Tuberculosis (TB) in Prisons and Immigration Removal Centres

Published Date: 27 Nov 2017

An evidence-based partnership approach to tackling Tuberculosis in Prisons and Immigration Removal Centres in London

Abstract

Background

The World Health Organization (WHO) have declared TB as a global emergency with 8.6 million cases of active TB and 1.3 million deaths. The incidence of TB in the UK remains high compared to most other Western European countries, with 8,751 cases reported in 2012, an incidence of 13.9 per 100,000 population. London accounts for the highest proportion of cases in the UK (39%) and the highest rate of disease (41.8 cases per 100,000). Left untreated, one person with pulmonary TB may infect around 10–15 people every year. People in prison and IRCs represent a population who are at particular risk. National estimates for TB prevalence in the prison population are 208 per 100,000 and amongst London's 10,000 or so prison population we would expect 20 cases, but we are seeing more than double. The cost of treating ‘normal’ TB is around £5000 and is much greater for more socially complex cases (estimated at £50,000–£70,000). There is considerable variation in the delivery of some aspects of TB services. A co-ordinated national TB strategy is required to support locally designed and implemented services, and monitor achievements against national standards. This paper presents findings from a partnership between NHS England, PHE and NOMS to tackle TB across its prison population in the overall approach to the overall TB strategy in London.

Aims/objectives

To establish whether national NICE guidance for TB in prisons and immigration removal centres is being met.

Methods

Target population included all 9 prisons and 3 IRCs for which NHS England (London region) are responsible. Methods used: 1) An organisational clinical audit during January 2014 using the NICE baseline assessment tool; 2) Stakeholder engagement through a steering group and a wider reference group.

Results

Effective stakeholder engagement contributed to a 100% completion rate. All establishments had referral pathways in place and a named contact within the local Multidisciplinary TB team and the local Public Health England health protection team. 2/12 establishments did not screen for TB within 48 hours of arrival. 3/12 did not have a local TB policy. 2/12 did not have a named TB lead. None of the DH funded x ray machines were being used in line with NICE recommendations. Latent TB was not being diagnosed or managed.

Conclusions

Active and systematic case finding is needed within a prison and IRC setting as well as more rigorous and standardised contingency and follow up care plans after release (or transfer).

Introduction

The World Health Organization (WHO) have declared TB as a global emergency with 8.6 million people with TB and 1.3 million deaths due to TB (World Health Organisation, 2013). The six point Stop TB Strategy (World Health Organisation, Europe, 2013b) explicitly addresses the key challenges facing TB with the goal to dramatically reduce the global burden of TB by 2015 by ensuring all TB patients benefit from universal access to high-quality diagnosis and patient-centred treatment. However, there have been challenges in developing and implementing program-wide interventions in both high income(Migliori, Sotgiu, Blasi, et al., 2011) as well as middle and low income countries(Cobelens, van Kampen, Ochodo, et al., 2012). England and Wales have responded to the need to tackle TB where the NHS and the Department of Health have developed a national Action Plan for ‘Stopping Tuberculosis in England’(Department of Health, 2004). NICE have also developed a set of National guidance for the identification and management of TB across a number of settings(NICE public health guidance, 2011) which highlights the need for a multi-agency approach. There has been little evidence evaluating the implementation of these guidance.

What is TB?

TB is caused by Mycobacterium tuberculosis, which spreads in airborne droplets when people with the disease cough or sneeze. Most people infected with M. tuberculosis never become ill as their immune system contains the infection. However, the bacteria remain dormant (latent) within the body, and a latent TB infection can cause active disease many years after the initial infection if immunity declines. The symptoms of TB include a persistent cough, weight loss, and night sweats.

The BCG vaccine (Bacillus Calmette-Guérin vaccine) protects against TB and it was thought possible to wipe out TB through a vaccination programme. The BCG vaccine is made from a weakened form of a bacterium closely related to human TB. Because the bacterium is weak, the vaccine does not cause any disease but it still triggers the immune system to protect against the disease, giving good immunity to people who receive it. In the past, the BCG vaccination programme was delivered to all teenagers in the UK but as TB is a difficult disease to catch because it requires prolonged exposure to an infected person, it was changed so that now only people inat-risk groups are given the vaccination. The vaccine is 70-80% effective against the most severe forms of TB, such as TB meningitis in children but It is less effective in preventing respiratory disease, which is the more common form in adults(Trunz, Fine & Dye, 2006). Even with the high coverage now achieved, BCG is unlikely to have any substantial effect on transmission.

Risk factors that seem to be of importance at the population level include poor living and working conditions associated with high risk of TB transmission, and factors that impair the host's defence against TB infection and disease, such as HIV infection, malnutrition, smoking, diabetes, alcohol abuse, and indoor air pollution. Preventive interventions may target these factors directly or via their underlying social determinants. The identification of risk groups also helps to target strategies for early detection of people in need of TB treatment(Lönnroth, Jaramillo, Williams, et al., 2009).

How common is TB in the UK?

It has been difficult to eradicate TB both globally and in the UK. Vaccination programs and improvements in housing, nutrition and access to treatment have been largely the reason for a global decrease in TB. However, TB is still rife in less developed countries where poor conditions are still present. Several strains of TB bacteria have developed a resistance to one or more anti-TB medications, making them much harder to treat. Theglobal epidemic of HIV that began in the 1980s has also led to a corresponding epidemic of TB cases. This is because HIV weakens a person's immune system, making them more likely to develop a TB infection. The rapid growth of international travel has allowed people to travel widely and this has helped to spread of the disease.

