What works to support disadvantaged groups towards employment? - Research report

Question 1

Executive summary

Accepted Answer

This report presents the results of a systematic review to better understand what works to support people from four disadvantaged groups to move towards and into employment. This includes care leavers, ex-offenders, people experiencing homelessness, and people with substance misuse issues.

Criminal convictions, substance misuse, and homelessness are often co-occurring: many people experience more than one of these disadvantages at the same time. Many interventions in the literature therefore addressed more than one of these groups. Interventions that explicitly aim to address multiple disadvantages appear, in general, to be more effective than those aimed at addressing only a single disadvantage. For example, housing interventions that integrated other elements, such as addiction or mental health support, were generally more effective than those that did not.

Individual Placement and Support (IPS), which involves rapidly placing people in work and then providing support to help them sustain employment, has been shown to be effective for a range of groups. The evidence is strongest for people with substance misuse issues, but there is also promising evidence for ex-offenders and people experiencing homelessness. There is also promising evidence that a variety of Employment Support programmes, which aim to prepare people for work by providing advice and guidance, can be effective. In particular, programmes that provide vocational skills or work experience for offenders, alongside wider support services, can be effective at helping people into work.

There is limited but promising evidence that recovery housing can be an effective way of helping ex-offenders and those with substance misuse issues into work. This typically involves people staying in small residences with peers who are in similar situations. Residents are required to maintain abstinence. However, these will need testing in a UK context.

For care leavers, there is less high-quality evidence than for the other groups. There is some evidence that certain Independent Living Programmes (ILPs) can be effective at improving housing stability and employment outcomes.

This systematic review should be read as providing a high-level overview of the evidence on a broad topic, and further work will be required to understand in more detail what is likely to work in the UK context. Most of the evidence included in the review is from the US, which limits the extent to which the findings are transferable to a UK context. Future work should start by reviewing the primary literature on interventions that show promise before conducting UK trials, if necessary. There are also many interventions being delivered in the UK that are not subject to evaluation and are therefore not included in this review. There may be value in evaluating these interventions.

Question 2

Acknowledgements

Accepted Answer

The authors would like to thank the extended team of researchers at NatCen who helped with screening, data extraction and analysis. This includes Stacey Link, Yasmin Spray, Eddie Cohen, Nevena Ilic and George Leeder. Additional thanks are extended to Nathan Hudson and Jasna Magic for providing their expertise during the scoping interviews.

Question 3

Author details

Accepted Answer

This report was produced by the Communities, Work and Income team at the National Centre for Social Research.

The authors of the report were Bernard Steen, Aisling Draper and Tommy Freegard.

Question 4

Glossary of terms

Accepted Answer

Interventions in the literature:

Individual placement and support (IPS): An intervention designed to help individuals with various disadvantages to secure and retain competitive employment. The IPS model is distinctive because it places individuals directly into employment and then provides the necessary support to sustain that employment.
Employment support: Interventions that provide support, or training, to help people prepare for work or move closer to work. This could include coaching, training, internships, work placements, or grants to set up business.
Inmate work programme: Interventions that provide prisoners with opportunities for low-paid employment, often involving manual labour.
Community and peer mentoring: Interventions that involve people with similar lived experiences providing mentoring or guidance. This also includes residential interventions where groups of people with similar disadvantages live together and provide support to one another.
Case management interventions: Interventions that involve a caseworker coordinating a range of services, typically for people in emergency circumstances and often involving a substantial mental health focus.
Education interventions: Interventions that provide general training, education or skills, as opposed to vocational or employment-focused training. This ranges from basic maths and language skills through to degree-level education.
Housing interventions: Any interventions providing housing support, accommodation, supported housing or temporary housing. These interventions often include additional support for mental health, substance misuse or other disadvantages.
Medication assisted treatment: Interventions that involve the use of medication alongside counselling and other activities to help people with opioid use disorder.
Independent living programmes for care leavers: Interventions which aim to equip young people with the necessary skills to live independently. These skills include communication, decision-making, emotional skills, as well as vocational skills.
Creating connections with family and other adults: A range of interventions aimed to foster positive relationships between young people currently in care and caring adults, with the goal of improving life stability and ultimately leading to better life outcomes such as education and employment.
Extending care arrangements past the age of 18: Interventions which allow care leavers to remain in their care placements beyond the legal adult age, which is typically 18. The goal of these interventions is to replicate the support structure that non-care experienced youth often receive from their parents.

Research terminology:

Control or comparison group: This is a group of participants who do not receive the experimental treatment or intervention. This group serves as a benchmark to compare against the group that does receive the treatment, helping researchers determine if any observed effects are due to the intervention itself.
Effect size: This is a measure of the impact or effectiveness of the programme on the outcomes of interest. It quantifies the extent to which the programme has made a difference.
Propensity score matching: This is a statistical technique used to create a fair comparison between two groups in observational studies. It involves pairing individuals with similar characteristics, based on a calculated probability (propensity score), to reduce bias and mimic a randomised experiment.
Quasi-experimental designs (QEDs): There are research methods that aim to evaluate the effect of an intervention or treatment without random assignment of participants to groups. While they attempt to establish causality, they are generally considered less rigorous than randomised controlled trials because the lack of randomisation can introduce biases or confounding variables.
Randomised controlled trial (RCT): A type of research study in which participants are randomly assigned to either a group receiving the treatment or intervention being tested, or a control group that does not receive the treatment. This method helps ensure that any differences observed between the groups are due to the treatment itself, rather than other factors. An RCT is typically considered the most robust way of determining whether an intervention has a causal impact on an outcome.
Systematic review: This is a comprehensive and structured synthesis of research studies on a particular topic or question. It involves systematically searching for, appraising, and summarising the evidence from multiple studies to provide a high level of evidence on the effectiveness or impact of interventions, policies, or practices.

Question 5

1.  Summary

Accepted Answer

This report presents the results of a systematic review to better understand what works to support people from four disadvantaged backgrounds to move towards and into employment. This includes care leavers, ex-offenders, people experiencing homelessness, and people with substance misuse issues.

1.1 Background

In November 2024, the Government published the Get Britain Working White Paper, which set out an ambition to achieve an employment rate of 80%. Part of this effort will involve supporting people with complex needs into work. This includes groups such as care leavers, ex-offenders, people experiencing homelessness, and people with substance misuse issues. These groups are less likely to be in employment than others. There is a clear need to understand what works to support these groups to move closer to employment.

1.2 Research aims

There is a large volume of published evidence on what works to support disadvantaged groups to move closer to employment, including several high-quality systematic literature reviews. However, the evidence tends to focus on a single population group, or on a single type of intervention. There was a need to bring this evidence together to provide a holistic overview of what works across disadvantaged groups, including those experiencing multiple forms of disadvantage. To achieve this, the Department for Work and Pensions (DWP), supported by the Ministry of Housing, Communities and Local Government (MHCLG) and the Ministry of Justice (MoJ), commissioned the National Centre for Social Research (NatCen) to conduct a systematic review.

The systematic review aimed to answer two main research questions (RQs):

RQ1. What works to help ex-offenders, care leavers, people experiencing homelessness, and those with substance abuse problems move closer to the labour market or enter work?
RQ2. How are interventions best delivered and by whom?

1.3 Research methods

This study used a systematic review methodology. This involves following a broad and comprehensive process for searching for literature, screening it for eligibility and quality, and extracting and synthesising information from it. Academic literature was located by running complex searches on several databases. Non-academic (grey) literature was located by searching the websites of a large number of relevant organisations, and by contacting expert stakeholders and academics.

To be included in the review, studies had to have been conducted in the OECD, have been published after 1990, and include samples of ex-offenders, care leavers, people with substance misuse issues, or people experiencing homelessness. They had to investigate the impact of interventions that explicitly aimed to improve employment outcomes, and/or had to report on employment outcomes for interventions that were not aimed at improving employment outcomes. Studies that looked at employment interventions but focused on other related outcomes, such as life stability, health and wellbeing, or education and training were also included.

All studies were quality appraised and the lowest quality studies were excluded. After final screening, 77 studies were included in the review, including 47 systematic reviews, 13 academic primary research studies, and 17 non-academic studies.

1.4 Overview of populations and outcomes

The study focused on four populations: care-leavers, ex-offenders, people with substance misuse issues, and people experiencing homelessness. In practice, there was a separation in the literature between care leavers and the other three groups. Interventions for care leavers were largely distinct, whereas many interventions explicitly targeted more than one of the other three groups, or explicitly targeted people with more than one of these disadvantages.

The key outcome of interest for the review was entry into employment, but more broadly the focus was on interventions that help move people closer to the labour market. Given this, a range of other outcomes besides entry to employment were in scope. These other outcomes are conceptualised as a pathway into work. This pathway starts with essential life stability outcomes, including housing, reduced criminal activity, and core life skills like financial management. It then includes health and wellbeing outcomes, including physical and mental health, or reduced substance misuse. It then includes education and skills outcomes. Lastly, it includes employment outcomes, like job search skills, entry into work, or earnings.

1.5 Findings and evidence map overview

The table below provides high-level overview of the evidence. Cells in green indicate that there was evidence that a category of intervention has a positive impact for a given group on the stated outcome. Cells in yellow indicate evidence of promise, meaning there is some evidence, but that we cannot be fully confident in the impact of the intervention. Cells in red indicate that there is evidence available, but it does not show an impact, whereas cells in grey show that there is no evidence available.

The table above provides a high-level overview of the evidence.

Cells in green indicate that there was evidence that a category of intervention has a positive impact for a given group on the stated outcome.

Cells in yellow indicate evidence of promise, meaning there is some evidence, but that we cannot be fully confident in the impact of the intervention.

Cells in red indicate that there is evidence available, but it does not show an impact.

Cells in grey show that there is no evidence available. The rest of this section provides more detail on these findings.

The contents of the table are also described below.

Individual Placement and Support

Ex-offenders, Substance misuse and Homeless - positive impact on employment outcomes
Care leavers – no evidence available

Employment support

Ex-offenders, Substance misuse - positive impact on employment outcomes
Homeless and Care leavers - evidence of promise for employment outcomes

Community and peer mentoring

Ex-offenders - evidence of promise for employment and substance misuse outcomes
Substance misuse: evidence of promise for employment outcomes; positive impact on substance misuse outcomes
Homeless and Care leavers - no evidence available

Case management interventions

Ex-offenders, Substance misuse and Homeless - no evidence of positive impact on employment outcomes
Ex-offenders and Substance misuse - no evidence of positive impact on health outcomes
Homeless - positive impact on housing options
Care leavers - no evidence available

Education interventions

Ex-offenders - positive impact on reoffending outcomes; evidence of promise for employment outcomes
Substance misuse and Homeless – no evidence available
Care leavers - evidence of promise for employment and education outcomes

Housing interventions

Ex-offenders, Substance misuse, Homeless and Care leavers: positive impact on housing outcomes; no evidence of positive impact on employment or health outcomes

Inmate work programmes

Ex-offenders - positive impact on reoffending outcomes; evidence of promise for employment outcomes
Substance misuse, Homeless and Care leavers - no evidence available

Medication assisted treatment

Substance misuse - evidence of promise for reoffending outcomes; no evidence of positive impact on employment or health outcomes
Ex-offenders, Homeless and Care leavers - no evidence available

Independent living programmes for care leavers

Care leavers - evidence of promise for employment, housing, reoffending and education outcomes
Ex-offenders, Substance misuse and Homeless - no evidence available

Creating connections with family and other adults

Care leavers: positive impact on adoption rates and social connectedness; evidence of promise for employment, health and reoffending outcomes
Ex-offenders, Substance misuse and Homeless - no evidence available

Extending care arrangements past the age of 18

Care leavers - evidence of promise for employment, education, housing, crime and health-related outcomes
Ex-offenders, Substance misuse and Homeless - no evidence available

1.6 What works for ex-offenders, people experiencing homelessness, and people with substance misuse issues?

The literature shows that a broad range of interventions can help to support these populations further along the pathway to work. For the earlier stages in the pathway, there is evidence for the positive impact of housing interventions, substance misuse interventions, peer mentoring interventions, educational interventions and others. For moving people directly into work, the evidence of positive impact is unsurprisingly strongest for interventions that explicitly aim to achieve this, such as Individual Placement and Support (IPS), and other Employment Support programmes.

A conclusion that emerges from this holistic review of the literature is that interventions that explicitly aim to address multiple disadvantages appear, in general, to be more effective than those aimed at addressing only a single disadvantage. For example, housing interventions that integrate other elements, such as addiction or mental health support, are generally more effective than those that do not. Substance misuse interventions that aim to reduce substance misuse by using Medically Assisted Treatment (MAT) tend to succeed on that front, but there is no evidence of positive impacts on employment or mental health outcomes. Similarly, there is evidence that 12-step substance misuse programmes (like Alcoholics Anonymous and Narcotics Anonymous) have a positive impact on substance misuse, but no evidence of impact on employment outcomes. By comparison, peer mentoring interventions that also provide housing show evidence of impact for both substance misuse and employment outcomes. There is strong evidence of impact on employment outcomes for IPS, which aims to rapidly place people into work and then provide a support package to help them sustain work. IPS has an explicit ‘zero exclusions’ policy, meaning support is available for a wide range of issues.

The overarching conclusion—that addressing multiple disadvantages is more effective than addressing single disadvantages—is based on a narrative synthesis, which involves a comprehensive reading of the literature to identify common themes and relationships among the findings. It is not, however, based on a quantitative meta-analysis, which would involve combining the data from multiple studies and using statistical methods to synthesise the findings. Both approaches are robust, but a meta-analysis would provide additional confidence in the finding and enable us to estimate the scale of the benefit of addressing multiple disadvantages, compared to addressing only single disadvantages.

This section provides short summaries of the evidence for each main category of intervention for these populations.

Individual Placement and Support (IPS)

IPS involves rapidly placing individuals into employment and then providing the necessary support to sustain that employment. This contrasts with other Employment Support programmes that aim to prepare individuals for work before helping them find jobs. IPS was originally developed for people with severe mental health issues and is well-evidenced in that context, but the evidence shows that IPS can also be effective for a broader range of populations. The evidence is strongest for people with substance misuse issues, but there is also promising evidence for ex-offenders and people experiencing homelessness. More UK-based trial evidence is required that specifically focuses on these populations. There is evidence from the UK that IPS interventions need to be highly tailored to local labour market contexts, such as the quantity and type of work available.

Employment Support programmes

The literature contained a diversity of interventions that aimed to prepare people for work or provide advice and guidance. Some involved coaching, others involved internships and work placements, and others involved grants to help people set up businesses. Overall, the interventions and the evidence are too diverse to draw clear overarching conclusions about what works best. But there is evidence from individual studies that some of these types of interventions can work for some groups. For example, there is evidence that a range of programmes that aim to support ex-offenders into work by providing vocational skills alongside wider support services can be effective. More intensive interventions that provide holistic services both pre- and post-release showed the most promise. There is also evidence that interventions that provide subsidised temporary work placements for ex-offenders can help them to transition into permanent employment. Further work is required to identify those interventions within this category that show the most promise and synthesise the primary evidence or conduct new trials.

Inmate work programmes

These interventions provide prisoners with opportunities for low-paid employment, often involving manual labour. The primary goal is to generate revenue for the prisons service, but a secondary aim is to help inmates develop positive work habits and job skills that assist them after their release. There is a reasonably strong body of evidence on these interventions, but all of it comes from a single US state that has a long history of similar programmes. While there is promising evidence that these programmes can have a positive impact on employment post-release, these findings should not be generalised beyond the highly specific context in which the interventions were tested.

Community and peer mentoring interventions

The literature included several “recovery housing” interventions, all in the US. These typically involved ex-offenders with substance misuse issues staying in small residences, such as “Therapeutic Communities” and “Oxford Houses”, with peers for potentially extended periods of time. These residences all require abstinence from drugs and alcohol. Some are run by staff and involve formal therapeutic elements (Therapeutic Communities); whereas others are entirely peer-run, with no professional involvement at all (Oxford Houses). There is a limited amount of evidence on these interventions, but the findings are promising and show that recovery houses can have a positive impact on employment outcomes. Oxford Houses, being entirely peer-run, are particularly cost effective. These interventions would need testing in the UK context to establish if these positive findings are maintained.

Other peer mentoring interventions included 12-step programmes. The evidence shows that these are effective for achieving abstinence, but there is no evidence, to date, on employment outcomes specifically.

Case management interventions

These interventions involve a caseworker coordinating a range of support services, typically for people in particularly difficult or emergency circumstances. Case management interventions have mostly been tested with homeless populations, and the interventions ranged from less intensive to highly intensive levels of support. Overall, there is evidence that case management interventions for homeless populations are an effective way of reducing homelessness, but there is only very limited evidence available on the impact for other outcomes areas (including employment), or on the impact for other groups.

Housing interventions

These interventions ranged from the simple provision of housing (whether short-term emergency housing or long-term solutions) with no additional support, to much higher levels of wrap-around support, approaching the level seen in case management interventions. This included Housing First, a widely tested intervention for chronic homelessness, particularly among individuals with mental health and substance misuse issues. Housing First provides immediate housing without preconditions such as abstinence or treatment compliance.

The evidence shows that housing interventions that involve additional support for co-occurring disadvantages (such as Housing First) are more effective than basic housing provision, when it comes to improving housing stability. However, there is not sufficient evidence to determine whether housing-focused interventions have positive effects on employment outcomes.

Correctional education programmes

These are educational programmes for individuals in prison. They ranged from basic maths and literacy skills through to degree-level qualifications. These have been comprehensively evaluated in the US. The evidence shows that correctional education programmes reduce rates of reoffending, and there is some promising evidence that they can improve employment outcomes.

Substance misuse interventions

Many interventions included components that aimed to help with substance misuse as part of a broader package of support. Some interventions, however, focused entirely on substance misuse, most notably Medication Assisted Treatment (MAT). This has been shown to be an effective treatment for substance misuse, but the evidence suggest it has limited broader impacts, including on employment or mental health outcomes.

1.7 What works for care leavers?

The evidence on what works for care leavers was considerably weaker than for the other three groups. The interventions in the literature took a wide range of different approaches to supporting care leavers. While some tried to help care leavers to live independently, others tried to help them remain in their existing care arrangements, and others tried to help them build connections to family members or other adults. The evidence for these three approaches is addressed in turn.

Independent living programmes (ILPs)

ILPs aim to equip young people with the necessary skills to live independently. These skills include communication, decision-making, emotional skills, as well as vocational skills, budgeting and common household tasks. These interventions were typically delivered when young people were in the transitionary period between the ages of 16 and 19. Interventions often involved mentoring or coaching alongside job readiness training. Other ILPs involved supported housing. This meant placing care leavers into safe housing in which they were supported by in-house carers. These carers helped young people develop everyday life skills such as cooking, cleaning and budgeting.

