among these, there were 11 cost-effectiveness analysis (CEA) modelling studies [Reference Marco, Dominguez-Hernandez and Casado16, Reference Martin18–Reference Sutton, Edmunds and Gill21, Reference Kwon23–Reference Nicolas Perez24, Reference Martin26–Reference Girardin27, Reference Liu30–Reference Sutton31], two posters on CEA modelling studies [Reference Wong34–Reference Manca35], one health technology assessment/CEA modelling study [Reference Castelnuovo32], three CEA and budgetary impact modelling studies [Reference Assoumou17, Reference Chhatwal25, Reference He29], and two epidemiological modelling studies [Reference Stone28, Reference Godin33]. Two studies did not use modelling. One of them was a CEA [Reference Chen22] based on retrospective observational cohort studies. Finally, the second non-modelling study [Reference Palmer15] had an ingredients-based costing approach to costing analysis.
Table 5 provides a visual representation grid of the community dividend-related outcomes and their distribution across the studies. Charting the data facilitated the collation of outcomes with 20 eventual items. Each item related to the community dividend and could be placed into three distinctive groups depending on the impact of the intervention: in the economic, the clinical, and/or the epidemiological domain(s). The green cells on the grid show where a positive outcome was demonstrated, the red cell where the community dividend of the outcome was refuted.
Table 5. Community dividend-related outcomes and their distribution across the included sources
ICER – incremental cost-effectiveness ratio, CHC – chronic hepatitis (CHC), QALY – Quality-Adjusted Life-Year, SVR – sustained virologic response, LC – liver cirrhosis, DC – decompensated cirrhosis, HCC – hepatocellular carcinoma, LT – liver transplant
Both individual-level benefits of prevention of disease progression and population-level benefits of prevention of HCV transmission were demonstrated in the included literature. The cost-saving outcomes that provided an economic rationale for implementing more intensive prison-based HCV testing and treatment were the reduction in HCV disease cost [Reference He29], the favourable cost of treatment initiation in the prison setting [Reference Palmer15] as opposed to community settings, and the positive net monetary benefit [Reference Ward19, Reference Kwon23, Reference Girardin27, Reference Manca35]. Budget impact analyses in the three studies [Reference Assoumou17, Reference Chhatwal25, Reference He29] provided an invaluable argument against budget holders’ reluctance to invest in correctional health. The positive economic impact was demonstrated by 17 of the 18 studies [Reference Marco, Dominguez-Hernandez and Casado16–Reference Girardin27, Reference He29–Reference Liu30, Reference Castelnuovo32, Reference Wong34–Reference Manca35] that completed an economic evaluation by measuring the cost-effectiveness of interventions from the healthcare provider’s or societal perspective. They found that the opportunity costs did not exceed the benefits of the interventions when the benefit was expressed as quality-adjusted life-years (QALYs) gained [Reference Marco, Dominguez-Hernandez and Casado16–Reference Mohamed20, Reference Kwon23–Reference Girardin27, Reference He29–Reference Liu30, Reference Castelnuovo32, Reference Wong34–Reference Manca35], sustained virologic response (SVR) achieved [Reference Chen22], chronic hepatitis (CHC) detected [Reference Sutton, Edmunds and Gill21, Reference He29, Reference Manca35], or the cost of screening per HCV-positive person linked to treatment [Reference Girardin27]. Cost-effectiveness was refuted in one study that was published in 2008 [Reference Sutton31].
The clinical benefit was measured across the studies by the changes recorded in the number of cases in specific liver disease stages. One third of the sources provided evidence of the reduction in one or more of the following: fibrosis, decompensated cirrhosis, hepatocellular carcinoma, liver transplant, and liver-related death [Reference Marco, Dominguez-Hernandez and Casado16–Reference Assoumou17, Reference Nicolas Perez24–Reference Chhatwal25, Reference He29–Reference Liu30, Reference Castelnuovo32]. These were lifetime cumulative measurements or projections for at least 30 years, which meant that the positive effects of testing and treating PLD well exceeded the length of their prison stay and would have impacted mostly on the community. Another outcome is the increase in the proportion of lifetime SVR [Reference Assoumou17]. Fourteen CEAs gave a value to the health benefit of their interventions in QALYs [Reference Marco, Dominguez-Hernandez and Casado16–Reference Assoumou17, Reference Ward19–Reference Mohamed20, Reference Kwon23–Reference Martin26, Reference He29–Reference Castelnuovo32, Reference Wong34–Reference Manca35]. They measured how many additional years of life a PLD and people in the community may gain due to different HCV test and/or treatment scenarios by being cured and/or by not getting infected.