Although the rates of TB have stabilised in the UK over the past seven years, following the increase in the incidence from 1990 to 2005, the incidence of TB in the UK remains high compared to most other Western European countries(Hayward, Darton, Van-Tam, et al., 2003). There were 8,751 cases reported in 2012, an incidence of 13.9 per 100,000 population (Health Protection Agency, 2013b). The majority of TB cases (73%) occurred among people born in high-incidence countries and are generally concentrated to large urban areas with a high proportion of people born outside the UK where the rate of TB among the non UK-born population is almost 20 times the rate in the UK-born (Health Protection Agency, 2013b).

London accounts for the highest proportion of cases in the UK (39%) and the highest rate of disease (41.8 cases per 100,000), followed by the West Midlands (12%; 19.3 cases per 100,000). Left untreated, one person with pulmonary TB may infect around 10–15 people every year (Department of Health, 2004).

TB in prisons

It is important to identify settings where the risk of TB transmission is particularly high. Groups at risk not only include people born in high prevalence areas (e.g. sub-Saharan Africa, South East Asia, Eastern Europe), but also people with reduced immunity (e.g. HIV, diabetes, renal failure), those with alcohol or drug problems and people who are homeless or living in overcrowded conditions (Story, Murad, Roberts, et al., 2007). These risk factors are over represented in prison populations with high levels of social and health needs. A systematic review on the incidence of TB in prisons globally (largely in the USA), showed that TB was about 26x higher than in the general population(Baussano, Williams, Nunn, et al., 2010). In the London prison population, the incidence of TB has been estimated at about 208 per100,000 (Story, Murad, Roberts, et al., 2007).

Figure 1: Incidence of TB in different locations and settings[A1]

Across the London prison and IRC estate, prisons are at or very close of operational capacity (ranging from 72% - 103%) and with the high churn rate which increases the risk of TB transmission and poses significant challenges for TB identification and management (see Table 1). In addition, a significant proportion of the prison population are of a foreign nationality (up to 44% in one prison) and on average, just over one quarter (27%) of the prison population are foreign nationals. The majority of the prison and IRC population are under the age of 39 years old, representing another TB risk factor.

Table 1: Summary of Prison Establishments

Establishment Name	Category	Operational Capacity[1]	Current Roll[2] (on 31st December 2013)	Current Capacity	Number of foreign national’s n (%)	Age 39 years or younger (%)[3]	DH DXR Machine at establishment
HMP Belmarsh	Male Cat A	910 (as of 25th July 2008)	801	88%	196 (24%)	74%	Yes
HMP Brixton	Male Cat C	798 (as of 24th August 2008)	757	95%	122 (16%)	71%	Yes
Colnbrook	IRC (Mixed)	308	unknown	unknown	100%	81%	No
Feltham YOI	Male YOI (16-17 & 18-21 yrs.)	762 (as of 2nd February 2009)	547	72%	121 (22%)	100%	No
Harmondsworth	IRC (Male)	615	unknown	unknown	100%	80%	No
HMP Holloway	Women (18 years and above)	501 (as of 23rd January 2008)	514	103%	172 (33%)	75%	No
Isis YOI	Male YOI (18 – 21 yrs.)	622 (as of 3rd June2011)	609	98%	84 (14%)	100%	No
HMP Pentonville	Male Cat B	1310 (as of May 2013)	1296	99%	386 (30%)	76%	Yes
HMP Thameside	Male Cat B	900	880	98%	260 (30%)	80%	Non-DH funded
HMP Wandsworth	Male Cat B	1877 (as of 29th May 2013)	1497	80%	665 (44%)	64%	Yes
HMP Wormwood Scrubs	Male Cat B	1279	1244	97%	408 (33%)	72%	Yes
Yarl’s Wood	IRC	405	unknown	unknown	100%	69%	No

Individuals at high risk for TB are typically unwilling or unable to seek and comply with medical care, and are therefore hard to reach. Individuals at high risk are also more likely to be diagnosed at a late stage of the disease and are less likely to adhere to treatment(Health Protection Agency, 2013b). In prison and IRC settings, overcrowding, late detection, barriers to adequate treatment, and poor implementation of infection control measures might also increase the TB transmission rate and improving prison conditions is a priority for any programme to control TB and reduce its spread back into the community (Levy, Reyes & Coninx, 1999).

TB has been identified as a key health concern where the need for greater TB control in the prison setting was highlighted in the Chief Medical Officer’s (CMO) action plan for England (Department of Health, 2004). The Department of Health (DH) announced that they were to fund the installation of static Digital X-Ray (DXR) machines in large local prisons receiving people from areas with a high prevalence of TB. This led to the installation of DXR machines in 5 London prisons (and 3 out of London). All participating prisons had their machines signed off and handed over by March 2012 but due to changes in commissioners and providers during the lifetime of the project, the impact of the programme to date has been variable. More recently, the new national partnership agreement(Anon, n.d.) just signed between Public Health England (PHE), NHS England (NHSE) and the National Offender Management Service (NOMS) also draws particular focus and commitment to the epidemiology of TB in prisons, particularly in those that have access to DXR machines. The agreement sets a priority for this year (2013-14) as:

“Priority 11: Improving the detection and management of TB among prisoners at or near reception.”