Several systematic reviews tried to determine the impact of ILPs generally, but the conclusions were mixed and conflicting. This is likely explained by the diversity of ILP interventions. The systematic reviews identified several high-quality studies that suggest certain individual ILPs are effective at improving employment outcomes, but drawing an overarching conclusion about all ILPs is not possible. An example of a robustly evaluated ILP programme was the YVLifeSet programme in the US, which involved both stable housing and community support.

Extending existing care arrangements

These interventions allow care leavers to remain in their care placements beyond the age of 18. The aim is to replicate the support structure that non-care experienced young people often receive from their parents, with a view to giving care leavers more time to complete their education, gain employment, and secure stable housing before they are fully independent. In the UK, the main example is the Staying Put programme.

There is some promising evidence that extended care may improve employment, housing, education and health outcomes, but in general, the quality of the available evidence is low, and it is not possible to draw strong conclusions about impact.

Creating connections with family and other adults

A range of interventions aimed to foster positive relationships between children currently in care and caring adults, with the goal of improving life stability and ultimately leading to better life outcomes such as education and employment. Some programmes focused on reconnecting young people with their biological families, but others focused on other non-parental adults such as grandparents, aunts, uncles, and other supportive figures.

There is some promising evidence that these programmes can increase the likelihood of adoption or reunification with family. And there is evidence that young people in care, who are later adopted, were more likely to enrol in non-compulsory education, to have higher earnings and to have criminal convictions. However, due to limitations in the scale and quality of the evidence, it is not possible to conclude that creating connections between young people in care and caring adults leads to positive longer-term outcomes.

1.8 The role of mental health

Mental health was discussed throughout the literature, whether as a cause or a consequence of other disadvantages, or simply as a co-occurring disadvantage, perhaps with similar underlying roots, such as childhood trauma. Overall, there was a consensus in the literature that a sensitive consideration of mental health was an essential component of almost any intervention aimed at any of the population groups considered in the review. Many of the interventions included a secondary mental health component alongside a primary service. Others were primarily focused on mental health but also included elements that focused on other needs, such as homelessness. There were also interventions, such as IPS, that were originally developed for people with mental issues and have since been adapted for use with other populations.

1.9 Limitations in the evidence

An absence of evidence in this report on particular interventions should not be read as implying that these interventions are not effective. There are many relevant interventions that have not been robustly evaluated, and others that have been robustly evaluated but have not been included in an existing systematic review. Most of the included evidence comes from the US and may not be directly transferable to the UK context. Many studies in the literature looked at populations experiencing multiple forms of disadvantage, but it was rare for studies to attempt to disaggregate the impact of interventions for different groups, such as whether an intervention was more effective for ex-offenders or for people with substance misuse issues, or to test whether interventions were either more or less effective when participants were experiencing multiple forms of disadvantage.

Question 6

2.  Introduction

Accepted Answer

This chapter provides an overview of the policy context in which this review was conducted and sets out the research questions it aimed to answer.

2.1 Policy Background

From October to December 2024, the economic inactivity rate for people aged 16 to 64 was estimated at 21.5%.^{[footnote 1]} In November 2024, the government published the Get Britain Working white paper, which set out an ambition to bring this down to 20%, or an employment rate of 80%.^{[footnote 2]}

Part of this effort will involve supporting people with complex needs into work. This includes groups such as ex-offenders, people experiencing homelessness, care leavers, and those with substance misuse issues. These groups are less likely to be in employment than others and often experience other disadvantages such as poor mental health, domestic abuse, or childhood trauma:

among care leavers in England aged 19 to 21 years old, 39% were not in education, employment or training (NEET), compared to an estimated 13% of all young people aged 19 to 21^{[footnote 3]}
35% of homeless adults were registered as unemployed at the time of their making their homeless application in 2023/24 ^{[footnote 4]}
in the year to March 2024, only 31% of ex-offenders in England and Wales were employed six months after release from custody ^{[footnote 5]}
amongst adults entering treatment for substance misuse just 27% were in employment in 2023 ^{[footnote 6]}

There is also substantial overlap between these groups:

analysis of user data from offending, substance misuse and homelessness services in England estimated that annually, over 250,000 people have contact with at least two of these services and at least 58,000 people have contact with all three ^{[footnote 7]}
around a quarter of people in prison have been in care ^{[footnote 3]}
among single households assessed as homeless and owed a duty of care: 9% had drug support needs, 7% had alcohol support needs and 13% had an offending background ^{[footnote 4]}
a fifth of people starting treatment for substance misuse (20%) are homeless ^{[footnote 8]}

2.2 Research Aims

There is a large volume of published evidence on what works to support disadvantaged groups into work, including several high-quality systematic literature reviews. However, the evidence tends to focus on a single population group, or on a single type of intervention. There was a need to bring this evidence together to provide a holistic overview of what works across disadvantaged groups, including those experiencing multiple forms of disadvantage. To achieve this, the Department for Work and Pensions (DWP), supported by the Ministry of Housing, Communities and Local Government (MHCLG) and the Ministry of Justice (MoJ), commissioned the National Centre for Social Research (NatCen) to conduct a systematic review.

The systematic review aimed to answer two main research questions:

1. What works to help ex-offenders, care leavers, people experiencing homelessness, and those with substance abuse problems move closer to the labour market or enter work?

a. What do these interventions involve?

b. What works for whom?

c. Is what is effective similar across different types of disadvantage or are different approaches needed for different groups?

d. What interventions are effective for those with overlapping needs and disadvantages?

e. Where there are multiple barriers, which is most important to address first?

2. How are interventions best delivered and by whom?

a. What does the evidence say on who is best placed to deliver support; does it matter who the delivery organisation is?

b. Does a multi-provider approach work best to support multiple barriers and complex needs?

c. What approaches work best in joining up different support services as people move closer to and into employment?

d. What are the most effective ways to encourage those from disadvantaged groups to engage with and take-up support to maximise the effectiveness of interventions?

2.3 Overview of the report

The next chapter, Chapter 3, provides an overview of the methods used in the systematic review. Chapters 4 and 5 then provide an overview of the four population groups included in the review, and an overview of the kinds of outcomes that the review sought evidence on. Chapters 6 through 15 then provide detailed findings on all the main types of interventions included in the literature. Chapter 16 then brings together findings from multiple types of interventions to reflect on multiple disadvantage specifically. Lastly, Chapter 17 outlines the main limitations of the review and what the implications are for future research needs.

Question 7

3.  Methods

Accepted Answer

The study used a systematic review methodology. This involved a rigorous and transparent process for searching, screening, quality appraisal, data extraction and synthesis of the evidence. This chapter provides an overview of the approach. A more detailed description of the methodology, including the search strings and PRISMA flowchart can be found in the accompanying technical report.

3.1 Scoping phase

The review began with a rapid scoping phase. The aim was to get a clear sense of the scale and nature of the evidence base. This began with a steering group made up of representatives from DWP, MoJ and MHLCG who decided the initial scope of the review. The initial literature search used Scopus (a large multi-disciplinary database of academic literature) and Google Scholar. It also involved reaching out to a range of academic and non-academic experts to ask for recommended literature. Lastly, a short series of conversations were conducted with charities that represent each of the four mains populations of interest. These conversations helped the research team to understand more about the current research and policy landscape and assist with identifying important literature.

3.2 Search Strategy

The search involved two separate elements:

Academic literature was searched for using complex search strings developed in partnership with an evidence search specialist. The strings were adapted and applied to seven academic databases (Scopus, Web of Science, EconLit, PsychInfo, ABI/Inform, Business Source Premier and Sociological Abstracts).
Grey literature was searched for by identifying relevant organisations and searching their websites using Boolean search terms. Academics and key stakeholders from the voluntary and charity sector were also contacted to signpost relevant sources.

The search identified an initial longlist of 8,231 academic papers and 108 non-academic papers (grey literature).

3.3 Screening and Quality Appraisal

To be included, the sources had to meet the following criteria:

Studies had to have been conducted within the OECD.
Studies had to include samples of ex-offenders, care leavers, people with substance misuse issues or people with experience of homelessness.
Studies had to investigate the impact of interventions that explicitly designed to improve employment related outcomes. Studies were also included if the interventions actively aimed to address underlying barriers to employment, such as housing interventions, if there was also consideration of employment outcomes. Because of this, outcomes such as housing stability, mental and physical health, social integration and education were classed as relevant.
Studies could be experimental or non-experimental in design or evidence reviews (excluding non-systematic literature reviews). Study protocols were not in scope.
Only studies published post 1990 in English, Spanish, French and German were included.

Screening was conducted at both title and abstract and full text. Based on the titles and abstracts, it was initially judged that over 900 academic papers were likely to meet the inclusion criteria. As a result, a decision was taken to focus primarily on existing systematic reviews, with a small number of primary research studies included to fill notable gaps.

Quality appraisal was completed alongside full text screening. For systematic reviews, quality was assessed using the AMSTAR-2 tool^{[footnote 9]}. All other types of evidence were graded from A-C using a bespoke tool using a range of existing guidance, frameworks, and checklists^{[footnote 10]}. Any academic studies achieving a C grade, indicating low quality, were excluded.

After full-text screening, 77 studies were included in the review, including 47 systematic reviews, 13 academic primary research studies, and 17 non-academic studies.

The next chapter, Chapter 3, provides an overview of the methods used in the systematic review. Chapters 4 and 5 then provide an overview of the four population groups included in the review, and an overview of the kinds of outcomes that the review sought evidence on. Chapters 6 through 15 then provide detailed findings on all the main types of interventions included in the literature. Chapter 16 then brings together findings from multiple types of interventions to reflect on multiple disadvantage specifically. Lastly, Chapter 17 outlines the main limitations of the review and what the implications are for future research needs.

3.4 Data extraction and Analysis

A thematic framework was devised based on key research themes to extract information on the aims, types of interventions and evidence of impact within included reviews. The research team extracted the findings as reported, capturing both quantitative and narrative summaries.

A narrative synthesis of the extracted data was conducted. Data was synthesised and organised according to the types of interventions, the evidence of impact and lessons for delivery for each of the four populations. Within each intervention chapter we have produced a summary of findings table. These show the volume of evidence, the evidence of impact, and recommendations for future research, for each of the four populations. Table 1 explains what the information in these tables means. More detail on how these assessments were made is available in the Technical Report.

Table 1: Interpretation of summary tables

Volume of evidence:

Possible values	Interpretation
Low / medium / high	Based on the volume and quality of the literature.
Not applicable	The intervention is simply not relevant to the population.

Evidence of impact:

Possible values	Interpretation
No evidence of impact on an outcome area	There is evidence available, but it does not show a positive impact.
Evidence of promise on an outcome area	There is evidence available, and it shows a positive impact, but there are reasons to be cautious, such as limitations in study designs, or some studies not showing evidence of impact.
Evidence of positive impact on an outcome area	There is consistently strong evidence of impact.
Not applicable	The intervention is not relevant to the population, or that there is simply no evidence available.

Future research recommendation:

Possible values	Interpretation
No more research needed	There is already enough evidence and the conclusions are clear.
Higher quality evidence needed	There is a small volume of high-quality evidence, or a high volume of medium/low quality evidence.
More directly relevant research needed	There is good evidence for similar or overlapping populations but not the precise population of interest, or there is good evidence on the population but in a very different place or context.
Not applicable	The intervention is not relevant to the population.

Question 8

4.  Overview of the four population groups

Accepted Answer

This chapter provides an overview of the four population groups examined in this review: care leavers, ex-offenders, individuals with substance misuse issues, and people experiencing homelessness. It provides an overview of the size of each group, the disadvantages they face, and the main interventions identified in the literature. It then summarises the best available evidence on the extent to which these groups overlap.

4.1 People experiencing homelessness

Population overview

Between July and September 2024, 88,690 households in the UK declared themselves as homeless and underwent an initial homelessness assessment.^{[footnote 11]} Of these, 81,370 individuals were assessed as owed a duty to prevent or relieve homelessness, indicating that over 7,000 households, despite describing themselves as homeless, did not meet the criteria for support. The most common employment status of lead applicants was ‘registered unemployed’, accounting for 34.3% of those owed a duty in July to September 2024.^{[footnote 11]} In September 2024, there were 9,079 people estimated to be sleeping rough over the month in England.^{[footnote 12]} It is also important to acknowledge the issue of ‘hidden homelessness’, referring to households that are homeless but not reflected in official statistics, with charities like Crisis highlighting the lack of comprehensive data on the scale of these households in their Homelessness Monitor reports.^{[footnote 13]} Homeless individuals frequently experience poor health, restricted employment opportunities, and social exclusion.

Key interventions in the literature

Unsurprisingly, the main types of interventions identified in the literature for homeless populations were housing interventions (Chapter 11). The literature highlights various housing interventions, including short-term shelters, transitional housing, and permanent supportive housing. Case management interventions also often targeted those who were homeless, aiming to provide comprehensive support, addressing not only housing but also healthcare, employment, and other social needs (Chapter 9). Homeless populations were also included in other interventions that were not specifically aimed at addressing homelessness, such as Individual Placement and Support (IPS).

4.2 Ex-offenders

Population overview

For the period of July to September 2024, 14,920 people were released from prison in the UK, an increase from April to June 2024.^{[footnote 14]} In the 12 months to September 2024, 74,107 offenders were sentenced to community sentence orders.^{[footnote 15]} Ex-offenders in the UK face challenges in reintegrating into society, with the Ministry of Justice highlighting an overall proven reoffending rate of 26.5% for the January to March 2023 offender cohort.^{[footnote 16]} In relation to employment, in the year to March 2024, just 19.3% of offenders were employed at 6 weeks post release from custody.^{[footnote 5]} A 2016 survey commissioned by DWP found that 50% of employers would not consider employing an ex-offender.^{[footnote 17]}

Key interventions in the literature

Ex-offender specific interventions included inmate work programmes, which aim to provide vocational skills and work experience to those involved in the justice system (Chapter 7). Correctional education programmes also supported ex-offenders focusing on improving literacy, numeracy, and other educational outcomes (Chapter 10). Other interventions included Individual Placement and Support (IPS) and employment support programmes which were typically focused on people with mental health issues or substance misuse issues, but often included ex-offenders with this (Chapter 6). These aim to support disadvantaged individuals gain and sustain employment.

4.3 People with substance misuse issues

Population overview

According to Public Health England there were 310,863 adults in contact with drug and alcohol treatment services between 1 April 2023 and 31 March 2024.^{[footnote 6]} Among them, 70% were receiving treatment for drug misuse, while 30% were being treated for alcohol misuse. The overall number of people in treatment (310,863) increased by 7% since last year, the highest amount since 2009 to 2010.^{[footnote 6]} A considerable number of individuals with substance misuse issues remain untreated; for instance, in 2019 to 2020, there were an estimated 608,416 adults with alcohol dependence and 129,584 individuals using opiates and crack in England.^{[footnote 18]} Moreover, in 2023 to 2024, 77.6% of individuals with alcohol misuse issues and 57.2% of those with opiate and/or crack misuse issues were identified as not receiving the treatment they needed.^{[footnote 19]} Substance misuse is likely to co-occur with a range of issues, including mental health problems, unemployment, and homelessness, requiring comprehensive treatment and support services.

Key interventions in the literature

For people with substance misuse issues, substance misuse specific interventions included medication assisted treatment (MAT). MAT combines medication with counselling and behavioural therapies to treat people with substance misuse issues (Chapter 12). Individual Placement and Support (IPS) was also commonly tested specifically on this population (Chapter 6). IPS is an employment support programme tailored for individuals with mental health and substance misuse issues. This group was also commonly supported through community and peer mentoring interventions designed to reduce substance misuse and improve life stability.

4.4 Care-leavers

Population overview

In 2023 to 2024, 13,112 young people in England aged out of the care system on their 18th birthday.^{[footnote 3]} Care leavers in the UK often struggle with the transition to independent living, including gaining employment. The Department for Education found that in 2022, 39% of care leavers aged 19 to 21 were not in education, employment, or training (NEET).^{[footnote 3]} Further evidence from the UK’s Children’s Commissioner highlights the challenges faced by care leavers. In a report published in January 2023, the Commissioner noted that care leavers often lack the necessary support and resources to make a successful transition to adulthood.^{[footnote 20]} The report indicated that many care leavers face issues, such as housing instability, mental health problems, and a lack of access to higher education and employment opportunities. The literature included in this review indicates that many care leavers have a history of trauma, abuse or neglect.

Key interventions in the literature

The main types of interventions identified in the literature supporting care-leavers into employment included independent living programmes, extending care arrangements, and programmes that create connections with family and other adults. These interventions are detailed in chapters 13, 14 and 15. Independent Living Programmes (ILPs) provide life skills training and support, extending care arrangements allow care-leavers to remain in care settings beyond the age of 18, and programmes creating connections with family and other adults focus on building supportive relationships.

4.5 Overlap across populations

Experience of one of the above disadvantages is unlikely to be the only issue faced by an individual. Figure 1, adapted from a 2015 report by Lankelly Chase,^{[footnote 7]} shows that approximately 58,000 individuals annually accessed support services for homelessness, substance misuse, and offending simultaneously.^{[footnote 21]} Additionally, 99,000 dealt with both substance misuse and offending, 31,000 faced homelessness and offending, and 34,000 encountered homelessness and substance misuse. Poor mental health is a common co-occurring disadvantage and was reported by 19% of all working-age adults in the wider population, but this figure increased to 40% to 46% among those who had experienced homelessness or involvement with the criminal justice system. Although this data is from 2010/2011, it remains the most comprehensive representation of severe multiple disadvantages in the UK.

Figure 1: Overlap between disadvantaged groups, England, 2010/2011

One disadvantage	Two disadvantages	Three disadvantages
Ex-offenders: 112,000	Ex-offenders and substance misuse: 99,000	Ex-offenders, substance misuse, and homeless: 58,000
Substance misuse: 189,000	Substance misuse and homeless: 34,000
Homeless: 63,000	Homeless and ex-offenders: 31,000

Source: Lankelly Chase (2015)^{[footnote 7]}

More recent data from the UK Government’s Changing Futures Programme also confirms the substantial degree of overlap. The report, which included data from 905 participants covering five distinct forms of disadvantage including homelessness, drug or alcohol problems, mental health issues, domestic abuse and contact with the criminal justice system, revealed that only 1.7% experienced a single form of disadvantage.^{[footnote 22]} The research showed that 32.6% of participants faced four forms of disadvantage, and 50.4% experienced five forms. These findings underscore the necessity for holistic approaches to address the complex needs of these populations where single-issue interventions may be insufficient.

