Provision of harm reduction and drug treatment services in custodial settings – Findings from the European ACCESS study

Abstract Aims: The aim of the study was to survey the availability, coverage and quality of harm reduction and drug treatment services delivered to drug users in prisons across Europe. Methods: A survey was conducted between 2012 and 2013 among the 29 European countries. An electronic semistructured questionnaire was sent to the national institutions responsible for prison services, and 27 countries responded. In addition, good practice interventions for drug offenders have been collated by 15 national experts covering 15 European countries. The interventions were described and assessed as to their quality through using European monitoring centre for drugs and drug addiction (EMCDDA) standard tools for reporting and quality assessment. Findings: Drug treatment including detoxification and opioid substitution treatment (OST) is available in prisons of most European countries. However, OST is unavailable in five countries. Almost all countries provide prison-based harm reduction measures to prevent and treat infectious diseases among prisoners. Especially, testing and treatment for HIV and tuberculosis are provided, while other measures, such as the distribution of condoms or bleach, and especially needle and syringe programmes are still rare. Conclusions: Access to and coverage of OST in prisons is higher in countries with a long history of OST provision, while in countries that introduced OST more recently the scale of OST is usually lower. Access to hepatitis C treatment is often limited in prisons due to the lack of drug abstinence or a health insurance.


Introduction
The main aim of this paper is to present an overview on harm reduction and drug treatment services implemented in the national prisons in Europe and delivered to drug users. As the overview provides general findings on the availability of prison-based services, a second aim is to highlight recent custodial harm reduction programmes that meet the good practice quality level.
The provision of harm reduction measures and drug treatment in prisons is important because problem drug users are overrepresented in prison populations (Fazel, Bains, & Doll, 2006). Studies among problem drug users in the community show that between one-third and three quarters of different samples of drug users have ever been in prison (EMCDDA, 2012). According to recent data of the Council of Europe, in 2012, one quarter of prisoners served a custodial sentence that lasted between one and three years (Aebi & Delgrande, 2014). Compared to people outside of prison, prisoners have a greater risk to become infected with HIV, HCV (hepatitis C virus), and tuberculosis. Further, they are disproportionally vulnerable to suffer from poor mental health and mortality such as suicide (EMCDDA, 2012;UNODC, 2012;WHO, 2014). For opioid users, the period after prison release is associated with a high mortality due to drug-related fatal overdose (Shaw, Appleby, & Baker, 2003;Verger, Rotily, Prudhomme, & Bird, 2003).
Harm reduction measures have been implemented in prisons since the early 1990s in order to respond to the risks of acquiring HIV and/or hepatitis C via unprotected sex and the sharing of injecting equipment. Even though the provision of drug services in European prison has been scaled up over the past years -especially with regard to opioid substitution therapy (OST) -prison drug services are still not equal to those available in the community (EMCDDA, 2012). Interventions that are delivered to drug users in prisons have the following three major aims: (i) the reduction of drug use, (ii) the prevention of reoffending and (iii) the reduction of viral infections related to risk behaviour.

Drug treatment and harm reduction in prisons
Due to the high proportion of drug users among the prison population, the World Health Organization (WHO, 2007) as well as the United Nations Office on Drugs and Crime (UNODC, 2012) repeatedly emphasised the need to implement evidence-based health interventions in prisons. The effectiveness of prison-based drug treatment programmes and harm reduction measures has been investigated in a number of primary studies and reviews. Randomised controlled trials (RCT) in the United Kingdom found similar effectiveness for detoxification with methadone and buprenorphine as regards the management of withdrawal symptoms (Howells et al., 2002;Sheard et al., 2009;Wright et al., 2011). However, due to the high rate of continued opiate use after detoxification, the authors recommend OST as the first-line drug treatment in prison. The effectiveness of OST in prison has been documented in a number of RCTs, experimental studies and cohort studies from different countries, such as the United States (Kinlock, Gordon, Schwartz, Fitzgerald, & O'Grady, 2009;Kinlock et al., 2007;McKenzie et al., 2012;Wilson, Kinlock, Gordon, O'Grady, & Schwartz, 2012), Australia (Dolan et al., 2003a;Dolan et al., 2005;Kinner, Moore, Spittal, & Indig, 2013;Larney, Toson, Burns, & Dolan, 2012) and Iran (Ahmadvand, Sepehrmanesh, Sadat-Ghoreyshi, & Zahiroddin, 2009). Further, there are three reviews of OST in prison settings (Hedrich et al., 2012;Larney, 2010;Larney & Dolan, 2009). Most of the primary studies were conducted among male prisoners and in these studies no other medication than methadone was examined. The reviews also considered treatment with buprenorphine. Research found clear evidence that prison-based OST is highly effective in reducing heroin use, drug injecting and sharing of needles and syringes if the daily dose is sufficient (over 60 mg) and if the treatment duration is long enough (more than 6 months). Further, research findings were equivocal that OST has an impact on reducing criminal behaviour and re-imprisonment (Dolan, et al., 2005;Kinlock, et al., 2009;Larney, Toson, Burns, & Dolan, 2012).