The commitment in the partnership agreement is to ensure that by April 2014, NHSE, NOMS and PHE will ensure that all fixed digital X-ray machines are fully operational and being used as part of an active care pathway in those prisons where they are currently installed.

TB in IRCs

Robust data relating to TB in IRCs is not routinely collected or available so there are no estimates of the incidence of TB in these settings. However, a sample of detainees in a single IRC within Southern England identified prevalence rates of 3% for TB (McLaren, Baugh, Plugge, et al., 2013) which is considerably higher than those found among the migrant population in England (Health Protection Agency, 2013b). Detainees at Harmondsworth and Colnbrook are men mostly aged 20-40 (see Table 1) and from disadvantaged areas of the world where TB is still rife. The average length of stay is around 2 weeks; although some have been detained for over 1 year.

Guidance for best practice

NICE have developed national guidance on the ‘Clinical diagnosis and management of tuberculosis, and measures for its prevention and control(NICE public health guidance, 2011), as well as more specific guidelines for identifying and managing TB among hard to reach groups (NICE public health guidance, 2012). This guidance, consistent with World Health Guidance(World Health Organisation, Europe, 2013a), aims to improve the way tuberculosis (TB) among hard-to-reach groups is identified and managed and makes specific reference to using prison and IRC settings to target these groups. NICE recommend that early identification and effective treatment of active TB provides the best outcomes, reduces onward transmission and reduces the development of drug-resistant forms of the disease. The identification and management of latent TB infection is also highlighted.

The NICE guidance is based on the evidence resulting from four large systematic reviews(NICE, 2012a, 2012b, 2012c, 2012d) which informed the key recommendations relating to TB in prisons and IRCs. These include the best ways to identify TB, manage TB, organisational factors and identifying and managing latent TB.

Identifying TB

There are several approaches to identify latent and/or active TB in different populations. The Mantoux test is a widely used test for latent TB. It involves injecting a substance called PPD tuberculin into the skin and those that are sensitive to PPD tuberculin will develop a hard red bump at the site of the injection, usually within 48 to 72 hours of having the test. This is indicative of a latent TB. A very strong skin reaction may require a chest X-ray to confirm if this is an active TB infection(NHS Choices, 2013).

The interferon gamma release assay (IGRA) is a newer type of blood test for TB that is becoming more widely available and can also help diagnosis latent TB. It can be used after a positive Mantoux test or as part of a screening or health check process.

An active TB infection is usually diagnosed from a chest X-ray and samples of mucus and phlegm which are checked under a microscope for the presence of TB bacteria. A CT scan, MRI and/or biopsy will also be taken if an extra-pulmonary TB is suspected.

A lack of information and awareness about TB services has been highlighted as a barrier to successful identification of TB (Brent Refugee Forum, 2007). Studies have highlighted that members of hard-to-reach groups frequently report incomplete or inaccurate information about the cause and transmission of TB with misconceptions included dirty or wet environment, sharing of domestic objects, and punishment from God (Brent Refugee Forum, 2007). Smoking(Brent Refugee Forum, 2007; Brewin, Jones, Kelly, et al., 2006; Gerrish, Ismail & Naisby, 2010), poor diet and malnutrition(Brewin, Jones, Kelly, et al., 2006; Gerrish, Ismail & Naisby, 2010), poverty (Brewin, Jones, Kelly, et al., 2006) however, were correctly perceived to affect susceptibility to TB. The fear of medical services as well as anxiety around the associations of TB with death have also been highlighted as barriers to diagnosing TB in high risk groups (Gerrish, Ismail & Naisby, 2010; Marais, 2007; Brent Refugee Forum, 2007).

Stigma is also highlighted as a major issue when diagnosing and screening for TB. Most studies with hard to reach groups describe a sense of shame and forced or voluntary isolation resulting from a TB diagnosis, although stigma was expressed differently in different groups. Homeless participants in London reported that being diagnosed with TB was embarrassing and rarely discussed among the homeless community because of the stigma attached to TB in this population(Whoolery, 2008). TB patients often face dual stigma—from their own communities and their wider communities. Most studies looking at the barriers to identifying TB have been conducted in immigrant groups in community settings and there is a lack of research into the barriers to identifying TB in prison or IRC populations. In addition, there are limited studies that focus on how to improve these passive case detection approaches or contact tracing approaches.

With the difficulties in identifying TB in these hard to reach groups, researchers have sought to evaluate the effectiveness of active screening for TB rather than a passive approach where it is up to the individual to make contact with health services. Active screening has been found to be an effective and cost-effective strategy in immigrants and new entrants (Laifer, Widmer, Simcock, et al., 2007; Monney & Zellweger, 2005; Verver, Bwire & Borgdorff, 2001), homeless and intravenous drug users(Watson, Abubaker, Story, et al., 2007) in identifying active TB cases are an early stage. In particular, the “Find&Treat” service, which is a Department of Health-funded initiative, aims to strengthen tuberculosis (TB) control among hard-to-reach populations through active case finding using a mobile X-ray unit (MXU)(Jit, Stagg, Aldridge, et al., 2011). In addition, the Find&Treat service follows up closely those on treatment and provides support in completing treatment. Although the service used to screen a large number of prisoners, it had mostly stopped since the introduction of DXR machines in prisons for active case finding in new inmates. On average, each year the 'find and treat' service identified 16 people with TB in the hard-to-reach population, who may not have been identified and treated and also managed and supports the treatment for a further 100 or more cases. Despite these studies, there is limited direct evidence for the best methods for screening for TB in prisons(NICE, 2012b).