Individuals experiencing severe and multiple disadvantage face marginalisation in the labour market, with employment rates among those with all three disadvantages low at only 6.4%.^{[footnote 7]} Many were either unemployed or engaged in casual employment, with a substantial portion suffering from long-term illness or disability. Financial difficulties were also widespread, with 86% of respondents receiving UK benefits and many struggling with money management or debt. The economic impact of severe and multiple disadvantage was conservatively estimated by the study authors at £10.1 billion per year, underscoring the profound societal and economic burden.^{[footnote 7]}

While discussions of multiple disadvantages, including the statistics above, often exclude care-leavers, evidence suggests considerable overlap between care-leavers and other disadvantaged groups. For instance, data from Barnardo’s suggested care-leavers are five times more likely to face criminal justice involvement compared to their peers. ^{[footnote 23]} Additionally, many care-leavers struggle to secure stable housing: in 2023, 6% of care leavers aged 19 to 21 were deemed to be living in unsuitable accommodation.^{[footnote 24]}

Question 9

5.  Overview of outcomes

Accepted Answer

This systematic review considers a wide range of outcomes, not only employment outcomes. This chapter explains the range of outcomes covered in this review.

5.1 The pathway into work

The main aim of the review was to understand what works to help the four disadvantaged groups move closer to employment. A key outcome of interest was therefore entry into employment, but the review also looked at a wider range of outcomes, particularly outcomes prior to entry to employment. For example, for people experiencing homelessness, it can be inappropriate or impossible to seek out employment before finding stable housing. For people with substance misuse issues, it can be challenging to access work without first achieving a period of abstinence or in stable treatment.

For this review, the outcomes of interest can be thought of as a pathway, starting with essential life stability outcomes, such as housing, and ending with stable and sustained employment. In between are a range of other outcomes of interest, as shown in Figure 2 and discussed in section 5.2 below. This is not meant to imply that all disadvantaged groups must achieve all these outcomes in precisely this order before they can find work. Rather, the pathway provides a useful framework for thinking about what interventions are trying to achieve, how they might work, and how they might ultimately contribute to employment outcomes, even if employment is not the explicit aim of an intervention or is not measured.

Figure 2: Pathway into work

Stage 1: Life stability interventions

Stable housing and independent living
Reduced criminal activity
Community integration and access to services

Stage 2: Health and wellbeing interventions

Physical health
Mental health and wellbeing
Reduced substance misuse
Actively using mental health and other services
Self-esteem and motivation

Stage 3: Education and skills interventions

Entry into education or training
Staying in education or training
Gaining qualifications and skills
Core life skills including financial management

Stage 4: Employment interventions

Improved job searching skills
Improved job application skills
Entry into work
Earnings
Staying in work longer term

Ideally, we would have good evidence for every outcome on this pathway, for every possible intervention, for each of the four population groups of interest. In practice, this is not the case. For all four groups, there is evidence on employment outcomes. But, understandably, the evidence on homeless populations focused much more on housing outcomes than on other types of outcomes. The evidence on ex-offenders focused more on reoffending, the evidence on those with substance misuse issues focused more on abstinence, and the evidence for care leavers focused more on independent living. Relatedly, most evaluations of interventions in the literature focused on only a small number of the possible outcomes, and not all reported employment outcomes.

5.2 Overview of outcome categories

This section provides more detail on the types of outcomes included under the four main categories in the pathway.

Life stability outcomes

Outcomes under the life stability umbrella all contribute to building a secure foundation that enables individuals to move closer to employment. This includes stable housing, independent living, reduced criminal activity, and community integration.

Stable housing and independent living. Stable housing refers to a living situation, where individuals have consistent, safe, and adequate shelter over a sustained period. The literature was not consistent in defining what counted as ‘sustained period’. Independent living outcomes relate to the capacity of individuals to manage their own lives and maintain self-sufficiency.

Reducing criminal activity involves steering individuals away from criminal behaviour, promoting positive life choices and decreasing the incidence, frequency, and severity of criminal behaviours.

Community integration and access to services. Community integration involves the successful re-entry of individuals into society through becoming an active and engaged member of their communities. Access to services involves the ability of individuals to obtain necessary well-being services within their community.

Health and wellbeing outcomes

Health and wellbeing outcomes aim to improve the overall health of individuals so that they are in a better position to enter employment. This includes physical health, mental health, reduced substance misuse and increased self-esteem and motivation.

Physical health outcomes involve measurable changes in the physical well-being of targeted populations. These may include improvements to unique health challenges faced by disadvantaged groups and the promotion of overall health.

Mental health and wellbeing refer to the measurable changes in the psychological well-being and mental health status of these populations. These may include improvements in emotional stability, reductions in symptoms of mental health disorders such as depression and anxiety, enhanced coping mechanisms, increased resilience, and overall improvement in quality of life.

Reducing substance use includes the decrease in the frequency, quantity, and harmful consequences of alcohol and drug consumption within targeted populations.

Self-esteem and motivation include measurable changes in an individual’s perception of their own worth and value, with improvements in self-respect and confidence. These can also relate to the degree to which individuals are driven to achieve personal goals and make positive changes in their lives.

Education and skills outcomes

Education and skills outcomes focus on educational achievement and skill development. This includes entry into education or training, remaining in these programmes, gaining qualifications and skills, and developing core life skills.

Entry into education or training, involves the process of enrolling in formal or informal educational programmes or training courses aimed at enhancing knowledge and skills.

Remaining in education or training involves the sustained participation in formal or informal educational programmes or training courses.

Gaining qualifications and skills involves the achievement of recognised certifications, diplomas, or degrees, as well as the development of specific competencies and proficiencies.

Developing core life skills involves the enhancement of essential personal and social abilities, such as communication, problem-solving, time management and financial management.

Employment outcomes

Employment outcomes aim to improve people’s job prospects and career development. This includes improving job searching and application skills, facilitating entry into work, increasing earnings, and ensuring long-term job retention.

Improving job searching and application skills refers to the improvement of abilities related to identifying job opportunities, crafting effective CVs and cover letters, and performing well in job interviews.

Facilitating entry into work involves the transition of individuals into any form of part-time, voluntary or full-time employment.

Long-term job retention involves the sustained ability to maintain employment over an extended period. There was no unanimous definition of a time-period for sustained employment in the literature.

5.3 What works for each stage of the pathway?

This section provides a summary of the findings of the review for each of the four main stages of the pathway. It includes cross references to the relevant sections of the report that contain more detail on each finding.

Life stability outcomes

The literature contained the following evidence of impact:

case management interventions for homeless populations can improve housing outcomes (Chapter 9)
housing interventions can improve housing outcomes for all four disadvantaged groups (Chapter 11)
correctional education programmes for people in prison can reduce rates of reoffending (Section 10.2)
inmate work programmes can reduce rates of reoffending (Chapter 7)

The literature contained the following evidence of promise:

extending care arrangements for care leavers past the age of 18 may improve housing outcomes and reduce rates of offending (Chapter 15)
creating connections with family and other adults for care leavers may reduce rates of offending (Chapter 14)
independent living programmes for care leavers may improve housing outcomes and rates of offending (Chapter 13)
Medication Assisted Treatment for people with substance misuse issues may reduce rates of reoffending (Chapter 12)

Health and wellbeing outcomes

The literature contained the following evidence of impact:

housing interventions can improve health outcomes for all four disadvantaged groups (Chapter 11)
community and peer mentoring interventions can reduce substance misuse for people with substance misuse issues (Chapter 8)

The literature contained the following evidence of promise:

extending care arrangements for care leavers past the age of 18 may improve health outcomes (Chapter 15)
creating connections with family and other adults for care leavers may improve health outcomes (Chapter 14)

Education and skills outcomes

The literature contained the following evidence of promise:

extending care arrangements for care leavers past the age of 18 may improve educational outcomes (Chapter 15)
independent living programmes for care leavers may improve educational outcomes (Chapter 13)
campus support programmes may improve educational outcomes for care leavers (Section 10.3)

Employment outcomes

The literature contained the following evidence of impact:

IPS can improve employment rates for ex-offenders, people with substance misuse issues, and people experiencing homelessness (Chapter 6)
some employment support interventions can improve employment rates for ex-offenders and people with substance misuse issues (Section 6.2)

The literature contained the following evidence of promise:

extending care arrangements for care leavers past the age of 18 may improve employment outcomes (Chapter 15)
creating connections with family and other adults for care leavers may improve employment outcomes (Chapter 14)
independent living programmes for care leavers may improve employment outcomes (Chapter 13)
some employment support interventions may improve employment rates for care leavers and people experiencing homelessness (Section 6.2)
correctional education programmes for people in prison may improve employment rates post-release (Section 10.2)
community and peer mentoring interventions may improve employment rates for ex-offenders and people with substance misuse issues (Chapter 8)
campus support programmes may improve employment rates for care leavers (Section 10.3)
inmate work programmes may improve employment outcomes post-release (Chapter 7)

Question 10

6.  IPS and Employment Support

Accepted Answer

This chapter summarises the evidence on employment-focused interventions. It first presents the evidence on Individual Placement and Support (IPS). This is followed by an examination of other employment support interventions.

6.1 Individual Placement and Support (IPS)

Summary of findings for IPS

The IPS model places individuals directly into employment and then provides the necessary support to sustain that employment. It was originally designed for populations with severe mental health issues and there is strong evidence that it has a positive impact on employment for this group.
It has since been tested with a wider range of groups. There is clear evidence that IPS can be effective for a broader range of groups beyond those defined solely by mental health issues.
Of the four disadvantaged groups considered in this review, the evidence is strongest for people with substance misuse issues. Those with a drug use disorder had almost triple the odds of obtaining employment, compared to standard addiction treatment.
Evidence shows that IPS is effective for ex-offenders and homeless populations as well, but these groups were included within broader studies so the impact of IPS cannot be isolated.

	Ex-offenders	Substance misuse	Homeless	Care leavers
Volume of high-quality evidence	Medium	High	Medium	Low
Evidence of positive impact	Evidence of positive impact on employment outcomes	Evidence of positive impact on employment outcomes	Evidence of positive impact on employment outcomes	NA
Future research recommendation	More directly relevant research recommended	No further research needed	More directly relevant research recommended	NA

IPS interventions in the literature

IPS is designed to help individuals with various disadvantages secure and retain competitive employment. The IPS model is distinctive in its approach as it places individuals directly into employment and then provides the necessary support to sustain that employment. This contrasts with other employment support models, discussed later in this chapter, that typically prepare individuals for work before placing them into jobs (Bond et al., 2019; Probyn et al., 2021).

There is strong evidence supporting the effectiveness of IPS in improving employment outcomes for populations with severe mental illness; the original target population for the intervention. Due to this strong evidence, IPS has been tested with other populations beyond those with severe mental health problems. Examples include ex-offenders, individuals with chronic pain, people with experience of substance misuse, people experiencing homelessness and NEET young adults (not in employment, education or training) (Whitworth et al., 2024; Probyn et al., 2021; Bond et al., 2019). Within these groups mental health issues are more common than among the wider population.

Because mental health issues often overlap with other disadvantages like substance misuse issues, homelessness, and having a criminal conviction, variations of IPS continue to be delivered according to the IPS fidelity manual^{[footnote 25]}, providing individualised mental health support alongside rapid job placement (Probyn et al., 2021, Whitworth et al., 2024). For instance, IPS was provided to people with substance misuse issues enrolled in a methadone treatment programme in Oregon, US, with an IPS specialist attending weekly treatment team meetings (Lones et al., 2017, cited in Magura and Marshall, 2020; Bond et al., 2019).

However, modifications to IPS do occur. For example, LePage et al, (2016) combined IPS with a group-based vocational programme called About Face for ex-offenders with substance misuse issues and in doing so, removed some of the core principles of IPS, maintaining ‘fair’ fidelity but including large caseloads, no integration with treatment and mandatory group attendance (cited in Whitworth et al., 2024; Magura and Marshall, 2020; Bond et al., 2019; Connell et al., 2023). The literature did not discuss the implications of modifications beyond reporting IPS fidelity scores.

Most examples of IPS are based in the US, but implementation in the UK has increased in recent years. In the UK, IPS was rolled out across 90 local authorities in England starting in 2018, based on a model developed in the US, with a focus on helping people with alcohol and drug dependence (Office for Health Improvement & Disparities, 2024; Marsden, 2024).

The scale and quality of the evidence

The strongest evidence for IPS’s effectiveness took a broad view, looking at any populations beyond those solely defined by problems with their mental health. For example, the Whitworth et al., (2024) review included any primary studies which looked at the impact of IPS for groups other than those with severe mental health issues. Within this evidence base, there was evidence on the impact of IPS for ex-offenders, homeless populations, and people with substance misuse issues.

Substance misuse. There is high-quality evidence on the impact of IPS interventions that are specifically focused on people with substance misuse disorders. For example, Hellström et al., (2021) conducted a meta-analysis of the effects of IPS on various subgroups including those with substance use issues. Additionally, recent applications of IPS in the UK for people with substance misuse issues have been evaluated through two recent high quality randomised controlled trials (RCTs) (Whitworth et al., 2024; Marsden, 2024). There is variability in the quality of primary studies, but overall, there is enough good evidence to be confident in our conclusions about substance misuse populations.

Ex-offenders and homelessness. There was no evidence on IPS interventions specifically focused on homeless populations or ex-offenders. Instead, people experiencing homelessness and ex-offenders are included within studies that have a wider focus (Probyn et al., 2021; Whitworth et al., 2024). Note that the included systematic reviews did not provide detail on the exact populations included in these studies, such as whether the studies included people currently experiencing homelessness, or at risk of homelessness, or who recently experienced homelessness. While we can draw some reasonably clear conclusions for people experiencing homelessness and ex-offenders, there is a need for evaluations of IPS interventions that specifically target these groups, and that take place in the UK.

Limitations of the evidence. Reviews often drew from the same pool of primary studies. For example, Lones et al. (2017), a study of IPS for people with substance misuse issues, is cited across many reviews, and therefore the volume of primary evidence should not be overstated, particularly as the risk of bias in this study was rated as high (Probyn et al., 2021).

Primary studies often had methodological limitations. For example, some did not use an experimental design, or had small samples. In some cases, it was not clear how closely an intervention had stuck to the principles of IPS (Bond et al., 2019). However, this variation in study quality was typically accounted for by systematic reviews, for example by checking whether findings changed when excluding lower quality studies. This means we can be confident in our overall conclusions.

Evidence of impact

The literature on IPS unsurprisingly centres around entry-to-employment outcomes. Consideration is also given to life stability outcomes, specifically housing and health (including mental health). The overall findings indicate that IPS improves employment outcomes for disadvantaged groups, including those with substance misuse issues, ex-offenders, and homeless individuals, compared to control groups and this improvement is statistically significant. Additionally, IPS has positive effects on health outcomes like self-esteem and mood.

Any groups other than those defined by mental health. Strong evidence indicates that IPS effectively improves employment outcomes for disadvantaged groups beyond those with severe mental illness. For example, a meta-analysis of 18 IPS interventions delivered to populations with a combination of disadvantages improved the job entry rate for people with various disadvantages compared to those who did not receive the interventions, with a statistically significant positive impact (Whitworth et al., 2024). The job entry rate for those taking part was high across all studies, ranging from 25% to 70% for people taking part in IPS. The delivery of IPS also resulted in more long-term employment than for control groups. However, follow-up times differed across interventions, ranging from 22 days to 2 years, meaning definitions of sustained employment varied (Probyn et al., 2021). One study, for example, found that veterans with PTSD who engaged in IPS worked an average of 656 hours over a one-year period, compared to 236 for the control group (Davis, 2012, cited in Probyn et al., 2021).

Substance misuse. The evidence suggests IPS has a significant positive impact on employment outcomes for people with substance misuse issues, although results vary by severity and type of substance use. Participants with any drug use disorder had 2.95 higher odds of obtaining employment following IPS compared to standard addiction treatment (Hellström et al., 2021). They also worked significantly more compared to those receiving standard treatment and returned to work faster: over an 18-month period follow-up period, the treatment group worked an average of 222 hours, compared to 144 hours for the control group. However, there was no evidence of impact on substance misuse outcomes: in Marsden (2024), IPS showed no statistically significant impact on alcohol consumption, opioid use, drug injecting days, cannabis use, amphetamine or cocaine use, or remission status.

Ex-offenders and homelessness. Evidence demonstrates that IPS is effective for ex-offenders and homeless populations, although it is often included within broader studies on multiple disadvantage (Hellström et al., 2021). Notably, Ferguson et al. (2012) found that homeless individuals with severe mental health and substance misuse issues in the IPS group worked significantly more months than the control group (cited in Whitworth et al., 2024): the IPS group worked an average of 5.2 months over a 10-month period, compared to 2.2 months for the control group. Similarly, LePage et al. (2016) reported that 46% of the IPS group (a population containing ex-offenders with substance use issues) gained employment, compared to 21% of the control group, with the IPS group also working more hours and earning higher wages (cited in Whitworth et al., 2024; Magura and Marshall, 2020; Bond et al., 2019; Connell et al., 2023).

Evidence also indicates that IPS has beneficial effects on health outcomes beyond employment, particularly in enhancing self-esteem, mood, and attention among homeless populations (Marshall et al., 2022).

Lessons for delivery

The literature did not give much attention to lessons for delivery beyond affirming the value of the rapid job searching and the zero-exclusion policy, which may successfully alleviate concerns held by employers about how ready individuals are for employment by immediately demonstrating their ability to work (Hellström et al., 2021).

Local barriers and facilitators significantly impacted IPS implementation, requiring case-by-case approaches as reported in a high-quality process evaluation of an IPS trial for alcohol and substance use in the UK (Hofman et al., 2024). This evaluation found that sites like Staffordshire and Derbyshire faced challenges implementing IPS across all treatment service teams due to geographical dispersion. Urban areas like Sheffield and Birmingham benefited from local job markets, while areas like Brighton faced high competition for entry-level jobs.

The evaluation also found that the recruitment, development, and retention of the IPS delivery workforce were identified as key factors for successful implementation. Employment specialists with the right skills and backgrounds were crucial for building personal relationships with keyworkers and clients. Sites that experienced staffing issues saw lower fidelity scores, meaning that the delivery of the service was less closely aligned with established IPS procedures. Training opportunities were essential and well-received. The study concluded that recruiting employment specialists with the right skills and providing ongoing training to improve these skills are prerequisites for effective IPS implementation.

6.2 Other types of Employment Support

Summary of findings for other employment support interventions

Employment support interventions involve preparing individuals for work in various ways. This can include vocational training, internships and work programmes, and funding to help people start businesses.
Existing systematic reviews tended to group together highly diverse interventions and as a result were unable to draw clear overarching conclusions about employment support interventions as a whole. Much of the primary literature was relatively low quality or found no evidence of impact.
However, within the existing systematic reviews there were examples of individual studies that showed positive evidence of impact for some interventions for some populations. For example, a small RCT of a ‘transitional employment’ intervention for ex-offenders found that 73% were employed during 1.5-year follow-up, compared to 43% of the control group.