To prevent opioid-dependent prisoners from dying of drug overdose shortly after prison release, the provision of takehome naloxone and respective training has been shown to be an encouraging approach. Studies demonstrate promising results for the uptake of naloxone-based overdose prevention training, the utilisation of naloxone in emergency cases and its effect to prevent fatal overdose (Barocas, Baker, Hull, Stokes, & Westergaard, 2015;Bennet & Holloway, 2012;Winter et al., 2015). Better results on the effectiveness of naloxone in post-prison overdose prevention can be expected from the pilot RTC ''N-ALIVE'' which preliminary involved 5600 drug users released from prisons in England (Farrell & Marsden, 2007).
The effectiveness of interventions aiming at the reduction of risk behaviour in prisoners has been investigated in a number of studies. For instance, prison-based needle and syringe programmes (PNSP) were reported to be effective in significantly reducing the sharing of injecting equipment among drug-injecting prisoners, and the effectiveness was confirmed for different prison settings (Dolan, Rutter, & Wodak, 2003b;Juergens, Ball, & Verster, 2009;Lines et al., 2004;Lines, Jurgens, Betteridge, & Stover, 2005;Stoever & Nelles, 2003). On the other hand, peer education or peer training was shown to have little effect on decreasing substance use and reducing sexual-related HIV risk behaviour (Braithwaite, Stephens, Treadwell, Braithwaite, & Conerly, 2005;Dolan, Bijl, & White, 2004). Psychoeducation, aiming to prevent HIV among drug users in prison, was reported to have limited impact on injection risk behaviour and an inconsistent effect on the reduction of sexual risk behaviour (Bauserman et al., 2003;Lubelczyk, Friedmann, Lemon, Stein, & Gerstein, 2002).
Various studies have addressed the effectiveness of vaccination, testing and treatment of hepatitis and HIV. Studies from Iran and the United Kingdom revealed that vaccination for hepatitis A and B is an effective measure to interrupt outbreaks of viral hepatitis in prisons, and large number of prisoners could be vaccinated in a short period of time (Asli et al., 2011;Gilbert et al., 2004;Sutton et al., 2006). The rate of testing for HCV and HIV has shown to be rather low despite comprehensive actions to motivate prisoners for confidential testing (Hickman et al., 2008;Perrett, 2011;Rosen et al, 2009;Skipper, Guy, Parkes, Roderick, & Rosenberg, 2003). Providing antiretroviral therapy for HIV and antiviral treatment for HCV in prison has been found to be feasible and effective, as the rates of treatment completion were satisfying (Farley et al., 2005;Saber-Tehrani et al., 2012). In drug-dependent prisoners, HCV treatment achieved acceptable sustained virological response (SVR) rates at the end of treatment and 6 months after treatment completion, and effectiveness of HCV treatment was also found for prisoners with HIV coinfection or co-occurring mental disorders (Allen et al., 2003;Chew, Allen, Taylor, Rich, & Feller, 2009;Lloyd et al., 2013;Maru, Bruce, Basu, & Altice, 2008;McGovern et al., 2005).
