Article Text

Developing best practice principles for the provision of programs and services to people transitioning from custody to the community: study protocol for a modified Delphi consensus exercise
  1. Tazeen Majeed1,
  2. Erica Breuer1,
  3. Layla Edwards1,
  4. Marc Remond1,
  5. Jo Taylor1,
  6. Reem Zeki1,2,
  7. Stephen Hampton2,
  8. Luke Grant3,
  9. Juanita Sherwood1,4,
  10. Eileen Baldry5,
  11. Elizabeth Sullivan1,2
  1. 1College of Health Medicine and Wellbeing, The University of Newcastle, Callaghan, NSW, Australia
  2. 2Justice Health and Forensic Mental Health Network, Matraville, New South Wales, Australia
  3. 3Office of the Deputy Commissioner, Corrective Services NSW, Sydney, New South Wales, Australia
  4. 4Jumbunna Institute for Indigenous Education and Research, University of Technology Sydney, Broadway, New South Wales, Australia
  5. 5School of Social Sciences, Faculty of Arts and Social Sciences, UNSW Australia, Sydney, New South Wales, Australia
  1. Correspondence to Professor Elizabeth Sullivan; e.sullivan{at}


Introduction There is a lack of standard nomenclature and a limited understanding of programmes and services delivered to people in prisons as they transition into the community to support their integration and reduce reoffending related risk factors. The aim of this paper is to outline the protocol for a modified Delphi study designed to develop expert consensus on the nomenclature and best-practice principles of programmes and services for people transitioning from prison into the community.

Methods and analysis An online, two-phase modified Delphi process will be conducted to develop an expert consensus on nomenclature and the best-practice principles for these programmes. In the preparatory phase, a questionnaire was developed comprising a list of potential best-practice statements identified from a systematic literature search. Subsequently, a heterogeneous sample of experts including service providers, Community and Justice Services, Not for Profits, First Nations stakeholders, those with lived experience, researchers and healthcare providers will participate in the consensus building phase (online survey rounds and online meeting) to achieve consensus on nomenclature and best-practice principles. Participants will indicate, via Likert scale, to what extent they agree with nomenclature and best-practice statements. If at least 80% of the experts agree to a term or statement (indicated via Likert scale), it will be included in a final list of nomenclature and best-practice statements. Statements will be excluded if 80% experts disagree. Nomenclature and statements not meeting positive or negative consensus will be explored in a facilitated online meeting. Approval from experts will be sought on the final list of nomenclature and best-practice statements.

Ethics and dissemination Ethical approval has been received from the Justice Health and Forensic Mental Health Network Human Research Ethics Committee, the Aboriginal Health and Medical Research Council Human Research Ethics Committee, the Corrective Services New South Wales Ethics Committee and the University of Newcastle Human Research Ethics Committee. The results will be disseminated via peer-reviewed publication.

  • public health
  • protocols & guidelines
  • qualitative research

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Strengths and limitations of this study

  • The modified Delphi methodology represents a rigorous and systematic technique to build consensus on what constitutes best-practice in programmes and services offered to people as they exit prison.

  • Our plan to explore a broad range of stakeholders’ perspectives is a significant strength of the study.

  • People with previous lived experience of prison will be able to participate.

  • This methodology could be used by other researchers seeking to develop best-practice statements and/or to investigate intervention programmes in Australia and globally.

  • One limitation of this study will be that people currently in prison will not be able to participate in this Delphi consensus-building process.


Incarceration, with its many direct and indirect impacts on individuals, families and communities, is a key social determinant of health and a major public health concern.1 2 The average daily adult prison population in Australia is approximately 42 000.3 4 During the past decade (2010–2020), Australia’s prison population has increased by 38%.5 Three-quarters of people entering prison in Australia have been in prison before, with previous offending history being the most important factor in predicting reoffending6–8 and almost half of the prison returnees having been incarcerated within the last 12 months.9 These data suggest that ‘surviving’ in the community is difficult for people with a history of incarceration, especially during the first 12 months post-release.

