An audit of hepatitis C service provision in a representative sample of prisons in England

Abstract

Background

Prisons are an important setting to address prevention, testing and treatment of hepatitis C virus (HCV) and other blood-borne viruses. This audit examined current practice against national standards in a representative sample of prisons in England.

Methods

The audit tool was developed based on best practice guidelines and piloted in one prison. In December 2012, the audit was conducted in a further 20 prisons, which were chosen to represent different types, sizes and geographical spread across England.

Results

Testing for HCV was offered in the majority of prisons audited (20 of 21), but only two-thirds had a written policy on testing and treatment; less than a third had a steering group to oversee the process. The nature of services varied greatly. There were inconsistencies across data sources on testing.

Conclusions

This audit found that while there were many areas of good practice, the quality and content of hepatitis C service provision varied. It highlighted the need to provide appropriate guidance for prisons in delivering a high-quality service, ensuring that relevant training is available for different staff and that adequate psychosocial support is provided to patients.

communicable disease, prisons, public health

Introduction

Hepatitis C virus (HCV) is the third most common risk factor for liver disease in the UK (following obesity and harmful drinking).¹ It is also preventable and can be cured by antiviral treatment. The hepatitis C strategy for England identified prisons as a specific setting in which hepatitis C testing and treatment should be delivered.^2,3 This is because many people with a history of injecting drug use (IDU), the main risk factor for hepatitis C in the UK, pass through the prison estate. Research from the Ministry of Justice on a sample of newly sentenced adult prisoners from 49 prisons in England and Wales found that 68% had used an illicit drug in the past year and 40% had injected a drug during the 4-week period prior to custody.⁴ Prison provides an ideal opportunity to identify, test and treat high-risk ‘hard-to-reach’ groups and to reduce the prevalence of hepatitis C both in prisons and in the wider community. This strategy will avoid downstream costs to the National Health Service (NHS) from untreated hepatitis C, including management of cirrhosis and liver cancer.

This is the first known audit of hepatitis C services in a representative sample of English prisons. An audit was previously conducted in a female prison in the North of England.⁵ This found that almost a quarter (69 individuals) of those arriving at the prison have contact with blood-borne virus (BBV) services; most reported a history of IDU (75%).⁵ Almost 80% agreed to be tested and half of them were positive for the hepatitis C antibody, the majority of whom (83%) had an active virus as identified by polymerase chain reaction (PCR) and were suitable for treatment.⁵ The average time from testing to treatment was 38 days.⁵ A study of hepatitis B and C testing (and vaccination coverage for hepatitis B) was published in 2012, which examined data from the 24 sentinel laboratories in England between 2005 and 2008.⁶ This found that, of the 39 prisons represented in the sentinel surveillance system, an estimated 2.4% of the prison population tested for anti-HCV (hepatitis C antibodies) and, of those tested, 24% tested positive for anti-HCV, but only 62% of these had an HCV RNA detection test, which is essential to understand whether the infection is current and to establish whether treatment is needed.⁶

In July 2012, a survey of hepatitis C services in prisons in England was published, which revealed variation in the structure, accessibility and quality of HCV services.⁷ These results prompted the development of a detailed audit of HCV services 1 year later.

Methods

Audit tool development

A working group of experts from the Offender Health team (OH), Department of Health (DH), Health Protection Agency (HPA), Her Majesty's Prison Service and The Hepatitis C Trust was formed to provide guidance and oversight of the audit. National BBV and HCV service standards^8–12 were reviewed to identify key criteria in delivering a good HCV service in prisons. The audit tool based on these criteria was then piloted by HMP Manchester to ensure usability and validity.

Sampling methodology

It was intended that the hepatitis C testing and treatment data from prisons would be triangulated with data from the Health Protection Units (HPUs) and HPA sentinel surveillance system. Thirty-one adult prisons have been included in the sentinel surveillance system since 2009, and 20 of these were selected to obtain a representative sample based on the security category, type and size, with a wide geographical distribution. As far as these criteria allowed, the prisons were selected randomly. The sampling strategy is presented in Table 1. The prison population at the time of the audit was 85 942, and the 20 prisons plus the pilot site included 16 772 prisoners (19.5% of the prison population or 17% of the total number of prisons).¹³ The working group also decided that a Foreign National Prison and two privately run prisons should be included to make the sample more representative of the adult prison estate.

