Interventions for the detection, monitoring, and management of chronic non-communicable diseases in the prison population: an international systematic review

Abstract

Background

High rates of health inequalities and chronic non-communicable diseases exist amongst the prison population. This places people in and/or released from prison at heightened risk of multimorbidity, premature mortality, and reduced quality of life. Ensuring appropriate healthcare for people in prison to improve their health outcomes is an important aspect of social justice. This review examines the global literature on healthcare interventions to detect, monitor and manage chronic non-communicable diseases amongst the prison population and people recently released from prison.

Methods

Systematic searches of EMBASE, MEDLINE, CINAHL, Web of Science, Scopus, and the Cochrane Library were conducted and supplemented by citation searching and review of the grey literature. The literature searches attempted to identify all articles describing any healthcare intervention for adults in prison, or released from prison in the past 1 year, to detect, monitor, or manage any chronic non-communicable illness. 19,061 articles were identified, of which 1058 articles were screened by abstract and 203 articles were reviewed by full text.

Results

Sixty-five studies were included in the review, involving 18,311 participants from multiple countries. Most studies were quasi-experimental and/or low to moderate in quality. Numerous healthcare interventions were described in the literature including chronic disease screening, telemedicine, health education, integrated care systems, implementing specialist equipment and staff roles to manage chronic diseases in prisons, and providing enhanced primary care contact and/or support from community health workers for people recently released from prison. These interventions were associated with improvement in various measures of clinical and cost effectiveness, although comparison between different care models was not possible due to high levels of clinical heterogeneity.

Conclusions

It is currently unclear which interventions are most effective at monitoring and managing chronic non-communicable diseases in prison. More research is needed to determine the most effective interventions for improving chronic disease management in prisons and how these should be implemented to ensure optimal success. Future research should examine interventions for addressing multimorbidity within prisons, since most studies tested interventions for a singular non-communicable disease.

Background

People in prison experience increased rates of chronic non-communicable diseases compared to the general population, including hypertension, diabetes, asthma, and arthritis, as well as various cancers [1]. These diseases cause significant morbidity and mortality, with cardiovascular illness and cancer causing 53% of deaths amongst the prison population in the United States (US) between 2001 and 2019 [2].

Under the principle of equivalence, people in prisons should not be discriminated from accessing healthcare [3,4,5]. Despite this, prior literature indicates barriers preventing them from achieving equitable health outcomes. In the United Kingdom, parliamentary inquires have demonstrated difficulties accessing prescribed medications, receiving timely intervention for health concerns, and attending internal and external healthcare appointments whilst incarcerated [6, 7]. Older persons detained in prison are disproportionately affected, with over twice as many outpatient appointments being missed or cancelled relative to non-imprisoned peers [7]. These difficulties highlight a need for research to determine the most effective methods of managing chronic illness within prisons, accounting for any limitations posed by such environments and prison regimes, so that people’s healthcare needs are not neglected.

Identifying and treating chronic non-communicable disease amongst prison populations is important. Firstly, this would reduce morbidity and mortality on a large scale. This is particularly apparent for people approaching release from prison, who experience a 3.5-fold increased mortality risk in the 1.9 years following discharge compared to the public, including elevated mortality from cancer, cardiovascular and hepatic disease [8]. Secondly, people in prison are part of wider society, and approximately 95% of such persons are eventually released from incarceration [9]. Adequately treating chronic diseases during imprisonment could decrease the burden on community healthcare resources and those caring for people released from imprisonment. Thirdly, people from deprived backgrounds are overrepresented in prisons with higher rates of homelessness, substance misuse, and mental illness, and often irregular contact with healthcare [10, 11]; imprisonment represents a period of stability where healthcare workers can intervene to reduce health inequalities. Finally, the ageing worldwide prison population means that chronic diseases are particularly prevalent in penal institutions [12,13,14]. 90% of older adults in prison have 1 or more chronic diseases [15], and they develop chronic illnesses earlier in life relative to their community peers [16].

Considering these health challenges faced by the prison population, this review aims to examine interventions to detect, monitor, and manage chronic non-communicable diseases amongst people residing in, or recently released from, prison. This research is timely given the rising global prison population [17], and the significant epidemiological, clinical, and patient burden of chronic disease [18].

