National consensus statement on opioid agonist treatment in custodial settings

In Australia, more than 40 000 people are incarcerated on any given day.1 Among incarcerated people, the prevalence of substance use and substance dependence is high — in large part due to the criminalisation of drug use.2 In Australia, surveys indicate almost two‐thirds of prison entrants had used illicit drugs in the previous year, over 50% of people who are incarcerated have a substance dependence, and over 30% have an opioid dependence.2,3,4

Opioid agonist treatment (OAT) is an effective and evidence‐based treatment for opioid dependence.5 Access to and retention in OAT reduces drug‐related harms, including blood‐borne virus acquisition and overdose, in prison and on release.6 However, access to and uptake of OAT in prisons is suboptimal and inconsistent.7 Access to OAT in custodial settings varies across Australia; custodial health care is overseen by jurisdictional authorities and delivered through a diverse array of both public and private health care providers. Barriers to providing OAT in custodial settings include jurisdictional or institutional policies that restrict provision of OAT, limited capacity or resources for provision of OAT, and negative societal attitudes to OAT.8 Access to health care in short term custody settings, including police cells or watch houses, can be particularly challenging given lack of resources for health care delivery, reliance on police staff to screen or identify health issues, and difficulty accessing necessary health information from community health services.9

The National Prisons Addiction Medicine Network was convened to address a gap in the national policy landscape relating to the provision of evidence‐based best practice medical care for incarcerated people with substance dependence, including OAT. The network comprises clinical, consumer and public health stakeholders from a range of jurisdictions with relevant experience in providing addiction and broader health services to incarcerated people.

This consensus statement aims to improve quality, consistency and continuity of OAT for people who are incarcerated in Australia by promoting a nationally coordinated and evidence‐based approach to OAT provision and identifying targets against which to monitor progress. In this statement, we adhere to the principles outlined in Rule 24 of the United Nations Standard Minimum Rules for the Treatment of Prisoners (the Mandela Rules) stipulating that prisoners are entitled to medical care that is equivalent to that which they could access in the community.10 This statement has been developed for application to all custodial settings, including adult and juvenile prisons, remand centres and police cells (or watch houses). The objectives are:

• to present a critical analysis of the evidence supporting the provision of OAT in custodial settings; and
• to develop consensus recommendations for the provision of OAT in custodial settings.

Methods

Expert panel and scope

TN conceived this project and the steering committee was established over the following months to support implementation. An expert panel was convened to represent a broad spectrum of expertise in the fields of addiction medicine and custodial health. (Box 1) An initial expert panel (KL, EM, MS, TT, CW, SA, BD, JH, PT) was formed from interested custodial health professionals attending the 2023 Health Care in Secure Settings meeting (https://jhfmhn.eventsair.com/hcss23/). The steering committee consulted with the initial expert panel on the scope of the consensus statement and composition of the expert panel (online meeting 1, 6 Nov 2023). Subsequently, further expert panel members were identified through professional connections and network recommendations to ensure inclusion of the desired range of expertise or experience. The final expert panel is comprised of 18 clinical (including medical, nursing and pharmacy), consumer and public health stakeholders representing all Australian jurisdictions (Supporting Information, appendix 1).

Literature review

JCh searched the published literature from January 2002 to November 2023. We identified relevant articles by searching PubMed using the key words “methadone” or “buprenorphine” or “opioid substitution treatment” and “prisons” or “prisoners” or “correctional facilities”. The search strategy is included in the Supporting Information, appendix 2. Expert panel members nominated additional relevant articles. We reviewed original research in English, including systematic reviews, randomised controlled trials observational studies and qualitative studies, relating to the provision of OAT in prisons, including transition from incarceration to the community.

