Pilot evaluation of detached youth work
| ISRCTN | ISRCTN80184427 |
|---|---|
| DOI | https://doi.org/10.1186/ISRCTN80184427 |
| Sponsor | Centre for Evidence and Implementation |
| Funder | Youth Endowment Fund |
- Submission date
- 30/06/2026
- Registration date
- 01/07/2026
- Last edited
- 13/07/2026
- Recruitment status
- Not yet recruiting
- Overall study status
- Ongoing
- Condition category
- Other
Plain English summary of protocol
Background and study aims
Detached youth work is a form of youth work that takes place in public spaces where young people choose to spend their time, such as parks, streets and town centres. Detached youth workers go to where young people are, rather than expecting them to come to a youth centre. Detached youth workers build trusting relationships with young people and offer support and provide informal learning. Young people take part voluntarily and can come and go as they want.
Despite a long history in England, there is very little strong evidence about whether detached youth work reduces youth crime and antisocial behaviour. This study aims to test whether a large-scale trial of the impact of detached youth work could work in future. It does this by running a smaller pilot trial first. The pilot has four aims: to test whether this kind of trial is feasible; to look for early signs of whether detached youth work affects local crime and antisocial behaviour; to understand how the work is delivered and experienced; and to estimate what it costs to deliver.
Who can participate?
The study focuses on small, defined public areas called 'patches' where young people are known to gather, rather than on individually recruited people. Forty patches have been identified across four regions of England.
In the patches that receive detached youth work, any young people aged 10 to 17 years can choose to engage with the detached youth workers. Engagement is always voluntary. Young people aged 16 years and over may also be invited to take part in surveys or creative research activities, with their informed consent. Youth workers, managers, residents and other local stakeholders may take part in interviews and focus groups as part of the study.
What does the study involve?
The 40 patches are randomly put into two groups: 20 receive detached youth work and 20 carry on with 'services as usual'. Allocating patches at random helps us understand whether any changes in crime and antisocial behaviour are because of the detached youth work.
In the delivery patches, trained youth workers spend at least four hours each week in public spaces. They build relationships with young people, offer informal support and activities, and respond to young people’s priorities. Before delivery begins, there is a 6-month period in which youth workers get to know the area and the community.
The study measures impact mainly by looking at police-recorded crime and antisocial behaviour in each patch. Researchers also carry out interviews and creative activities with young people, youth workers, managers and community members to understand how the work is delivered and what difference it makes. A separate cost analysis looks at how much it costs to deliver the service. The team will also analyse historical police data from other areas to estimate the likely size of any effect, which is needed to plan the needed sample size of a future impact study.
What are the possible benefits and risks of participating?
Young people in the delivery patches will have access to detached youth work, including relationship building, informal education and signposting to other services. The wider benefit is better evidence about whether and how detached youth work helps young people and communities.
Young people in the control patches will not receive detached youth work during the study but will continue to receive any other services available in their area. Control areas have committed to not delivering detached youth work for the study period, but they can deliver other forms of youth work to address arising risks/needs of young people.
While researchers will not ask sensitive questions, there is a small risk that people taking part in research activities may bring up sensitive topics voluntarily, and they might find some questions uncomfortable. Taking part is voluntary, and they can stop at any time. All detached youth workers and researchers are trained in safeguarding, and the team has robust data protection and ethical procedures in place. The police data used is anonymised and cannot be linked to individuals.
Where is the study run from?
The study is run by the Centre for Evidence and Implementation (CEI), working with Bryson Purdon Social Research, Dr David McLennan (DJM Social Research) and Professor Alex Sutherland (University of Oxford). Detached youth work is delivered across four regional consortia in England: Greater Manchester; Wiltshire and Swindon; Leeds, Wakefield and Barnsley; and Middlesbrough and Stockton (UK).
When is the study starting and how long is it expected to run for?
The study began its co-design phase in October 2025. Set-up runs from April to September 2026; randomisation will take place in July 2026 after study registration. A six-month reconnaissance period begins in October 2026, followed by 12 months of detached youth work delivery from April 2027 to March 2028. Final analysis and reporting take place from mid-2028, with findings expected to be shared by March 2029. In total the study runs for about three and a half years.
Who is funding the study?
