2.1 Search Strategy

An initial limited search of MEDLINE (via Ovid) was undertaken with analysis of text and index terms used to describe key articles. These were developed into a comprehensive search strategy with assistance from a research librarian which was used and adapted as necessary across the databases, see Appendix S1. The reference lists of all included articles were hand searched for additional studies.

The electronic databases Medline (Ovid), Embase (Ovid), CINAHL, APA PsycINFO (Ovid) and CINAHL (EBSCoHost) were searched in November 2023 for studies published from 2000 onwards to capture the range of experiences across changing treatment availability; interferon-based treatments were available from 2000 and direct-acting antiviral treatments (DAAs) from 2011 onwards [7]. Google Scholar was searched for grey literature using the same date range and adapted search strategy.

Studies in all languages were considered. Non-English texts were translated using DeepL Translator software (www.DeepL.com/translator) with further professional translation if required.

2.2 Screening of Studies

Studies identified from the searches were uploaded into Covidence [19] and de-duplicated. Two reviewers (HE and PS) independently screened all titles and abstracts, with discrepancies resolved through discussion with a third reviewer (JL-B). Full texts of potentially eligible studies underwent the same process.

2.3 Data Extraction and Critical Appraisal

Extracted data related to study characteristics and study findings (with accompanying illustrations) were entered into a previously piloted spreadsheet (Appendix S2). Following a piloting of extracted data for five studies by two reviewers (HE and JL-B), the remaining studies were extracted by one reviewer (HE). Study findings were verbatim extracts of the study author's analytical interpretation of the results of the data. Each finding was assigned a level of credibility (unequivocal, credible and not supported) dependent on the congruency between the data and the accompanying illustration [16]. Studies were critically appraised using the JBI Checklist for Qualitative Research [16]. Appraisal was initially conducted by two reviewers (HE and JL-B) for five studies, with a high level of agreement being seen, therefore the remaining studies were critically appraised by one reviewer (HE).

2.4 Data Synthesis and Confidence in the Findings

Findings assigned as either unequivocal or credible were grouped into categories based on similarity in meaning and concept through discussion by four reviewers (HE, FA, EM and JL-B). Findings rated as not supported were not synthesised [16]. Further joint discussions (HE, JL-B and CE) refined the categories and aggregated them into synthesised findings which could be used for evidence-based practice. Findings from each separate participant group were initially analysed separately. However, joint discussions (HE, JL-B and CE) identified strong similarities in meaning across the different groups. Hence, the review findings are presented as an overall synthesis in which the findings from different groups are combined. A collaborative process was used to establish a level of confidence in each synthesised finding using the JBI ConQual approach [20]. The confidence ranking was derived from the dependability of the studies and the credibility of individual included findings (full details in Table 6).

3.1 Study Inclusion and Quality Assessment

The database search returned 662 records, of which 256 records were duplicates. Title and abstract screening were performed in 407 studies, of which 45 were potentially relevant and underwent full-text screening. Hand-searching reference lists of these returned one further study. Twenty-eight studies were eligible for inclusion in the review (Table 1). The remaining 17 studies were excluded based on ineligibility of study design (8 studies), context (1 study), phenomena of interest (7 studies) or participants (1 study) (Appendix S3).

Most studies (n = 22) were assessed as being of moderate to high quality, scoring between 8 and 10 [21, 23-27, 29, 30, 32-35, 38, 40, 42, 44, 46, 48, 50-53], with six studies of lower quality [22, 28, 31, 45, 47, 49] (Table 1; Appendix S4). A key weakness across many studies was limited reporting of reflexivity.