Modified Delphi method

The steering committee drew on the literature identified in the reviews to develop a draft set of recommendations, which were disseminated to the expert panel alongside a summary of the literature. The expert panel provided input on the draft recommendations via an online meeting (online meeting 2, 8 Feb 2024) and email feedback (response period from 8 Feb to 12 Mar 2024). The recommendations were further refined through two rounds of anonymous surveys (survey 1 in May 2024, survey 2 in July 2024), developed and distributed using the Qualtrics XM platform. Both surveys contained 19 recommendations and response options to indicate agreement (agree v disagree). In the first survey, comments for each recommendation were collected through a free‐text entry option. The surveys were piloted among the steering committee before dissemination to the expert panel. Consensus was defined a priori as a greater than 80% agreement from respondents. Recommendations that reached consensus were included in the final document. The survey responses were collated to categorise each recommendation based on the level of agreement, in which “U” denotes unanimous (100%) agreement, “A” 90–99% agreement, and “B” 80–89% agreement. The methods were based on recently published consensus statements.11,12,13

This study was not prospectively registered. This consensus statement has been endorsed by the organisations listed in the Supporting Information, appendix 3. We reported this study in accordance with the Accurate Consensus Reporting Document (ACCORD) guideline (https://www.ismpp.org/accord) for consensus methods in biomedicine (Supporting Information, appendix 4).

Delphi study results

Among a national multidisciplinary expert panel, the Delphi process yielded 19 recommendations across five domains for the provision of OAT in custodial settings in Australia. In the first survey, consensus was achieved for all but one of the 19 recommendations — nine recommendations achieved unanimous agreement, seven recommendations reached 90–99% agreement, and two recommendations reached 80–89% agreement (Supporting Information, appendix 5). In the second and final survey, 14 recommendations achieved unanimous agreement (U), four recommendations reached 90–99% agreement (A), and one recommendation reached 80–89% agreement (B) (Box 2).

Consensus recommendations and supporting literature

The consensus recommendations are shown in Box 2. A summary of the literature relevant to the recommendations are presented below, organised within the five domains: induction or continuation of OAT, OAT options and administration, transition of care to the community, special populations, organisational support.

Induction or continuation of OAT (R1.1–1.5)

The provision of OAT is associated with a range of positive outcomes for individuals with opioid dependence both during incarceration and in the high risk period immediately after release. There is consistent evidence for significant reductions in illicit opioid use, injecting drug use and syringe sharing during imprisonment from systematic reviews of trials and observational studies.14,15 Systematic reviews of trials indicate provision of OAT in prison, particularly methadone, improves post‐release outcomes, including decreasing mortality, increasing engagement with community‐based treatment, reducing opioid use, and reducing injecting drug use.14,16,17

Given the high prevalence of opioid dependence and opioid‐related harms among incarcerated people and the availability of an effective treatment, we recommend screening for opioid dependence among all people entering custodial settings. Evidence suggests that screening for substance dependence is feasible and reliable in the prison setting.18,19,20 Screening tools that have been evaluated in the prison setting include the Alcohol, Smoking and Substance Involvement Screening Test (ASSIST) and the Simple Screening Instrument.18,19 Providing timely access to adequate doses of OAT is important to prevent and manage opioid withdrawal. Importantly, persistent vomiting and diarrhoea can lead to severe fluid loss and electrolyte abnormalities that result in haemodynamic instability and, in rare instances, death.21 Delays in accessing OAT in custodial settings may also affect subsequent uptake of OAT.7

In addition to screening on entry, individuals should be provided with the opportunity to engage with treatment throughout their period of incarceration. Evidence indicates that, despite being prohibited, illicit drug use is common in prison settings.22 Furthermore, for individuals with a history of injecting drug use, rapid resumption of injecting drug use following prison release is common.23 This resumption in drug use is associated with a high risk of mortality — attributable primarily to opioid overdose — particularly in the first two weeks following release.24,25 Evidence supports initiation of OAT in custodial settings rather than initiation at or after release.14,26 In addition to evidence for positive outcomes during imprisonment, initiation in prison improves treatment retention and reduces illicit opioid use after release compared with initiation at release or counselling only.27

OAT options and administration (R2.1–2.3)

Evidence on the relative benefits of different OAT options in custodial settings is limited. Three small randomised controlled trials directly compared sublingual buprenorphine to methadone in prison. They reported improved treatment retention and fewer side effects with sublingual buprenorphine, but comparable clinical effectiveness in reducing illicit opioid use.28 Preliminary evidence for the relatively newer long‐acting buprenorphine depot indicates that it is safe, acceptable and lower cost compared with other treatment options in prison.29