Seen as prison-based interventions have been shown to be effective, a study was conducted to survey the drug treatment programmes and harm reduction services available in national prisons across Europe (Stöver & Zurhold, 2014). The research included the collation of current best practice programmes implemented in the criminal justice system in 15 European countries. The research was part of the broader European ACCESS study 1 , which aimed to identify, research and disseminate best practice in the delivery of harm reduction within prisons. The main approaches within the ACCESS study were research, training of prison staff, experiential learning and dissemination of evidence (https://www.frankfurtuniversity.de/fachbereiche/fb4/forschung/forschungsinstitute/isff/ personen/prof-dr-heino-stoever/access.html)

Methods
For the prison survey, an electronic semistructured questionnaire was developed, covering the following issues: (a) assessment of drug use and health problems at prison entry; (b) the legal framework for health care; (c) cooperation with external agencies; (d) type and coverage of interventions provided and (e) assessment of services lacking in prison. The section on types of interventions was adopted from the UNODC ''comprehensive package'' for HIV prevention, treatment and care in prisons (UNODC, 2012). The survey aimed to cover all 27 European Union Member States (Croatia was not a Member at that time), Norway and Switzerland. Therefore, the partners involved in the ACESS study identified the national institutions responsible for prison services and nominated the respective contact person whom they most often know personally. Accordingly, the questionnaire along with information about the research purpose was initially sent by email to the contact persons in the Ministry of Justice, the Head of Prison Administration or the Ministry of Health. In case of non-response reminders were sent regularly, asking to nominate another expert who may be able to provide the requested national data. Using this procedure, 165 different individuals were contacted between February 2012 and January 2013, demonstrating that an inventory of harm reduction services in European prisons is a huge challenge.
Finally, representatives from 27 European countries completed the questionnaire, which is a response rate of 93%. Most respondents were from the prison administration or prison service in charge for health services (n ¼ 17), followed by respondents from community drug services connected to the national prison system (n ¼ 7). Another three respondents were from national health departments responsible for prison and probation services. The correctness of the information provided by the national experts was assessed by the research team, and in case of ambiguous national information, a further expert from the respective country was consulted for validation.
The survey includes 25 EU Member States, Norway and Switzerland. Despite follow-up efforts, no information was provided from Luxembourg and Malta. The data from United Kingdom and Germany are aggregated for the analysis; the United Kingdom covers England, Scotland and Northern Ireland, and Germany covers the six (of 16) responding federal states.
The second part of the research included the collation of examples of good practices in harm reduction and drug treatment targeted at drug offenders. The definition of good practice was adopted from the Exchange on Drug Demand Reduction Action (EDDRA) best practice portal. To collect good practice examples, 15 national experts were appointed and each one represented one European country. The experts were selected due to their profound knowledge of both the national prison services and the drugs problems. The expert group was heterogeneous, consisting of (a) prison-based advisors for rehabilitation (Estonia, Hungary, Lithuania, Latvia, Slovakia), medical care (Romania, Italy) and harm reduction (Scotland, Ireland); (b) staff members of community drug services (Bulgaria, Greece and Portugal); and (c) specialists in public departments for prevention (Poland), substance misuse (England/Wales) or health care (the Netherlands).
In every country, there had to be identified at least three examples of good practice that had been implemented in the criminal justice system between 2008 and 2012. Each intervention identified was described comprehensively and assessed as to its quality through completing the following standardized reporting forms: the good practice report form, the quality criteria for good practice and the quality assessment form. The quality criteria comprised a defined list of methodological questions the intervention has to be tested against. The quality assessment corresponded to the method of EDDRA for determining the quality level (http:// www.emcdda.europa.eu/themes/best-practice/examples/qualitylevels). EDDRA defines three quality levels, ranging from 1 ''promising practice'', 2 ''good practice'' and 3 ''top level practice''.
A total of 38 intervention programmes were reported by the experts, with most of them related to rehabilitation programmes (n ¼ 8) and medication-assisted treatment (n ¼ 8). Second most often were peer approaches (n ¼ 5). Twenty-four of the reported programmes achieved the good practice level, 12 programmes showed promising quality, and two programmes from England met the criteria for top level quality. For this paper, four of the 24 programmes that met the good practice level are briefly presented.
The ACCESS study did not require ethical approval. Participants gave their consent to use the data in an aggregated manner for scientific purposes. The study followed the principles of the Declaration of Helsinki.