Table 1

Sampling strategy used in the national audit of hepatitis C services in prisons in England, 2013

Prison security category and population sizea requirements	Total number
Category B (at least one of each size)	4
Category C/D (at least two of each size)	7
Female (any size)	1
High security (any size)	1
Local (only large and medium in sample frame)	7
Total number of prisons in sample	20

^aPrisons categorized into three sizes based on actual prison size reported on 27 June 2012.

Data collection

A letter was e-mailed to the Governing Governor and Health Care Manager of all prisons in the sample on behalf of the Chief Executive of the National Offender Management Service (NOMS) and the Director of Offender Health, DH. Evidence of auditable information was requested along with information provided in the audit tool including: Data on testing were also requested from the relevant HPU, HPA Prison Infection Prevention (PIP) team, HPA Sentinel Surveillance and Prison Health Performance Quality Indicators (PHPQIs) dataset. The full dataset are now available in the published audit of the Public Health England (PHE) website.¹⁴

• Written HCV documentation used such as care pathway(s), policy and standard operating procedures.

• Induction program, i.e. a list of activities carried out during the prisoner's first week in the prison.

• Pre- and post-test checklists.

• Health promotion material used (both locally and nationally produced).

Ethical approval

As this is a service audit, it did not require ethical approval. However, prior to any research being carried out by the DH, a proposal has to be approved via the Review of Central Returns (ROCR) process to ensure that the data being collected are appropriate and necessary. The ROCR process is concerned with supporting the DH and its Arm's Length Bodies to implement the government's policy in ‘reducing the burden’ of data collections from the NHS. In order to obtain the ROCR approval, the audit also had to be approved by the equalities team at the DH and by a Minister. ROCR approval was gained to proceed with this audit (reference: ROCR/12/2179VOLU).

Results

The participation rate for the audit was 95% (19 of 20 prisons). However, another prison from the same region and of the same category was used to replace the non-responding one. In the analysis, the pilot prison was also used resulting in a denominator of 21 for the questions. Data were available for the required period from the sentinel surveillance system for 19 of 21 prisons and for 20 of 21 prisons in the PHPQI. The results are summarized in Table 2, and full results are available in the audit report.¹⁴

Table 2

Summary of audit results

	Indicator	Number of prisons with a positive response	Percentage out of eligible prisons
Steering group and leadership	Steering group	6	29
	Clinical lead	17	81
	Clinical team	10	48
Policy and governance	Policy	13	62
	Prison presented some testing and/or treatment data	15	71
	Reported testing data to PHPQI	20	95
	Sentinel surveillance testing data available	19	90
	Prison on HPZonea	18	86
	PIP log has data on HCV case from prison	8	38
Prevention	Leaflets	21	100
	DVDs	6	29
	Posters	16	78
	BBV information available at induction	11	52
	Disinfectant tablets available	17	81
Testing	Offered	20	95
	Carried out by GUM	15	71
	Carried out by drug treatment services	13	62
	Carried out by prison healthcare services	20	95
	Venous blood test available	21	100
Service delivery	Most common model—hospital outpatients	11	52
	Most common model—hospital in-reach	9	43
	Most common model—prison GP led	1	5
	A form of psychosocial support available	13	52
Continuity of care	Medical hold—prisoners always placed once treatment commenced	10	48
	Medical hold—prisoners sometimes placed once treatment commenced	8	38
	Medical hold—prisoners never placed once treatment commenced	3	14
	Follow-up once released into the community—contact local service providers	11	52
	Follow-up once released into the community—contact local service providers and GP	5	24
	Follow-up once released into the community—contact local service providers/GP	1	5
	Follow-up once released into the community—prisoner takes copy of SystmOneb record to GP	3	14
	Follow-up once released into the community—no detail provided	1	5
	Follow-up on transfer to another prison—SystmOne only	6	29
	Follow-up on transfer to another prison—SystmOne and contact by phone/fax	13	61
	Follow-up on transfer to another prison—SystmOne and contact by phone/fax and contact consultant treating patient	2	10
Training on BBVs	For healthcare staff	17	81
	For prison officers	10	48
	For drug workers	12	57

^aPrison on HPZone in 18 of 21 prisons but hepatitis C cases only recorded in 11 of these prisons.

^bSystmOne is an electronic patient management system used by prisons and other healthcare providers in the UK.

Steering group and clinical leadership

Most prisons audited had a clinical lead for hepatitis C (17 of 21, 81%), but far fewer had a steering group to oversee testing, treatment and care strategy in prisons (6 of 21, 29%). Less than half (10 of 21, 48%) had a clinical team who would oversee the operational side of treatment and caring for prisoners with HCV.