People transitioning to the community after any period of incarceration are confronted with a range of challenges that may include health issues, barriers to accessing health services, issues in securing housing, lack of skills necessary to gain employment and loss of family connections.1 10–12 In addition, social disadvantage, and complex needs such as comorbid substance-use disorders and mental illness, acquired brain injury, homelessness and unemployment can exacerbate these challenges.7 11 13 Therefore, people exiting correctional facilities require substantial physical, psychological, social and economic support to transition into the community and avoid returning to custody.6 7 11 14–16

To support the transition into the community and desistance from crime, a number of programmes and services are generally offered to people in prison and during their postrelease time. A diverse range of nomenclature is used to describe these programmes and services, including the terms: ‘throughcare’, ‘aftercare’, ‘throughcare and aftercare (TCAC)’, ‘community re-entry programmes’, ‘prisoner re-entry programmes’, ‘transitional interventions’, ‘transitional support programmes’, ‘resettlement/ reintegration programmes’ or ‘bridging support programmes’. Despite a lack of general consensus regarding the terminology and definition surrounding such programmes and services, they have been described in the literature as constituting intensive, continuous, coordinated and integrated management of people in prison from their first point of contact with correctional services to their successful integration into the community and completion of their legal order.10 The WHO used the term ‘throughcare services’ to describe ‘…essential elements of efforts to reduce relapse and re-offence’.17

Internationally, these programmes and services have been provided for many years and are widely recognised as a ‘best-practice’ approach to working with people in prison to improve community integration and reduce reoffending.18 However, in Australia, such programmes and services are a relatively recent initiative within correctional settings, being first adopted into policy in the late 1990s.19–22 Despite being labelled best practice, there is no standardised definition of these programmes and services and no agreed terms to use.23 Furthermore, there is no Australian or international consensus as to what the aims and outcomes of such programmes and services should be, what are their essential components and attributes, how they should be delivered in practice, and how they should be funded, monitored and evaluated.23–26 In Australia, this may, in part be due to significant variation regarding the legislative context for the provision across Australian States and territories.26 This problem is exacerbated by different terms being used by different agencies and service providers and the diverse range of government and non-government services that deliver these programmes around Australia. The result is that provision of such programmes and services in Australia is often fragmented, not gender-specific,10 27 and not culturally appropriate or relevant for Aboriginal and Torres Strait Islander people.15 28

In New South Wales (NSW), Australia, one of the few voluntary programmes operating within both correctional and community environments is the Connections Programme (hereafter referred to as ‘Connections’). Connections is offered at all Adult Correctional Centres across New South Wales by Justice Health and Forensic Mental Health Network a Statutory Health Corporation established under New South Wales legislation. It aims to improve the continuity of care for adults with a history of problematic drug users who are being released into the community.29 30 Recently, Sullivan and colleagues published the protocol for a project evaluating the Connections programme.7 As part of the Connections evaluation project, a systematic review of evaluations of programmes for people with problematic drug use and mental health disorders was conducted.23 This review highlighted the lack of a common nomenclature and understanding of such programmes, including their scope and attributes.

To address the health and social needs of people exiting prison, a nationally endorsed template for such programmes and services is required. The involvement of regional and national experts in developing any such best-practice model through a Delphi consensus project is likely to lead to widespread endorsement and uptake. This article describes the protocol for a modified Delphi consensus exercise to identify best-practice principles for programmes and services offered to people who are in prison and who are transitioning out of prison to live in the community.


The aim of this study is to develop expert consensus on the nomenclature for and provision of programmes and services offered to people transitioning from custody to the community. Specifically, we aim to develop a set of best-practice principles for such programmes that describes: the aims and outcomes of the programme; who should be eligible to access them; service components; attributes of service providers; monitoring and evaluation; funding; and how these programmes can target individual needs while being responsive to age, gender, ethnicity, disability status and culture (see box 1 for the full list of Research questions).

Box 1

Delphi research questions

With respect to programs and services offered to people transitioning from custody to the community in Australia:

  1. What is the nomenclature used to describe the services/programs, and how to define them?

  2. What are the aims and intended outcomes?

  3. Who should be eligible to participate?

  4. What services should be provided?

  5. What are some key attributes of service providers?

  6. At what point in an individual’s contact with the justice system should these be provided?

  7. Who should be responsible for funding, provision and coordination?

  8. How should they be monitored and evaluated?

  9. How do we ensure that they are targeted to individual needs and responsive to age, gender, ethnicity, health status, disability status and culture?