Youth Endowment Fund (YEF) (UK)
Who is the main contact?
Amy Hall, amy.hall@ceiglobal.org
Contact information
Principal investigator, Public, Scientific
Nest58
58 Victoria Embankment
London
EC4Y 0DS
United Kingdom
| Phone | +44 (0)7355649122 |
|---|---|
| katherine.young@ceiglobal.org |
Study information
| Primary study design | Interventional |
|---|---|
| Allocation | Randomized controlled trial |
| Masking | Open (masking not used) |
| Control | Active |
| Assignment | Parallel |
| Purpose | Treatment |
| Scientific title | Pilot area-level randomised controlled trial of detached youth work |
| Study objectives | |
| Ethics approval(s) |
Not yet submitted |
| Health condition(s) or problem(s) studied | Youth involvement in violent and non-violent crime and anti-social behaviour |
| Intervention | Detached youth work, as defined by the co-designed shared practice model. The shared practice model describes a model that constitutes a middle ground between delivery best practice (namely, the flexibility inherent to detached youth work) and trialability. The model is summarised here: The principles of detached youth work emphasise that interactions between youth workers and young people take place in public spaces where young people choose to gather – to some extent ‘owning’ the territory – and that young people's voluntary engagement is negotiated on an ongoing basis. These principles also state that detached youth work is relational and youth-centred so that detached youth workers do not set the rules for engagement but rather focus on the young person's strengths and seek to provide positive experiences and challenges. Finally, the principles stress that detached youth work is flexible and responsive to young people's needs, as well as being equity-driven. Project leads use local data and knowledge and community engagement in selecting areas and will undertake a six-month reconnaissance period to gain knowledge and integrate into the community. Each detached youth work team should include two detached youth workers, plus a supervisor or manager depending on the size of the delivery team. The lead detached youth worker will have 2+ years of experience and will oversee delivery and provide regular supervision. Detached youth workers will have undertaken or be working to a level 2 certificate in youth work training and hold a defined set of skills (outlined in the SPM) Detached youth work involves detached youth workers being present for four hours a week in each patch and focusing their engagements with young people on building relationships. These engagements typically take the form of unstructured street-based contact and informal education: initiating conversations, spending time with peer groups, and offering one-to-one and group support or challenge where a young person is open to it. It can also include semi-structured activities that include practical skills; teamwork and collaboration; and skill building, if that aligns with young people’s priorities. Detached youth work delivery will take place across a minimum of 12 months. Nearing the completion of the 12 months of on-street delivery, the ending of the project will be planned to ensure that young people are forewarned of a change or end to the provision. Randomisation process: During the mobilisation phase, the ten eligible patches per consortium (40 patches in total) will be randomly allocated to a study arm, five to detached youth work and five to control (business as usual) using stratification. The randomisation will be stratified to mitigate baseline qualitative and quantitative differences on factors that are predictive of outcomes between intervention and control groups. The optimal stratification dimensions will be best refined once we have full information on the nature of included evaluation patches. However, this section systematically outlines our preferred approach. Stratified randomisation will be conducted by location (i.e., regional consortium), local authority (if necessary – applies to two consortia), median split crime counts in the previous 12 months (from publicly available police.uk data), and, if possible, size of patch measure in metres squared. Specifically, our approach will proceed in the following steps: Stratification by regional consortium: this stage recognises inherent differences in how the shared practice model is delivered locally, ensures comparability between different local contexts, and ensures an equal distribution of DYW delivery across areas. Stratification by local authority: in regional consortia that cover more than one local authority and where at least two patches have been identified per local authority, randomisation will be stratified by local authority to ensure at least one patch is randomised to the intervention group. This is deemed important to ensure local authority commitment across the evaluation period. This level of stratification has been agreed with the Youth Endowment Fund evaluation manager. Stratifying by local authority has been prioritised over other potential stratifiers (deprivation and local area demographics) as it assures the buy-in required from participating local authorities at this stage in the evaluation process by guaranteeing that they will receive delivery in their area, despite constraining the ability to control for variation on these other dimensions. Stratification by median-split crimes