We support using a patient‐centred approach that allows choice of OAT in custody. Person‐centred care and shared decision making are recommended in standards and guidelines, both locally and internationally.30,31 Although there is limited research on how patient preference affects outcomes of OAT treatment for incarcerated people, systematic reviews from the broader mental health literature indicate incorporating patient preference improves treatment engagement and patient satisfaction, with mixed results (positive or no effect) for treatment outcomes.32 Qualitative research indicates that many people who are incarcerated prefer OAT options based on their effectiveness at treating cravings, route of delivery, side effects and structures of medication delivery in the community.33 People have also described switching OAT options to mitigate risk of violence from peers to divert their prescribed OAT.7

Restrictions to the availability of OAT options in prisons are often decided based on risk of diversion or non‐medical use. Diversion refers to the selling, trading, sharing or giving away of medications to recipients other than the prescribed person.34 Evidence indicates that both methadone and sublingual buprenorphine are diverted in prison; however, data on rates of diversion are scarce, and available evidence indicates that patterns of non‐prescribed pharmaceutical opioid use in prison are heterogenous dependent on location and context.35 The few studies that have investigated the relative risk of diversion of different formulations in custodial settings indicate sublingual buprenorphine is more commonly diverted than methadone.36 The long‐acting buprenorphine depot, administered subcutaneously weekly or monthly, has been suggested as a possible solution to limit potential for diversion.29 No diversion was self‐reported from a recent Australian trial of the buprenorphine depot.29

In addition to the reduced risk of diversion and improved safety profile, the long‐acting buprenorphine depot is preferred in the prison setting due to its longer dosing interval. Monthly dosing enables a period of coverage post‐release, which is a time of high overdose and mortality risk. This may be particularly important for individuals on remand, who do not benefit from post‐release planning. Furthermore, monthly rather than daily dosing reduces the resources required for OAT delivery, increasing capacity to enrol more individuals in the program, potentially reducing waitlists and improving access.37,38,39

Importantly, we recommend against the discontinuation of OAT or dose reduction as a disciplinary measure.40 Forced withdrawal increases the risk of substance use during incarceration and decreases treatment engagement on release, thereby increasing the risk of death.41,42 People who use drugs may also be less likely to initiate OAT in the community for fear of losing access during periods of incarceration and undergoing forced withdrawal.43

Transition of care to the community (R3.1–3.4)

The period after release from custodial settings is characterised by a markedly increased risk of mortality, especially from opioid overdose.24,44,45 Retention in OAT following release from custodial settings mitigates the risk of opioid overdose by maintaining opioid tolerance and reducing demand for illicit opioids. Retention in OAT also improves rates of primary health contact and reduces rates of ambulance or emergency department use.46,47 However, discontinuation from OAT following release from prison is common.48,49

Peer or patient navigators have the potential to support OAT retention by helping navigate complex health care and social services systems. Despite strong evidence for patient navigation across a range of health domains,50,51 there is limited and mixed evidence on the effect of patient navigation on OAT retention after release from prison.52,53 Qualitative studies support the role of peer and patient navigators following release from prison.54,55,56

A systematic review of qualitative evaluations of prison release programs identified factors associated in program success, including access to structural supports, particularly housing and employment, and continuity of care, through fostering the formation and maintenance of a therapeutic relationship throughout the pre‐release and post‐release periods.57

Given the high risk of overdose mortality after release from prison, we also recommend the provision of take‐home naloxone before or on release, alongside training in how to use naloxone, and information about where to receive additional training and resources in the community. An evaluation of a national prison‐based take‐home naloxone program using observational data in Scotland reported a continuous decrease in overdose‐related mortality within the four weeks after release.58,59 Studies also indicate widespread support for naloxone training in custody and distribution at release among both people recently released from prison and key prison stakeholders.60,61

Special populations (R4.1–4.2)

Literature on OAT for incarcerated Aboriginal and Torres Strait Islander people and pregnant people was reviewed. We were unable to identify evidence related to culturally and linguistically diverse populations or youth populations.