3.2 Characteristics of Included Studies

All studies were conducted in high-income countries (Table 1). Two used mixed methodology with the others solely qualitative. The majority (n = 20) employed a form of grounded theory or thematic analysis with five studies using framework analysis. Twenty studies reported interviews with PIP (10 male and female [21, 23, 31-33, 35, 45, 46, 51, 53], nine male only [26, 28, 30, 34, 38, 42, 44, 48, 52] and one female only [50]), nine reported interviews with prison healthcare staff [22, 25, 27, 31, 40, 47, 49, 52, 53] and five reported interviews with custodial staff [24, 28, 29, 31, 40]. Six studies reported interviews with more than one category [28, 31, 40, 47, 52, 53]. Study data were collected between 1996 and 2020. The sample sizes of the studies ranged from 6 to 156.

Four studies focused on generalised viral hepatitis harm reduction measures [28, 31, 44, 50]. Twenty-four studies focused on screening and treatment for HCV, nine of these when interferon-based treatments were available [27, 33, 34, 45-48, 51, 53] and 15 when direct-acting antiviral treatments were being used in the setting [21-26, 29, 30, 32, 35, 38, 40, 42, 49, 52]. No studies were specific to HBV.

3.3 Review Findings

The 28 studies in this review included 104 findings from the experiences of PIP, 36 findings from the experiences of prison healthcare staff and 24 findings from the experiences of custodial staff (Appendix S5). The findings were aggregated into 20 categories which were further interpreted into four synthesised findings. The categories underpinning the synthesised findings are presented in Tables 2–5 and Appendix S5.

Synthesised Finding 1: Accurate, up-to-date knowledge of HBV and HCV disease and treatment among PIP and staff is a facilitator to uptake in viral hepatitis programmes, particularly when imparted by a trusted source.

This finding was synthesised from five categories, which included 17 findings from interviews with PIP, seven from healthcare staff (HS) and six from custodial staff (CS) (Table 2; Appendix S5). The synthesised finding highlights the importance of viral hepatitis disease and treatment knowledge, in which accurate, up-to-date knowledge of HBV and HCV disease and treatment among PIP and staff is a facilitator to intervention uptake, particularly when imparted by a trusted source.

Poor awareness of viral hepatitis disease, its prevalence and long-term health effects along with testing and treatment availability limited intervention uptake among PIP [52, 53]. When PIP's hepatitis knowledge was inaccurate, particularly about hepatitis acquisition, it led to fear and prejudiced behaviour from enacted stigma which discouraged testing [30]. Poor or out-of-date knowledge of treatments led to fear of treatment initiation since previous interferon-based regimes were longer, more invasive and with greater side effects than modern DAA programmes [23]. The method of imparting knowledge was also important. PIP were more likely to believe information when from a credible source, which may be healthcare or trusted peers but rarely custodial staff [46, 48].

Poor staff knowledge and skills were also barriers to programme participation. When staff members lacked up-to-date, accurate knowledge appropriate interventions were not promoted [48]. Viral hepatitis was regarded as a complex health issue requiring specialist skills such as blood taking which the available workforce may lack [25, 47].

Synthesised Finding 2: Personal subjective and relative views have a bearing on viral hepatitis programme participation.

The finding was synthesised from six categories comprising 30 findings from interviews with PIP, three from healthcare staff and two from custodial staff (Table 3; Appendix S5). This synthesised finding incorporates personal subjective and relative views which have a bearing on programme participation. These encompass the person in prison's perceptions of their own identity and self-worth, health beliefs, ability to act on intention, previous experiences, competing priorities and enactment of choice. For staff, this incorporates personal views of health risks. Each person's own value system is imported into the prison environment and reflected in their behaviours.

The association of hepatitis with injecting drug use may lead a person to fear they will be thought less of by others if they have acquired the illness, which in turn leads to internalised feelings of fear and shame. Categorised as felt stigma, this acts as a barrier to testing uptake [26, 52].