Puisis et al conducted an innovative program of high speed radiographic screening for pulmonary tuberculosis (TB) at a large American correctional facility. The case finding rate for active disease with radiographic screening was approximately double the rate previously achieved with Mantoux skin testing. (Puisis, Feinglass, Lidow, et al., 1996). However, the findings are unclear how much of the difference in prevalence is caused by the different screening strategies and how much reflects different baseline disease prevalence. Another retrospective cohort study, compared the potential impact of limiting screening with mobile X-ray units to prisoners in the UK with symptoms of TB, compared with universal screening regardless of symptoms. Restricting screening just to prisoners with any of the five symptoms would have missed 36.7% of TB cases and more cases of TB would have been missed if screening was limited to a smaller range of symptoms. (S Yates; A Story; AC Hayward, 2009). The study is limited because although these symptoms may have been present at the time of screening, it is not known if professionals would have screened for TB based on these symptoms in real practice. Mobile X-ray unit (MXU) screening in those that are homeless, drug users or in prison have also been found to reduce diagnostic delay compared with passive case-detection and cases were less likely to be contagious on diagnosis compared with passive case-detection (Watson, Abubaker, Story, et al., 2007). However, the main limitation of this study is that results for different sub-populations were not reported separately, so it remains unclear whether any one hard-to-reach group benefited significantly from mobile x-ray screening. Chest X-ray screening has also been shown to be more cost-effective than the Mantoux test in immigrants and in prisoners(Jones & Schaffner, 2001). However, the start-up costs of implementing the miniature chest radiograph screening were not taken into account. Considering the technology and training necessary to implement such a tool in a prison setting, this information could have had an effect on the costs.

Active screening seems to increase identification of latent and active TB infection across hard-to-reach groups who are at high risk of infection, compared with passive case-detection, and leads to earlier diagnosis and reduced infective periods in those with active TB. Although the effectiveness and cost effectiveness of mobile X-ray screening is limited in prisons settings, NICE recommend that in prisons housing populations from high incidence areas and where the start-up costs had been largely funded by the DH, it was judged that X-ray screening would be cost effective. For other prisons, initial, symptom-based screening was adequate(NICE public health guidance, 2012).

Managing and treating active TB

Although TBcan be a very serious disease, it is possible to make a full recovery from most forms of TB with treatment. TB can usually be cured by taking several powerful antibiotics daily for several months. However, the emergence of antibiotic-resistant bacterial strains and the poor adherence to treatment has kept TB high up on the international health agenda with WHO declaring a crisis of multidrug resistant TB (World Health Organisation, 2013). The Health Protection Agency has found that only 79% of people with TB in the UK completed treatment which is below the World Health Organisation target of 85% (Health Protection Agency, 2013c). The mix of drug regimes, treatment isolation and length of time of treatment presents a number of challenges to ensure patients adhere to treatment regimes. Adherence can be particularly difficult in those with multiple needs, e.g. homeless and seeking substance abuse treatment (Whoolery, 2008).

Directly Observed Therapy Short course (DOTS) is one method used to increase adherence to TB treatment. DOT is not just the direct supervision of therapy but also considers distinct elements of political commitment; microscopy services; drug supplies; surveillance and monitoring systems and use of highly efficacious regimens (World Health Organisation, Europe, 2013b). It can be difficult to evaluate the effectiveness of DOTS as a complete strategy to increase adherence and the focus of studies have evaluated the direct supervision of therapy. For example, significantly more people adhered to more than six months of treatment when they received DOT in substance misuse(Alwood, Keruly, Moore-Rice, et al., 1994) and in foreign born individuals (MacIntyre, Goebel, Brown, et al., 2003). However, there have been limited studies into prison populations with some suggestions of improved adherence with DOT(Rodrigo, Caylà, García de Olalla, et al., 2002) and other findings showing no significant differences (Dèruaz & Zellweger, 2004). The effectiveness of DOT across prison and IRC populations still remains unclear.

The views of treatment and management of TB can be particularly important when considering adherence. For example, the views on traditional and modern medicine can also vary between different groups which can impact the management of TB. The Brent Refugees Forum reported that Somalis in the UK reported a preference to try traditional medicine as a first choice (Brent Refugee Forum, 2007) whereas Somalis in New Zealand would prefer modern medicine first in response to their experiences of TB related deaths in their home country(van der Oest, Chenhall, Hood, et al., 2005). Some groups preferred approaches to treatment that included both traditional and modern medicines(NICE, 2012a). Talking to the patient to find out their preferences can help patients to make decisions about their treatment based on an understanding of the likely benefits and risks rather than on misconceptions (Nunes V, Neilson J, O’Flynn N, Calvert, N, Kuntze S, Smit, hson H, Benson J,, et al., 2009). Very little is known of the impact of TB treatment on jobs, family and children (NICE, 2012a) and in particular, the psychological impact of isolation. TB treatment should be provided on a voluntary basis and the WHO highlight the importance of “engaging with patients as partners in the treatment process and respecting their autonomy and privacy” (World Health Organisation, 2013). This can be a particularly important issue when concerned with isolating an individual with a suspected or confirmed case of TB, which should be undertaken on a voluntary basis and involuntary isolation should only be used as a last resort.