The most reported outcomes in the epidemiological domain were the changes in HCV prevalence [Reference Ward19, Reference Kwon23, Reference Chhatwal25–Reference Martin26, Reference Stone28, Reference Godin33] and incidence [Reference Ward19, Reference Chhatwal25–Reference Martin26, Reference Stone28, Reference Godin33] computed for the overall population, not just the prison population. Reduction was found in point prevalence in all the studies that measured the proportion of the population with HCV at a specific future time [Reference Kwon23, Reference Chhatwal25–Reference Martin26, Reference Stone28]. Studies reporting on cumulative prevalence or period prevalence [Reference Ward19, Reference Godin33] – quantifying the change in the HCV-infected proportion of the population over a period of time – also described reduction. Reduction was seen in incidence due to interventions described by five studies [Reference Ward19, Reference Chhatwal25–Reference Martin26, Reference Stone28, Reference Godin33]. They modelled the difference in the number of new HCV cases in a population over a given interval of time (1 year) attributed to interventions and their comparators. The prevention effect on the prevalence and incidence of different testing (T) – treatment (Tx) – linkage to care (LtC) strategies – varying in intensity and their coverage of the cascade of care – was compared in a study [Reference Godin33] by the cumulative percentages of new first chronic infections that could be prevented between 2018 and 2030. The prevention of new cases per person treated in prison for the different scenarios also captured the community dividend [Reference Godin33]. The same prevention benefit was expressed in a converse way in one source [Reference He29] where the number of new HCV cases resulting from untreated positives being released into the community was given as an epidemiological outcome.
The main author and a health economist completed the quality assessment of the included sources using the CHEERS criteria list [Reference Husereau36] (Table 6). Overall, 11 studies were assessed as good quality, one study was of moderate quality and 6 studies were assessed as high quality. Almost all the studies identified their studies as an economic evaluation except for two studies [Reference He29–Reference Liu30]. Only one study [Reference Castelnuovo32] indicated whether a health economic plan was developed. All the studies except two [Reference Mohamed20–Reference Sutton, Edmunds and Gill21] provided characteristics of the study population, including age range, demographics, socioeconomic, and clinical characteristics. All the studies provided details of alternative interventions or strategies compared. All studies stated the perspective adopted and why it was chosen except one [Reference Chen22]. Most of the studies stated the time horizon adopted and why it was appropriate except two studies [Reference Chen22, Reference Girardin27]. Only a few studies [Reference Marco, Dominguez-Hernandez and Casado16, Reference Martin18, Reference Chhatwal25–Reference Girardin27, Reference He29–Reference Castelnuovo32] described how costs were valued. All the studies described methods to characterize any sources of uncertainty in the study. All included studies reported mean values of costs and outcomes except one study [Reference Nicolas Perez24], effects of uncertainty from analytic judgements and input parameters on findings. All studies reported key findings and limitations of study.
Table 6. Quality assessment using CHEERS
Discussion
HCV is an important global health issue, and it is particularly significant for people in places of detention, as health inequalities disproportionately affect them [Reference Larney2]. Several studies have shown the effect on the general population of HCV testing and treatment interventions carried out in places of detention by focusing on some of the measurable outcomes. These studies have all been included in our review. Our study is the first to introduce the concept of community dividend to the research and stakeholder community, synthesising and describing all the outcomes that are related to it.
Our results demonstrate that the community dividend of testing for HCV in places of detention and treating chronic HCV-infected incarcerated individuals comprise 20 measurable outcomes within three major domains: economic, clinical, and epidemiological. It is a cost-effective public health strategy and increasingly so with the recent availability of DAAs. Case-finding and treatment are a good investment of taxpayers’ money and result in savings in long-term health expenditure. The cost is amply compensated by individual and collective benefits. Cost-effectiveness was refuted by only one study [Reference Sutton31]. However, the study took place in 2008 before the availability of well-tolerated DAAs that have a shorter treatment duration, and the researchers themselves noted that improved treatment acceptance and adherence would ensure more cost-effectiveness.
All studies evidenced that testing and treating HCV in PLD reduces the incidence of HCV-related liver complications, increases survival, improves quality of life for both the prison population and the general population, and reduces infection transmission. Most of the benefits are realized in the community following release.