	Ex-offenders	Substance misuse	Homeless	Care leavers
Volume of high-quality evidence	Medium	Medium	Low	Low
Evidence of positive impact	Evidence of positive impact on employment outcomes	Evidence of positive impact on employment outcomes	Evidence of promise on employment outcomes	Evidence of promise on employment outcomes
Future research recommendation	Higher quality evidence needed	Higher quality evidence needed	Higher quality evidence needed	Higher quality evidence needed

Employment support interventions in the literature

As discussed above, employment support interventions involve preparing individuals for work by providing pre-employment skills.

The literature covers an extensive range of interventions which can be broadly categorised into vocational training programmes, internship and work programmes and business venture programmes. However, unlike IPS which follows a standardised model, the employment support interventions found in the literature were almost all unique. While some were robustly evaluated, others were not. The variation in both the interventions and the evaluations of them means systematic reviews in this area generally struggled to reach clear overarching conclusions.

Vocational Training Programmes

Vocational training programmes target disadvantaged populations by integrating various forms of job-coaching to prepare individuals for work. Within the included systematic reviews, there was often little description of what the interventions entailed. Some examples included:

a combination of work-related cognitive behavioural therapy (CBT) and vocational skills training, job placement and housing provision (Marshall et al., 2022)
The Fairbridge programme, which targeted marginalised young people, including care-leavers, to gain skills and stabilise their lives, aiming to move them into education, employment, or training (Gibb and Edwards, 2017)
a series of ex-offender programmes in the US which included holistic and vocational support. There were three main examples in the literature:
- The Milwaukee Safe Streets Prisoner Reintegration Initiative (PRI), had several stages: first, there was a 6-month programme prior to release in which prisoners had access to a social worker who assessed their skills and needs; second, they were given access to vocational skills training, a restorative justice programme, and drug/alcohol treatment; and third, 30 days before release they were met by a coordinated care team who helped them to organise plans for release, such as housing, transport, documents, and job searching (Cook et al. 2014)
- another example is the OPTIONS intervention in Oregon, which involved 5 weekly group sessions of 6 to 7 male inmates focused on job preparation: obtaining information about jobs, practicing interviewing skills, and learning how to utilise social support (Cook et al., 2014)
- The Minnesota Comprehensive Offender Re-entry Plan (MCORP) involved both pre- and post-release elements. Pre-release, case workers helped inmates to establish a ‘transition accountability plan’. Post-release, agents helped ex-offenders to access services for employment, vocational training, education, housing, substance misuse, faith-based services and income support (Duwe 2012, cited in Connell et al., 2023)

Internship and work experience

These interventions provided work experience with the aim of providing individuals with practical skills to improve their employability. There were two specific examples in the literature:

transitional employment was an intervention for ex-offenders in which participants were placed in temporary subsidised jobs, given case workers, and required to attend therapy sessions. Local employers were engaged to provide work, including construction companies, kitchens and warehouse/retail companies. There was support to help participants find permanent work after the temporary jobs ended, but no guarantee of work (Atkin-Plunk, 2023)
simulated employment programmes were experimental interventions for populations struggling with substance misuse, where those delivering the programme acted as employers in a controlled setting (Walton and Hall, 2016)

Business venture programmes

Business venture programmes were designed to help individuals generate their own business ideas to move them into self-employment. These included social enterprise interventions (SEIs) which engaged individuals experiencing social disadvantage in developing business ventures. The aim of these interventions was to generate income and equip participants with self-employment skills (Marshall et al., 2020).

The scale and quality of the evidence

Because employment support interventions beyond IPS were so varied, interventions and outcome measures were often too heterogeneous to draw substantial conclusions about the overall impact on employment.

Ex-offenders. The evidence base is strongest for ex-offenders, where interventions were more similar. These interventions were evaluated through robust RCTs (Connell et al., 2023, Cook et al., 2014). However, the evidence comes from a highly specific US context and as such has limited generalisability to the UK context.

Substance misuse. For people with substance misuse issues, interventions were evaluated within high-quality systematic reviews, but the quality of the underlying evidence was often still mixed. We can draw some reasonably strong conclusions for some interventions, but further high-quality research is needed. For example, for work-related cognitive behavioural therapy (CBT) and simulated employment, there was a high-quality evidence base in systematic reviews based on RCTs (Marshall et al., 2022; Walton and Hall, 2016). However, studies about transitional employment indicated the need for additional research due to small sample sizes (Atkin-Plunk, 2023).

Homelessness. Evidence on interventions specifically targeting homeless populations was limited. For example, the quality of evidence on social enterprise interventions was mixed and when RCTs were employed, these were often lacking in rigour (Marshall et al., 2022). A substantial amount of the evidence about care-leavers relied on small-scale qualitative studies and as such cannot be generalised (Furey and Harris-Evans, 2021, Gibb and Edwards, 2017, Harrison et al., 2023)

Evidence of impact

Evaluations of employment support interventions primarily focused on employment outcomes, though some also assessed re-offending and health outcomes like substance misuse. Overall, there was some positive indication that certain employment support interventions can increase employment for ex-offenders and those with substance misuse issues. However, these findings were limited to either US contexts or specifically to internships and work placements.

Vocational training programmes

There was evidence of impact for the ex-offender programmes in the US:

The Milwaukee Safe Streets Prisoner Reintegration Initiative described above, which was an intensive and holistic intervention, found that 81% of the treatment group had earnings in first 12 months after release compared to 59% of the control group (Cook et al., 2014)
the OPTIONS programme, a much less intensive programme, did not measure employment outcomes but instead looked at a range of psychological metrics including career-search self-efficacy, problem-solving, and hopefulness. There were modest but statistically significant improvements on these metrics for the treatment group, as compared to the control group (Cook et al., 2014)
The Minnesota Comprehensive Offender Re-entry Plan (MCORP), which largely involved helping ex-offenders to access a range of existing services, had an impact on employment outcomes. After 6 months, 55% of the treatment group were employed, compared to 39% of the control group (Connell et al. 2023)

Across all other populations no overall conclusion can be made about the effectiveness of vocational training programmes for helping disadvantaged groups into work. Work-related CBT increased job search activities for homeless populations with substance misuse issues but not the number of hours worked (Marshall et al., 2022). For combined substance use and vocational training programmes the evidence is too mixed to conclude there was an impact.

For care-leavers the qualitative data suggests a positive impact on employment for interventions such as the Fairbridge programme. However, these findings are not robust, due to the lack of comparison groups and reliance on qualitative data (Gibb and Edwards, 2017).

Internships and work experience

There was evidence of impact on employment outcomes for internships and work experience programmes, although this finding should be interpreted with some caution due to the relatively small scale of the evidence it is based on. A small RCT found positive outcomes for transitional employment for ex-offenders, with 73% of the treatment group employed during the 1.5-year follow-up, compared to 43% of the control group (Atkin-Plunk, 2023). No impact on re-offending was reported but authors attributed this to the small sample size.

Simulated employment programmes showed significant benefits for people with substance misuse issues. Participants receiving employment interventions were more likely to secure and retain employment. Most studies found a relationship between employment and reduced substance use among individuals recovering from substance misuse issues, consistent across different populations (Walton and Hall, 2016).

Business venture programmes

The literature on Social Enterprise Interventions (SEIs) focused on community integration for people experiencing homelessness, rather than explicit employment outcomes. However, there were too few studies with mixed results to conclude the impact of SEIs on social integration for homeless populations.

Lessons for Delivery

Despite limited attention to lessons for delivery within the included literature, a general conclusion was that successful interventions often employ holistic, wrap-around services addressing multiple needs beyond employment, such as housing, mental health, and substance use treatment, for example the Milwaukee Safe Streets PRI’s success was attributed to comprehensive support both pre- and post-release (Cook et al., 2014). Individualised support and relationship-building were crucial in care-leaver programmes, where emotional support and personalised learning were key to success (Furey and Harris-Evans, 2021).

Question 11

7.  Inmate work programmes

Accepted Answer

This chapter summarises the evidence on inmate work programmes for ex-offenders. These were programmes that supported inmates to gain vocational skills and experience through low paid employment while in prison or through early supervised release.

7.1 Summary of findings

Inmate work programmes provide prisoners with employment, typically manual labour. The aim is to generate revenue for the prisons service, but research has also investigated the impact on employment and reoffending after release.
All the evidence came from a single US state with a history of inmate work programmes, meaning the findings cannot be easily transferred to the UK context.
However, there was some promising evidence that these programmes can improve employment outcomes. In one programme, 84% of inmates found work within two years of release compared to 45% of the control group. There were also positive impacts on rates of rearrest and reconviction.

	Ex-offenders
Volume of evidence	Medium
Evidence of impact	Evidence of promise on employment outcomes
Evidence of positive impact on reoffending outcomes
Future research recommendation	More directly relevant research needed

7.2 Interventions in the literature

Inmate work programmes are initiatives that provide prisoners with opportunities for low-paid employment, often involving manual labour. Some programmes offer the possibility of early supervised release, contingent upon finding work. The primary goal of these programmes is to generate revenue for the prisons service, but a secondary aim is to help inmates develop positive work habits and marketable job skills that can assist them in securing employment after their release.

All three inmate work programmes identified in the literature were assessed in the US, in a single State (Minnesota), and were only open to inmates considered to be low risk. These programmes involved inmates producing different goods or working on construction projects (Duwe and McNeeley, 2020, Bohmert and Duwe, 2012). The Minnesota Correctional Industries programme (MINNCOR) was established in 1993 and involves inmates producing furniture, prints and clothing for state operated companies at low pay. The Minnesota Affordable Homes programmes began in 1998 and involved inmates building and renovating homes for $1 per hour, which were then sold to low-income families. The Minnesota Prison Work Release programme started in 1967 and offered early supervised released to prisoners into secure residential facilities. Prisoners were expected to find work or participate in an approved vocational programme (Duwe, 2014).

7.3 The scale and quality of the evidence

Evidence on the effectiveness of inmate work programs was of reasonably high quality. However, as all the evidence comes from a single US state, their findings may not be generalisable to the UK.

All three identified papers were quasi experimental impact evaluations that generated a comparison group using propensity score matching (Duwe and McNeeley, 2020, Bohmert and Duwe, 2012, Duwe, 2014). Participants and non-participants were matched based on race, criminal record, and disciplinary history. This meant the authors were able to control for the impact of these factors on their findings.

However, Minnesota has an established history of inmate work programmes. It is likely that local employers are more familiar with the approach than elsewhere, which may limit the transferability of the findings.

7.4 Evidence of impact

Identified studies showed promising evidence that inmate work programs improved outcomes related to employment and reoffending. However, there was evidence that the impact on employment was lower for inmates with additional disadvantages and was specific to the type of work inmates were gaining experience in.

For example, inmates who took part in one programme were found to be 24% more likely to secure a job in the first year following release compared to the non-participating comparison group (Duwe and McNeeley, 2020). Similarly, 84% of participants in another programme had found work within two years of release compared to 45% of the comparison group. However, employment rates for inmates from minority ethnic groups and inmates with suicidal tendencies were significantly lower than the group average.

One programme involved inmates working on construction sites, and the evaluation found that participation improved chances of participants getting a job in the construction industry, but not in other sectors (Bohmert and Duwe, 2012). However, there were some issues with the comparability of the comparison group in this study.

Studies also found that inmates participating in work programmes had a reduced likelihood or rearrest and reconviction compared to comparison groups. For example, the Minnesota Work Release Programme reduced chances of rearrest by 16%, reconviction by 14%, and reincarceration by 17%, with follow-up periods ranging from 24 to 72 months.

Question 12

8.  Community and peer mentoring

Accepted Answer

This chapter summarises the evidence on community and peer mentoring interventions. This included five main types of interventions: residential interventions for ex-offenders with substance misuse issues; non-residential peer mentoring interventions for ex-offenders; peer recovery support interventions specifically targeting substance misuse, often involving 12-step programmes; peer support communities for people experiencing homelessness; and community-based emergency housing for homeless youth.

8.1 Summary of findings

There were three kinds of community and peer mentoring interventions that showed promise:

12-step recovery programmes involve a group of peers helping each other to maintain abstinence (e.g. Alcoholics Anonymous and Narcotics Anonymous). Three RCTs found they helped to improve abstinence, but not other outcomes such as employment
‘recovery housing’ interventions are residential settings for ex-offenders emphasising abstinence from drugs/alcohol. Evidence was limited but showed promising results for employment outcomes. Oxford Houses, which are entirely peer-led, were found to be particularly cost effective in one study
‘credible messenger mentoring’ involves older peer mentors for young people involved in the criminal justice system. The available evidence was relatively low quality, but there was some evidence of promise for rates of reoffending

Volume of high-quality evidence

Ex-offenders	Medium
Substance misuse	Medium
Homeless	Low
Care leavers	Low

Evidence of positive impact

Ex-offenders	Evidence of promise for substance misuse and employment outcomes
Substance misuse	Evidence of promise for employment outcomes. Evidence of positive impact for substance misuse outcomes

Future research recommendation

Ex-offenders	Higher quality and more relevant evidence needed
Substance misuse	Higher quality and more relevant evidence needed
Homeless	Higher quality and more relevant evidence needed
Care leavers	N/A
Homeless	N/A
Care leavers	N/A

8.2 Interventions in the literature

The literature discusses five main types of community and peer mentoring interventions. Some interventions aimed at care leavers included a peer mentoring component, but only as part of a broader independent living programme, and as such are discussed in Chapter 13.

Peer support for substance misuse (often involving 12-step programmes)

The literature included a wide range of peer recovery support services for people with substance misuse issues, mainly in the US (Eddie et al., 2019). Several interventions involved peer volunteers from 12-step programmes (such as Alcoholics Anonymous and Narcotics Anonymous) who helped connect patients receiving outpatient treatment for substance misuse issues to 12-step programmes in the community.

One intervention involved one brief peer education session, written advice, a referral list, and a follow up call. Another involved “Citizenship Training”, which consisted of a series of classes to support social participation and community integration, combined with peer support and standard clinical treatment.

Residential interventions for ex-offenders with substance misuse issues

Known generally as “recovery housing”, these interventions were all in the US, and there were several distinct varieties. In all versions, participants were provided with housing alongside peers and were required to maintain abstinence from drugs or alcohol. Most of these interventions in the literature were specifically focused on ex-offenders who had substance misuse issues. There were the following varieties:

Therapeutic Communities. These are professional, staff-run residential settings. They vary widely in their size, level of staffing, staff training, and included services. They often incorporate mutual self-help elements. Staff and residents share responsibilities, and residents typically stay between 6 to 15 months (Jason et al. 2015 cited in Connell et al., 2023, Reif et al., 2013, Magura and Marshall, 2020).
Oxford Houses. In contrast to Therapeutic Communities, Oxford Houses are entirely self-run. There are no formal substance misuse interventions, and there is no onsite access to health care or substance misuse professionals. Residents live together in moderately sized, single-sex, single-family homes, and can remain indefinitely. They provide each other with an abstinent social support network. Oxford Houses are democratically run, abstinence is mandatory and residents who relapse may be asked to leave. They were first introduced in the 1970s and as of 2015 there were c.1,700 across the US (Jason et al. 2015 cited in Connell et al., 2023, Reif et al., 2013, Magura and Marshall, 2020).
Sober Living Houses. These are similar to Oxford Houses. They are small, peer-run residences where individuals are expected to pay rent, work and maintain household responsibilities (Reif et al., 2013).

Non-residential peer mentoring for ex-offenders

There were two non-residential peer mentoring interventions for ex-offenders discussed in the literature, both in the US. Credible Messenger Mentoring (CMM) focuses on mentoring young people involved in the juvenile legal system, using mentors with lived experiences, such as prior incarceration, to reduce violence and crime and to shift young people’s thinking and behaviour (Lesnick et al., 2023). For example, the Arches programme was a 6- to 12-month programme for youth sentenced to probation in the community, with a high risk of reoffending. It included one-to-one support alongside group sessions facilitated by credible messenger mentors, which included a CBT journalling exercise.

The Mentors Empowering Now to Overcome Recidivism (MENTOR) programme in Philadelphia also follows a peer mentoring model. Although all mentors have some relevant experience, not all are ex-offenders: for example, some are attorneys. Mentors support participants through regular contact and meetings within a “problem-solving court”, a specialised court that tries to divert people away from the criminal justice system and towards community-based treatment (Taylor, 2021).

Peer support for people experiencing homelessness

One evidence review looked at peer support communities for people experiencing homelessness in the US. Individuals were recruited to form the community, and within this, a subgroup of homeless individuals who were willing to provide peer support to others were identified. Group sessions were then delivered by an occupational therapist and a peer support worker over 9 months (Marshall et al., 2020).

Community-based emergency housing for homeless youth

One grey literature study evaluated the Nightstop service in the UK. This is a same-night emergency accommodation service that links young people in crisis with a trained volunteer host, who accommodates them in their own home, run by Depaul UK. If a young person requires accommodation for more than one night, they might be moved to another host or stay with the same host for a period of time. The service works with the young person to help find them more permanent accommodation. As of 2017, there were 31 Nightstops in the UK (McCoy and Kempton, 2018).

8.3 The scale and quality of the evidence

The evidence on peer recovery support interventions for substance misuse was reasonably strong. These interventions were covered in a single systematic review, which drew on several RCTs. However, the review was narrowly focused on the impact of interventions on substance misuse outcomes, and did not include any evidence on employment outcomes (Eddie et al., 2019).

Recovery houses, including Therapeutic Communities and Oxford Houses, were discussed in three systematic reviews (Reif et al., 2013, Connell et al., 2023 and Magura and Marshall, 2020), although all focused on the same small set of studies. The reviews cited Jason et al. (2015), an RCT that compared substance misuse and employment outcomes for ex-offenders who participated in Therapeutic Communities and Oxford Houses to those who received usual services. In this study, 270 individuals who had been released from prison within the last two years were recruited through inpatient substance misuse facilities and randomly assigned to either a Therapeutic Community, Oxford House or usual services.

Credible Messenger Mentoring interventions were also covered in a single systematic review. This review included several primary studies, only one of which was a quantitative impact evaluation with a comparator group (the Arches programme discussed above). The remaining studies included a mix of correlational quantitative findings and detailed qualitative findings (Lesnick et al., 2023).

The MENTOR programme was evaluated in a single study. This did use a quasi-experimental design, but the sample was small (only 38 in the treatment group), there were substantial differences between the treatment and comparison groups, and the treatment group had low engagement with the employment services offered. Conclusions about impact on any outcomes are unlikely to be reliable (Taylor, 2021).

The peer support community for people experiencing homelessness was covered in a systematic review, but findings came from a single RCT with a small sample size (n=18), meaning the findings cannot be generalised (Marshall et al., 2020).

Lastly, the evaluation of Nightstop did not include a control or comparison group. Instead, it reported descriptive findings from a survey and qualitative interviews with participants and other stakeholders, meaning that conclusions about impact are suggestive but not conclusive, and more high-quality research would be needed to be confidence that the intervention was effective (McCoy and Kempton, 2018).

8.4 Evidence of impact

Most of the community and peer mentoring interventions in the literature focused on health outcomes (specifically, substance misuse). However, some studies looked at impacts on employment outcomes and life stability outcomes (specifically, reoffending).