Methods

The protocol for this systematic review is available on PROSPERO (CRD42022309518) [19].

Search strategy and selection criteria

To be eligible for inclusion in the review, studies must have reported interventions for adults (aged 18 + years) residing in any category prison or who had been released from prison in the past year. Any type of intervention to detect, monitor, and/or manage any chronic non-communicable physical disease was considered. Chronic diseases were defined as per the National Centre for Chronic Disease Prevention and Health Promotion (NCCDPHP): conditions lasting 1 year or more and requiring ongoing medical attention or limiting activities of daily living [20]. Studies must have described the effects of interventions to allow ascertainment of their acceptability or effectiveness, although no specific outcome measures were pre-specified. No control groups were required. All publication types reporting original data were considered.

Exclusion criteria were studies: not reporting original data; focusing on mental, communicable, or acute illnesses; reporting interventions occurring pre-imprisonment or more than 1 year following prison discharge; involving adolescent and/or juveniles; situated solely in immigration detention centres; not reporting the effects of health interventions; focusing on chronic symptoms rather than disease/s; those published before 01/01/2000; and those published in non-English languages.

Systematic searches of EMBASE, MEDLINE, CINAHL, Web of Science, Scopus and the Cochrane Library were conducted covering literature published up to 10th May 2023. Searches were restricted to articles published in English from 01/01/2000 onwards to capture the most relevant interventions to modern day clinical practice and prisons. The search strategy included terms relating to imprisonment, chronic non-communicable disease, and healthcare services or interventions (Additional file 1). Grey literature was searched by reviewing the first 100 articles retrieved from Google and Google Scholar and the websites of relevant organisations including the Ministry of Justice, Howard League for Penal Reform, and Prison Reform Trust (Additional file 2). Backward citation searching was performed by manually reviewing the reference lists of included studies.

All studies were independently screened by two authors. Both authors initially screened articles by reading their titles and/or abstracts, before then reading their full text. Any disagreements regarding article screening were resolved by consensus or seeking third reviewer opinion.

Data analysis

Data were independently extracted from all studies by two authors using standardised templates. The following information was extracted: study type, setting, participant demographics, intervention/s reported, outcome measures, key findings.

Risk of bias was independently assessed by two authors using standardised quality appraisal tools including the Critical Appraisal Skills Programme (CASP) checklists for randomised controlled trials (RCTs), case control studies, cohort studies, and economic evaluations [21]; the National Heart Lung and Blood Institute (NHLBI) quality assessment tools for observational cohort and cross-sectional studies, and case series [22]; the mixed methods appraisal tool (MMAT) [23]; and the Joanna Briggs Institute critical appraisal checklist for quasi-experimental studies [24]. Disagreements regarding quality ratings were resolved by consensus or consulting a third reviewer.

Due to the heterogeneity of clinical interventions, chronic non-communicable diseases and healthcare outcomes studied, collected data were narratively synthesised. Interventions for detecting, monitoring, and managing chronic non-communicable diseases were described and compared between studies. The effects of different interventions were contrasted, considering patterns in the direction and size of effect. Reported barriers and facilitators to implementing healthcare interventions were summarised and compared between diseases and patient groups.

Results

Seventeen thousand two hundred fifteen articles were identified from databases and 1,846 articles were identified from citation searching and the grey literature (Fig. 1). Following the removal of duplicates and non-relevant titles, the abstracts, and full texts of 1058 and 203 articles, respectively, were reviewed for eligibility and 65 articles were included in the review (Table 1). Reasons for article exclusion are detailed in Additional file 3.

Fig. 1

PRISMA flow-diagram demonstrating article screening processes

Table 1 Summary characteristics of included studies

The 65 studies included in this review were conducted in the USA (n = 34), UK (n = 10), Australia (n = 6), Canada (n = 4), France (n = 3), Brazil (n = 2), India (n = 1), Spain (n = 1), South Korea (n = 1), Italy (n = 1), Germany, (n = 1), and Malawi (n = 1). The most common research designs were quasi-experimental (n = 14), case series (n = 11), mixed methods (n = 9), and cross-sectional studies (n = 7).

