Service Evaluation of Scotland's Take-Home Naloxone Programme

Appendix 2: Rapid Literature Review

Naloxone training and use - a review of evidence

Purpose and scope

1. The purpose of this brief evidence review was to set the context for Blake Stevenson's service evaluation of Scotland's National Take-Home Naloxone (THN) programme, and to guide the development of the research framework for the stage 2 fieldwork.

2. A recent review by McAuley et al (2012) sets out a detailed description of the establishment of the Scottish National THN programme, detailing key research, advocacy and policy contributions that facilitated its development. The evaluation of the THN Demonstration Project in Wales (Bennett & Holloway, 2011) outlines the key learning, outcomes and process issues from the pilot project as well as offering a review of 10 evaluations of naloxone distribution programmes (six in US, two in England and two in Scotland - Lanarkshire & Glasgow). Given that these studies have recently been undertaken and that each involve a review of key evidence sources in relation to naloxone training and distribution programmes, this evidence review does not duplicate this activity, but summarises some of the key points of interest and examples cited within these publications, as well as drawing on other grey literature on the Scottish National THN Programme.

Setting the Scottish Programme in context

Prevalence and nature of drug-related deaths in Scotland

3. Drug-related death is a major public health problem across the globe, with rates in Scotland currently higher than any other UK region and amongst the highest in Europe (McAuley et al, 2012). In 2012, 581 drug-related deaths were registered in Scotland. This was three (0.5 per cent) fewer than in 2011. This was the second highest number ever recorded, and 199 (52 per cent) more than in 2002.

4. Scotland's current national drugs strategy^[34] (2008) focuses on recovery from problem drug use but also stresses the specific need for action to prevent DRDs. The strategy highlights that it is possible to identify people more likely to die from their drug-use, which presents an opportunity for preventative action to reduce DRD incidence through activities such as increased general health care, the provision of routine liver function tests, and a range of education and awareness raising measures. Giving people the confidence to know when to intervene, and what to look for and do in the case of an overdose is highlighted as a possible way of preventing DRDs, as is the training and the provision of relevant information to staff and service users, family and friends.

5. A national DRD database was established in 2009 to increase knowledge and understanding in respect of DRDs. Findings from it have confirmed earlier research (Zador et al; 2005) that those most vulnerable to DRDs are male, live in deprived areas, and are aged 25-44. Furthermore, it has established that the majority of DRDs are 'accidental', involve opioids, are witnessed (highlighting potential for intervention) and two thirds involving someone in contact with a drug treatment service prior to their death. According to the recently published report 'National Drug Related Death Database (Scotland) Report: Analysis of Deaths occurring in 2012'^[35], around half of the cohort (47%) had been in prison at some point in their lives prior to death. Over one in ten (12%) had spent time in prison in the six months prior to death, a decrease compared to 2011 (18%).

Development of a Scottish National Naloxone Programme

6. Following successful pilots in NHS Greater Glasgow and Clyde, NHS Lanarkshire (2007) and in the Inverness area of NHS Highland (2009), in November 2010 Scotland established a National THN programme aimed at reversing the upward trend in DRDs.

7. Naloxone hydrochloride is an opioid antagonist which reverses the effects of respiratory depression caused by opioid overdose. It has no obvious potential for abuse and a strong safety profile. It is highly effective, but short acting and therefore multiple doses may be required to fully reverse the effect of opioid overdose. It is only effective in tacking opioid overdose and has no effect on other drugs that have potential for overdose. However in a polydrug overdose situation it removes the opioid element and reduces the potential for fatal overdose, providing additional time for emergency services to address the other toxicities. (Lenton & Hargreaves, 2000).

8. In the UK, naloxone is licensed for intravenous, intramuscular and subcutaneous administration. Scotland's national THN programme currently supplies those at-risk of opioid overdose with naloxone for intra muscular (IM) injection in line with the majority of similar programmes across the globe explored through this review. IM is the favoured route for peer administration because of the ease of site identification for injection and the relatively slow onset compared to intravenous injection (McAuley et al 2012).

9. In 2005, a change in the legal status of naloxone permitted any member of the public to administer it legally in an emergency situation. This facilitated the implementation of naloxone distribution and training programmes in a way that had not been possible previously and paved the way for the first pilots in Scotland out of which the national programme has developed.