Peer support for substance misuse

The systematic review of peer recovery support interventions for substance misuse concluded that overall, the interventions were effective. Three RCTs, all cited in Eddie et al. (2019) looked at 12-step programmes. It is not possible to ethically randomise access to 12-step programmes, so these studies instead randomised whether participants were referred to a 12-step programme, who they were referred by, or how ‘intensively’ the referral was made:

Timko et al. (2006) found that intensive referral to 12-step programmes, compared to standard referral processes, was associated with greater attendance of the programmes and better drug and alcohol use outcomes over a 6-month follow-up period. Standard referral involved giving participants a schedule for local 12-step meetings and encouraging them to attend. Intensive referral involved counsellors arranging a meeting between the participant and a member of the local Alcoholics Anonymous or Narcotics Anonymous group, who then offered to attend the meeting together with the participant. Participants also completed journals to check meeting attendance.
Timko et al. (2011) repeated the above study for individuals with both substance misuse and mental health issues, finding again that intensive referral led to greater 12-step programme attendance in addition to less drug use and fewer psychiatric symptoms.
Manning et al. (2012) compared referral by a peer with shared experiences, referral by a doctor, and no referral, and simply looked at the impact on attendance of a 12-step programme. The study found that longer term attendance was highest when participants were referred by a peer.

Residential interventions for ex-offenders with substance misuse issues

The evidence on residential interventions for ex-offenders with substance misuse issues comes largely from a single RCT. But that RCT did show some evidence of impact on employment outcomes. Jason et al. (2015) found that participants in recovery housing worked more days than those in usual services, although the difference was relatively modest. The impact on employment of participating in an Oxford House increased as the length of stay increased. To illustrate, after 6 months, participants in Oxford Houses worked an average of 1.3 days in the month, compared to 1.9 for usual services, but by 18 months, participants in Oxford Houses worked an average of 10.2 days in the month, compared to 6.7 for usual services.

A cost benefit analysis found that Oxford Houses had a more positive cost benefit ratio than Therapeutic Communities. The authors calculated a net benefit per participant by subtracting the costs of the intervention from the benefits. Costs included the residence itself, and the costs of in-patient treatment or illegal activities committed. The benefits included the earned income of participants. The authors concluded that the net benefit for Oxford Houses was $12,700 per participant, and for Therapeutic Communities was $7,500 per participant.

Overall, this evaluation of Oxford Houses and Therapeutic Communities provides some promising evidence of impact, especially for Oxford Houses, but should be treated with caution. This is for three main reasons. Firstly, it is a single study. Secondly, there were some statistically significant age differences between the conditions, even if these were controlled for in the analysis. For example, individuals in Therapeutic Communities were older than those in Oxford Houses. And thirdly, the authors have conducted several evaluations of recovery houses and have some professional connection to the service.

Non-residential peer mentoring for ex-offenders

The evidence on the impact of CMM was relatively weak. The one quantitative impact evaluation with a comparator (the Arches evaluation) showed significant reductions in some but not all indicators of reoffending, particularly for younger participants, for whom the rate of serious criminal convictions for participants was half that of non-participants after 12 to 24 months. Most other quantitative studies found some positive associations between participation and reduced reoffending and other indicators, but as these were correlational studies it is not possible to draw causal conclusions. The qualitative findings all suggested that the interventions were successful, with mentees forming close, supportive relationships with mentors (Lesnick et al., 2023).

The single evaluation of the MENTOR programme showed no evidence of impact on employment outcomes (Taylor, 2021). The single evaluation of peer support communities was not robust enough to draw any conclusions about impact (Marshall et al., 2020).

Community-based emergency housing for homeless youth

The evaluation of Nightstop reported a wide range of positive impacts. The qualitative evidence, in particular, showed that participants viewed the service overwhelmingly positively. Participants who used a Nightstop for only a few nights reported a range of short-term outcomes, such as better sleep, personal care, hygiene, and mental health. Those who used the service for a longer time reported a broader set of positive impacts, such as improved social and emotional capabilities, practical skills like cooking and cleaning, and engagement with education, training and employment. However, given the methodological limitations of the study, this should be treated as tentative evidence of impact (McCoy and Kempton, 2018).

8.5 Lessons for delivery

Most of the systematic reviews did not include extensive detail on why some interventions worked and others did not, nor lessons for how interventions can best be delivered.

Eddie et al., (2019), who looked at peer recovery support interventions for substance misuse, emphasised the need to determine the optimal conditions for peer interventions, including the appropriate training and certification for peer workers. They suggest that stablishing best practice models could enhance the effectiveness and consistency of peer support services. Relatedly, Taylor, 2021 speculated that the diverse backgrounds of mentors in the MENTOR programme, which included non-peer mentors, may have influenced the outcomes and experiences of participants.

Question 13

9.  Case management interventions

Accepted Answer

This chapter summarises the evidence on standalone case management interventions. Any case management interventions that were paired with the provision of housing are discussed in Chapter 11. Case management interventions involve coordinated and structured support services to help vulnerable individuals achieve stability and address specific needs, with a designated ‘case manager’ who coordinates a range of intensive support services tailored to the individual’s needs.

9.1 Summary of findings

There is good evidence that case management interventions for homeless populations are an effective way of reducing homelessness.
More intensive approaches, involving a wide range of additional services, were more effective at reducing homeless. For example, Assertive Community Treatment (ACT) provides 24/7 holistic support with a multi-disciplinary team focused on severe mental illness.
But the evidence on employment outcomes was too limited and too mixed to draw conclusions, across all four population groups.

Volume of high-quality evidence

Ex-offenders	Low
Substance misuse	Medium
Homeless	High
Care leavers	Low

Evidence of positive impact

Ex-offenders	No evidence of positive impact on employment or health outcomes
Substance misuse	No evidence of positive impact on employment or health outcomes
Homeless	Evidence of positive impact for housing outcomes. No evidence of positive impact on employment
Care leavers	N/A

Future research recommendation

Ex-offenders	Higher quality and more directly relevant evidence needed
Substance misuse	Higher quality and more relevant evidence needed
Homeless	No further research needed
Care leavers	N/A

9.2 Interventions in the literature

The most common types of case management were tested with people experiencing homelessness with multiple disadvantages such as substance misuse issues, ex-offenders and persistent mental health problems. These included:

Standard Case Management (SCM) which coordinates ongoing support to access services and achieve stability
Intensive Case Management (ICM), usually given to individuals with the highest needs, and involves a case manager having more frequent contact and smaller caseloads so that they can provide more targeted support
Assertive Community Treatment (ACT) which uses a multidisciplinary team to provide 24/7 comprehensive support
Critical Time Intervention (CTI) which offers structured, time-limited assistance during critical transitions, such as moving into permanent housing

Interventions targeting substance misuse specifically included an additional form of case management, known as coordinated case management with five phases: outreach and assessment, planning and engagement, care coordination and advocacy, aftercare and relapse monitoring, and crisis management (Magura and Marshall, 2020).

Holistic interventions for ex-offenders with personality disorders, such as the Good Lives Model (GLM) and Multi-disciplinary Forensic Psychiatric Services (MSUs), addressed mental health and social reintegration, incorporating drug-treatment elements to reduce reoffending and improve overall well-being (Connell et al., 2017).

In the US, the Substance Use Treatment and Re-entry (STAR) programme combined case management with cognitive behavioural therapy for juvenile ex-offenders with substance misuse issues (Hunter and Huang, 2014).

9.3 The scale and quality of the evidence

Homelessness. The evidence on case management interventions for people experiencing homelessness was robust, examined through three systematic reviews and experimental studies. De Vet et al., 2013 analysed 41 papers from the US and the UK, including RCTs and quasi-experimental studies. Weightman et al., (2023) conducted a large-scale meta-analysis comparing five case management models from the US and Canada. Marshall et al., (2022) focused on employment-based interventions for homeless individuals in Canada.

Substance misuse. The volume of evidence for people with substance misuse issues specifically was lower than for people experiencing homelessness but of reasonable quality, appearing in one systematic review (Magura and Marshall, 2020). However, reviews on homeless populations often covered those with substance misuse issues as a co-occurring disadvantage.

Ex-offenders. For ex-offenders, evidence was less robust and drawn from observational studies rather than RCTs (Connell et al., 2017). Common challenges across the studies included high attrition rates, reliance on self-reported data, and limited generalisability due to the specific populations studied (Hunter and Huang, 2014).

9.4 Evidence of impact

Although some case management interventions included employment outcomes, the literature primarily focused on life stability outcomes, particularly housing and health outcomes including substance use.

The impact of case management on homelessness

Overall, there is evidence that case management interventions for homeless populations are an effective way of reducing homelessness, but the evidence is too limited for drawing any conclusions about other outcome areas (including employment), and of impact for other population groups.

Case management interventions were generally more effective in improving housing stability among homeless populations compared to usual care (De Vet et al., 2013, Weightman et al., 2023). A meta-analysis found that case management of any type improved long term (1 year or longer) homelessness outcomes, but because the included primary studies used a wide range of different outcome measures, the meta-analysis report the impact in standard deviations: participation improved outcomes by half a standard deviation compared to control (Weightman et al., 2023). Across these interventions, more intensive approaches, such as ACT, were the most effective (De Vet et al., 2013, Weightman et al., 2023). This was particularly evident for those with dual diagnoses or severe mental health issues who require additional targeted support (De Vet et al., 2013).

The impact of case management on wider outcomes

For ex-offenders, people experiencing homelessness and people with substance misuse issues, the evidence was too limited and mixed to confirm positive effects on employment outcomes.

Evidence on physical health outcomes for homeless populations is limited, with only one study showing a reduction in physical medical problems through Standard Case Management (De Vet et al., 2013). Mental health outcomes were inconsistent, though Assertive Community Treatment and Critical Time Intervention showed some positive effects (De Vet et al., 2013). Substance use outcomes also varied, with Standard Case Management being somewhat effective, while Intensive Case Management and Assertive Community Treatment had limited impact (De Vet et al., 2013).

The impact of case management for ex-offenders

Evidence for the effectiveness of case management interventions for ex-offenders was limited. The GLM intervention showed no impact on social functioning due to significant study design limitations (Connell et al., 2017). An observational study reported increased university enrolment, but the small sample size limits the robustness of this finding (Connell et al., 2017). Similarly, the MSUs intervention did not impact social functioning as measured by the “Work and Social Adjustment Scale” after two years (Connell et al., 2017).

9.5 Lessons for delivery

There was limited attention given to delivery lessons for case management interventions. One review by Weightman et al., (2023), underscored the importance of minimising conditionality, such as abstinence requirements, and highlighted the value of including case managers or peer mentors with lived experience. Consistency and continuity in the individual case manager assigned to a participant was identified as crucial for success. However, these claims were speculative based on the discussion of findings.

Question 14

10.  Education interventions

Accepted Answer

This chapter summarises the evidence on educational interventions, of which there were two main types in the literature: correctional education programmes for incarcerated individuals; and interventions to help care leavers enter and/or remain in education or training.

10.1 Summary of findings

Correctional education programmes have been comprehensively evaluated in the US. The evidence shows a strong positive impact on rates of reoffending. A meta-analysis found that participants reoffended at 64% the rate of non-participants. The evidence on employment outcomes was less robust but showed some promise.
The evidence on educational programmes for care leavers was generally qualitative or low quality. But findings all pointed in the direction of positive impacts on care leavers feeling supported and confident.

Volume of high-quality evidence

Ex-offenders	High for reoffending outcomes. Medium for employment outcomes
Substance misuse	Low
Homeless	Low
Care leavers	Low

Evidence of positive impact

Ex-offenders	Evidence of positive impact for reoffending outcomes. Evidence of promise for employment outcomes.
Substance misuse	N/A
Homeless	N/A
Care leavers	Evidence of promise for employment and education outcomes

Future research recommendation

Ex-offenders	Higher quality and more directly relevant evidence needed
Substance misuse	Higher quality and more directly relevant evidence needed
Homeless	Higher quality and more directly relevant evidence needed
Care leavers	Higher quality and more directly relevant evidence needed

10.2 Correctional education programmes

Interventions in the literature

The literature discussed a wide range of correctional education programmes for incarcerated individuals, all in the US (Davis et al., 2013, Wilson et al., 2000). These included:

adult basic education (ABE), which focuses on basic skills instruction in arithmetic, reading, writing, and English as a second language (ESL) if necessary
adult secondary education (ASE) aims to help individuals complete high school or prepare for a high school equivalency certificate, such as the General Education Development (GED) test
vocational education or career and technical education (CTE) provides training in general employment skills and specific job or industry skills
post-secondary education (PSE) offers college-level instruction that enables individuals to earn college credits toward a two-year or four-year degree.

The nature and scale of the evidence

These interventions have been comprehensively evaluated in the US context. Davis et al., (2013) conducted a meta-analysis of 58 studies in the US looking at interventions from 1980 to 2011, all of which were at least Level 3 on the Maryland Scientific Methods Scale (meaning they used a control or comparison group). Wilson et al., 2000 conducted a meta-analysis of 33 experimental and quasi-experimental evaluations from 1975 onwards.

Evidence of impact

Reoffending outcomes. Both meta-analyses found positive impacts on reoffending outcomes. Davis et al., (2013) found that on average, inmates who received correctional education interventions reoffended at 64% the rate of those who did not. They found that the overall effect was not driven by lower quality studies. Wilson et al., (2000) found that post-secondary education reduced reoffending from 50% to 37%, vocational training reduced it to 39%, and adult basic education and GED programmes reduced it to 41%.

Employment outcomes. There were promising findings in both studies for employment outcomes. Davis et al., (2013) found that inmates who received correctional education interventions had 13% higher odds of achieving employment. However, given the low rates of post-release employment, this translated to only a modest improvement in overall employment rates. The authors also noted that only one study that included employment outcomes had a high rating for quality. Wilson et al., (2000) also found positive impacts, with post-secondary education showing the most impact on employment outcomes but also flagged variability in the quality of the literature.

10.3 Educational interventions for care leavers

Interventions in the literature

The literature discussed several educational interventions for care leavers, all of which were in the US. These were mostly campus support programmes that offered social, personal and informational support to help with the transition from high school to university, or from university onwards (Woodgate et al., 2017). For example, Foster Care Alumni Creating Educational Success was an initiative that offered services to foster care leavers to encourage employment after graduation from university. Another example (the Chafee Educational Training Voucher Program) provided up to $5,000 per year toward the higher education expenses of care leavers, including housing costs and tuition fees.

One UK-based initiative was discussed in the literature. Virtual Schools are a statutory service under the Children and Families Act 2014, which requires local authorities (LAs) in England to appoint a Virtual School Head to fulfil the LA’s duty to promote education achievement of children in care.

The nature and scale of the evidence

Campus support interventions were discussed in a single scoping review by Woodgate et al., (2017), which identified a large number of relevant primary studies. All these primary studies were small, none used experimental or quasi-experimental methods, and most relied entirely on qualitative interviews with participants.

The Virtual Schools service was discussed in the study that was based on qualitative interviews with care leavers and professionals within LAs. The study looked at three LAs in which the Virtual School was continued beyond the age of 18.

Evidence of impact

For campus support interventions, findings were generally positive: participants felt they had learned new skills, or felt supported, or went on to achieve employment. But because of the methodological limitations, none of these studies provide evidence of impact.

The Virtual Schools service was considered by interviewees to be a critical element in supporting care leavers to access further and higher educational opportunities. Participants explained that some care leavers are only ready to engage in support to enter employment, education or training once their post-care circumstances, including accommodation, have stabilised, and that extending the Virtual Schools system acknowledged that fact. However, given the nature of the evidence, it is not possible to draw conclusions about the impact of this service.

Question 15

11.  Housing interventions

Accepted Answer

This chapter summarises the evidence on housing interventions. These are interventions designed to improve access to safe and stable housing for disadvantaged groups.

11.1 Summary of findings

The evidence shows that housing interventions involving additional support for co-occurring disadvantages were more effective than more basic housing provision for improving housing stability outcomes. An example of the former is Housing First, which provides immediate housing and access to a range of support services. This was robustly demonstrated using a meta-analysis of RCTs from the US, Canada, and the UK
The quality of evidence for other outcomes, including employment outcomes, was much weaker and was typically excluded from meta-analyses as a result.

Volume of high-quality evidence

Ex-offenders	Low
Substance misuse	Medium
Homeless	High for housing outcomes. Medium for employment outcomes
Care leavers	Medium

Experience of positive impact

Ex-offenders	Evidence of positive impact on housing outcomes. No evidence of positive impact on employment or health outcomes.
Substance misuse	Evidence of positive impact on housing outcomes. No evidence of positive impact on employment or health outcomes
Homeless	Evidence of positive impact on housing outcomes. No evidence of positive impact on employment or health outcomes.
Care leavers	Evidence of positive impact on housing outcomes. No evidence of positive impact on employment or health outcomes.

Future research recommendation

Ex-offenders	More directly relevant evidence needed
Substance misuse	More directly relevant evidence needed
Homeless	More directly relevant evidence needed
Care leavers	Higher quality and more directly relevant evidence needed

11.2 Interventions in the literature

Understandably, housing interventions were targeted at people experiencing homelessness, but often included people experiencing homelessness with other disadvantages, including people with substance misuse issues, criminal convictions, and mental health problems. Interventions within the literature can be classified as providing low, medium, or high support.

Low support housing interventions involved providing accommodation without additional services. Examples range from short-term hostels to longer-term housing (Keenan et al., 2021).

Moderate support interventions, all in the US and Canada, included financial assistance for housing costs, such as short-term rental subsidies, furniture provision, and long-term support for independent living, particularly for those with mental illness who had experienced homelessness (Aubry et al., 2020). Other examples were rent subsidies and life skills training for single mothers (Lakrouf et al., 2024).

High support interventions addressed the complex needs of individuals with multiple disadvantages. Examples included the Homelessness Prevention Therapeutic Community (HP-TC), where residential groups supported each other’s addiction recovery and received interventions for parenting, employment, and housing stabilisation. Another example was an intervention providing three months of assistance for rent and utility bills, along with six months of substance misuse and mental health counselling, and skill-building services (Lakrouf et al., 2024).

Further examples in this category were split into the following:

Housing with case management

Housing interventions were often combined with case management. See Chapter 9 for more detailed explanations of standalone case management. Examples include:

Supportive Housing for Families with Intensive Case Management (SHF with ICM): This intervention involves permanent housing combined with mental health interventions, employment, and vocational support. Participants work with case managers to develop a family care plan and meet regularly to address their needs (Krahn et al., 2018).
Assertive Case Management (ACM): ACM involves shelter workers referring homeless individuals to case managers trained in psychosocial rehabilitation. The focus is on building rapport, conducting relationship-based outreach, and providing mental health support. The intervention aims to assist in finding housing and improving overall well-being (Marshall et al., 2020).