Of the 65 studies included in the review, both reviewers independently selected the same quality rating for 47 studies (72.3%). Of the remaining 18 cases, consensus was achieved between both reviewers, after discussion, in 14 cases (77.8%) and third reviewer opinion was obtained in 4 cases (22.2%). Most studies were rated moderate in quality (n = 36), whilst 18 and 11 articles were respectively rated as ‘poor’ and ‘good’ (Additional file 4). Common study limitations include a lack of control groups, non-randomised study designs, lack of control of confounding variables, small sample sizes, and reliance on subjective participant self-report.

The total sample size across all studies is 18,311 participants, although five studies were excluded from this calculation as their sample sizes were not explicitly stated or overlapped with another study [41, 62, 69, 71, 79].

Thirty-five studies focused on the management of chronic diseases whilst incarcerated [45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79], 12 on chronic disease screening [25,26,27,28,29,30,31,32,33,34,35,36], one on monitoring chronic illness care [80], and nine on managing chronic disease upon release from prison [81,82,83,84,85,86,87,88,89]. Eight studies described both screening and management interventions [37,38,39,40,41,42,43,44]. Numerous chronic non-communicable conditions are represented in the included literature, encompassing diabetes (n= 11) [45, 49,50,51, 53, 55, 56, 59, 60, 63, 67], gynaecological diseases (n= 10) [26, 27, 29,30,31,32,33, 38, 43, 70], cardiovascular disease (CVD) (n= 6) [34, 37, 42, 61, 62, 83], chronic kidney disease (CKD) (n= 5) [35, 52, 66, 72, 73], dermatological conditions (n= 4) [64, 74, 78, 75], ophthalmological conditions (n= 3) [57, 58, 77], respiratory illnesses (n= 3) [48, 54, 68], oral cancer (n = 1) [25], and colorectal cancer (n = 1) [28]. Approximately one third of studies (n= 22) covered several diseases and/or general long-term prisoner health [36, 39,40,41, 44, 46, 47, 65, 69, 71, 76, 79,80,81,82, 84,85,86,87,88,89]. A minority of research focused on specific populations with prison, including women (n= 12) [26, 27, 29,30,31,32,33, 38, 42, 43, 51, 70], older adults (n= 8) [39,40,41, 44, 87,88,89], and people of Aboriginal Australian ethnicity (n= 1) [35].

Screening for chronic disease in prisons

Several studies investigated screening uptake amongst prison populations [28, 29, 31, 32, 34, 35, 42]. Packham et al. (2020) found that 76·4% of people in prison accepted NHS cardiovascular healthchecks, exceeding uptake in the general population [34]. Similarly, screening uptake and treatment completion for a cardiovascular health programme were significantly higher for incarcerated than nonincarcerated women [42]. Uptake of urinalysis screening for CKD was also high amongst Aboriginal persons in prison [35]. Another study found that between 70.2% and 79.1% of prisoners completed faecal immunochemical testing for colorectal cancer [28]. In contrast, low rates of cervical screening amongst prisoners were reported in three studies varying from 13.5% to 26.9% [29, 31, 32].

Four studies evaluated interventions promoting cervical screening engagement [31, 38, 43, 70]. A prison women’s health clinic significantly increased cervical screening uptake from 15 to 54% and improved healthcare access and experiences [38]. Nurse-led Pap testing clinics and information sessions similarly increased screening uptake, albeit non-significantly by 5.9% [31]. A cervical health promotion intervention involving educational sessions improved women’s knowledge about Pap testing and confidence navigating health systems [43]; in a larger study of the same intervention, statistically significantly increased self-efficacy for cervical screening and follow-up were demonstrated [70].

High rates of diagnosed comorbidities were generally reported following screening interventions. Chaudhari et al. (2013) educated people in prison about detecting precancerous oral lesions and reported 92.2% sensitivity of this screening method, compared to 96.6% for clinical examination. [25] Oral precancerous lesions were found in 6·4% of people in Indian prisons, exceeding the national prevalence of 0·4% [25] Similarly, high rates of CKD were detected amongst Aboriginal people in prison with 25.1% of screenings being positive [35], whereas rates of cardiovascular disease amongst the prison population undergoing NHS healthchecks were comparable with general communities [34]. Rates of reported cervical screening abnormalities in prisons varied from 3% to 16.4% [26, 27, 38]. In a study of colorectal cancer screening over two years, 13.5% and 21.3% of completed FITs screened positive each year [28].