10. The Scottish National Programme allows for the distribution of naloxone to those at risk of opioid overdose including prisoners on liberation (following receipt of specialist training on its use). Use of a PGD supports this as it allows the supply of a prescription only medicine (POM), in this case naloxone, to be provided without the need for a prescription written by a doctor. Training is also available to family, friends, carers, and others likely to be in the vicinity of a person at risk if an overdose occurs (e.g. healthcare addictions staff). Kits are supplied in community health and care settings as well as in prisons at the point of release. Supplies of Prescription Only Medicines are restricted and can only be made to named patients. .Although naloxone is a POM, the Lord Advocate issued Guidelines (Scottish Government, 2011b) in 2011 which enables services in contact with people with problem drug use who are at risk of opioid overdose (e.g. those working in homeless hostels, needle exchanges etc) to receive supplies and to hold stocks of naloxone within their service for use in an emergency (not for onward distribution). The Lord Advocate's guidance provides immunity from prosecution for staff making a supply of a POM to services as this is outside the normal legal requirements of a POM supply.

11. The Programme is Scottish Government-funded, managed by a National Coordinator (based at the Scottish Drugs Forum) and overseen by a National Naloxone Advisory Group comprising experts from statutory and third sectors. Other key programme elements include: a national monitoring and evaluation programme based at NHS ISD Scotland, a naloxone Peer Educator initiative^[36], national training and information resources and guidelines, and specific support to Health Boards to deliver the programme in prisons. There are currently local programmes supported and coordinated by 29 of Scotland's 30 Alcohol and Drug Partnerships (ADPs); 13^[37] of the 14 territorial Health Boards are participating, and as of June 2011, so are all 16 Scottish prisons^[38].

Reflection on the nature of published evidence on the effectiveness of naloxone distribution programmes

12. The evaluation report from the Welsh Demonstration project (Bennett & Holloway, 2011) reflects on the fact that the number, as well as the quality of studies on the effectiveness of THN and similar distribution programmes is limited^[39]. Many of the studies are characterised by the absence of control groups, small sample sizes and low follow up rates.

13. Information Services Division (ISD) is currently gathering detailed monitoring data from across Scotland using a national dataset (agreed with the National Naloxone Advisory Group (NNAG)) on THN in Scotland). The data is published annually and will add to the evidence base around effectiveness of the Scottish THN programme. The ISD dataset evidences the reach of the programme (across prisons and community settings) in terms of:

• Number of kits issued (in community/prisons)
• Participation across NHS Boards/Prisons
• Whom kits are supplied to (age, gender, person at risk)
• First supply/repeat supply
• Kits supplied to 'persons at risk' - gender and age of recipient

14. ISD are also measuring the impact of increased naloxone availability on the number of (opioid) Drug-Related Deaths (DRDs) in Scotland, including the number and percentage of these occurring within four weeks of prison release. A baseline has been established using calendar years 2007-2010 and performance against this baseline will be measured for calendar years 2011-2015.

15. A Medical Research Council funded NALoxone InVEstigation Randomised Controlled Trial (N-ALIVE) ^[40] (in England only) began in late 2011 and is likely to address some of the methodological limitations associated with the THN literature highlighted above. The project is a large prison-based randomized controlled trial, designed to test the effectiveness of giving naloxone-on-release to prisoners with history of heroin use to prevent fatal opioid overdoses. The project has two stages: the pilot randomized trial (involving 5,600 participants) and the subsequent main randomised trial. A total of 56,000 participants are planned to be recruited in total during the study. Treatment groups in the trial will be provided with overdose prevention training and naloxone, and the control groups will be offered the training alone.

16. The principal questions^[41] the study is seeking to address are:

Pilot trial:

• What happens to the Naloxone and the participants, in terms of heroin use and overdoses (witnessed or experienced) within 4 and 12 weeks after release?
• Do 75% of prisoners assigned to Naloxone carry it with them in the first 4 weeks after release?
• Do prisons and prisoners participate in the numbers expected and required for the main trial?
• Do the N-ALIVE procedures work well logistically in the National Offender Management Service, or will they need to be changed for the main trial?
• If changes are necessary, what needs to be done?

Main trial

• Does giving Naloxone on release to prisoners with a history of heroin injection reduce heroin overdose deaths by 28% in the first 12 weeks after release?

Summary of outcome findings relating to naloxone training and use

17. To date, evidence from the studies reviewed demonstrates positive and encouraging outcome findings with regard to naloxone distribution programmes, although there are some methodological weaknesses (highlighted above) and none of the findings yet come from Randomised Control Tests. The main outcomes findings in relation to THN (based on Bennett & Holloway's summary (2011, p11)) are explored below under the following headings:

a. Impact on knowledge, skills and behaviour following training

b. Number of naloxone kits administered

c. Number of naloxone Kits used

d. The number of lives saved

e. Other harm-reduction outcomes and unintended outcomes

Impact on knowledge, skills and behaviour following training

18. Across the studies reviewed, naloxone training tends to take the format of group sessions where participants learn skills to prevent, recognise and respond to opioid overdose, including calling for emergency services, performing CPR and resuscitation techniques, as well as administering naloxone. Sometimes two facilitators deliver the training - this has been found to be helpful for managing behaviour and larger numbers of participants. Training is a pre-requisite for receiving a kit, and target audiences include people with problem drug use alongside a buddy/carer/friend/family member.