Few studies have explored the potential benefits that patients may experience when seeking TB treatment. A small number of homeless participants reported that TB treatment helped make further lifestyle changes that improved their health in general. For example, improved living conditions and regaining relationships with family (Whoolery, 2008). Another study reported that immigrants reported a ‘social responsibility’ to seek TB treatment although this was anxiety provoking (Brewin, Jones, Kelly, et al., 2006). Evidence suggests that discussing with the patient why they might benefit from the treatment can improve patient engagement and adherence(Nunes V, Neilson J, O’Flynn N, Calvert, N, Kuntze S, Smit, hson H, Benson J,, et al., 2009).

Organisational factors

Delays in identifying and successfully managing TB can be the result of individual and service provider factors. The provisions used to deliver care and support can determine how services should be structured to manage people with TB in hard-to-reach groups. These organisational factors can include the settings used to identify and manage TB as well as the type and needs of the healthcare worker.

A lack of specialist services and coordination of care can be a major difficulty in TB service provision, since most GPs see few cases of TB a year (Belling, McLaren, Boudioni, et al., 2012; Gerrish, Ismail & Naisby, 2010). The complex social and clinical interactions surrounding a patient with TB can be a challenge to participation and adherence and there is a need for TB link workers to facilitate coordination of services (Brent Refugee Forum, 2007; Belling, McLaren, Boudioni, et al., 2012). Healthcare workers may find it challenging to meet the complex care needs of hard-to-reach groups with TB, especially where there are cultural and language barriers that make it difficult to interpret symptoms and explain about the disease and its treatment (Moro, Resi, Lelli, et al., 2005). In addition, service providers can also be afraid of the consequences of contracting TB, including becoming stigmatised. Non clinical healthcare workers may also have limited knowledge about TB, the need for screening and the implications of a positive test result (Joseph, Shrestha-Kuwahara, Lowry, et al., 2004).

There is considerable variation in the delivery of some aspects of TB services and more research is needed in the UK on the effectiveness and cost-effectiveness of different service structures to manage TB(NICE, 2012d). In addition to the NICE national guidance, Public Health England have produced London specific guidelines on the management of TB in prisons (Health Protection Agency, 2013a) based on pilot work across a number of London prisons. This guidance aims to minimise the risk of transmission of TB within the prison environment through efficient systems to detect cases early and ensure effective treatment, and by ensuring continuity of care when patients move around the prison estate and/or leave the prison. However, it is unclear as to whether the guidance available to prisons and IRCs in developing best practice is implemented successfully. The recent restructuring of the NHS has now transferred healthcare responsibility from Primary Care Trusts (PCTs) to NHS commissioning boards and the changes to the structure of the NHS aims to create a more integrated approach to providing health services and address variations that exist in accessing quality healthcare. The impact in providing healthcare within criminal justice settings is particularly important for groups with multiple health needs, mental health conditions and those from low socioeconomic backgrounds. Analysis of the current service provisions for tackling TB in prisons and IRCs therefore is not only essential for informing and developing future services but it timely to coincide with the NHS new commissioning strategy and priorities.

Aims

To establish whether the evidence based recommendations outlined in the national NICE guidance for TB in prisons and immigration removal centres are implemented and to what extent.

Objectives

1. Conduct an audit of the services and structures in place to tackle TB across the London prisons and IRCs.
2. Identify gaps in the implementation of national NICE guidance.
3. Understand the roles of partner organisations in London and assess the need for local Service Level Agreements to define roles and responsibilities especially at transition points when people enter or leave a detention setting, in line with the national partnership agreement (PHE, NHSE, NOMS).
4. Provide recommendations to improve the identification and management of TB in these settings.

Method

Research design

A single baseline service audit of prison and IRC healthcare service across the London region was conducted. This audit did not include external organisations or agencies that may provide support for prisons and IRCs (e.g. health protection teams). Since prison and IRC healthcare teams provide the primary point for identifying and managing TB, it was felt important to audit this part of the service. In addition, healthcare teams may have made particular reference to other individuals and teams as part of the audit and it was deemed important to evaluate the knowledge of and access to these external agencies.

This audit was completed by a designated member of each healthcare team which was facilitated by an interviewer (Anita Mehay and Thara Raj). The interviewer’s role was to ensure successful completion of all components of the audit, answer any questions and probe for any additional information.

The NICE guidance for TB identification and management in hard to reach groups was chosen as the benchmark with which to evaluate current practice since the guidance is based on rigorous systemic reviews of the evidence, are applicable to all services (including prisons and IRCs) with a statutory requirement placed on healthcare teams to follow these.

Settings

Target settings included all nine prisons and three IRCs for which NHS England (London region) are responsible. Five adult male prisons currently have a DH funded DXR Machine for TB screening (HMP Belmarsh, HMP Brixton, HMP Pentonville, HMP Wandsworth and HMP Wormwood Scrubs). In addition, HMP Thameside have a non-DH funded DXR machine for TB screening.