per hectare over the previous 12 months: patches within each local authority will be rank-ordered by crime rate and divided into two categories split by the median for the sample of patches in the study. Patches will be randomised 1:1 (1 treatment: 1 control) using a random number generator in a reproducible statistical programme and/or randomisation platform that logs allocations and has a random seed function. Police.uk is easily accessible, publicly available crime data which can be mapped onto the shapefiles generated for each patch. Using the readily available police.uk data as the source of a stratifier (rather than youth-crime data from police forces that will be used for outcome measurement) will allow for the random allocation to be completed very quickly during the mobilisation phase. This step will mitigate baseline differences in crime. Police.uk data uses ‘snap points’ for the location of incidents. For small patches, this may introduce noise and uncertainty to the data. If this challenge is not easily navigable, this step will not be conducted. Stratification by patch size: if deemed necessary and possible upon assessment of the evaluation patches, we will stratify by geographical size. Stratifying by many variables with a small number of units risks problems with ‘sparseness’ – that is, if divided by several variables, it is likely that some combinations will have few or no patches in them. This can lead to severe imbalances that make it infeasible. This step will therefore only be undertaken if the features of patches make it feasible. This also has implications for degrees of freedom in the analysis. While this is manageable within a pilot of this scale, it is noted as a consideration for the design of any subsequent efficacy trial with a larger number of patches. We note that stratifying by crimes per hectare in the previous 12 months incorporates patch size so it may be that it is not necessary/collinear to include patch size as well. The final approach will be established prior to the randomisation based on the assessment of evaluation patches; the evaluator team will conduct the randomisation only once (June 2026). The randomisation will be carried out within SPSS, as it can log allocations and has a random seed function by the pilot trial statistician. |
| Intervention type | Behavioural |
| Primary outcome measure(s) |
|
| Key secondary outcome measure(s) |
|
| Completion date | 31/03/2028 |
Eligibility
| Participant type(s) | |
|---|---|
| Age group | Mixed |
| Lower age limit | 10 Years |
| Upper age limit | 18 Years |
| Sex | All |
| Target sample size at registration | 40 |
| Key inclusion criteria | Randomisation at the 'patch' level, i.e., pre-defined geographic areas of between 10 and 100 hectares. Patches will be selected prior to randomisation and will not change for the duration of the study. Four consortia will each identify 10 patches, providing a total of 40 patches. Any young people (aged 10-18 years) who engage with the detached youth workers will be considered eligible within the patches. Patches must meet these eligibility criteria: 1. No detached youth work has been delivered within the patch from October 2025 to the start of the reconnaissance period (October 2026). 2. There is evidence of an issue of youth crime/ASB that detached youth work might help to reduce. This criterion does not have a numeric threshold, acknowledging variation across contexts and communities. However, consortia were provided with historical data about national averages to help justify what might be considered ‘high’ or ‘normal’ rates of incidents per hectare. 3. The patch is an appropriate size for DYW teams to cover it within one session while also being large enough to allow for some movement of young people, for example, between indoor and outdoor spaces that are favoured differently by young people depending on seasonal changes. Consortia have been asked to identify patches of between 10 and 100 hectares in size. If larger patches are proposed, then consortia are expected to provide evidence that these patches can be covered effectively by the detached youth work team each week. 4. The patch covers an area where a reachable population of young people are known or observed to gather frequently. Consortia are expected to consider young people’s movement between meeting places, and patches should therefore be designed to accommodate some movement. 5. The consortia have sufficient local knowledge, relationships and commitment of relevant organisations to deliver, monitor and influence what takes place in the patch throughout the duration of the project. |
| Key exclusion criteria | Patches that do not meet the eligibility criteria |
| Date of first enrolment | 01/04/2027 |
| Date of final enrolment | 31/07/2027 |
Locations
Countries of recruitment
- United Kingdom
- England
Study participating centre
58 Victoria Embankment
London
EC4Y 0DS
England
Results and Publications
| Individual participant data (IPD) Intention to share | No |
|---|
Study outputs
| Output type | Details | Date created | Date added | Peer reviewed? | Patient-facing? |
|---|---|---|---|---|---|
| Protocol file | 23/06/2026 | 13/07/2026 | Yes | No |
Additional files
- ISRCTN80184427_Protocol_v.3_23June2026.pdf
- Protocol file
Editorial Notes
13/07/2026: Protocol uploaded.
01/07/2026: Study's existence confirmed by the Youth Endowment Fund.