Self-perception of health and personal relevance of hepatitis influences the initiation of programme participation. For the PIP, this encompasses self-perception of their current health status, possibly not starting treatment if they considered their HCV was not severe enough or considering treatment pointless if they were continuing drug use [53]. HCV infection could be viewed as benign or lacking in relevance to the individual due to a lack of symptoms [26, 32, 45]. Other PIP did not participate in interventions due to a fatalistic view of treatment not being worthwhile and a high probability of reinfection [38]. Custodial staff were more likely to promote and engage with programmes when they perceived their personal health was at risk by exposure to blood-borne viruses [40].

A person's decision to participate in a programme was influenced by self-perception of their ability to take part which in turn is influenced by personal experience. Being confident in their ability to participate is a facilitator and strengthened by a straightforward, low-hazard programme [42]. An unacceptable emotional burden associated with the programme was a barrier [42]. Positive personal experiences acted as a facilitator. Those engaging in modern HCV treatments often found them easy to take, with positive clinical and nonclinical outcomes, leading to continued involvement [26]. However, negative personal experiences, usually based on older interferon-based regimes, were a barrier to participation. The side effects of older treatments led to perceived personal weakness and potential vulnerability within the custodial setting [21].

Competing priorities such as comorbidities, substance use, coping with the stress of incarceration or family commitments may take precedence over treatment for some individuals [53]. For others, programme participation was a priority since they valued the benefits which improving their own health would bring to their family [21].

Incarceration involves punishment through the removal of liberties and the right to refuse healthcare can be one of the few freedoms available. The PIP may decide to not participate in preventative or nonurgent healthcare, despite negative long-term consequences, for the current opportunity of expressing choice [30].

Synthesised Finding 3: Social interactions and relationships both within the community of PIP and between them and staff groups influence participation in viral hepatitis programmes.

This synthesised finding is underpinned by four categories, developed from 21 findings taken from interviews with PIP, six from healthcare staff and four from custodial staff (Table 4; Appendix S5). This finding explains how the social interactions and relationships both within the community of PIP and between them and staff groups influence intervention participation. It demonstrates the influence of social capital, trust and respect, enacted stigma and duty of care.

Social capital embodies the ‘features of social organization such as networks, norms and social trust that facilitate coordination and cooperation for mutual benefit’ [54]. This review found the concept of bonding social capital captures the beneficial feelings a PIP gains by being part of a group and wanting to conform or behave like others in the group. This bond served as a facilitator when the group promoted a programme or participation strengthened friendship [38] but was a barrier when strong bonds within prison drug-injecting networks promoted continued use of nonsterile equipment [35]. Trust and respect within a relationship can also be utilised for effective therapeutic encounters either between PIP and healthcare staff or between PIP and fellow peer educators [48, 51]. Discrimination resulting from HCV's association with injecting drug use led to enacted stigma, subsequent confidentiality concerns and reduced programme participation [45]. Conversely, nonjudgemental interactions promoted programme participation [42].

Without explicit protocols, social bonds within staff groups and between them and PIP caused duty of care conflicts which led to altered programme participation. Staff experienced conflict between maintaining patient confidentiality and reducing a security or infection risk when security staff were present at health interviews [25] or when blood contact occurred [29].

Synthesised Finding 4: The organisational structure of the prison and healthcare services within it affect participation in viral hepatitis programmes.

The finding was synthesised from five categories comprising 27 findings from interviews with PIP, 19 from healthcare staff and 11 from custodial staff (Table 5; Appendix S5). This synthesised finding incorporates how the organisational structure of the prison and healthcare services within it affect programme participation. Complementary custodial and healthcare systems facilitate passage through a viral hepatitis care pathway, increasing participation.

The routines of imprisonment gave relative stability to PIP who led chaotic, stressful lives outside of prison, affording them time and motivation to address health issues [23]. Prison routines became a barrier to healthcare when the priorities of each conflicted. Security requirements always take precedence over healthcare, so programme access only occurs if security is controlled [25, 29].

The organisation of prison healthcare also influenced programme access. Streamlined, on-site care improved pathway navigation, increased awareness and removed barriers such as escorted hospital visits [24]. Poor linkage of care between prison sites, and with outside prison services, interrupted care pathways for the often-transient prison population [25, 32]. Healthcare access was obstructed by restrictive application procedures or inconsistent availability, both within and between facilities [25, 33, 50].