Audit Tool

To facilitate implementation of national guidelines, NICE provide a standardised baseline assessment tool that can be used to evaluate current service provisions with the national guidelines[4]. This standardised tool can be used to compare services across each prison as well as provide a tool for prisons to regularly evaluate their services at different time points. Each prison and IRC were asked to self-rate whether they met the criteria’s set out in each main recommendation of the guidance (either fully, partially, not met, or not applicable). Any additional comments within the audit were also captured and noted.

As this assessment tool is designed to evaluate a range of services across different settings, including prisons and IRCs, only four of the 16 recommendations were relevant to prisons and IRCs. Therefore, only these four recommendations where included in the audit. An additional two criteria around prison staff awareness and screening which were set out in the general NICE TB clinical guidance were also included in the audit as they made specific references to prisons and IRCs.

Procedure

Stakeholder engagement was established through a steering group and a wider reference group where during December 2013, Thara Raj presented the project proposal to all heads of health in prisons and IRCs. It was decided that to ensure successful completion of all stages of the project, heads of health and their teams would be provided with extra support to complete the baseline assessment. All Heads of health were subsequently contacted by Anita Mehay, through an email reiterating the aims and objectives of this project and a request to arrange a day and time to facilitate the completion the baseline assessment.

Face to face or telephone appointments were arranged for some time in January 2014. Heads of health had the opportunity to invite or delegate the task of completing a baseline assessment audit to other relevant members of the team (e.g. Nurse Lead, Primary Care Manager). For those that did not respond to the initial email request, a subsequent follow up email was sent and further contact attempts were made, including by phone and through support from Clinical Commissioning Managers.

The baseline assessment itself took between 30mins to 1 hour to complete and any additional contextual information was recorded. Soon after each meeting, an electronic version of the completed baseline assessment was sent to each heads of health for quality assurance and approval.

Limitations of approach

The main limitation to this approach is the reliance on provider-led information therefore any information given was supplemented by written evidence and documents where necessary (e.g. written TB policy).

Analysis

Descriptives of categorical data was analysed for each prison and IRC and presented both individually as well as a collective. No statistical analysis is provided due to the audit nature of the study. Descriptives are presented to reflect comparisons rather than provide any statistical significance. In addition, qualitative data was subjected to content analysis to provide further contextual understanding of the key themes that evolved.

Results

The audit was completed in 100% of establishments within the London region (See Appendix I for timelines) reflecting the acceptability of a facilitated self-audit approach. Four establishments completed the audit by telephone due to either time constraints or convenience in avoiding obtaining gate passes to the prison. The remaining eight establishments completed the audit through face to face meetings.

Overall findings

Of the 26 criteria set out in the four recommendations, just over half were fully met by the prisons and IRCs in the London region and nearly a third were not met at all:

Figure 2: Audit result for all recommendations

By analysing each recommendation in more detail, key gaps are identified (Figure 3):

Figure 3: Audit results by recommendation

The majority (90%) of the criteria’s set out in recommendation 12 (identification and management of Latent TB) are not being met across all the prisons and IRCs. Although most of the criteria’s in Recommendation 8, 9 and 10 are reported as fully met by prisons and IRCs, there appears to be important gaps in meeting some of these criteria’s, where up to 40% of the criteria’s in Recommendation 8 are either partially, unsure or not being met at all.

Key Findings

• 2 of the 12 establishments did not screen for TB within 48 hours of arrival
• 3 of the 12 did not have a local TB policy
• 2 of the 12 did not have a named TB lead
• None of the DH DXR machines were being used in line with NICE recommendations
• Latent TB was not being diagnosed or managed

Further descriptions and analysis into each recommendation with an exploration of qualitative findings highlight the difficulties and successes in meeting the NICE recommendations:

 

Organisational factors in identifying and managing active TB

Criteria descriptions
1	Multidisciplinary TB (MDTB) teams, prison and immigration removal centre healthcare services should have named TB liaison leads to ensure they can communicate effectively with each other.
2	Prison and immigration removal centre healthcare services should develop a TB policy by working with the MDTB team and the local Public Health England unit.
3	MDTB teams, in conjunction with prison and immigration removal centre healthcare services, should agree a care pathway for TB to ensure any suspected or confirmed cases are reported to, and managed by, the MDTB team.
4	MDTB teams, in liaison with prison or immigration removal centre healthcare providers, should manage all cases of active TB. Investigations and follow-up should be undertaken within the prison or immigration removal centre, wherever practically possible
5	Healthcare workers providing care for prisoners and remand centre detainees should be aware of the signs and symptoms of active TB. TB services should ensure that awareness of these signs and symptoms is also promoted among prisoners and prison staff. (1.9.3 clinical guideline 117)
6	Prison service staff and others who have regular contact with prisoners (for example, probation officers and education and social workers) should have pre- and on-employment screening at the same level as for healthcare workers with patient contact (see sections 1.9.1 and 1.9.2 NICE clinical guideline 117).

Over 60% of the criteria’s in Recommendation 8 where fully met by London prisons and IRCs (See figure 3). Further analysis of criteria’s within this recommendation reveals that criteria 6 (screening for prison staff) was an area of uncertainty (See figure 4). Nearly all prisons and IRCs were either unaware of TB screening and procedures for prison staff or stated that this is not within the healthcare responsibility:

Figure 4: Results by criteria for Recommendation 8

Criteria’s 1-3 highlight the need for TB leads and MDTB teams with policies and care pathways. Although on the whole, these criteria’s were fully met, establishments varied in how they met these criteria’s (see Figure 4, below). For example, Isis YOI, met very few criteria’s and stated that did not have a TB policy as they had not developed the MDTB links to develop this since they only opened in 2010 and were still a new prison. They also thought that the service provider, Harmoni, should have a policy in place (although if they had, they were not aware of it). Isis YOI were also unsure whether TB was an issue for YOI and if a TB policy was required.