Findings
Expert responses indicated that in most of the 27 European countries, there is an initial assessment of drug use problems at prison entry (n ¼ 23; 85%). In 15 countries, data were collected on the number of problem drug users in their national prisons. According to the data, the proportion of prisoners with drugs problems varied considerably between 8% (France) and 66% (Portugal). In most of the participating countries, an initial screening for infectious diseases such as HIV and HCV was performed (n ¼ 24; 89%). However, in Bulgaria, Hungary and Romania, neither an assessment of drugs problems nor of infectious diseases was conducted at prison entry.

Availability of drug treatment
In all participating European countries, specialised drug treatment is provided in prison, although the availability of different types of drug treatment varies considerably across countries (Figure 1). Most countries provide detoxification with opiate agonists (n ¼ 20) and the continuation of OST for those prisoners who were enrolled in community treatment before entering prison (n ¼ 22). OST has not been introduced in Bulgaria, Cyprus, Greece, Hungary, Lithuania and Slovakia as doctors are not allowed to prescribe long-term substitution treatment in prisons. While OST is officially allowed in prisons since 2001, it has not yet been implemented.
In order to assess the accessibility to prison-based drug treatment, the survey questionnaire asked for the national coverage of those services available. Coverage was defined as the proportion of prisons out of all nationwide prisons that provide a service. However, only in half of the countries, the coverage of available drug treatment programmes was specified. In four of the 12 reporting countries, detoxification was available in almost all prisons (Austria, Norway, Slovenia, United Kingdom), while in another four countries, the coverage was below 50% (Czech Republic, Estonia, Poland, Romania). Furthermore, in seven out of 17 reporting countries, the continuation of OST was possible in all prisons (e.g. in Austria, Ireland, Norway, Slovenia, United Kingdom). In four countries -Czech Republic, Romania, Latvia and Poland -continued OST was accessible in less than half of the existing prisons. The provision of naloxone was not very common in the European prison system. Only in England/Wales and Scotland, naloxone was offered in all prisons, and from Spain, it was reported that naloxone was available in the prisons in Madrid.

Availability of harm reduction measures to combat infectious diseases
All participating 27 European countries delivered testing and treatment for HIV/Aids and treatment for tuberculosis to prisoners. All but two countries also offered testing and treatment for hepatitis C in prisons ( Figure 2). However, it remains unclear how many of the infected prisoners received treatment. Post-exposure prophylaxis (PEP) after needlestick injuries was available in all but three European countries. The data did not specify if PEP was only provided to prison staff or as well to prisoners.
The coverage of testing and treatment for infectious diseases was rather high in many countries. Treatment for HIV/AIDS was available in the entire prison system in 13 of 18 reporting countries and treatment for HCV was available in all prisons in 10 of 16 reporting countries. Treatment for tuberculosis was provided in all prisons in 14 of 17 countries.
A limited availability of testing and/or treatment for infectious diseases demonstrated three countries -Estonia, Greece and Lithuania.
Prevention measures, such as the vaccination for hepatitis A and B, have been implemented in the majority of the 27 European countries (Figure 3). On the other hand, condoms or bleach were rarely distributed in European prisons (countries: n ¼ 18 and n ¼ 16). Only in four of nine countries, which specified the coverage, condoms were distributed in all national prisons (Poland, Romania, Slovenia and United Kingdom). The distribution of bleach to clean injecting equipment was only available in France and Scotland on a nationwide level. Furthermore, PNSP were still rare despite the evidence that they reduce the spread of infectious diseases among people who inject drugs. PSNP has been implemented in a few prisons in Spain, Switzerland, Portugal, Germany and Romania. In Romania, PNSP is provided in eight of the 44 national penal institutions, and in Germany, PNSP is available only in one out of 190 prisons.
In general, harm reduction measures aiming at the reduction of infectious diseases have been implemented in many prisons of the European countries. A high coverage of  several harm reduction measures existed in England, Scotland, Spain and France and to a slightly lower degree in Ireland and Czech Republic.

Availability of information and education related to prevention
Information and education aim at raising the awareness of prisoners for risks related to sexually transmitted infections, HIV, viral hepatitis and tuberculosis. With the exception of Sweden, education for the prevention of drug-and sex-related infections was available in prisons of all participating European countries. For instance, in the Netherlands, a new programme on health education for prisoners, including education on drugs, drug use, tattooing, and piercings was established in 2013. Education to prevent overdoses after prison release was reported to be available in 24 countries. In many European countries, measures, such as education on health risks, information on the prevention of infectious diseases, and overdose prevention, were available throughout the whole prison system. Full coverage of these interventions was reported from Cyprus, Ireland, Romania, Latvia, Lithuania, Poland, Slovakia, Slovenia and United Kingdom.