Six studies assessed people’s experiences and perceptions of screening interventions in prison [26, 27, 30, 32, 33, 36]. Williams et al. found that awareness of NHS healthcheck results was variable but generally poor within prison. [36] Similarly, 52.5% of people in prison were unaware of their cervical screening outcomes [26]. Barriers to screening uptake included perceived lack of opportunity within prisons, restrictive prison regimes, difficulties accessing healthcare, lack of standardised processes, males conducting the screening, prioritisation of health emergencies and some patients needs over others, poor treatment by authorities and health professionals, and costs [26, 30, 33]. For cervical screening specifically, women experienced pain, fear, and embarrassment during the procedure and differential screening uptake existed across different groups, with females serving longer sentences, those aged 35–64 years, and those involved in working activities in prison being most likely to engage [26, 29, 30, 33]. One study found that women with lower education levels and fewer than five aliases were more likely to be re-screened for cervical cancer [32]. DeLuget et al. and Giuseppe et al. demonstrated that only 48% and 36.4% of female prisoners, respectively, were aware of the link between cervical cancer and human papillomavirus infection; [27, 29] this knowledge was associated with improved adherence with cervical cancer screening [29].

Tracking chronic healthcare needs in prisons

Four articles assessed the implementation of the Older Prisoner Health and Social Care Assessment and Plan (OHSCAP) [39,40,41, 44]. This tool involves assessing and reviewing the health and social care needs of older adults in prison, creating care plans, and actioning referrals. Patients and staff rated the OHSCAP to be appropriate, beneficial, and feasible [44], but no benefits were seen regarding the number of unmet health and social care needs amongst older adults in prison [39, 41]. This was in part attributed to prison staffing shortages and poor fidelity of implementation [40, 41].

Telemedicine in prisons

Nine studies examined telemedicine in prisons: a remote method of delivering healthcare, which removes the need for hospital transport and associated security risks [55, 59, 62, 71, 74, 75, 77,78,79]. Three studies found telemedicine to be cost effective per session [62, 77], or care plan created [78], when compared to face-to-face appointments. However, this was only applicable if the number of patients seen exceeded a minimum threshold [62, 78]. The time taken to break even on costs incurred from introducing telemedicine varied from 32 months to 275 teleconsultations per year per prison [71, 78]. Two studies assessed telemedicine for diabetes, one reporting reductions in HbA1c (glycated haemoglobin) for 56·9% of their sample [59], and the other finding that 29% attained HbA1c levels below 7% [55]. Yogesan et al. found that, of six patients seen for ophthalmological complaints, only two required face-to-face appointments after telemedicine. [77] Three studies assessed teledermatology, finding that 86·7% of patients experienced clinical improvement at follow-up [78], that teledermatology improved completion of treatment plans [74],and shortened treatment delays [75]. Rates of clinical follow-up after telemedicine appointments ranged from 37.4% to 72% [55, 59, 74]. Low follow-up rates were linked to patient refusal, prison transfer or parole, improvement of disease, financial barriers, or death.

Health education in prisons

Seven studies assessed educational interventions in prisons [46,47,48,49,50,51, 61, 64]. In an RCT, statistically significant improvements were observed in the resting pulse and physical endurance of patients with chronic illnesses completing a 12-week health education and exercise programme in prison [46], but no significant differences in levels of psychological distress were detected [47]. In another study, good knowledge scores were demonstrated amongst patients completing educational group sessions about chronic diseases and healthy lifestyles, although no pre-intervention comparison data were available [61]. Other reported interventions included a skin care manual teaching people about dermatological conditions in prison, deemed by patients and staff to be valid and appropriate [64], as well as staff training which improved prison nurse’s inhaler technique for managing respiratory diseases [48]. Three studies assessed educational programmes for diabetes [51,52,53]. A pharmacist-led diabetes workshop increased patients knowledge of diabetic medications, leading to better management and decreased HbA1c levels compared to controls [49]. Two studies assessed diabetes programmes combining education with calorie reduction or tracking, producing varied results [50, 51]; one study found a significant reduction in HbA1c levels compared to controls [51], while the other found significantly reduced weight in the intervention group [50].