  10. How do we ensure an equitable approach to delivering?

Methods and analysis

Study timeline

The study is currently in the preparatory phase, with the research team developing a Delphi questionnaire. The expected end date of the study is December 2022.

Study design

This study will use an online Delphi approach to develop an expert consensus on best-practice for programmes and services offered to people transitioning from custody to the community. The Delphi approach is a structured and systematic method to reach a consensus between a group of heterogeneous experts or informed respondents.31–33 This method has been extensively used in health and intervention research. The participants take part anonymously in sequential questionnaires that constitute different rounds, with each round designed to refine results from the previous one. After each round, the group responses are reported back to the participants, who are then able to reconsider their views in light of the views of the entire group of experts. The Delphi approach maintains that the opinions of many outweigh those of the individual; thus, any consensus generated may be considered a valid expert opinion.34 35

As there is a lack of consensus around what should constitute best-practice principles for programmes and services offered to prisoners as they prepare to transition into the community, a Delphi study was considered a suitable research tool to achieve unanimity in opinion.36 The four distinct characteristics of the Delphi technique—anonymity, iteration, controlled feedback and statistical ‘group response’—are also suited to the aims of the project.37

The Delphi approach

The Delphi will occur in two phases—the preparatory phase and the consensus building phase (figure 1).

Figure 1

Delphi project activities—preparatory phase and the consensus building phase.

Preparatory phase: qualitative evidence synthesis for the development of the Delphi questionnaire

In the preparatory phase, we systematically used four search strategies to find published literature and online sources relating to programmes and services offered to people as they exit prison. These aimed to efficiently identify and retrieve relevant literature. Specifically, (1) international systematic reviews; (2) Australian primary studies published in peer-reviewed journals; (3) Australian grey literature; and (4) international guidance documents related to programmes and services for people exiting prison. These search strategies are summarised below and in table 1. Additional details are available in online supplemental material appendix tables 1 and 2.

Table 1

Systematic search strategy for the preparatory phase of modified Delphi study

For all four of the strategies, one reviewer (LE) searched the data sources (table 1). To find international systematic reviews (strategy 1) and Australian peer-reviewed publications (strategy 2), LE exported the titles and abstracts were imported into Endnote,38 removed duplicates and conducted a title and abstract screen using predefined inclusion and exclusion criteria outlined in table 1. Any manuscripts that did not satisfy the inclusion criteria (ie, not relevant to study topic or search terms or population) were excluded. LE then obtained the full text for the remaining articles, which were reviewed by a second reviewer (MR or EB) against the inclusion and exclusion criteria. To find Australian grey literature (search strategy 3) and international guidance documents (strategy 4), LE searched the data sources, retrieved full-text documents from relevant websites, evaluated these against the inclusion and exclusion criteria and then imported those documents which met the inclusion criteria into Endnote. One other reviewer (MR or EB) checked the final list of included documents.

Figure 2 presents a Preferred Reporting Items for Systematic Reviews and Meta-Analyses diagram describing the results of the search and review strategy used to identify and retrieve relevant information. The final group of 53 documents, comprising 5 primary studies and 48 secondary manuscripts, were imported into the data analysis software NVivo39 for analysis.

Figure 2

Results of the phase I search and review strategy—Preferred Reporting Items for Systematic Reviews and Meta-Analyses diagram.

Patient and public involvement

No patients were involved in this protocol study.


The technique of ‘Framework synthesis’ was used to develop an a priori framework (based on our research questions) to synthesise the qualitative evidence retrieved from included documents.40 41

Development of questionnaire

The research team analysed and synthesised the literature to develop framework categories and developed a set of potential best-practice statements for each category. These statements were then collated to develop the first Delphi questionnaire (Questionnaire 1). Questionnaire 1 has been reviewed multiple times by the research team and has been converted into an online survey via the Research Electronic Data Capture application (REDCap).42 43 Currently, it is being piloted by the core research team, including representatives from Corrective Services NSW (CSNSW) and the Justice Health and Forensic Mental Health Network (JHFMHN) prior to administration to the expert panel.

Consensus building phase: identification of an expert panel

The first step of the consensus building phase will involve identifying national and international experts and inviting them to participate in the project. Those who wish to participate and agree to the online consent process will form an expert panel to conduct the Delphi consensus building process.