Our scoping review provides invaluable evidence that can significantly contribute to evidence-based policymaking and the design of interventions aimed at scaling up HCV testing and treatment in detention facilities. For a healthcare intervention to be thoroughly appraised and recommended for practice, evidence must encompass not only effectiveness but also appropriateness and feasibility [Reference Evans37]. We have identified major gaps in the existing evidence base as only one aspect of feasibility—budgetary impact—was covered in the studies reviewed, and none of the included sources considered appropriateness. These gaps could be addressed by various research methods, for example, observational, interpretive, and descriptive studies, focus groups, action research, case studies, expert opinion, and so forth. As for improving the evidence of effectiveness, taking a broader perspective when carrying out future economic research could present HCV interventions as more cost-effective. Cost-consequence analysis, preferably as a supplement to CEA would allow consideration of non-health-related or difficult to quantify outcomes such as equity. Social return on investment analysis would be able to account for broader value and outcomes, social, economic, and environmental benefits [38].
The main limitation of this scoping review is that the main search, study selection, and data charting and collating results were conducted by only the main author as part of their master’s dissertation project although co-authors viewed a sample of included studies. Bias was minimized by adherence to JBI guidance, completion of PRISMA ScR Checklist (Table 7) and frequent meetings and discussions.
Table 7. Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) Checklist
JBI = Joanna Briggs Institute; PRISMA-ScR = Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews.
* Where sources of evidence (see second footnote) are compiled from, such as bibliographic databases, social media platforms, and Web sites.
† A more inclusive/heterogeneous term used to account for the different types of evidence or data sources (e.g., quantitative and/or qualitative research, expert opinion, and policy documents) that may be eligible in a scoping review as opposed to only studies. This is not to be confused with information sources (see first footnote).
‡ The frameworks by Arksey and O’Malley (6) and Levac and colleagues (7) and the JBI guidance (4, 5) refer to the process of data extraction in a scoping review as data charting.
§ The process of systematically examining research evidence to assess its validity, results, and relevance before using it to inform a decision. This term is used for items 12 and 19 instead of “risk of bias” (which is more applicable to systematic reviews of interventions) to include and acknowledge the various sources of evidence that may be used in a scoping review (e.g., quantitative and/or qualitative research, expert opinion, and policy document).
Tricco et al. [Reference Tricco, Lillie, Zarin, O’Brien, Colquhoun and Levac41].
Although we aimed to include a wide range of sources, the current review can be characterized by homogeneity in study types, mainly providing evidence of effectiveness via economic evaluations. The validity and trustworthiness of the evidence however is supported by the high-ranking research methods of the included sources as well as the quality assessment of the included papers, which was good overall. The quality assessment was performed by two independent reviewers, the main author and a health economist, to reduce bias.
Another limitation of the findings is the geographical spread of the located studies. Most sources were from the United Kingdom and the United States and a few other high-income countries (Spain, Australia, Taiwan, Canada, Switzerland, Ireland). Middle- and low-income countries were not represented at all. This has been previously documented in prison research [Reference Plugge39]. Investment in research and elimination programmes remains low in many parts of the world due to lack of political commitment and domestic and international financing [Reference Pedrana40]. It must be noted that some of the factors influencing the community dividend cannot be separated from characteristics of local political systems, economies, and epidemics such as the degree of penalization of drug use, average length of prison stay (which might determine eligibility), drug availability, the efficiency of drugs used, prevalence and incidence of HCV infection and of intravenous drug use, and other HCV risk factors in prisons and in the community. Therefore, we expect that research aimed at providing evidence of effectiveness, appropriateness, and feasibility would vary geographically.
The authors have demonstrated that HCV testing and/or treatment interventions in PLD greatly benefit general population health by resulting in long-term health expenditure savings, reduction in HCV-related disease sequalae, increase in survival, improvement in quality of life, reduction in infection transmission, and most of these benefits are seen in the community. Understanding the considerable impact that testing and/or treating HCV in PLD have on general population health will inform stakeholder decisions to invest in testing and treatment services for PLD, which will be crucial for achieving HCV elimination and for reducing inequalities both within and outside the prison wall.
Data availability statement
All sources that support my findings are included in the reference list and can be accessed individually. All data extracted from the 21 sources that met inclusion criteria for the scoping review can be found in the attached Excel table entitled charting table.
Acknowledgements
Nick De Viggiani – dissertation supervision (superscript no 4)
Author contributions
Formal analysis: C.E.; Conceptualization: E.P.; Writing – review & editing: E.P.; Supervision: N.D.V.; Writing – original draft: E.K.
Competing interest
The authors declare none.
Open access