Staff-led specialist services in prisons

Five studies evaluated staff-led services to manage chronic diseases in prison [37, 45, 60, 63, 76]. Two studies assessed pharmacist-led diabetes care involving providing consultations, follow-up evaluations, medication, and health education [45, 60]. Both evaluations reported decreased HbA1c levels from baseline to follow-up. A pharmacist-led anticoagulation clinic in prison also increased the frequency of people with international normalised ratios (INR) at goal by 94% [45]. Three studies assessed the impact of nurse practitioners/specialists in prison healthcare teams [37, 63, 76]. This role was valued by staff for being safe and reducing treatment delays but did not impact patient compliance and satisfaction [76]. The employment of a cardiovascular specialist nurse increased the prison’s hypertension register numbers by 30%, and achieved high patient and staff satisfaction [37]. Mills (2013) demonstrated statistically significant improvements in the glycaemic control of people with diabetes following implementation of a nurse-led diabetes service in prison [63]. The numbers of patient experiencing severe hypoglycaemia, undergoing hospital admissions, and missing healthcare appointments also declined post-implementation of this service [63].

Equipment/device-related interventions in prisons

Three studies described providing specialised healthcare equipment within prisons to manage chronic diseases [53, 72, 73]. Implementing haemodialysis for three patients with end-stage renal disease (ESRD) minimised hospital travel from prison and produced estimated annual cost savings of £100,000 [72]. Similarly, Sankaranarayan et al. demonstrated feasibility of self-performed peritoneal dialysis within prisons, which was described as safe and effective but with higher rates of hospitalisations for peritonitis compared to general population data (160 vs 100 hospitalisations for peritonitis per 1000 patient years). [73] Providing glucose meters to patients in prison with diabetes slightly, but not statistically significantly, decreased their HbA1c levels at 8 months follow-up with no safety issues reported [53].

Multi-faceted interventions and care models in prisons

Nine studies assessed specific care models or programmes, including integrated healthcare services and multi-faceted interventions [47, 49, 51,52,53, 60,61,62, 69]. Raimer and Stobo (2004) examined the ‘Texas correctional managed health care system’, a collaboration between the criminal justice system, healthcare teams and medical schools involving using standard disease management guidelines, patient and clinician education, chronic care clinics, telemedicine and electronic medical records to deliver care in prisons [69]. The system increased overall clinical performance measures for six chronic diseases from 40.1% to 96.8% and produced estimated cost savings of $215 million over 6 years [69]. Ha and Robinson (2011) evaluated the chronic care model (CCM) in prison, especially for asthma [53]. The CCM promotes evidence-based guidelines, clinical information systems, and patient and clinician education, whilst also involving planning for prison release. The CCM produced estimated cost savings of $15 million over 3 years and was perceived positively by patients and staff with fewer patient complaints about treatment [53]. Pimentel (2019) described a ‘patient-centred medical home model’ involving identifying and tracking people in prison with asthma, patient and clinician education, and allocating physicians responsible for coordinating each person’s care [68]. Following implementation, visits to the prison treatment area and hospital for exacerbated asthma markedly reduced, although statistical significance was not tested [68]. Jenkins et al. (2012) implemented consultant diabetologist and diabetes nurse sessions in prisons, case note reviews, joint specialist clinics, dietetic clinics, and staff education and training. [56] This intervention produced estimated cost savings of £24,639 compared to traditional hospital-based care and achieved improvements in all quality indicators except for numbers undergoing retinal screening [56]. Cost savings of $635.65 per person were also reported from integrating nursing services in courts to provide diabetic assessments and treatment [67]. Two studies evaluated renal transplant programs for patients in prison which achieved one-year post-transplant survival rates of 100% and 1-year graft survival rates of 94% and 100% [52, 66] Annual cost savings 2–3 years post-transplant varied from $50,644 to $60 749 and the median waitlisted time for people in prison was similar to non-incarcerated persons [52, 66].