Participant selection and recruitment

Study population

The success of a Delphi study depends, in part, on the participants' ability to provide relevant input. In the absence of specific guidelines defining what constitutes ‘experts’ for Delphi studies, this study will adopt a definition proposed by Blaschke et al44: Delphi experts are those ‘individuals who possess both knowledge and experience representative of the capacity to articulate informed opinion and provide relevant input about a given topic’.

Adopting this definition, experts from Justice Health and Forensic Mental Health Network, Corrective Services, New South Wales, other relevant organisations will be identified and invited to participate, along with national and international researchers and local and international stakeholders providing these programmes and services in Australia and globally. To ensure that a broad range of perspectives are incorporated in the consensus-building exercise, we will also include the voices of those with lived experience of incarceration, advocacy groups, Not for Profits and First Nations stakeholders. These organisations and key stakeholders will be identified: (1) through mapping of stakeholders (currently being conducted by members of the research team with input from Corrective Services, New South Wales and Justice Health and Forensic Mental Health Network); (2) from websites of agencies and advocacy organisations (Australian and international); and (3) by asking individuals who agree to participate in the Delphi study to recommend up to five experts in the field whom we may contact.

Inclusion criteria

Potential participants will comprise individuals who have experience in implementing, managing, funding, advocating for, participating in or researching programmes or services related to people in or exiting prison in Australia and globally. Participants must be willing and able to participate in all three rounds of the Delphi and the online consensus meeting.

Sample size

There are no published specific recommendations regarding appropriate sample sizes for Delphi studies,45 but many published Delphi studies report between 10 and 100 participants.46 Owing to the multisectorial nature of these programmes, we will endeavour to ensure that representatives from stakeholders and organisations involved in all aspects are included. Nonetheless, it will be necessary to ensure that the consensus building meetings are small enough for meaningful participation from all participants. For this reason, we aim to include approximately 60 participants with representation from all relevant stakeholder groups (see table 2). Furthermore, we will endeavour to recruit representatives across gender, ethnic and cultural groups (specifically including Aboriginal and Torres Strait Islander people and people of Culturally and Linguistically Diverse backgrounds) and people living with a disability. In addition, we aim to include people who have lived experience of prison, who have disclosed their experience in the public domain, and who currently use these experiences to inform and strengthen their work as advocates, academic researchers or employees of Non-Government Organisations or governmental organisations.

Table 2

Potential participants


Potential participants (both organisations and individuals) will be approached directly by the research team via email and/or via automatically generated emails from the REDcap application. All potential participants will receive a soft copy of the Participant Information Sheet outlining the purpose of the study, methodology, potential risks and benefits of participation, anticipated study outcomes and the expected time contribution of participation. The Participant Information Sheet will state that participation in the study is entirely voluntary, and it will reiterate that participation in the research will confer no benefit or risk to their employment. Potential participants will be given at least 3 days to decide whether they would like to participate in the research. It will be made clear that even if they consent to participate in the research, they are free to withdraw at any time without consequence.

Participants who are recruited as described above will also be invited to assist with snowball sampling. That is, when they complete the first online questionnaire for the study (described below), they will be asked to nominate other experts they believe should be included in the study. These nominated experts will be forwarded an ‘expression of interest’ email via a separate and confidential REDcap generated form. Using this snowball sampling technique, we expect to include experts from specific groups, including Aboriginal and Torres Strait Islander people, people from Culturally And Linguistically Diverse backgrounds and people living with a disability.

The basic principles of confidentiality, privacy and anonymity of nominated experts/individuals will be maintained throughout this process. Once an individual decides to participate, they can click on a hyperlink provided within the invitation email and in the Participant Information Sheet which will be redirected to the online consent form. Once the online consent form is signed, participants will be redirected to Questionnaire 1. If participants wish to decline to answer any specific questions or if they wish to withdraw their consent or data from the study, they can do so at any time by closing down the browser, exiting the online meeting (see below) or contacting the research team.