Housing First

The Housing First (HF) model was identified as a leading intervention for chronic homelessness, particularly among individuals with mental health and substance misuse issues. HF provides immediate housing without preconditions, such as sobriety or treatment compliance, differing significantly from traditional models, which require abstinence before participants can be housed. It includes a consumer choice element, which means that participants are actively involved in choosing their housing arrangements. The approach integrates case management with other supportive services (Weightman et al., 2023; Marshall et al., 2020; dePass et al., 2023). Housing First interventions for care leavers include immediate housing access and ongoing one-to-one support to prevent future homelessness (Dixson et al., 2021).

Permanent Supportive Housing

Permanent Supportive Housing, like Housing First, provides housing without requiring adherence to treatment or abstinence from substance use. This intervention is often combined with case management (Aubry et al., 2020). The approach aims to support independent living, while addressing mental health and substance use issues.

11.3 The scale and quality of the evidence

There was substantial evidence on housing interventions for homeless populations and those with multiple disadvantages. For instance, the systematic review by Keenan et al., (2021) was based on two existing evidence gap maps commissioned by the Centre for Homelessness Impact.

Within the literature, there was strong evidence on housing interventions published in high-quality systematic reviews. Reviews and meta-analyses by Keenan et al., (2021) and Aubry et al., (2020) synthesised experimental studies from the US, Canada, and the UK. High-quality RCTs, such as Slesnick et al. (2023) with 240 participants, compared low, medium, and high support housing interventions (Lakrouf et al., 2024).

Despite quality evidence reviews, a considerable portion of the primary academic literature within these reviews was of weaker quality. Inconsistent definitions, population heterogeneity, and varied outcome measures hindered the comparability and generalisability of primary evidence (Bassuk et al., 2014, Krahn et al., 2018). Systematic reviews highlighted challenges, such as small sample sizes and lack of comparison groups, leading to mixed results (Marshall et al., 2020, Bassuk et al., 2014).

The strongest evidence base focused on the impact of housing interventions on housing outcomes. Evidence quality for secondary outcomes, like employment or reoffending, was weaker and often excluded from meta-analyses (Keenan et al., 2021).

11.4 Evidence of impact

Most literature on housing interventions focused on housing as a key life stability outcome. Employment, substance misuse and health outcomes were also considered but were often explored as secondary outcomes and given less attention.

Evidence consistently showed that providing high support housing, where housing is given unconditionally, was more effective at improving housing stability than simply providing basic housing. For instance, one systematic review found that those who received high support housing had a moderate improvement in housing stability compared to those who received no intervention. The primary studies included in the meta-analysis used a wide range of different outcome measures, so the systematic review standardised these to generate overall effect size estimates. The impact of high support housing on housing stability was assessed to have an effect size of 0.62, indicating a moderate positive impact. Furthermore, when comparing high support housing to basic housing interventions that also provided housing unconditionally but without additional support, the improvement was even greater, with an effect size of 1.10, indicating a large positive impact. This suggests that high support housing is significantly more effective at improving housing stability than basic housing interventions (Keenan et al., 2021). Additionally, one review found that stable housing combined with support also led to a reduction in criminal activity (Carnemolla and Skinner, 2021).

Housing First was effective for helping individuals secure and maintain stable housing (Marshall et al., 2020; Saldanha et al., 2022; dePass et al., 2023; Dixson et al., 2021). The consumer choice element, which gives participants an active role in choosing their housing arrangements, was positively associated with housing stability. The findings suggest that the core elements of the model—the unconditional provision of housing—leads to better outcomes (Saldanha et al., 2022).

High support interventions often included some form of employment support, although this was rarely defined and not a major focus of the systematic reviews. The evidence was insufficient to confirm a relationship between housing interventions and employment. Positive impacts on employment were noted, but the evidence was too weak for a definitive conclusion (Bassuk et al., 2014; Woodgate et al., 2017). One meta-analysis found that only two papers had statistically significant results and showed a negative impact for the interventions (Milby et al., 2000; Poremski et al., 2016 cited in Keenan et al., 2021). There was also no evidence, beyond tentative qualitative findings, on the benefits of Housing First on employment.

High support interventions, such as housing, combined with case management or therapeutic support showed positive effects on mental health (Krahn et al., 2018; Keenan et al., 2021). However, permanent supportive housing did not improve social integration (Marshall et al., 2020), Evidence on Housing First’s impact on reducing substance use, social integration, and mental health was also mixed and inconclusive (Saldanha et al., 2022; dePass et al., 2023; Marshall et al., 2020).

The evidence on housing subsidies was too inconsistent to support any effect on income or employment outcomes (Aubry et al., 2020).

11.5 Lessons for delivery

Reviews predominantly focused on outcomes rather than lessons for delivery, but some explanations and contextual factors for successful housing interventions were considered. These were often not strongly evidenced and instead were formed by the views of the authors, based on the findings of their review. For example, Marshall et al. (2019) suggested that Housing First needed to be combined with strategies addressing broader structural issues, like poverty and discrimination, and incorporating peer support to improve community integration outcomes (Marshall et al., 2020).

The effectiveness of housing interventions was heavily influenced by factors such as access to affordable housing and social welfare systems, evidenced by a synthesis of six process evaluations (Keenan et al., 2021). For example, shortages in the availability of affordable housing caused frustration and delays in securing permanent homes, affecting the success of the interventions (Keenan et al., 2021). Key enablers included clear identification of suitable participants, effective referral routes, meaningful user engagement, and a non-judgemental, flexible support approach (Keenan et al., 2021). The role of staff was pivotal; their ability to communicate safely and without judgement, as well as their commitment and capacity to coordinate with other agencies, was crucial for tailoring support to individual needs (Keenan et al., 2021).

Recommendations for practice included incorporating additional support that focuses on reducing harm, offering employment services, and considering specific drug use patterns in Housing First programmes (Saldanha et al., 2022). Furthermore, the importance of community-based drop-in centres was emphasised for improving social connections by giving individuals direct and frequent access to support (Carnemolla and Skinner, 2021).

Question 16

12.  Medication assisted treatment

Accepted Answer

12.1 Summary of findings

Medication assisted treatment (MAT) involves the use of medication alongside counselling and other activities to help people with opioid use disorder. It has been proven to be an effective treatment for opioid use.
One systematic review looked at whether MAT has a wider impact on other outcomes on the pathway to work. Overall, there was no evidence of wider impacts of MAT.

Substance misuse

Volume of evidence	High
Evidence of impact	Evidence of promise for criminal activity
No evidence of impact on employment and mental health outcomes
Future research recommendation	No further evidence needed

12.2 Interventions in the literature

The literature included a wide range of interventions that either directly or indirectly aimed to address substance misuse. These are addressed in other chapters: for example, 12-step programmes like Alcoholics Anonymous are addressed in Chapter 8 on peer mentoring interventions. This chapter focuses on an intervention for substance misuse that was distinct from the others: medication assisted treatment (MAT). MAT involves a comprehensive treatment plan that includes approved medication, counselling, behavioural therapies, and patient monitoring (Maglione et al., 2018). It is offered to people with opioid use disorder.

12.3 The scale and quality of the evidence

MAT has been proven to be an effective treatment for opioid use. One systematic review looked at whether MAT has a wider impact on other outcomes included in the pathway to work (Maglione et al., 2018). It included 12 RCTs and 3 observational studies, looking at interventions in the US, Canada and Europe with people with opioid use disorder. The review includes meta-analyses for certain outcome measures.

12.4 Evidence of impact

Overall, while MAT is established as effective treatment for opioid use, it appears to have limited wider impacts. None of the studies included in the Maglione et al., (2018) found statistically significant impacts on employment and mental health outcomes. There was mixed evidence on the effectiveness of MAT in reducing criminal activity, with some primary studies finding a positive effect and others not, and a meta-analysis showing no overall effect.

Question 17

13.  Independent living programs for care leavers

Accepted Answer

This chapter summarises the evidence on independent living programs (ILPs) for care leavers. This includes a range of approaches taken to building life skills and supporting care leavers to live independently, including mentoring, coaching, job readiness training, and supported living.

13.1 Summary of findings

Overall conclusions about ILPs generally are not possible due to mixed and conflicting evidence, which is likely explained by the diversity of different kinds of ILPs.
However, there was some promising evidence that some ILPs can improve a range of outcomes, including employment outcomes. For example, the MyLife programme in the USA involved coaching to identify and reach independence goals, alongside identification of career opportunities. RCTs demonstrated that participants were 23% more likely to be in employment 9 months after leaving the programme and were also more likely to engaged in independent living activities like paying rent and utility bills.

Care leavers

Volume of evidence	Medium
Evidence of impact	Evidence of promise on employment, housing, criminal activities and education outcomes
Future research recommendation	Higher quality of evidence needed

13.2 Interventions in the literature

Independent living programmes (ILPs) aim to equip young people with the necessary skills to live independently. These skills include communication, decision-making, emotional skills, as well as vocational skills, budgeting, common household tasks, and how to obtain support from services. ILPs specifically focus on developing these skills for care leavers between the ages of 16 and 19, so that they can better cope with independent life after leaving their care.

ILP’s identified in the literature mostly took place in the UK and US. ILP’s took three main approaches: mentoring/coaching, job readiness programmes and supported living. Although some similar interventions are discussed in other chapters, the interventions included here all had a specific focus on care leavers.

Mentoring and coaching ILPs matched care leavers with older peer mentors with similar life experiences or coaches with specific training in supporting transition to independent living. Mentors worked closely with mentees to identify life goals relevant to them and then use their experience to support them to achieve them (Greeson et al., 2020, Leathers et al., 2023 and Bonella et al., 2019).

ILPs that included job readiness training focused on improving soft skills relevant to employment, such as group working and identifying career opportunities and pathways to reach them. This was achieved through intensive short courses involving group activities as well as opportunities to take part in short term internships and spending time with work coaches (Leathers et al., 2023, Greeson et al., 2020 and Gibb and Edwards, 2017).

Most ILPs included both a mentoring/coaching component and a job readiness component. For example, the MyLife programme in the US involved both coaching to identify and reach independence goals alongside identification of career opportunities and pathways. The LV LifeSet programme is an intensive programme in which specialist staff meet with care leavers in person weekly, and remain in contact with them throughout weekly, available at all times. They help care leavers navigate their transition, including helping them to obtain housing, education and employment.

Supported living ILPs involved placing care leavers into safe housing where they were supported by in-house carers who help them develop everyday life skills such as cooking, cleaning, budgeting and managing tenancies. These are different to extended care policies that allow care leavers to remain in placements they are already in past the age of 18. These interventions often offer some form of mental health support such as access to a clinical psychologist or staff trained in trauma-focused therapy at the same time (Greeson et al., 2020, Neagu and Dixon, 2020b, Neagu and Dixon, 2020a and Mitchell-Smith et al., 2020).

13.3 The scale and quality of the evidence

There was a large volume of literature investigating ILPs, including six evidence reviews, six pieces of grey literature and one academic paper. The quality of these papers varied with some reporting on high quality randomised controlled trials while others relied on observational studies with very low sample sizes.

This meant that certain ILPs had more convincing evidence than others. Studies investigating the impact of ILPs that involved mentoring and coaching and job readiness training were of higher quality than those investigating ILPs involving supported living. For the former, there were multiple RCTs, whereas for the latter the evidence involved low quality impact evaluations with small samples.

13.4 Evidence of impact

Evaluations of ILPs looked at their impact on a wide range of outcomes, including employment, education, criminal behaviour, housing and social connectedness.

Several systematic reviews tried to determine the impact of ILPs generally and included a wide range of highly diverse interventions. Overall, the conclusions from these reviews were highly mixed, with some concluding there was a consensus in the literature on the positive impact of interventions (Montgomery et al., 2006, Heerde et al., 2016 and dePass et al., 2023), and others disagreeing (Taylor et al., 2024).

It seems likely that this may be explained by the diversity of interventions included in these reviews. The evidence contains several high-quality evaluations, which suggest that certain ILPs are effective, but overarching conclusions about all ILPs should be avoided. As such, the remainder of this section draws out specific findings relating to different types of ILPs.

Mentoring, coaching and job readiness training

There was strong evidence that ILPs that involve mentoring, coaching and job readiness training can have a positive impact for care leavers. However, it is not possible, based on the evidence, to disentangle which approach (mentoring/coaching, or job readiness training) was the most effective.

Three RCTs assessing this family of approaches found care leavers taking part in interventions had significantly higher rates of employment or worked significantly more hours a week compared to those in the control groups (Leathers et al., 2023; Greeson et al., 2020). For example, the MyLife programme involved 12 months of coaching to help care leavers identify and reach their goals, as well as connecting care leavers to career opportunities. One RCT (Greenen et al., 2013, cited in Greeson et al., 2020) found that care leavers taking part in the MyLife programme were 23% more likely to be in paid employment 9 months after the start of the programme compared to care leavers in the control group.

These ILPs were also shown to be effective for a range of other outcomes on the pathway to work, including reducing rates of arrest, increasing the chance of finishing high school, and increasing the chance of living independently (Leathers et al., 2023; Greeson et al., 2020). For example, Leathers et al., 2023 found that just 5% of care leavers, who took part in the Adult Connections Team (ACT) programme, reported having been arrested at follow up compared to 20% of those in the control group. Similarly, one RCT (Powers et al. 2012, cited in Greeson et al., 2020) found care leavers who took part in the MyLife intervention were substantially more likely to be engaging in independent living activities (paying rent and utilities) and to have graduated high school (72% vs 50%) post-intervention compared to those in the control group. A qualitative evaluation of the Fairbridge programme, which involved job readiness training and coaching, left care leavers with a heightened sense of work preparedness. Programme staff observed that most care leavers had gained increased confidence and improved teamwork skills (Gibb and Edwards, 2017).

Supported living

There was some evidence to suggest that supported living ILPs reduced care leavers’ risk of becoming homeless. But there was no strong evidence for the impact on other outcomes, such as education, employment and criminal activity.

A single RCT study conducted by Jacobs et al. (2015) from Greeson et al., 2020 was the only quantitative research with sufficiently robust findings to report. This study compared care leavers, who were randomly assigned to participate in the YVLifeSet program, which provided stable housing and community-based support from the Youth Villages charity in Tennessee, with care leavers who continued to receive typical support. Jacobs et al. (2015) found that significantly fewer care leavers who participated in the intervention experienced homelessness after a one-year follow-up compared to those in the control group (21% vs. 27%).

Qualitative findings from grey literature, however, do indicate some promise for creating stable environments that encourage continued engagement in education, employment or training. Care leavers felt these programmes had led to increased stability in their life, as well as greater feelings of social connectedness, due to good quality relationships built with their in-house carers (Neagu and Dixon, 2020b; Neagu and Dixon, 2020a; Mitchell-Smith et al., 2020). Both carers and care leavers felt that the programmes had helped them to gain greater independence skills, as well as creating an environment that encouraged continued engagement in education employment or training.

13.5 Lessons for delivery

Most of the identified grey literature included qualitative process evaluations, involving interviews with delivery staff that shed light on the factors that facilitated delivery of independent living programmes (Neagu and Dixon, 2020b, Neagu and Dixon, 2020a, Mitchell-Smith et al., 2020, Gibb and Edwards, 2017, Davey et al., 2023 and Bonella et al., 2019). These papers identified four factors that contributed to success of their interventions:

Co-production. Giving care leavers opportunities to provide feedback and influence the continued development of the intervention encouraged greater engagement (Neagu and Dixon, 2020b, Neagu and Dixon, 2020a; Mitchell-Smith et al., 2020). An example of this was the Staying Close programme, involving care leavers in the recruitment process of new caring staff.
Flexible approaches. Caring staff or mentors being empowered to tailor support to care leavers’ needs, rather than following a strict method, allowed for development of goals and skills that were relevant to their individual needs (Gibb and Edwards, 2017; Neagu and Dixon, 2020b; Davey et al., 2023).
Multi agency working. Having good connections and collaborations with different supportive providers was considered crucial to be able to support the varying needs of different care leavers (Gibb and Edwards, 2017; Neagu and Dixon, 2020b; Neagu and Dixon, 2020a; Davey et al., 2023). For example, the Staying Close programme was able to employ a clinical psychologist to train in house carers evidence-based approaches to supporting care leavers’ mental health.
Consistency. Staff delivering the projects need to be consistent and demonstrate commitment to the care leavers they are supporting. This facilitates the development of high-quality relationships with care leavers and prevents emotional distress due to their particular sensitivity to abandonment. (Bonella et al., 2019; Davey et al., 2023). For example, care leavers whose mentors consistently showed up to support them, felt a greater sense of life stability (Bonella et al., 2019).

Question 18

14.  Creating connections with family and other adults

Accepted Answer

This chapter summarises the evidence on interventions designed to create connections between children in care with family and other caring adults. It also includes discussion of the long-term impact of adoption on adult life outcomes.

14.1 Summary of findings

There was some evidence that these interventions can improve the chances of adoption or reunification with biological family.
There was also evidence that young people who are adopted out of care have better longer-term outcomes, including employment outcomes. However, the evidence is not strong enough to conclude that adoption itself is the cause of these outcomes.

Volume of high-quality evidence
Care leavers	Low

Evidence of positive impact
Care leavers	Evidence of impact on rates of adoption and social connectedness. Evidence of promise for employment, health and criminal activity

Future research recommendation
Care leavers	More directly relevant evidence needed

14.2 Interventions in the literature

A range of interventions aimed to foster positive relationships between children currently in care and caring adults, with the goal of improving life stability and ultimately leading to better life outcomes such as education and employment.

Some programmes focused on reconnecting young people with their biological families, but others focused on other non-parental adults such as grandparents, aunts, uncles, and other supportive figures. The interventions were evaluated in the US, and the UK, and tended to be focused mostly on children and teenagers rather than adult care leavers aged 18 and above (Greeson et al., 2020).

Some examples of these interventions included (all took place in the US):

Wendy’s Wonderful Kids. This programme aimed to connect youth to non-parental caring adults with the ultimate aim being adoption.
Family Finding Programme. This programme involved establishing contact with caring family members, while also providing planning and support to maintain more permanent relationships.
Creating Ongoing Relationships Effectively. This was a trauma-informed programme for youth aged 14 to 19 leaving foster care, with the aim being to develop relationships with non-parental caring adults.
On the Way Home. This programme attempted to reunify young people with their biological parents through a 12-month family engagement process.

14.3 The scale and quality of the evidence

There were two systematic reviews that investigated these interventions. One study (Greeson et al., 2020) synthesised the results of RCTs of the four interventions listed above but only looked at the impact of the interventions on the extent of young people’s connections with caring adults, rather than outcomes on the pathway to work. The other study (Lindner and Hanlon, 2023) simply synthesised observational studies that looked at the different outcomes for young people who were adopted out of care compared to those who were not, meaning it is not possible to determine whether differences are due to being adopted or due to other factors. Overall, therefore, there is limited evidence on whether these interventions support young people along the pathway to work.