HMP Thameside also stated they had not developed the relevant MDTB links to progress with any formal TB policy since they were newly opened (2012).

Figure 5: Results for meeting each criteria set out in recommendation 8

In addition, HMP Brixton reported that they do not have a formal TB policy. They have pathways in place that are ‘person dependent’ and are reliant on the Health Protection Team nurse. They were in the process of writing some guidelines on the use of the DH DXR machine but not a general TB policy, however, they are considering drawing up a TB policy in light of the audit. Colnbrook IRC also stated that although they have pathways in place, they do not have a formal TB policy in place.

HMP Wandsworth partially met many criteria’s in this recommendation as although they reported to have TB pathways in place, this was a “bit person dependent” and lacked a more formal TB policy. They are currently developing a TB policy but have had some setbacks in achieving this due to short staffing, uncertainty surrounding the commissioning of healthcare services, the opening of new Category C wing and their new status in accepting younger adults.

Case Examples

HMP Belmarsh reported to have robust and effective systems in place to identify and manage TB. They ‘fully met’ the majority of the recommendations set out in the NICE guidelines – more than any other establishment. They have clear written policies in place which key members of the MDTB teams who work well to manage TB. The procedures do not appear to be person specific.

They overcome some of the logistical challenges of liaising quickly and efficiently with MDTB teams where meetings are facilitated by the use of a teleconference. The TB policies also link in well with prison cell allocation policies whereby prisoners are only allocated a cell after the health screen at reception which means greater control over possible TB identification and onward transmission.

HMP Brixton however, do not have a TB policy and are currently identifying and managing TB through individual staff interests with unwritten procedures. There is a reliance on specific members of staff with allocated days for TB screening which prevents adequate and timely identification and management of potential cases. In addition, there were little reported signs that they are seeking to improve procedures and pathways or that there was indeed a need, since it was felt that being a resettlement prison, prisoners should have been screened prior to arriving at HMP Brixton.

Identifying active TB

Criteria Description
1	On arrival at the prison or immigration removal centre, healthcare professionals should ask all prisoners and detainees (including those being transferred from other establishments) if they are taking TB medication, to ensure continuity of treatment.
2	Healthcare professionals in prisons and immigration removal centres should follow the recommendations for prison screening set out in NICE's clinical guideline on tuberculosis – but should ensure prisoners and detainees are screened within 48 hours of arrival.
3	Prisons with Department of Health-funded static digital X-ray facilities for TB screening should X-ray all new prisoners and detainees (including those being transferred from other establishments) if they have not received a chest X-ray in the last 6 months. This should take place within 48 hours of arrival.

Criteria 3 make specific reference to establishments with DH DXR machines. Of the 12 establishments, five have a DH DXR machine and another one establishment has a privately owned DXR machine. None of these establishments are using these machines in line with NICE recommendations.

Figure 6: Results by criteria for recommendation 9

Overall, the key themes reported by these establishments which prevented them meeting these criteria included:

• lack of healthcare staff to conduct through health screening
• lack of trained staff to conduct DXR
• lack of operational prison staff to transport prisoners to DXR machines
• location of DXR situated too far from reception
• lack of time to conduct all health screening
• Technical problems with the DXR (incl. IT problems in communicating X-rays to hospital, power outages, lack of staff technical knowledge)
• Lack of space to conduct screening
• Lack of need for TB screening

An in-depth account of each establishment provides further details into the challenges in meeting these criteria:

HMP Belmarsh:

Reportedly, it has taken about 3 years to get the DXR machine at HMP Belmarsh fully operational and they described great persistence and hard work to get to it fully functioning. Guided by the Health Protection teams who have shaped their policy on the use of the DH DXR machines, the DXR machine is not used to screen of all new prisoners but instead, they use the initial health screen questionnaire to highlight potential TB cases that require further investigation through a DXR. They state that screening all new prisoners through a DXR would be a very “onerous task, considering they have around 3600 prisoners coming through the estate in a single year”. At present, they anecdotally report to conduct about 20 DXRs a week and they believe that by identifying high risk TB cases and targeting those for DXR is just as accurate at identifying TB cases as part of a wider screening programme.

HMP Wandsworth

HMP Wandsworth also reported several years to get the DXR machine operational due to technical problems. However, it was currently working at the time of the audit. However, they were unable to use the machine in line with NICE guidance due to a lack of staff (in particular, they were lacking a deputy head of health and senior members of staff), the uncertainty surrounding tendering for a service, the opening of a new wing and the inclusion of younger offenders into the prison population. These competing pressures had relegated TB although it was still seen as an important priority that they hoped to develop again.