Aboriginal and Torres strait Islander people are disproportionately incarcerated in Australia. Despite making up only 4% of the Australian population, Aboriginal and Torres Strait Islander people make up 33% of all people in prison.1 Furthermore, Aboriginal and Torres Strait Islander people are at increased risk of OAT discontinuation on release from prison.48

Aboriginal and Torres Strait Islander communities have a strong history of providing holistic, culturally appropriate health care and leading responses to reduce drug‐related harms.62 In line with the United Nations Declaration on the Rights of Indigenous Peoples principle of self‐determination, we recommend collaborating with Aboriginal and Torres Strait Islander communities to ensure optimal OAT treatment outcomes.63 The Winnunga Holistic Health Care Prison Model in the Australian Capital Territory and the South Australian Prison Health Service Model of Care for Aboriginal Health and Wellbeing provide examples of models of prison health care developed in consultation with the community.64,65 Facilitating in‐reach services from Aboriginal Community Controlled Health Organisations (ACCHOs) may improve health care delivery to Aboriginal and Torres Strait Islander people in prison and facilitate continuity of care on release.66 However, OAT programs are rare among ACCHOs.67

Women make up a small but increasing proportion of the Australian prison population. Most incarcerated women are of child‐bearing age and, in 2021, 7% of women in prison were pregnant.68 Prevalence of substance use is high among incarcerated women, including pregnant incarcerated women.69

OAT is strongly recommended for pregnant people with opioid dependence. In addition to reductions in opioid use and risk of opioid overdose, systematic reviews of observational studies indicate that treatment with methadone or buprenorphine (including buprenorphine–naloxone) during pregnancy is associated with improved adherence to antenatal care and lower incidence of preterm birth and stillbirth.70,71,72 Neither methadone nor buprenorphine appear to be teratogenic; however, pregnant people need to be counselled about the risks of neonatal opioid withdrawal syndrome.73 Within prison contexts, a 2019 study recommends screening women for both pregnancy and opioid dependence upon intake to ensure that women receive timely treatment and avoid experiencing withdrawal.74 This is supported by data from Australia indicating that most incarcerated women (88%) described their current pregnancy as unplanned and half were unaware they were pregnant before incarceration.75

Organisational support (R5.1–5.5)

Despite strong evidence for the effectiveness of OAT against a range of health outcomes, in‐prison coverage remains suboptimal. The National Opioid Pharmacotherapy Statistics Annual Data collection reports substantial differences in the pattern of OAT prescribing by state, indicating inconsistencies in service availability and provision.76 Although quantitative estimates of unmet demand are not available, qualitative data indicate there are delays and challenges in accessing OAT due to OAT prison policy and service delivery limitations.7

There have been two recent reviews of qualitative studies examining the barriers to implementation of OAT within prison settings.8,77 Common barriers identified by both reviews included: stigma associated with OAT, a lack of knowledge about benefits of OAT among prison stakeholders, preference for abstinence‐oriented treatment, lack of resources including qualified staff, a lack of appropriate policies and protocols, and poor continuity of care on transfer or after release.8,77 A scoping review published in 2020 highlighted the interrelated nature of these barriers. For example, the societal stigma associated with OAT leads to unfavourable institutional policies, limited resourcing and resultant poor experiences, including forced withdrawals. These poor experiences then reinforce negative attitudes towards OAT among incarcerated individuals.77

There have been comparatively fewer studies examining facilitators to implementation of OAT in prison settings.77 The 2020 scoping review assessed three intervention studies, indicating that training can improve knowledge among custodial staff as measured by surveys pre‐ and post‐intervention; however, an intervention that linked health care providers reported greater improvements in staff attitudes and referral intentions compared with information provision alone.77

Conclusion

This consensus statement, developed by a national multidisciplinary expert panel with robust representation from custodial health practitioners and supported by academic and consumer input, offers evidence‐based and actionable recommendations to enhance OAT provision in custodial settings.

Although this study benefited from the lived experience of drug use through a representative from the Australian Injecting and Illicit Drug Users League, our study did not include representation from people with experience of incarceration. Another limitation of our study was that only 61% of the expert panel participated in the final survey; however, all panel members participated in at least one component, and 89% of the panel participated in at least one survey. All but one of the recommendations (R2.1) achieved consensus across both surveys, indicating stability agreement for the majority of recommendations.

In alignment with the Mandela Rules, the recommendations aim to ensure that incarcerated individuals receive treatment equivalent to that available in the community. Through improvement in OAT provision, both in prison and after release, we anticipate significant gains in health outcomes and a reduction in post‐release morbidity and mortality among this highly marginalised population.