14.4 Evidence of impact

There is some promising evidence that programmes that aim to create connections between young people in care and caring adults can increase the likelihood of adoption or reunification with family. There is also evidence that young people in care, who are adopted, have better employment, health and criminal activity outcomes in adulthood. However, it is not possible to conclude that creating connections between young people in care and caring adults leads to positive outcomes, of any kind, due to limitations in the evidence.

Greeson et al., 2020 found mostly consistent positive results for interventions that aimed to increase connections with caring adults. For example, children who took part in Wendy’s Wonderful Kids were 1.7 times more likely to be adopted than those in the control group. Of those who took part in the On The Way Home programme, 91% were able to stay in home and community settings with family members, compared to 65% in the control group.

Lindner and Hanlon, 2023 found that individuals, who were adopted out of care, had significantly better life outcomes related to education, employment, health and criminal activity, when compared to individuals who were not. For example, young people who were adopted were more likely to have enrolled into non-compulsory education, to have higher base earnings and to have not been incarcerated.

However, children who are adopted and those who are not are systematically different. Most obviously, children who are adopted tend to be adopted in relatively more affluent families. This means we cannot conclude that adoption itself leads to more positive life outcomes for young people in care.

Question 19

15.  Extending care arrangements past the age of 18

Accepted Answer

This chapter summarises the evidence on interventions that involve keeping young people in care in their existing placements beyond the age of 18, rather than supporting them to live independently.

15.1 Summary of findings

There was some promising evidence that extended care arrangements may improve employment, housing, education and health outcomes.
However, it is not possible to draw strong conclusions about impact due to the low quality of the evidence available.

Care leavers

Ex-offenders	High for reoffending outcomes. Medium for employment outcomes
Substance misuse	Low
Homeless	Low
Care leavers	Low

15.2 Interventions in the literature

Extended care interventions allow care leavers to remain in their care placements beyond the legal adult age, which is typically 18. The goal of these interventions is to replicate the support structure that non-care experienced youth often receive from their parents. By providing this extended support, the interventions aim to give care leavers more time to complete their education, gain employment, secure stable housing, and avoid pitfalls, such as homelessness, incarceration, and substance abuse. Extended care differs from independent living programmes, discussed in Chapter 13, which aim to help care leavers build the necessary skills to live independently and often involve the provision of housing.

The impact of extended care policies has been mostly assessed in the United States (Jones 2019; Taylor et al., 2024; Nunez et al., 2022). However, the Staying Put arrangement is a UK initiative that allows care leavers to remain with foster carers past the age of 18 (evaluated by Munro et al., 2012).

15.3 The scale and quality of the evidence

Overall, the quality of evidence was low. Primary studies tended not to have comparison groups, and those that did had a high risk of bias due to potential confounding factors.

The evidence consisted of two systematic reviews (Jones 2019; Taylor et al., 2024) and one grey literature evaluation (Munro et al., 2012). Jones 2019 synthesised the findings of 13 cross sectional and longitudinal papers comparing outcomes for young people in extended care to youth who left care at 18. Taylor et al., 2024 included two studies that employed a quasi-experimental method to control for systematic differences in the two groups. However, these two papers were assessed by the review authors to have high risk of bias.

Finally, Munro et al., 2012 employed a process and impact evaluation of the Staying Put programme using qualitative interviews and administrative data to compare outcomes of care leavers who stayed with their foster carers and care leavers who did not. However, the sample size was small, with only 21 care leavers taking part, which means the generalisability of the findings is limited.

15.4 Evidence of impact

The evidence looked at outcome measures related to employment, housing, education and access to health care. There is promising evidence that remaining in extended care may improve all of these outcomes. However, it is not possible to establish a causal relationship with confidence without more robust trials.

Jones 2019 found that youth who remained in the care system beyond the age of 18 were more likely to have obtained high school diplomas and vocational certificates, to be employed, and avoid homelessness or arrest by the age of 19, compared to those who left the care system at the age of 18. Similarly, Taylor et al., 2024 reported consistent positive impacts of extended care, noting higher rates of high school completion, fewer instances of homelessness, and fewer interactions with the criminal justice system.

Jones 2019 also found evidence that those in extended care were more likely to have access to state-funded healthcare (85% vs. 63%) and to be receiving mental health and substance misuse support at the age of 19 compared to those who did not stay in extended care. Note that the study authors interpreted this as a positive finding, showing that young people were better able to access needed support.

The qualitative impact evaluation of Staying Put found that care leavers reported having greater feelings of home life stability and larger support networks compared to those who left care at 18 (Munro et al., 2012). Those who did not experience extended care tended to return to their biological family and in most cases had to leave for alternative accommodation quickly. Additionally, both care leavers and foster carers argued that without extended care, additional challenges, such as finding stable housing, managing finances and self-care would have made it more difficult for them to complete their education. While the qualitative evidence from this study was consistent and compelling, given the small sample size and absence of a control or comparison group, these conclusions are highly tentative.

15.5 Lessons for delivery

The evaluation of Staying Put identified two factors that may hinder the impact of extended care (Munro et al., 2012). Interviews with professional support staff working on the programme suggested that foster carers were often treating adult care leavers as children by enforcing strict curfews. Delivery staff felt that, for extended support to be successful, care leavers should be treated as adults and given the freedom to make their own decisions and learn from them.

Additionally, the evaluation found that some local authorities delivering Staying Put arrangements were only allowing care leavers already engaged in education, employment, or training (EET) to remain with their foster carers. The authors argued that this restriction excluded some of the most vulnerable care leavers from stable accommodation and support from a caring adult.

Question 20

16.  Multiple disadvantage

Accepted Answer

This chapter brings together the findings from the review on multiple disadvantage. It looks at the scale and quality of the available evidence, the key findings, and the key limitations.

Most interventions target multiple disadvantages

As discussed in Chapter 4, it is very common for the four disadvantages covered in this review to co-occur. In particular, criminal convictions, substance misuse, and homelessness are often co-occurring: many people experience more than one of these disadvantages at the same time. Care leavers are more likely than other young people to experience homelessness, substance misuse, or to come into contact with the justice system. All four of the population groups are more likely than the wider population to experience mental health issues.

Given this, most interventions in the literature targeted multiple disadvantages. They did this in one of two ways:

Many interventions had a main focus, such as homelessness, or substance misuse, but included additional elements aimed at addressing other co-occurring needs. For example, the Adult Connections Team programme, an ILP for care leavers, included a mentoring component for developing independence-related skills alongside targeted employment support (Leathers et al., 2023). Another example is Housing First programmes that provide permanent housing for those experiencing homelessness, alongside intensive case management to identify and address additional support needs (Krahn et al., 2018).
Other interventions were explicitly intended to help those experiencing multiple disadvantages. For example, community residential interventions such as Oxford Houses provide safe housing with a supportive sober community for ex-offenders who struggle with addiction (Connell et al., 2023). Similarly, case management interventions are typically aimed at helping people experiencing homelessness with additional disadvantages such as substance misuse by providing a combination of coordinated support to address their specific needs (De Vet et al., 2013).

Only a small number of interventions in the literature specifically and narrowly targeted a single disadvantage. One example is Medication Assisted Treatment (MAT), which aims to treat opioid use disorder. These interventions could, however, be used alongside other interventions aimed at other disadvantages.

Detailed evidence on multiple disadvantage was limited

Despite the fact that most interventions targeted multiple disadvantages, the evidence included in the review did not tend to disaggregate the impact of interventions for different groups. Studies of interventions that included both ex-offenders and people with substance misuse issues, for example, did not tend to assess whether the intervention was more effective for one group or the other, nor whether it was more effective for people experiencing both disadvantages.

For example, Whitworth et al., 2024 carried out a meta-analysis to determine the impact of IPS on different outcome measures. However, only a small number of studies in the meta-analysis had samples that included people experiencing homelessness or those experiencing substance misuse, and the analysis did not distinguish between the two groups when looking at impacts.

Some studies looked at whether more ‘intensive’ interventions were more effective than less intensive ones, and this typically meant interventions that provided holistic and wrap-around services, including for other disadvantages. For example, one systematic review compared housing interventions that provided basic accommodation for people experiencing homelessness to those that included accommodation, alongside targeted support (Keenan et al., 2021).

Overall, this means we cannot draw firm conclusions about precisely which interventions work best for which groups, at least for ex-offenders, people with substance misuse issues, and people with experience of homelessness.

Interventions that address multiple disadvantages appear to be more effective than those that do not

A conclusion from the review is that interventions that explicitly aim to address multiple disadvantages appear to be more effective than those aimed at addressing only a single disadvantage. However, this conclusion is tentative due to the limitations of the evidence discussed earlier and the fact that only a small number of interventions were focused on a single disadvantage limiting a direct comparison.

For example, housing interventions offering additional individualised support had a larger effect on improving housing stability for people experiencing homelessness compared to interventions offering just housing support alone (Keenan et al., 2021).

Interventions that narrowly aimed to reduce substance misuse tended to succeed in that but did not have wider impacts (Maglione et al., 2018; Eddie et al., 2019). Medication Assisted Treatment (MAT), which involves use of approved medication with counselling to treat opioid use disorder, had no wider impacts on employment or mental health outcomes. Similarly, 12-step programs like Alcoholic Anonymous were not shown to have an impact on employment outcomes. By contrast, peer mentoring interventions that also provided housing, such as Therapeutic Communities and Oxford Houses, showed evidence of impact on both substance misuse and employment outcomes.

One of the most well-evidenced interventions in the literature was IPS. The evidence suggests that part of the reason IPS is effective may be because it provides a broad support package tailored to the participants’ needs, which may include help to find stable accommodation, support with managing substance use and treatment for mental illness. IPS also has a ‘zero-exclusions’ policy, meaning no one is excluded for having a substance misuse issue, or due to a criminal record. By design, therefore, IPS includes people with multiple disadvantages and actively engages in supporting them with their issues.

Mental health issues are a pervasive co-occurring disadvantage

Mental health was discussed throughout the literature, whether as a cause or a consequence of other disadvantages, or simply as a co-occurring disadvantage, perhaps with similar underlying roots, such as childhood trauma. Overall, there was a consensus in the literature that a sensitive consideration of mental health was an essential component of almost any intervention aimed at any of the population groups considered in the review.

Many of the interventions in the literature included a secondary mental health component. For example, ILPs that focused on providing supported living for care leavers included access to in-house carers trained in providing mental health support (Neagu and Dixon, 2020a; Neagu and Dixon, 2020b). Similarly Housing First interventions tended to include clinical support for managing poor mental health once participants are settled into stable accommodation (Saldanha et al., 2022; dePass et al., 2023; Marshall et al., 2020).

Other interventions in the literature were primarily focused on mental health but included secondary components aimed at homelessness, or substance misuse. For example, the Substance Use Treatment and Re-entry (STAR) programme for youth offenders offered both cognitive behavioural therapy alongside case management to address additional needs, some of which could include support with managing substance use (Hunter and Huang, 2014).

Lastly, there were interventions in the literature that started as interventions for people with mental health issues and have since been adapted for use with other populations. The clearest example here is Individual Placement and Support (IPS).

Question 21

17.  Strengths, limitations and directions for future research

Accepted Answer

This chapter summarises the strengths and limitations of the evidence base and this systematic review with some considerations for future research.

Constraints of the meta-review approach

Due to the large number of papers included during title and abstract screening, an addendum was made to the review protocol. The focus shifted to including existing systematic reviews and high-quality grey literature. This created a risk that the review would include less evidence for some populations than others. To mitigate this, after full-text screening, we conducted a mapping exercise to highlight any populations and outcomes not covered by our selected reviews. We then systematically selected a range of primary studies to cover these gaps. Additionally, incorporating grey literature captured a broader range of evidence not available in academic databases.

An important limitation of this ‘meta-review’ approach is that systematic reviews, by their nature, do not include as much detail and nuance as individual primary research studies. This means there was limited evidence for Research Question 2, regarding how interventions are best delivered, by which organisations, and what lessons there were for delivery going forward.

The advantage of the ‘meta-review approach’ is that it enabled the review to maintain a broad scope, providing a synthesis of evidence across the four target populations and a wide range of outcomes. The review should be seen as a starting point, rather than an end point. For interventions that have been highlighted in this report as promising, a valuable next step would be a more detailed review of the primary literature. For example, a specific search for studies on recovery housing, such as Oxford Houses, could uncover relevant research not identified in this review.

The review did not explicitly search for process evaluations. The systematic reviews included tended to focus on impact, which limited the discussion of delivery lessons across interventions. There are likely to be relevant published process evaluations covering some of the interventions discussed in this report. Therefore, for interventions with evidence of positive or promising impact, such as IPS, a more thorough search for this body of evidence would be beneficial.

Like many systematic reviews and meta-reviews, there are limitations due to the delay between the publication of primary studies and their inclusion in systematic reviews. Our search strategy included reviews published up to September 2024. Consequently, primary studies relevant to our research questions published after this period may have been overlooked.

US-centric evidence

Most of the evidence across interventions originated from the US, raising questions about the generalisability of these findings to a UK context. This is especially important for interventions such as IPS, which are being more widely tested in the UK. The success of IPS has shown variability even within different UK contexts, which suggests that results in the UK may differ to those in the US (Hofman et al., 2024).

The healthcare and justice systems in the US and the UK differ fundamentally, affecting access to care, treatment protocols, and patient outcomes. This is especially important for health-related outcomes from interventions around substance misuse. For instance, IPS and other employment support interventions showed no significant effect on opioid use disorder, which is less prevalent in the UK (Marsden, 2024). Evidence on ex-offenders was particularly specific to the US context, with interventions like inmate work programmes being specific to states such as Minnesota (Duwe and McNeeley, 2020, Bohmert and Duwe, 2012, Duwe, 2014).

Future research should seek additional evidence from the UK or countries with a similar healthcare and justice systems to ensure that the conclusions are relevant and applicable to the UK setting.

Overlapping populations

The review identified areas where research, though acceptable in quality, lacked direct relevance to target populations. Specific populations such as homeless individuals, ex-offenders, those with substance misuse issues, and care-leavers were often not separately reported. For instance, while there was substantial evidence on housing interventions for homeless populations, these studies often overlapped with other groups like those with substance misuse issues. However, no specific attention was given to overlapping disadvantages or subgroup analysis in some of the included studies.

Similarly, some of the literature on IPS stressed that findings should not be generalised for ex-offenders and people experiencing homelessness. Many studies tended to aggregate data across these diverse groups, which obscures the unique challenges and needs each subgroup faces. For instance, while IPS might be effective overall beyond those suffering with severe mental health issues, it is unclear whether it is equally beneficial for a homeless individual versus an ex-offender, or if specific adjustments are needed to address their distinct circumstances. The risk is that conclusions about the positive impact of IPS lead to it being uncritically applied to diverse groups, overlooking the need to tailor the intervention to each individual’s specific needs.

Moreover, the compounded impact of multiple disadvantages is often not addressed in the studies. Individuals who face multiple challenges, such as homelessness and substance misuse simultaneously, may experience a different set of barriers compared to those dealing with a single issue. Without distinguishing the effects and interactions of these multiple disadvantages, it is difficult to develop comprehensive and effective employment interventions. Future research should prioritise disaggregating data and providing detailed findings for each specific group, as well as exploring the compounded effects of multiple disadvantages, to inform more precise and effective policy and practice recommendations

Quality of evidence

This review’s methodology was robust, employing comprehensive search strategies and transparent inclusion and exclusion criteria.

The use of AMSTAR 2 for systematic reviews and a bespoke appraisal tool for other evidence types ensured the inclusion of only high-quality evaluations.9 While this increased the credibility of the synthesis, it also meant that some relevant but unevaluated interventions were excluded. Scoping interviews with third sector organisations revealed that many UK organisations conduct a wide range of interventions that either lack robust evaluations or have not been evaluated at all. The absence of evaluations in this review does not imply a lack of successful interventions in practice.

The included reviews highlighted the need for higher quality evidence in several areas. The evidence across interventions and populations within systematic reviews was limited by small sample sizes, lack of comparison groups, non-experimental designs and heterogeneous outcome measures. For example, employment outcomes could be measured by days, weeks and months (Marsden, 2024). Additionally, extended care and independent living interventions for care leavers required higher quality evidence beyond qualitative interviews.

Priority areas for further investigation

Based on the results of this review, there are several possible priority areas for further investigation. There are some areas that would benefit from a closer review of the primary literature:

a review of process evaluations of IPS interventions to understand why and how these interventions bring about positive outcomes, and to understand the key considerations when applying IPS to groups other than those with severe mental health issues
a more detailed review of the primary literature on Employment Support interventions, particularly those aimed at ex-offenders. See Cook et al. (2014) and Connell et al. (2023) as starting points
a more detailed search for and review of the literature on recovery housing. See Connell et al. (2023), Reif et al. (2013), and Magura and Marshall (2020) as starting points
a further review of the evidence on independent living programmes. See Jacobs et al. (2015), and Greenen et al. (2013), both cited in Greeson et al., (2020) as a starting point

There are also some areas that show evidence of promise but that may benefit from further evaluation in a UK setting. For example: IPS for ex-offenders or homeless populations; recovery housing; inmate work programmes; and independent living programmes for care leavers. Finally, there may be benefit to evaluating interventions that are already being delivered in a UK context, particularly those being delivered by third sector organisations.

Question 22

18.  ​​References

Accepted Answer

Atkin-Plunk, C. (2023). Examining the effects of a transitional employment program for formerly incarcerated people on employment and recidivism: a randomized controlled trial during COVID-19. Journal of Experimental Criminology, 1-29.

Aubry, T., Bloch, G., Brcic, V., Saad, A., Magwood, O., Abdalla, T et al. (2020). Effectiveness of permanent supportive housing and income assistance interventions for homeless individuals in high-income countries: a systematic review. Lancet Public Health, e342-360.

Bassuk, E. L., DeCandia, C., Tsertsvadze, A., & Richard, M. (2014). The effectiveness of housing interventions and housing and service interventions on ending family homelessness: A systematic review. American Journal of Orthopsychiatry.

Bond, G. R., Drake, R. E., & Pogue, J. A. (2019). Expanding Individual Placement and Support to Populations With Conditions and Disorders Other Than Serious Mental Illness. Psychiatric Services.

Bonella, B., Beirwolf, K., Coleman, L., Sterger, C., Pulli, K., Anguiano, C., & Barton, K. (2019). Mixed methods evaluation of formal mentoring: Journey UP for aging out of foster care. Global Social Welfare, 7, 121-130.

Bretherton, J., & Pleace, N. (2019). Is Work an Answer to Homelessness? Evaluating an Employment Programme for Homeless Adults. European Journal of Homelessness, 59-83.

Carnemolla, P., & Skinner, V. (2021). Outcomes Associated with Providing Secure, Stable, and Permanent Housing for People Who Have Been Homeless: An International Scoping Review. Journal of Planning Literature, 467-635.