Data collection

Data collection for the Delphi consensus building will consist of three rounds (Rounds 1–3). For online Questionnaire 1, participants will use a five-point Likert scale (Strongly Agree, Agree, Neither Agree nor Disagree, Disagree and Strongly Disagree) to rate how strongly they agree or disagree with each proposed best-practice statement about programmes and services offered to people transitioning from custody to the community. Other questions will ask participants to choose a preferred option from a list of statements, or to choose if options are Essential, High Priority, Medium Priority, Low Priority, or Not a Priority. Within each category, there will be free text options for participants to provide additional comments/feedback.

Once Questionnaire 1 is completed, participant responses will be analysed by the research team. Any items assessed to have reached ‘positive consensus’ (at least 80% agreement) will be removed from the questionnaire and included in the final best-practice statement, while any items which reach ‘negative consensus’ (at least 80% agreement) will be removed from the study. Any items that do not achieve positive or negative consensus will be flagged for discussion at the online meeting.

Participants will then be invited to attend an online meeting to discuss the results from Questionnaire 1. Overall descriptive statistics for each item in Questionnaire 1 will enable participants to understand how their rating of an item compared with the average rating of all participants. Participants will be given the opportunity to discuss these items and the ratings with each other, with facilitators asking them to particularly focus on items that did not reach a positive or negative consensus. Participants may also merge or change the wording of items. Qualitative data from the online meetings will be transcribed via online meeting platforms and will be used to verify the discussions in the meetings. They will not be used for further analysis and will be discarded once Round 2 starts. All items that did not reach positive or negative consensus during Round 1, or the merged or modified version of these items agreed to during the online meeting, will be included in Questionnaire 2.

In Round 2, Questionnaire 2 will be circulated online to all participants, who will be asked to rate each item using the previously described Likert scales. Responses will be analysed as described above, and only items that do not reach a positive or negative consensus will be included in Questionnaire 3 if needed.

In case Round 3 is needed, Questionnaire 3 will be circulated online to all participants, who will be asked to rate each item using the previously described Likert scales. After Questionnaire 3, any items that reached positive consensus will be included in the best-practice statement, while any items that reached negative or no consensus will be removed. If there remains a lack of consensus among participants for over a 1/3 of the items in this final questionnaire, an additional online consensus meeting will be held with the expert panel to resolve these items.

Data storage and security

The online questionnaire is being developed in REDCap.43 REDCap is a secure web platform for building and managing online databases and surveys hosted by the Hunter Medical Research Institute.47 The Hunter Medical Research Institute’s REDCap and Information management policy is available here ( During the active phase of the project, copies of data will be stored on REDCap with access restricted to the research team. REDCap is continuously backed up to the Hunter Medical Research Institute servers. Data entered into the REDCap questionnaires will be downloaded for analysis and stored in a secure file share environment (OneDrive) at the University of Newcastle, with access restricted to the research team. Only deidentified participant data will be exported from REDCap. Once all three rounds of data collection have been completed, the data will be permanently deidentified by replacing the names with randomly generated codes and deleting participant email addresses. These data will be continuously backed up to the University’s cloud environment. Once the project has been completed, project data will be stored in the University of Newcastle’s Data Repository, Cr8it. These data will be retained for 5 years from the completion of the project and then destroyed in line with the University of Newcastle’s Research Data and Materials Management Guideline, the State Records Act 1998 and the General Retention and Disposal Authority—Higher Education and Further Education Records (GA47).

Data analysis

Analysis of these data collected during Rounds 1–3 will comprise the calculation of descriptive statistics to quantify the proportion of participants who choose each category of the Likert scale for each item in the questionnaire. We will define ‘positive consensus’ for an item as occurring if >80% of respondents32 48 indicate that they ‘Strongly Agree’ or ‘Agree’ with a statement or that they believe that a factor is ‘Essential’ or ‘High Priority’. We will define ‘negative consensus’ for an item as occurring if >80% of respondents indicate that they ‘Disagree’ or ‘Strongly Disagree’ with a statement, or that they believe that a factor is ‘Low Priority’ or ‘Not a Priority’.48 For statements that only allow for a Yes/No response, ‘positive consensus’ will be defined as >80% of participants selecting ‘Yes’ while ‘negative consensus’ will be defined as >80% of respondents selecting ‘No’.

Data subject rights

Participation in this study is voluntary. Participants can withdraw from the study at any time without adverse consequences. Any data collected before a participant withdraws will be included in a deidentified analysis and kept confidential by the research team. In addition, we will collect data on gender, Aboriginal and Torres Strait Islander and disability status, and type and length of experience in the field.