HMP Wormwood Scrubs

HMP Wandsworth also had some initial problems setting up the DXR machine. It took about 6-8months to get the machine operational which also required the DXR machine to be relocated to a more suitable area in the prison. Due to the size of the prison, they are often overwhelmed with health screening at reception and have a lack of staff and staff skills mix and time. They had five healthcare staff trained in using the machine but with staff turnover, only three now remain within the healthcare team. They still have IT issues relating to the DXR machine and report that “doing the DXR is fine but we need better support with the technical aspects of DXR machines – the IT part of communicating the reading to the hospital. Phillips have not been very helpful”. They are keen to identify and manage TB better as part of their work on communicable diseases.

HMP Brixton

There is no written TB policy at Brixton and pathways are dependent on individual staff “knowing what to do”. They are not screening all receptions through the DXR machine as they report that they do not have the workforce or time to do this. They highlighted a particular problem of space and the lack of rooms in the healthcare unit. Since there are a number of additional clinics this puts limits and pressures on how much TB DXR screening can be achieved. For example, they stated that the Physiotherapist uses the same room that the DXR machine is located in on set days and a radiographer also uses the room on Wednesdays (not for TB). Therefore, they have decided to conduct any TB screening and investigations with the machine on a set day per week (Fridays) when they know there is capacity and room availability. DXRs may be carried out if someone has been highlighted through the reception health questionnaire although the DXR may be conducted over the recommended 48 hour period. The healthcare staff stated that as they are a Category C prison and that most prisoners are transferees, they should have been screened prior to arriving at Brixton.

HMP Pentonville

Similar to HMP Brixton, DXR screening occurs on a set day per week (Friday) for those highlighted as potential risk at the reception health questionnaire (which again, may have been undertaken up to a week before the set DXR day). A radiographer from the Whittington Hospital comes to HMP Pentonville and carries out the DXR on site. There are three nurses at the prison trained to use the DXR, however, it is very time consuming to train more and get them skilled to a standard where they can undertake DXR screening independent of the radiographer. They describe that nurses need to watch and conduct ten DXR with the radiographer before they are able to do them alone. However, it is not always possible to schedule staff shifts with the radiographer’s Friday visits so the training process can be a very time consuming. They also state that the “IT computer bit is confusing with all the jargon”.

HMP Thameside

HMP Thameside have a non-DH funded DXR machine which was installed when the prison first opened as part of the new build. The DXR machine is not in use for TB screening and has not been operational although staff have been trained to use this. There are currently some technical problems with the N3 connection in providing readings.

Figure 7: Results for meeting each criteria set out in recommendation 9

In addition, all establishments ask prisoners or detainees if they are taking TB medication (criteria 1) and most screen them within 48 hours of arrival (criteria 2). The majority of establishments screen all new receptions with a basic health questionnaire which is usually conducted within 48 hours and identifies those with possible signs and symptoms of TB who may require further investigations.

HMP Wandsworth however, partially met these criteria as they stated that health screening did not always happen within the 48 hours (although they did aim for this). This is a particularly pertinent issue at HMP Wandsworth since they are the largest prison in the UK and as a result, can experience a high turnover of prisoners entering the establishment. They reported that “reception health screening can sometimes take them up to 3am”. They also reported that short staffing is placing extra strain on the healthcare system.

HMP Brixton did not meet this recommendation as although they believe that they “might” screen for TB at reception, they did not necessarily use this information to inform a TB pathway. As stated previously, HMP Brixton felt that receptions were usually transferees from other prisons therefore they were unsure whether they needed to screen.

Case Example

HMP Holloway recently described two complex cases of TB; one involving a pregnant women and another woman who was multi-resistant. Both were challenging cases and were described as “learning experiences” for the healthcare staff involved. Neither case were identified at an early stage despite being screened for TB through a reception health questionnaire. A root cause analysis post-cases were conducted to learn from the cases and it was concluded that the risk factors that are recorded at reception were not always linked together to consider TB risk (e.g. the first case was a women who was a drug user and a migrant from a high incident country but she was not identified as a high risk). The second case raised some concerns around communication between hospital and prison healthcare where an initial DXR was conducted at reception which showed suspicions of TB but this was not reported back to prison healthcare teams. This miscommunication became apparent 8months later once the woman started to experience symptoms of TB and was investigated further.

As a result of these two cases and the learning opportunity undertaken, the prison healthcare team have developed some key action plans to deal with future TB risk and cases:

1. Developed a working group/MDTB team to manage TB risk and cases.
2. Improving and standardising health screening questionnaires at reception and developing templates which have local relevance and clinical input.
3. Arranged provisions for a mobile DXR unit to visit and screen prisoners once a month. This has been negotiated with the prison who are able to provide prison staff as it avoids the need to send prisoners off-site for investigations. Anecdotally, the prison healthcare reported that 4 women were identified for further investigation after 1 mobile DXR visit.

Managing active TB

Criteria description
1	Everyone with X-ray changes indicative of active TB, and those with symptoms who are awaiting X-ray, should be isolated in an individual room or cell. Prisoners and detainees should be retained on medical hold until they have: - proven smear negative and had an X-ray that does not suggest active TB or - had a negative risk assessment for multi-drug resistant (MDR)-TB and completed 2 weeks of the standard treatment regimen.
2	Prison and immigration removal centre health staff should report all suspected and confirmed TB cases to the local MDTB team within 1 working day.
3	MDTB staff should visit every confirmed TB case in a prison or immigration removal centre in their locality within 5 working days.
4	If a case of active TB is identified, the local Public Health England unit, in conjunction with the MDTB team, should plan a contact investigations exercise. They should also consider using m