Policy and governance

Around two-thirds (62%) of audited prisons had a written HCV policy or equivalent document, although these documents varied in quality and content (see Table 2). Almost all provided information on testing criteria but not all provided information on delivering and managing positive and negative results.

Prevention

All prisons audited had leaflets and booklets available in prison, and 52% (11 of 21) of these made them available during the induction and reception period. The majority of those audited also had posters (16 of 21, 76%) and DVDs (6 of 21, 29%). Disinfectant tablets were available in the majority of prisons, although the route of obtaining them varied.

Testing

Testing was offered in 20 of 21 prisons (95%), and in all these prisons, the testing was carried out by resident healthcare staff. (One prison did not offer testing but prisoners may be referred or self-referred for testing, in which case an in-reach testing service is provided.) Testing was also carried out by genito-urinary medicine (GUM) services (15 of 21, 71%); Integrated Drug Treatment Service/Counselling, Assessment, Referral, Advice and Throughcare teams (13 of 21, 62%) and prison primary care services (2 of 21, 10%).Venous blood testing was carried out by all prisons (21 of 21, 100%). It was the main test used in all but one prison. Around two-thirds of prisons audited stated that blood samples testing positive for HCV antibody were automatically tested in the referral laboratories for the presence of the virus by PCR. Around three-quarters of prisons audited had a documented pretest and post-test discussion.

Some testing and treatment data in 2011 was reported by 15 of 21 (71%) of prisons. Ninety-five percent reported testing information to the PHPQIs for 2011/2012. (PHPQI data are collated annually from April to March.) According to prison data, the number tested in 2011 ranged from 0 to 465 per prison. In contrast, according to the PHPQI data, the number tested in 2011/2012 ranged from 0 to 370 and according to sentinel data 1 to 416 in 2011. There is a poor concordance between the prison reported data, PHPQI data, PIP team records and data recorded by the HPUs on (HPZONE is a web-based case management and surveillance tool designed by the HPA to support their health protection work).

Service delivery for assessment and treatment

The most common service model to assess infected people for antiviral treatment was hospital outpatient (52%, 11 of 21), followed by hospital in-reach (43%, 9 of 21) and general practitioner (GP) in-reach service (5%, 1 of 21). Most prisons were not able to provide further investigations beyond blood tests for hepatitis C. Two of 21 (10%) and 3 of 21 (14%) of prisons, respectively, had on-site access to transient elastography (Fibroscan^®) or ultrasound scans. Some form of psychosocial support was available in 52% (13 of 21) of prisons, but only around half of these were provided by mental health services. There was insufficient data in this snap-shot audit to determine rates of instigation, completion or efficacy of antiviral treatment in this setting.

Continuity of care

Eighty-six percent (18 of 21) of prisons reported that they either always or sometimes place prisoners on medical hold (held at the same location for the duration of their treatment) if they had started treatment and 3 of 21 (14%) prisons stated that they did not place prisoners on medical hold for this purpose. The extent of follow-up, when the prisoner was transferred to another prison or released into the community, was very variable. Some prisons did not have a procedure in place because it was considered a rare event.

Training

Most prisons, 81% (17 of 21), had some form of training on BBVs for healthcare staff; 48% (10 of 21) had training for prison officers and 57% (12 of 21) had training for drug workers. This varied considerably in frequency.

Discussion

Main findings of the study

This audit measured a representative sample of 21 English prisons (17% of prisons) against identified national HCV good practice guidelines, which were relevant for the prison estate and found great variation in quality. Some prisons have a comprehensive and well-established HCV service whereas in others the service is much more limited.

There was not always a clear policy available in prison covering the key areas of testing, treatment and care. Prisoners should ideally only be tested for infection if there is a clear care pathway in place, to ensure those who test positive receive high-quality care in an appropriate and timely fashion. Ensuring that there is adequate training for healthcare staff and wing officers is important. This would vary depending on the person's role, but all staff working with prisoners play an important part in promoting health and encouraging prisoners to be tested for HCV if considered at risk. Psychosocial support appears to be another area that requires development. There are a number of different services and organizations that can provide support, including The Hepatitis C Trust free phone line, drug and alcohol services and mental health service. Current provision is variable and requires better coordination. The audit also highlighted the need to improve the completeness of recording of testing and treatment information.