Ethics approval

The protocol for the study has been submitted to, and approved by the Justice Health Human Research Ethics Committee (HREC) (JH File No.G217/16), Corrective Services Ethics Committee (CSEC; G217/16) and the Aboriginal Health & Medical Research Council (AH&MRC; HREC 1187/16). Ethics ratification was granted from the University of Newcastle HREC for expedited review (H-2022-0039).

Discussion and conclusion

In its current usage, many different terms are being used in Australia and globally to describe a range of fragmented activities, programmes and services offered and provided to people exiting custodial settings. ‘Throughcare’, ‘aftercare’, ‘throughcare and aftercare (TCAC)’, ‘community re-entry’ and ‘transitional programmes’ are some of the more commonly used terms. Moreover, there is no general expert consensus as to how these programmes should be defined, what are the minimum standard of services provided under these programmes, what best practice should comprise or how they should be funded, provided, monitored and evaluated. The absence of an agreed nomenclature for programmes and services offered to people exiting prisons results in the use of a variety of different terms that are open to differing interpretations and usage. This phenomenon may be associated with the fragmented programmes and services observed and the lack of a best-practice model for such programmes and services. This study will address this gap by achieving expert consensus on nomenclature and developing a set of best-practice principles for these programmes and services via a modified Delphi consensus process.

One main limitation of the Delphi process is that any questionnaire consensus can be substantially influenced by the choice of the rating scale and consensus criteria.49 However, open discussions during ‘Online meetings’ with experts will ensure the authenticity and strength of consensus. The strengths of this Delphi approach are that it is designed to elicit knowledge from a diverse range of experts, enabling the identification of (dis)agreements between them, while also aligning them with other existing work. Using a rigorous evidence-based Delphi approach and including national and international experts and stakeholders in our surveys and online meeting/s (during the consensus building phase), will enable us to establish and generate a list of agreed best-practice statements that will address the needs of people exiting custody in Australia and internationally. The use of REDCap to operationalise this Delphi method also entails advantages. Globally, researchers are increasingly using REDCap to develop and deploy electronic data collection tools to support their translational research. In contrast to conventional in-person interviews or hard-copy surveys, this method provides increased convenience for experts, facilitates remote participation, improves inclusion and enrolment/retention, and provides effective and efficient outreach.42 43

It is anticipated that the findings from this study will help to consolidate the research and understanding of service provision by providing expert consensus on key issues surrounding programmes for people exiting prisons in Australia and globally. A common and expert understanding of the best-practice principles of these programmes will help to set core standards for service provision and research in this field by providing a concrete framework within which to design such programmes using evidence-informed best-practice guidelines. While this research will be conducted in Australia, results should be of international interest as common aims, elements and attributes underpin the delivery of these programmes across disparate jurisdictions. Results will be published in peer-reviewed journals and presented in a variety of organisational forums and conferences.

Ethics statements

Patient consent for publication


This paper and the research behind it would not have been possible without the exceptional support of our research team and all the experts. Thank you to our colleagues at all the involved institutions for their helpful feedback during the drafting of this paper. We are also grateful for the insightful comments offered by the anonymous peer reviewers.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • Twitter @Dr_TazeenMajeed

  • Contributors ES and EB conceived the study, and EB drafted the initial protocol. EB, LE and MR were responsible for identifying the eligible studies. EB, LE, MR, TM and ES worked on framework synthesis and Questionnaire development. TM prepared the first draft of the manuscript, and LE reviewed it, EB, LE and MR continued to provide comments on revisions, with ES, SH, JT, RZ, JS, EB and LG providing an expert review. The manuscript was critically reviewed by TM and all coauthors (LE, MR, EB, RZ, LG, JT, SH, JS, EB and ES), who read and approved the final manuscript.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests SH is the Executive Medical Director at Justice Health and Forensic Mental Health Network, LG is the Deputy Commissioner at Corrective Services NSW Department of Communities and Justice, RZ is a Senior Research Officer at Justice Health and Forensic Mental Health Network, and ES is the Custodial Health Research lead at Justice Health and Forensic Mental Health Network.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting or dissemination plans of this research.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.