The optimum timing for programme participation was dependent on the routines of the individual facility and cooperation between custodial and healthcare services. Reduction programmes initiated at integrated prison admission healthcare were viewed as routine and less stigmatising [23]. However, the prison admission process at some facilities can be stressful for PIP with programme uptake improved by delayed scheduling [23, 30].

Limited custodial and healthcare resources restricted programme access, through limited staff or programme availability. For example, when insufficient staff were available for individualised healthcare [48] or to enable prisoner movement [42], scarcity of harm reduction measures led to continued unsafe injecting drug use and treatment unavailable due to funding restrictions [25, 48].

3.4 Confidence in the Review Findings

Confidence in the review findings (as assessed by ConQual) was moderate for Synthesised Findings 1, 2 and 3, relating to knowledge, personal views and social interactions and low for Synthesised Finding 4, relating to organisational structure as detailed in Table 6.

4.1 The Importance of Knowledge

This review found that knowledge of viral hepatitis disease, the potential associated health harms and possible treatments among PIP and staff promotes programme uptake with inaccurate knowledge serving as a barrier. This shares similarities with recent reviews exploring hepatitis programme uptake in community populations at risk of HCV and during the transition away from custodial settings [55, 56].

The accuracy of information and who imparts it are important considerations. Newer, streamlined treatment programmes with reduced side effects became available from 2011 [7]. Without formal training, imparted knowledge is potentially out-of-date, describing more severe consequences than currently occur. PIP are known to frequently acquire health knowledge from other incarcerated individuals so they should receive viral hepatitis education in addition to health and custodial staff [55]. Formal peer education programmes have been utilised to impart many types of knowledge within prison systems [57] and two included studies reported on such schemes [48, 51]. A systematic review of formal prison peer education has found it to be effective [57] although it is of interest that studies in our review showed informal peer knowledge transfer to have more mixed outcomes [24, 30, 33]. The use of peer educators has raised confidentiality concerns [26] which could be mitigated by formal oversight of these schemes. The timing, format and content of educational materials for PIP were also raised as important for knowledge transfer and different to community environments, with consideration of literacy level and potential consequences of being seen to engage with education required [50].

The findings of this review suggest a need for widespread availability of accurate, up-to-date knowledge of HBV and HCV (how they are acquired, health implications and testing and treatment regimens) for PIP and all prison staff. The information must be accessible, in appropriate language and format and available in a nonstigmatising manner. The information should be imparted by a trusted source. If peer education is used, it should be a formal, monitored programme.

4.2 Personal Subjective and Relative Views

The personal subjective and relative views of the PIP have a bearing on programme participation. These are multifaceted with each having the potential to be either a barrier or facilitator, depending on the individual's own perceptions and past experiences. Prison demographics may give broad generalisations of characteristics from which associations are applied but a person's reason for accepting or denying a programme will be personal to them. A person's imported value system and influences prior to prison entry continue to shape their behaviour once incarcerated [58].

Globally, at least one in five PIP is detained for illegal drug use [8]. A previous review exploring experiences of HCV testing and diagnosis among people who inject drugs in the community drew conclusions comparable to this synthesised theme, with programme participation being dependent on shifting priorities, lifestyle chaos level and personal risk perception [59]. While incarceration can bring stability and aid participation, an individual's unique motivations should be recognised within the wider programme framework.

The findings of this review suggest that any HBV or HCV programme framework includes some flexibility to allow recognition of the individual and a degree of personalised care. Awareness of the significance of PIP personal views on participation should be incorporated into custodial and healthcare staff training.