19. Naloxone training has shown to result in changes in knowledge and behaviour of participants including increases in knowledge about preventing, recognising and responding to an overdose, as well as increased knowledge about how to use naloxone and increased willingness/confidence to do so. Naloxone training can also lead to greater harm-reduction knowledge and practices. The Welsh evaluation identifies a positive correlation between the measured strength of programme input and the strength of the programme outcomes in respect of skills, knowledge and confidence. (Bennett & Holloway, 2011). Examples from the evidence review of the impact of training on knowledge, skills and behaviour evidence include:

• trained participants being more able to recognise opioid overdose incidents accurately compared to non-trained participants;
• trained participants reporting improved knowledge and confidence in managing overdose situations using naloxone
• trained participants being able to manage their own naloxone supply responsibly
• myths about how to treat overdoses dispelled
• increases in confidence and self-esteem among trained participants
• trained participants reporting increased skills in use of life-saving techniques including CPR, putting someone into a recovery position, resuscitation
• trained participants reporting having the tools and confidence to save lives

The number of kits supplied

20. As a key aim of naloxone distribution programmes, the number of kits supplied is measured by many of the studies reviewed as an indicator of programme reach. McAuley et al (2012) suggests 'reach' of two Scottish pilots was better estimated in terms of the numbers of people with problem drug use supplied.

21. The evidence suggests a number of challenges around recruitment of clients to naloxone training and distribution programmes which can impact on the number of kits supplied. For example, the Welsh demonstrator project evaluation identifies few problems with recruitment in the prison service as all prisoners were told about programme as part of induction. However, they experienced greater problems recruiting clients in community settings with agencies having to be very proactive to ensure good throughput of clients (Bennett & Holloway, 2011). Although the National Treatment Agency for Substance Misuse (NTA) naloxone Carer pilot (2011) found the opposite, in that many prisoners were refusing the kit on liberation.

22. Recruitment methods used include signing people up at initial assessment, advertising in needle exchanges, spreading the word through outreach workers, recruiting directly through large agencies, and making naloxone training compulsory for any person on the agency prescribing programme. The following suggestions were to improve recruitment: peer-led training, improving advertising, paying incentives to attend training, expanding the number of training outlets, and shortening the training.

23. Views from a small number of service users participating in the Welsh evaluation suggest recruitment could be an issue for the following reasons:

• Fear among people with problem drug use that naloxone needs to be injected intravenously;
• Cost - incentives were suggested as a way to improve recruitment (one participant suggested they felt even £5 could make a difference); and
• Lack of knowledge about the programme.

The number of kits used

24. 'Number of kits used' is used as an indicator of effectiveness of naloxone distribution programmes in many of the studies reviewed. In most cases administration of naloxone is reported as trouble free, without adverse effects, and in nearly all cases the casualty survives. The evidence (Bennett & Holloway, 2011) also offers informative findings on the circumstances of naloxone use, for example:

• all overdose occurred in the company of someone else and in most cases naloxone was administered by a friend or relative
• other life-saving actions were taken alongside administration of naloxone
• in most cases the recovery position is used and an ambulance called
• service users seem comfortable with the prospect of injecting naloxone

The number of lives saved

25. It is difficult to determine conclusively the impact of THN programmes on DRDs for a number of reasons, including the fact there are currently no sizeable cohort studies that look at comparative survival rates for use/non-use of naloxone; and it is unknown what proportion of overdose events where naloxone was administered were potentially fatal or potentially recoverable without intervention.

26. Overall the literature suggests that overdose casualties nearly always survive if administered naloxone, concluding that naloxone therefore saves lives. The Welsh evaluation points out that "research is less clear about whether alternative actions taken at this first stage could be equally life-saving" (Bennett & Holloway, 2011) but concludes that evidence to date supports the continuation of developing and implementing methods for wider dissemination of naloxone.