What is already known on this topic

In the UK, the most common mode of transmission of hepatitis C is IDU.¹⁵ A large proportion of the prison population in the UK are former or current drug users and, therefore, more likely to be at risk of exposure to HCV. Prisons in the UK were, therefore, highlighted in the HCV strategy first published in 2002 and again in 2012 as a specific setting for HCV testing and treatment.^2,3 The first national survey of HCV services published in 2012 found that all prisons surveyed offered some form of testing; however, only 40% (44 of 110) were routinely testing for HCV RNA on antibody-positive samples, despite the fact that 99% were venous blood samples, which can be used to test for both anti-HCV and HCV RNA.⁷ 82 of 110 prisons (74%) of those surveyed have a written care pathway in place to describe what happens to prisoners who test positive for HCV infection.⁷ According to the PHPQI data, 6.2% of the prison population in 2012 were tested for HCV.¹⁶

What this study adds

There appears to be some improvement in the HCV services in the prisons audited compared with when the survey was conducted 1 year before. For example, some of the prisons were automatically testing for HCV RNA where they were not previously. Although this is a policy more relevant to laboratory practice than prisons, nonetheless the service specification for any testing for HCV is recommended to include PCR tests.¹² One of the main findings of the audit is that there is no entirely accurate source of HCV testing and treatment data. In some cases, the sentinel surveillance data recorded more prisoners being tested than the prison data. This may be because other services, such as the GUM clinic, undertook the testing and did not necessarily report back to the prison. On other occasions, the sentinel surveillance figures were lower, potentially because the prison may count the same prisoner more than once if he/she was repeatedly tested. There was also inconsistent recording by the HPUs. Currently, only acute cases of hepatitis C need to be recorded by the local health protection team under the health protection (notification) regulation 2010 (regulation 2).¹⁷ The data from the five sources published in this report need to be better coordinated to inform commissioning and service development in order to drive forward quality improvement.

Medical hold practice is also variable. Whilst ideally prisoners should be placed on medical hold if they have started treatment,¹² this is not always practical for a variety of reasons such as security and transfer to be nearer their family or to serve their sentence in a lower security prison. It is therefore important that a policy of HCV services in prisons includes information on medical hold and what to do when this is not possible.

Limitations of the Study

This audit has some limitations. It is based on self-reporting from the prison and is therefore reliant on the responses being accurate. An audit of prisoners' notes may be more accurate, but this was not feasible to carry out in the timescale. To address any ambiguities, the audit was piloted to make sure whether the questions were clear and covered all the relevant points. Some of the items were open to interpretation. To try and overcome this, documents were requested as evidence, but not all prisons were able to provide this supplementary information. 10 of 21 submitted some form of supplementary information, and in 8 of 21 of these prisons, a HCV policy was included. The audit returns were checked to ensure they were filled in correctly and any answer that looked inconsistent was verified with a follow-up call where possible. To validate the outputs for this audit, data were requested from the sentinel surveillance scheme, PIP team and HPUs. However, there was insufficient concordance to rely on these comparisons. In addition, the data requested were for 2011, in order to have a complete year, and so it does not capture recent changes to testing arrangements in the prison post-2011. The audit only captures 17% of prisons, although they were selected to be as representative as possible of the types of prison estate. This audit was designed to measure the key features of a good HCV service in prisons, but due to time constraints, it was not possible to audit every detail of the service.

Conclusions

Similarly to the previous survey conducted by the DH/HPA, this audit has found varying quality of HCV services in England.¹⁴ It has identified areas of good practice and other places where there is a need for improvement. The audit measured the prisons against good practice guidance some of which may be aspirational given the many demands and constraints on the prison service. However, HCV along with other BBVs are found more commonly in the prison population than in the community, which provides an important opportunity to not only test and treat but also give preventative advice. This audit focused on testing, treatment and health promotion for prisoners with hepatitis C. However, it is clear that this should be part of a BBV service and be closely linked with drug treatment services working to tackle the main risk factor for acquiring HCV. A future study may explore these links in more detail. The findings of the audit are being taken forward by the Health and Justice, PHE team working together with NOMS to ensure that an effective hepatitis C service, encompassing testing, treatment and prevention, is delivered in the prison setting, taking into consideration the constraints of individual prisons.

Note

This audit was carried out by the Offender Health team, Department of Health (OH, DH) and the Prison Infection Prevention Team, Health Protection Agency (PIP, HPA). However, since 1 April 2013, the functions of both agencies have now been subsumed into the responsibilities of the Health and Justice Team within Public Health England (PHE). There are several references to the HPA and DH OH within the document, but it is important to note that the recommendations made in the report are supported by PHE.

Acknowledgements

We gratefully acknowledge the work of all those who have contributed to the audit including the hepatitis C audit working group, prison healthcare staff, PHE sentinel surveillance of hepatitis testing team and NHS South.