Examples of good practice in drug treatment and harm reduction
From the 24 programmes, which were reported by the experts and that achieved the good practice level, four harm reduction programmes have been selected for this paper. These programmes had been selected as they represent effective and innovative approaches to address main health problems of prisoners; the post-release overdose deaths, infectious diseases and risk behaviours in prisons.
An example of good practice in drug treatment is the Scottish national naloxone programme. Since November 2010, all prisons in Scotland provide ''take-home naloxone'' for individuals at risk of opioid overdose after prison release. To facilitate the delivery of the programme, a national coordinator and training team was established. This team assists all Health Boards across Scotland to embed the naloxone programmes in their communities and in the Scottish Prison Service. Prisoners at risk of overdose receive a naloxone kit after training on how to administer it safely and quickly. Between 2011 and 2012, a total of 715 naloxone kits were supplied by prisons.
Two examples of good practice in combating infectious diseases are related to the identification of tuberculosis (TB) as this bacterial disease affects a considerable number of prisoners. In England, diagnosis of TB is available to all prisoners where it is recommended according to NICE guidelines. Eight large prisons serving populations with a high prevalence of TB were provided with new digital X-ray machines, which are linked to a national centre of excellence for reading and reporting the images. However, all prisons in England have either direct access (through prison-based X-ray machines) or indirect access (via NHS acute hospital trusts) to X-rays for the diagnosis of TB. In the Netherlands, mobile X-ray units are weekly available in all penal institutions since 2009. After a pilot testing of a new TB guideline (2010) in three prisons, the TB screening of newly detained offenders through mobile X-rays units was modified. Previously, each detainee was screened for TB upon reception, since 2011 X-rays for TB diagnosis are used for those detainees who are assessed by a nurse to at risk of TB. In case of a positive X-ray, the prison doctor makes further medical examinations in order to verify the result and to initiate treatment with antibiotics. Between 2008 and 2011, almost 12,000 prisoners were diagnosed using X-ray. Since the modification of the TB diagnosis procedure in 2011, there has been a 45% reduction of indicated X-rays.
The last example of good practice is related to a peer education training that has been implemented in Portugal in the Guarda prison in November 2010. Among the prison population of Guarda, with a capacity of 175 prisoners, there was a high prevalence of risk behaviours (drug injecting, piercings, tattoos), drug use and comorbidity, which led to the peer approach. The aim of this approach was to train prisoners as peers, who promote knowledge to prisoners on how to prevent infectious diseases. Ten trainers provided a comprehensive and certified peer education training that includes 15 different modules performed in 185 hours. The training was evaluated through tests of knowledge improvement, role-play and more. Peers, who successfully completed all modules, received a certification that enabled them to work in harm reduction outreach teams after their release. Peer activities in Figure 3. Harm reduction provided in national prisons. DOI: 10.3109/09687637.2015.1112363 prison comprised group sessions and individual contacts to prisoners in order to improve their knowledge on drug-and sex-related harms. All peer activities were monitored by the social service team, and according to the monitoring results, a high number of prisoners has been reached by the peer educators (n ¼ 185 in a two years period).

Conclusions
Within the ACCESS project, a prison survey on available drug treatment harm reduction measures was conducted between 2012 and 2013. A semistructured questionnaire was addressed to the national institutions responsible for health care in prisons, such as the Ministry of Justice or the National Prison Service. The response rate to the survey was very high as 27 of the 29 European countries submitted the requested national data.
The findings from the survey indicate that the majority of the European countries provide detoxification (n ¼ 20), allow to continue community OST when entering prison (n ¼ 22), and initiate OST at prison entry (n ¼ 17). Appropriate provision of prison-based OST had been demonstrated to be effective in reducing opioid use, injecting and sharing of injecting equipment. Therefore, it is highly recommended to introduce OST in prison in those countries which still do not provide OST (such as Cyprus, Greece, Hungary, Lithuania).