In contrast to the positive outcomes from care models described above, patients in prison who received glaucoma care at an academic referral centre were found to have fewer clinic visits compared to non-incarcerated controls. They were also more frequently lost to follow-up, with only 26.6% of repeat consultations occurring within the recommended time-frame [57, 58]. Oladeru et al. also reported poor follow-up rates for patients in prison with cancer undergoing radiation therapy at safety net hospital (where healthcare is provided regardless of insurance status or ability to pay). [65]

Assessing chronic disease care in prison

One study evaluated a 34-item self-administered tool for assessing chronic illness care in prison (ACIC-P) based on the CCM [80]. Prison staff generally perceived the tool as useful, representing an ideal target for healthcare, although amendments were required to improve its relevancy to prisons [80].

Post-release healthcare interventions

Eight studies investigated community-based healthcare programmes for people leaving prison, often termed ‘transitions clinics’ [81, 82, 84,85,86,87,88,89]. These programmes typically involve primary care by physicians in the first two weeks of release, referrals to community organisations, and case management from community health workers (CHW) with histories of incarceration. One study found that 34% of people attended post-release appointments at a designated healthcare clinic [85], while Shavit et al. (2017) reported one month engagement rates varying from 15–77% across transitions clinic sites. [86] Retention in primary care at six months ranged from 38 to 45% [81, 82, 87]. Two studies found positive effects of recruiting CHWs including increased patient enrolment [87], and retention in care at 6 months [81]. Conversely, Wang found no significant difference in primary care utilisation between transitions clinic clients provided with primary care and a CHW, and those receiving primary care [88]. In two studies, lack of transport hindered access to healthcare for people released from prison [81, 85]. Two studies reported reduced acute care utilisation amongst transitions clinic users compared to normal primary care [88, 89], while one study found increased acute care utilisation amongst people engaging in transition clinics within one month of release from prison, compared to those engaging later [86]. One study assessed the effect of transitions clinics on disease outcomes, finding that 35% and 14% of patients with hypertension and diabetes respectively reached their disease outcome goals [82]. Another study found that providing such services returned 2.55 US dollars per dollar spent [84].

Fuller et al. (2021) described a mobile app where peer mentors assisted people released from prison with cardiovascular screening and linkage to health resources. [83] Most mentors and patients rated the app as navigable and useful for supporting community re-entry [83].

Discussion

This review has described numerous interventions to detect, monitor and treat chronic non-communicable illness amongst the prison population using evidence from 65 studies and 12 countries.

Screening interventions for CVD and CKD had high uptake whilst people’s engagement with cervical cancer screening in prison was poor. This discrepancy is likely due to barriers specific to cervical screening, such as fears of embarrassment, detecting cancer, and/or experiencing pain [90]. Furthermore, females in prison experience disproportionately high rates of sexual trauma [91], which may decrease their engagement with Pap testing [92]. This suggests that screening interventions must be sensitively advertised and explained to prison populations, whilst considering trauma-informed approaches and addressing population-specific barriers to non-participation.

Telemedicine was the most frequently studied intervention for treating chronic non-communicable disease in prison. This generally reduced the need for face-to-face hospital appointments and associated transport for people living with chronic diseases. This is likely to significantly improve healthcare availability given the high frequency of missed hospital appointments across the prison estate [93]. Telemedicine was generally more cost-effective than face-to-face healthcare provided that a minimum number of teleconsultations took place [62, 78]. Clinical outcomes from telemedicine were positive across multiple conditions including diabetes, ophthalmological and dermatological diseases [55, 59, 74, 77, 78] These findings are consistent with prior research; in a systematic review of telemedicine in prisons, Edge et al. found that telemedicine provided equivalent or improved care quality, increased convenience, reduced stigma of accessing healthcare, reduced costs, and improved security. [94] Telemedicine may also upskill prison staff in disease management through remote exposure to multidisciplinary specialists and ‘telementoring’ [94]. Despite these benefits, telemedicine may not be appropriate for all situations, with patient preference, abilities to engage with technology, staff burden, and requirements for face-to-face examination warranting consideration [95].