Connell, C., Birken, M., Carver, H., Brown, T., & Greenhalgh, J. (2023). Effectiveness of interventions to improve employment for people released from prison: systematic review and meta-analysis. Health and Justice, 1-18.

Connell, C., Furtado, V., McKay, E., & Singh, S. (2017). How effective are interventions to improve social outcomes among offenders with personality disorder: a systematic review. BMC Psychiatry, 1-11.

Cook, P. J., Kang, S., Braga, A. A., Ludwig, J., & O’Brien, M. E. (2014). An Experimental Evaluation of a Comprehensive Employment-Oriented Prisoner Re-entry Program . Journal of Quantitative Criminology, 355-382.

Coren, E., Hossain, R., Pardo, J., & Bakker, B. (2016). Interventions for promoting reintegration and reducing harmful behaviour and lifestyles in street‐connected children and young people: a systematic review. Campbell Systematic Reviews, 12(1), 1-198.

Davey, C., Elsby, A., Erskine, C., Hill-Newell, M., Mon, L., Palmer, H et al. (2013). Evaluation of the Care Leavers Social Impact Bong (SIB) programme: Final evaluation report. Retrieved from https://assets.publishing.service.gov.uk/media/64b14dac07d4b8000d3472ea/Evaluation_of_the_Care_Leavers_Social_Impact_Bond__SIB__programme.pdf

Davis, L. M., Bozick, R., Steele, J. L., Saunders, J., & Miles, J. N. (2013). Evaluating the Effectiveness of Correctional Education: A Meta-Analysis of Programs That Provide Education to Incarcerated Adults. Retrieved from https://www.rand.org/pubs/research_reports/RR266.html

De Pass, T., Dada, O., Lund, J., John, J., & Kidd, S. A. (2023). A scoping review of housing stabilization interventions for youth experiencing homelessness. Children and Youth Services Review.

de Vet, R., Luijtelaar, M. J., Brilleslijper-Kater, S. N., Vanderplasschen, W., Beijersbergen, M. D., & R. L. M., J. (2013). Effectiveness of Case Management for Homeless Persons: A Systematic Review. American Journal of Public Health, e1-1915.

Department for Work & Pensions. (2014). Evaluation of day one mandation of prison leavers to the Work Programme. Retrieved from https://assets.publishing.service.gov.uk/media/5a7dfee0e5274a2e8ab45288/rr897-evaluation-day-one-mandation.pdf

Dixon, J., Quilgars, D., & Wenham, A. (2021). Relationships First? The initial two years of Haringey Housing First Project for Care Leavers. Retrieved from https://centrepoint.org.uk/sites/default/files/2023-06/evaluation-of-haringey-housing-first-for-care-leavers.pdf

Doucet, M. M., Greeson, J. K., & Eldeeb, N. (2022). Independent living programs and services for youth ‘aging out’ of care in Canada and the U.S: A systematic review. Children and Youth Services Review.

Duwe, G. (2015). An outcome evaluation of a prison work release program: Estimating its effects on recidivism, employment, and cost avoidance. Criminal Justice Policy Review, 26(6), 531-554.

Duwe, G., & McNeeley, S. (2017). The effects of prison labor on institutional misconduct, postprison employment, and recidivism. Corrections, 5(2), 89-108.

Eddie, D., Hoffman, L., Vilsaint, C., Abry, A., Bergman, B., Hoeppner, B et al. (2019). Lived Experience in New Models of Care for Substance Use Disorder: A Systematic Review of Peer Recovery Support Services and Recovery Coaching. Frontiers in Psychology, 1-12.

Everson-Hock, E. S., Jones, R., Guillaume, L., Clapton, J., Duenas, A., Goyder, E et al. (2011). Supporting the transition of looked-after young people to independent living: a systematic review of interventions and adult outcomes. Child Care Health Dev.

Feather, J., Allen, D., Crompton, R., Jones, Z., Christiansen, A., & Butler, G. (2023). Transitional support interventions for care leavers: A scoping review. Children & Society, 1419-1879.

Foster, R., Small, L., Foster, S., Oonagh Skrine, G. H., & Turnbull, P. (2013). Evaluation of the Employment and Reoffending Pilot: Lessons learnt from the planning and early implementation phase. Retrieved from https://assets.publishing.service.gov.uk/media/5a7ce859e5274a724f0be3e1/employment-reoffending-pilot.pdf

Furey, R., & Harris-Evans, J. (2021). Work and resilience: Care leavers’ experiences of navigating towards employment and independence. Child & Family Social Work, 404-414.

Gibb, J., & Edwards, A. (2017). From Care to Independence: Findings from research supported by The Big Lottery, conducted in partnership with The Prince’s Trust. Retrieved from https://assets.ctfassets.net/qq0roodynp09/1MEpRQkUrs3mfhESU2yakN/2dc48439c49e91c54d0201562bf72f4a/From_Care_to_Independence_full_report.PDF

Greeson, J. K., Antonio R. Garcia, F. T., Chacon, A., & Ortiz, A. J. (2020). Interventions for youth aging out of foster care: A state of the science review. Children and Youth Services Review.

Gypen, L., Vanderfaeillie, J., Maeyer, S. D., Belenger, L., & Holen, F. V. (2017). Outcomes of children who grew up in foster care: Systematic-review. Children and Youth Services Review, 74-83.

Häggman-Laitila, A., Salokekkilä, P., & Karki, S. (2018). Transition to adult life of young people leaving foster care: A qualitative systematic review. Children and Youth Services Review, 95, 134-143.

Harrison, D. N., Dixon, J., Sanders-Ellis, D., Ward, J., & Asker, P. (2023). Care leavers’ transition into the labour market in England. Retrieved from https://www.education.ox.ac.uk/wp-content/uploads/2023/01/CareLeaversLabourMarket.pdf

Harrison, J., Krieger, M. J., & Johnson, H. A. (2020). Review of Individual Placement and Support Employment Intervention for Persons with Substance Use Disorder. Substance Use & Misuse, 636-643.

Heerde, J. A., Hemphill, S. A., & Scholes-Balog, K. E. (2018). The impact of transitional programmes on post-transition outcomes for youth leaving out-of-home care: a meta-analysis. Health and Social Care in the community, e15-e20.

Hellström, L., Pedersen, P., Christensen, T. N., Wallstroem, I. G., Bojesen, A. B., Stenager, E et al. (2021). Vocational Outcomes of the Individual Placement and Support Model in Subgroups of Diagnoses, Substance Abuse, and Forensic Conditions: A Systematic Review and Analysis of Pooled Original Data. Journal of Occupational Rehabilitation, 699-710.

Hofman, J., Picken, N., Hutchinson, J., Harris, A., Stewart, K., & Disley, E. (2024). Process evaluation of the Individual Placement and Support for Alcohol and Drug Dependence (IPS-AD) trial.

Hornberger, A. (2014). A Randomized, Controlled Microtrial of an Ex-Offender-Focused Job Search Motivation Intervention. The George Washington University.

Hunter, S. B., & Huang, C. Y. (2014). Substance Use Treatment and Reentry (STAR) Program: Final Evaluation Report.

Jason, L. A., Olson, B. D., & Harvey, R. (2015). Evaluating Alternative Aftercare Models for Ex-Offenders. Journal of Drug Issues, 53-68.

Jones, L. (2019). Remaining in Foster Care After Age 18 and Youth Outcomes at the Transition to Adulthood: A Review. Journal of Language and Social Psychology.

Keenan, C., Miller, S., Hanratty, J., Pigott, T., Hamilton, J., Coughlan, C et al. (2021). Accommodation-based interventions for individuals experiencing, or at risk of experiencing, homelessness. Campbell Systematic Reviews, 1-93.

Kim, M., Byrne, A. M., & Jeon, J. (2022). The Effect of Vocational Counseling Interventions for Adults with Substance Use Disorders: A Narrative Review. International Journal of Envrionmental Research and Public Health, 1-11.

Krahn, J., Caine, V., Chaw-Kant, J., & Singh, A. E. (2018). Housing interventions for homeless, pregnant/parenting women with addictions: a systematic review. Journal of Social Distress and Homelessness.

Lakrouf, R., Roy, L., Arbaud, C., & Stich, C. (2024). Interventions for pregnant youth and young mothers experiencing homelessness: A systematic review. Children and Youth Services Review, 1-12.

Leathers, S., Holtschneider, C., Ludington, M., & Ross, E. (2023). Mentoring, employment assistance, and enhanced staff outreach for older youth in care: Outcomes from a randomized controlled trial. Children and Youth Services Review, 153, 107095.

Lesnick, J., Abrams, L. S., Angel, K., & Barnert, E. S. (2023). Credible messenger mentoring to promote the health of youth involved in the juvenile legal system: A narrative review. Current Problems in Pediatric and Adolescent Health Care.

Lindner, A. R., & Hanlon, R. (2024). Outcomes of youth with foster care experiences based on permanency outcome – Adoption, aging out, long-term foster care, and reunification: A systematic review. Children and Youth Services Review.

Lohmeyer, B. A., McGregor, J. R., Crittenden, Z., & Hartung, C. (2024). Mentoring for care-experienced young people: A rapid review of program design. Children and Youth Services Review.

Maglione, M. A., Raaen, , L., Chen, , C., & Gulrez Azhar, N. (2018). Effects of medication assisted treatment (MAT) for opioid use disorder on functional outcomes: A systematic review. Journal of Substance Abuse Treatment, 28-51.

Magura, S., & Marshall, T. (2020). The Effectiveness of Interventions Intended to Improve Employment Outcomes for Persons with Substance Use Disorder: An Updated Systematic Review. Substance Use & Misuse, 2230-2236.

Marsden, J., Anders, P., Shaw, C., Amasiatu, C., Collate, W., Eastwood, B et al. (2024). Superiority and cost-effectiveness of Individual Placement and Support versus standard employment support for people with alcohol and drug dependence: a pragmatic, parallel-group, open-label, multicentre, randomised, controlled, phase 3 trial. eClinicalMedicine, 1-16.

Marshall, C. A., Boland, L., Westover, L. A., Goldszmidt, R., Bengall, J., Aryobi, S et al. (2022). Effectiveness of employment-based interventions for persons experiencing homelessness: A systematic review. Health and Social Care, 2142-2169.

Marshall, C. A., Boland, L., Westover, L. A., Marcellus, B., Weil, S., & Wickett, S. (2020). Effectiveness of interventions targeting community integration among individuals with lived experiences of homelessness: A systematic review. Health & Social Care in the Community.

McCoy, S., & Kempton, O. (2018). More than bednights: An evaluation of the Nightstop Service. Retrieved from https://www.depaul.org.uk/wp-content/uploads/2020/07/More-than-bednights_full-report.pdf

Medlock, E. (2009). Preparing inmates for community re-entry: An employment preparation intervention. University of Oregon.

Ministry of Housing, Communities and Local Government. (2018). STRIVE Evaluation: Final report. Retrieved from https://assets.publishing.service.gov.uk/media/5b6c3fe9ed915d31035ccbb4/Strive_Evaluation_2018.pdf

Mitchell-Smith, D. Z., Caton, D. S., & Potter, D. A. (2020). Shared Lives 16+(Pilot). Retrieved from https://assets.publishing.service.gov.uk/media/5fa411b7e90e07041dd0fbff/Shared_Lives.pdf

Mollidor, C., Bierman, R., & Akhurst, E. (2020). Evaluation of the Care Leavers Partnership: Southwark Council and Catch 22: Evaluation Report. Retrieved from https://assets.publishing.service.gov.uk/media/5fa014268fa8f57f3abf0d41/Catch22Southwark-_Ipsos_MORI_final_report.pdf

Montgomery, P., Donkoh, C., & Underhill, K. (2006). Independent living programs for young people leaving the care system: The state of the evidence. Children and Youth Services Review, 1435-1448.

Munro, E. R., Lushey, C., Maskell-Graham, D., Ward, H., & Holmes, L. (2010). Evaluation of the Staying Put: 18 Plus Family Placement Programme: Final Report. Retrieved from https://assets.publishing.service.gov.uk/media/5a7b06bbed915d429748c477/DFE-RR191.pdf

Neagu, M., & Dixon, J. (2020b). The Fair Ways Staying Close Project: Evaluation report. Retrieved from https://www.education.ox.ac.uk/wp-content/uploads/2024/07/Staying_Close_Fair_Ways.pdf

Neagu, M., Centre, R., & Dixon, J. (2020a). The Portsmouth Aspiration Staying Close Project: Evaluation report. Retrieved from https://www.york.ac.uk/media/spru/projectfiles/Staying_Close_Portsmouth.pdf

Newton, D., Day, A., Margaret Giles, J. W., Graffam, J., & Baldry, E. (2016). The Impact of Vocational Education and Training Programs on Recidivism: A Systematic Review of Current Experimental Evidence. International Journal of Offender Therapy and Comparative Criminology, 187-207.

Northcutt Bohmert, M., & Duwe, G. (2012). Minnesota’s affordable homes program: Evaluating the effects of a prison work program on recidivism, employment and cost avoidance. Criminal Justice Policy Review, 23(3), 327-351.

Nuñez, M., Beal, S. J., & Jacquez, F. (2022). Resilience factors in youth transitioning out of foster care: A systematic review. Psychological Trauma: Theory, Research, Practice and Policy , S72-S81.

O’Donnell, R., Hatzikiriakidis, K., Mendes, P., Savaglio, M., R. G., Kerridgec, G et al. (2020). The impact of transition interventions for young people leaving care: a review of the Australian evidence. International Journal of Adolescence and Youth, 1076-1088.

Office for Health Improvement and Disparities. (2024). IPS for alcohol and drug dependence: data linkage outcomes report.

Probyn, K., Engedahl, M. S., Rajendran, D., Pincus, T., Naeem, K., Mistry, D., & Martin Underwood, R. F. (2021). The effects of supported employment interventions in populations of people with conditions other than severe mental health: a systematic review. Primary Health Care Research & Development, 1-12.

Reif, S., George, P., Braude, L., H, R., Dougherty, A. S., Ghose, S. S., & Delphin-Rittmon, M. E. (2014). Recovery Housing: Assessing the Evidence. Psychiatric Services, e401-402.

Saldanha, S., Tavitian, N., Lehman, E., Carey, L., & Higgs, P. (2024). Effectiveness of the Housing First model among substance users: a scoping review. Housing Care and Support.

Sinclair, D., Chantry, M., De Ruysscher, C., & Magerman, N. (2024). Recovery-supportive interventions for people with substance use disorders: a scoping review. Frontiers in Psychiatry, 15, p.1352818.

Talbot, E. C., Völlm, B., & Khalifa, N. (2015). Effectiveness of work skills programmes for offenders with mental disorders: A systematic review. Criminal Behaviour and Mental Health.

Taylor, C. (2022). Does Mentoring Work with High-Risk Adult Probationers?: The Implementation and Outcomes of an Adult Mentoring Court. American. Journal of Criminal Justice, 48(3), 635-655.

Taylor, D., Albers, B., Mann, G., Jane Lewis, R. T., Mendes, P., Macdonald, G., & Shlonsky, A. (2024). Systematic Review and Meta-Analysis of Policies and Interventions that Improve Health, Psychosocial, and Economic Outcomes for Young People Leaving the Out-of-Home Care System. Trauma, Violence & Abuse, 3441-4293.

Thomas, Y., Gray, M., & McGinty, S. (2011). A Systematic Review of Occupational Therapy Interventions With Homeless People. Occupational Therapy in Health Care, 38-53.

Walton, M. T., & Hall, M. T. (2016). The Effects of Employment Interventions on Addiction Treatment Outcomes: A Review of the Literature. Journal of Social Work Practice in the Addictions, 358-384.

Weightman, A. L., Kelson, M. J., Thomas, I., Mann, M. K., Searchfield, L., Willis, S et al. (2023). Exploring the effect of case management in homelessness per components: A systematic review of effectiveness and implementation, with meta-analysis and thematic synthesis. Campbell Systematic Reviews.

Whitworth, A., Baxter, S., Cullingworth, J., & Clowes, M. (2024). Individual Placement and Support (IPS) beyond severe mental health: An overview review and meta-analysis of evidence around vocational outcomes. Preventive Medicine Reports.

Wilson, D. B., Gallagher, C. A., & MacKenzie, D. L. (2000). A Meta-Analysis of Corrections-Based Education, Vocation, and Work Programs for Adult Offenders. Journal of Research in Crime and Delinquency.

Woodgate, R. L., Morakinyo, O., & Martin, K. M. (2017). Interventions for youth aging out of care: A scoping review. Children and Youth Services Review, 280-300.

Volume of evidence	Low
Evidence of impact	Evidence of promise on employment, education, housing, crime, and health related outcomes
Future research recommendation	Higher quality evidence needed

Cookies on GOV.UK

Executive summary

Acknowledgements

Author details

Glossary of terms

1. Summary

1.1 Background

1.2 Research aims

1.3 Research methods

1.4 Overview of populations and outcomes

1.5 Findings and evidence map overview

Individual Placement and Support

Employment support

Community and peer mentoring

Case management interventions

Education interventions

Housing interventions

Inmate work programmes

Medication assisted treatment

Independent living programmes for care leavers

Creating connections with family and other adults

Extending care arrangements past the age of 18

1.6 What works for ex-offenders, people experiencing homelessness, and people with substance misuse issues?

Individual Placement and Support (IPS)

Employment Support programmes

Inmate work programmes

Community and peer mentoring interventions

Case management interventions

Housing interventions

Correctional education programmes

Substance misuse interventions

1.7 What works for care leavers?

Independent living programmes (ILPs)

Extending existing care arrangements

Creating connections with family and other adults

1.8 The role of mental health

1.9 Limitations in the evidence

2. Introduction

2.1 Policy Background

2.2 Research Aims

2.3 Overview of the report

3. Methods

3.1 Scoping phase

3.2 Search Strategy

3.3 Screening and Quality Appraisal

3.4 Data extraction and Analysis

Table 1: Interpretation of summary tables

4. Overview of the four population groups

4.1 People experiencing homelessness

Population overview

Key interventions in the literature

4.2 Ex-offenders

Population overview

Key interventions in the literature

4.3 People with substance misuse issues

Population overview

Key interventions in the literature

4.4 Care-leavers

Population overview

Key interventions in the literature

4.5 Overlap across populations

Figure 1: Overlap between disadvantaged groups, England, 2010/2011

5. Overview of outcomes

5.1 The pathway into work

Figure 2: Pathway into work

5.2 Overview of outcome categories

Life stability outcomes

Health and wellbeing outcomes

Education and skills outcomes

Employment outcomes

5.3 What works for each stage of the pathway?

Life stability outcomes

Health and wellbeing outcomes

Education and skills outcomes

Employment outcomes

6. IPS and Employment Support

6.1 Individual Placement and Support (IPS)

Summary of findings for IPS

IPS interventions in the literature

The scale and quality of the evidence