The availability does not reflect if drug treatment is provided in all national prisons or only in a limited number of prisons. The survey questionnaire examined the coverage of each of the available service; coverage was defined as the number of penal institutions providing a service. Based on this definition, the experts of countries such as Austria, France, United Kingdom and Ireland reported a high coverage of OST in their penal institutions. In the prison report of the European Centre for Disease Prevention and Control (ECDC) huge differences in the scale of prison-based OST were found between the European countries (Fazel et al., 2006). The differences were explained with the tradition in providing this treatment. Countries with a long history in providing OST in prisons generally have a higher availability and coverage of this treatment. If countries have introduced OST into their prisons more recently (Czech Republic (2007), Estonia (2008), and Romania (2008)), the scale of OST is usually lower 2 . However, the scale of OST does not allow drawing conclusions as to the percentage of the opioid-dependent prisoners receiving this treatment.
The availability of harm reduction measures to prevent mortality after release and to reduce infectious diseases among prisoners varies considerably in Europe. Seventeen countries reported to provide naloxone to drug-dependent prisoners upon release in order to prevent fatal overdose. In most countries, only few prisons distribute naloxone, whereas in England and Scotland, a naloxone programme has been implemented in all prisons. All participating 27 countries provide testing and treatment for HIV/AIDS, and treatment for tuberculosis in prison. With the exception of Latvia and Bulgaria, all countries also offer testing and treatment for hepatitis C in their prisons. Some countries reported restrictions in accessing treatment for hepatitis C. For instance, in Finland and Ireland, HCV treatment is provided under the condition that drug users are either stable on methadone or achieved abstinence for a period of time. The ECDC report (2013) mentioned the lack of a health insurance as an important barrier in providing HCV treatment. Prisoners often have no health insurance that is required for HCV treatment in countries such as Bulgaria and Czech Republic. Even though HCV treatment is generally available in European prisons, access to this treatment might be rather limited. As treatment for HCV in prison has shown to be effective in curing an infection with hepatitis C and in preventing further transmission, access to treatment needs to be scaled up in many European countries, especially in Lithuania and Bulgaria.
With regard to other harm reduction measures, the findings from the survey suggest some shortcomings. The provision of condoms as a measure to reduce transmission of infectious diseases through promoting safer sex varies notably across the European countries. While in 18 countries, there is a procedure for distributing condoms, condoms were unavailable in nine countries. However, countries offering condoms to prisoners have different distribution procedures in place. Most often condoms are only available on request at medical services or the prison health service. This method was reported from Slovenia, Finland, Belgium, Portugal, the Netherlands, France and Poland. Requests for condoms are rare as this method is lacking confidentiality and anonymity. On the other hand, in Switzerland and Belgium, condoms are distributed discretely through machines, allowing a lowthreshold access. Prison-based needle and syringe programmes (PNSP) are still rarely implemented in European prisons. At present, PNSP is available in Spain and Switzerland, and in few prisons, in Romania, Portugal, Luxembourg and Germany. Despite evidence for the feasibility and effectiveness of PNSP, there is a controversial discussion about this harm reduction measure in most of the European countries. The responses to the survey demonstrated that a variety of arguments hamper the introduction of PNSP. Accordingly, the experts mentioned rational arguments (low proportion of drug injecting), security issues or they neglected that drugs are available in prison.
In conclusion, the findings from the prison survey show that most European countries provide the main important measures to treat drug addiction and to prevent infectious diseases. However, access to health care in prison and equity of services to those being efficient in community is a human right for prisoners (Michel, Carrieri, & Wodak, 2008;WHO, 2014). In this respect, improvements in availability and access are still necessary in order to make evidence-based drug treatment and harm reduction available and to ensure access to these services to all prisoners in need. Some efforts towards a more efficient harm reduction policy in European prisons reveal the 38 reported interventions that were recently implemented in the penal system and which demonstrated good practice or at least promising practice.
The findings from the survey have several limitations. First of all, the data rely completely on the information provided by the respondents, even though they were national experts in the field of prison health services. Further, in large countries with many prisons, their knowledge on availability and coverage of health care services might be limited. There was also a different understanding of the questions on coverage among the respondents. Partly, the respondents indicated in how many prisons a specific intervention is available, and partly, it was reported if an intervention is available to all prisoners. However, the survey mainly reflects the general availability of core interventions rather than their complete implementation or accessibility.