Educational interventions were effective at increasing patient’s and staff’s knowledge and skills in chronic disease management in prison, improving disease outcomes in some studies. These findings are akin to research demonstrating efficacy of therapeutic patient education amongst general communities [96]. The review also highlighted evidence of effectiveness for specialist staff roles in prison, such as nurses and/or pharmacists with expertise managing long-term conditions, including improved disease outcomes and detection of morbidity. These findings are similarly echoed in the wider literature, where clinical nurse specialists are associated with improved patient, family, and healthcare team outcomes [97].

Transitions clinics involving discharge planning and early contact with primary physicians generally supported engagement with healthcare for people released from prison, although engagement rates varied between studies and clinic sites [81, 82, 85,86,87]. Given the heightened risk of mortality upon discharge from prison [8], more research is needed to explore these differences and understand facilitators and barriers to continuity of care.

Few studies measured similar healthcare outcomes, making it difficult to compare the effectiveness of different clinical interventions, especially across different patient groups. Some studies reported disease-specific clinical outcomes, such as glycated haemoglobin levels in diabetes, whilst other studies focused on costs, patient and staff experiences, disease knowledge, self-efficacy, follow-up rates and patient engagement. All these outcomes are useful and often considered by policymakers to determine the most efficacious, effective, and acceptable healthcare interventions. Agreeing and implementing a framework for defining high quality management of non-communicable illness in prison could help to better track the quality of care delivered in such settings. Comparing health outcome data across different diseases and patient groups within prisons could also identify areas of pressing need where health interventions should be specifically targeted and/or tailored.

To our knowledge, this is the first systematic review of interventions to manage chronic non-communicable diseases amongst prison populations. The review is reported as per PRISMA guidelines. The inclusion of both published and grey literature and all study types increased the pool of evidence, permitting understanding of a broad range of interventions for numerous diseases. Several countries are represented in the included literature, improving the generalisability of the review findings; however, Western countries are over-represented, and resources may differ between individual prisons.

There are several limitations of this systematic review. Firstly, the review is limited by the low quality of evidence from several studies, with many lacking control groups and utilising non-randomised or observational designs. Longitudinal follow-up of disease outcomes was generally lacking, limiting understanding of disease trajectories. Including a wide range of chronic diseases and healthcare interventions in this review largely increased clinical heterogeneity; however, this allowed a broad overview of different clinical practices being employed to manage chronic non-communicable illness within prisons, including variations in treatment approaches between different diseases and patient groups. A further limitation is that the literature search strategy did not include disease-specific terms, owing to the large range of different chronic non-communicable diseases, which may have resulted in some relevant research not being identified.

Determining the prevalence of chronic diseases in prisons is important to ensure that illness monitoring, and treatment are aligned to patient-need. To achieve this, health screening programmes are required to consistently detect morbidity amongst people in prison, as well as robust systems for recording, storing, and transferring health information. English prisons have recently implemented a primary-care patient registration system (GMS1) allowing lifelong electronic health records to be transferred into and out of prison; such systems support continuity of care, and their impact should be formally evaluated.

Robustly designed, longitudinal studies with control groups are needed to explore the most effective interventions for monitoring and managing chronic non-communicable diseases in the longer-term in prisons. This research is essential for determining which interventions achieve the most progress towards equitable health outcomes. Research would also be useful to understand the context and mechanisms by which chronic disease interventions succeed or fail in prisons, as well as factors affecting differential uptake and success of interventions such as telemedicine, disease screening, and ‘transitions clinics’ between different patient groups and locations. Learning from the pandemic, when telemedicine uptake increased throughout penal institutions, will be essential for guiding how technology influences chronic illness care in prisons in the future. Reviewing the experiences of patients and prison staff regarding disease management in prison could also provide insights into relevant challenges and innovative practice, whilst allowing service-user co-design of healthcare interventions.

Conclusions

This review highlights numerous types of interventions available to manage chronic non-communicable diseases in prison settings, many of which were associated with positive clinical outcomes. The quality of the evidence, however, is limited by a lack of longitudinal follow-up of patients and lack of control groups. Future studies should directly compare the effectiveness of different clinical interventions in prisons to detect, monitor, and manage chronic non-communicable diseases and multimorbidity. This will help to inform policy decisions regarding the design of healthcare systems to manage chronic illness in prison.