4.3 Social Interactions and Relationships

The social interactions and relationships both within the community of PIP and between them and staff groups can be either a facilitator or barrier to HBV or HCV programme participation. Prisons are complex environments with multiple interwoven relationships and influences which can be difficult to separate [58]. This review found the concepts of social capital and trust underpinned successful healthcare interactions, therefore focusing on these within viral hepatitis programmes will enhance participation.

The sociological concept of ‘prisonization’ incorporates the impact of PIP taking on ‘the general culture of the penitentiary’ and adhering to the ‘inmates code of behaviour’ [15]. While this can lead to a contraculture among PIP [58], knowledge of its influence, intertwined with social capital, can be utilised for the benefit of healthcare. HBV and HCV programme participation will be facilitated if it is a feature of normative behaviour within the environment.

Trust is an essential component of effective therapeutic encounters [60] and has been shown to facilitate engagement with HCV care in prison settings [34]. This was a frequent concept within included papers but despite its significance to PIP, mistrust was pervasive within the environment [46].

The findings of this review suggest that therapeutic relationships within viral hepatitis programmes are built from a foundation of trust and respect. Programme participation should be integrated into normative behaviour to minimise stigmatising actions.

4.4 Organisational Structures of Prison and Healthcare Services

The organisational structure of the prison and the healthcare services within it affect participation in HBV and HCV programmes. Complimentary organisational structures facilitate passage through the viral hepatitis care pathway.

The United Nations has minimum rules for the treatment of PIP which includes the state being responsible for access to the same standard of healthcare as available in the community [61]. The predominant function of prisons to punish individuals and protect society from harm can conflict with healthcare provision and reduce the services available. Proven harm reduction measures such as needle exchange programmes are particularly lacking in prison environments [62]. The different management structures, funding and outcome goals of prison healthcare and security within and between prisons of the same country also contribute to organisational conflict and reduce programme uptake [25].

Ease of movement through the care pathway also influences programme involvement. Community studies have demonstrated the importance of a streamlined care pathway to improving hepatitis programme uptake [57] which was borne out in this review. Point-of-care testing [22] and healthcare telementoring [49] are examples of facilitators via process streamlining. Continued focus on the complete care pathway is required to tackle the risk of HCV transmission due to care interruption from frequent prisoner movement [63].

The findings of this review reinforce the importance of collaboration between custodial and healthcare services to build streamlined care pathways which are appropriate to the facility and population and integrate with community services.

4.5 Strengths and Limitations

Previous reviews have focused on specific aspects of hepatitis interventions in prisons, such as testing and treatment uptake or linkage to community care [39, 55, 59, 64, 65] and incorated few studies of modern HCV treatments [55, 64, 65]. This review has expanded on current knowledge by adopting a comprehensive search across all HBV and HCV care-related interventions, enabling a broad review focus which we believe enhances the relevance and transferability of the findings to the diverse range of custodial hepatitis programmes available.

This review has brought together the experiences of PIP, healthcare and custodial staff. Almost two-thirds of findings were from experiences of PIP, with a fifth or less being from each staff group. The findings were synthesised together due to their similarities and the multidimensional perspective this will give to the conclusions, but we acknowledge the depth of experience expressed within each group is reduced.

This review is limited by the included studies all being from high-income countries and reflecting their prison and healthcare systems. Prison systems vary greatly across the globe. Those within the global south are poorly resourced in comparison to the north with austere, punitive regimes and minimal health promotion [66, 67]. The people incarcerated generally have little input from custodial staff, often organising their own daily routines and discipline with encounters underpinned by negotiation more than conflict [66]. Further research is required in diverse locations, particularly sub-Saharan Africa where custodial health promotion is atypical [67], to assess the relevance of these findings to more diverse prison regimes.

The findings are limited by potential bias among the sampled populations of included studies. Experiences of men and women are included but reporting of ethnicity when it occurred was varied. The difficulties with researchers gaining access to conduct prison studies meant that included studies frequently employed purposive sampling and PIP deemed to pose a risk to researchers were excluded. Therefore, the findings may not reflect the complete prison population.