Other harm-reduction outcomes/unintended outcomes

27. The evidence review highlights a number of less commonly cited and unintended outcomes resulting from naloxone training and distribution programmes. Including:

• Impact on access to services: The Welsh evaluation (Bennett & Holloway, 2011) identified the potential harm reduction benefits that might occur by bringing problem people with problem drug use into contact with treatment agencies. The evaluation showed that almost one-third (29%) of users recruited for training who responded to a survey were not currently in contact with any treatment agency - it is possible that at least some of these maintained contact with the agency.
• Impact on access to hepatitis/HIV testing: THN data collection in Wales involved collection of data about whether problem people with problem drug use had ever had a hepatitis B or C, or HIV test and whether they would like one. Just under 20% said they had never had a Hep B/C test, and over half (52%) said they would like one; just over 20% had never had a HIV test and just under half (48%) said they would like one.
• Participants re-evaluate own heroin use: For example, one respondent in the Welsh evaluation said the training had encouraged them to be more responsible in their own heroin use.
• Potential adverse effects of administering naloxone: While the evidence to date suggest that naloxone has a high safety profile, concern was raised by some participants in the Welsh evaluation study that multiple doses may result in major withdrawal. Other concerns raised in the evidence reviewed by McAuley et al (2012) were speculations that the perceived safety net may encourage increased drug use and potentially increase the risk of overdose (Ashworth and Kidd, 2001), and concerns over whether ambulances may still be called if naloxone appears to have successfully resuscitated victims (Sporer, 2003).

Summary of process issues raised by the evidence review

28. In addition to highlighting the identified outcomes of naloxone programmes, the evidence review raised a number of issues in relation to the process of delivering a THN programme that informed the design of the research tools used in the stage two case study research. These include:

Issues to explore with stakeholders

• What are the experiences of THN implementation in Scotland at a local level, and how do the national protocol, training resources and guidelines support local delivery of the programme?
• What is the extent of partnership working in Scotland and how does it assist the programme? Which agencies are involved, and how are they involved in different areas? In particular, how, if at all, are General Practitioners, ambulance and police services involved?
• What is the Scottish experience of recruiting programme participants from both community and prison settings? What approaches are being taken, what is most effective, and what are the challenges?
• How is training delivered and what works well/less well about this? How effective is the cascading model of training (training for trainers)?
• How do the experiences of THN implementation compare across community settings and prison?
• How is momentum for THN programme maintained locally and nationally?
• Are THN trainees still using other harm-reduction/life-saving methods?
• What are the possible impacts of naloxone ending up in the hands of those who haven't been trained?
• What are perceptions in Scotland in relation to the current method for administration (i.e. by injection)?
• Are concerns about inappropriate use of naloxone prevalent?
• How do local stakeholders feel about the level of resource available in Scotland to support local THN programmes?

Issues to explore with service users

• How do Scottish participants feel about the kit (including shape, size and appearance)?
• Are those supplied with kits carrying them regularly? Where are kits stored? Is there any reluctance to carry kits? Why/why not?

• How do participants feel about the content, length, time, location, delivery of training and size of session group?

References

Ashworth, A.J., & Kidd., A. (2001). Take-home naloxone for opioid addicts. Apparent advantages may be balanced by hidden harms. British Medical Journal, 323, 935.

Bennet, T., & Holloway., K. (2011) Evaluation of the Take-Home Naloxone Demonstration Project

ISD (2013). National Naloxone Programme Scotland - naloxone kits issued in 2012/13

Lenton, S., & Hargreaves, K. (2000). Should we conduct a trial of distributing naloxone to heroin users for peer administration to prevent fatal overdose? Medical Journal of Australia, 173, 260-263.

McAuley et al, 2013, From evidence to policy: The Scottish national naloxone programme

McAuley, A., & Best, D. (2012). A quantitative exploration of risk factors associated with drug-related deaths involving heroin, alcohol or methadone in the West of Scotland. Addiction Research and Theory, 20, 153-161.

McAuley, A., Lindsay, G., Woods, M., & Louttit, D. (2010). Responsible Management and use of personal take-home naloxone supply: A pilot project. Drugs: Education, Prevention and Policy, 4, 388-399.

National Treatment Agency for Substance Misuse. (2011). The NTA overdose and naloxone training programme for families and carers

Scottish Government (2011b). Lord Advocate's guidelines on allowing the supply of naloxone to extend to staff working for services in contact with people at risk of opioid overdoses.

Scottish Government (2008). The Road to Recovery: A New Approach to Tackling Scotland's Drug Problem

Sporer, K.A. (2003). Strategies for preventing heroin overdose. British Medical Journal, 326, 442-444.

Zador, D., Kidd, B., Hutchinson, S., Taylor, A., Hickman, M.,Fahey, T., Baldacchino, A. (2005). National Investigation into Drug-Related Deaths in Scotland, 2003. Edinburgh: Scottish Executive.