A research synthesis commissioned by the Youth Endowment Fund in England and Wales is the first to examine systematically whether Communities That Care reduces youth violence. Its results: around ten per cent less violence in CTC areas, an estimate that narrowly misses statistical significance; no measurable effect on other offending; and a set of clear recommendations for implementation, together with a verdict on commissioning in the UK that deserves to be read in full.
Why this review is different
Many studies have examined the effectiveness of Communities That Care, above all in preventing substance use. What was missing was a systematic synthesis of the research with a focus on violence and youth offending. That gap is now closed by a review published in November 2025 on behalf of the Youth Endowment Fund (Valdebenito et al., 2025).
The Youth Endowment Fund was established in 2019 with a £200 million endowment from the UK Home Office. Its mission is to find out what actually works in preventing youth violence and to bring that knowledge into practice. Within that remit, the Fund commissioned a research team led by Sara Valdebenito to evaluate the international CTC research systematically.
The decisive point: the review examined only whether CTC reduces violence and offending among young people. Other important areas in which CTC has effects, such as substance use or mental health, were not part of the analysis. The results are therefore an important piece of the picture, not the picture of CTC’s effectiveness as a whole.
The method: what was examined, and how
The research team combined two methods.
A quantitative synthesis of the impact studies
The team systematically searched the scientific databases for studies that compared CTC areas with comparison areas and recorded violence or offending as an outcome. The outcomes counted were:
- young people’s self-reports of violent behaviour, as perpetrators or as victims
- police-recorded offences such as assault, robbery or criminal damage
- arrests or convictions for violent offences
The studies identified were pooled statistically using a multi-level meta-analysis. This method takes into account that several publications often rest on the same data, for instance when a longitudinal study is published repeatedly over the years. It prevents such dependent findings from distorting the overall result.
A narrative review of the implementation studies
In addition, the team analysed 24 publications on the practical implementation of CTC. This qualitative analysis was meant to establish which conditions favour success, which obstacles arise, and which strategies have proved themselves.
The evidence base: four independent studies from two countries
Despite an extensive search, only four independent impact studies met the inclusion criteria. Those four studies were, however, documented in a total of 13 publications, because their long-term effects were reported over several years. None of the studies came from the United Kingdom: the review’s own summary table records eleven publications from the United States and two from Australia for the impact strand, and nineteen from the United States, three from Australia and three from the Netherlands for the implementation strand.
The (Hawkins et al., 2008 onwards): the methodologically strongest study, a randomised controlled trial. Twenty-four small towns in seven US states were assigned at random to CTC or to a comparison group. The study has been running since 2003 and supplied six of the 13 publications included.
The Pennsylvania study (Chilenski et al., 2019): a quasi-experimental design in which CTC areas were compared with statistically similar control areas.
The Bronzeville study in Chicago (Gorman-Smith et al., 2024): the only study from a large-city setting with high social deprivation. Here CTC was adapted to the particular conditions of a predominantly African-American neighbourhood.
The Victoria study (Toumbourou et al., 2019): the only study outside the United States. It examined whether the CTC approach transfers to the Australian context.
The central results
What a research synthesis does, and what that means for the results
Individual studies can show positive effects, and these did. Why, then, is the overall result of the synthesis not clear-cut?
A research synthesis pools several studies and tests whether an effect holds up across different contexts. That is a stricter test than any single study: if CTC works in a similar way in Pennsylvania, Chicago and Australia under very different conditions, the finding is more robust.
That is also where the difficulty lies. The four studies differ substantially, in their settings (rural against urban), in how CTC was implemented, and in their results. The review puts total heterogeneity at 73 %, and its prediction interval for a future study runs from a relative risk of 0.65 to 1.32, which is to say from a large benefit to the possibility of none. That spread makes the pooled effect statistically more uncertain, even though the individual studies produced positive findings on their own.
The pooled result does not contradict the individual studies. What it says is that, across four very different implementations, the review could not establish that the effect on violence is reliably different from zero, and that how much CTC achieves appears to depend on the conditions on the ground.
Overall effect: seven per cent less violence and offending
Pooling all the studies, young people in CTC areas had a risk of violence and offending about seven per cent lower than in comparison areas: a relative risk of 0.93, with a 95 % confidence interval from 0.84 to 1.02, based on 41 effect sizes from 13 publications. Because of the variation between the studies, this result does not reach the conventional threshold for statistical significance, which, given the differences in context, was to be expected.
Violence against general offending: different effects
Looking at the two outcomes separately reveals an instructive pattern.
Violence outcomes: here the risk reduction was more pronounced, at around ten per cent (a relative risk of 0.87, confidence interval 0.75 to 1.01, from seven studies and 22 effect sizes). Despite the variation between the studies, this finding missed the significance threshold only narrowly (p = 0.07).
General offending: for behaviour such as theft, vandalism or antisocial behaviour there was no measurable effect (a relative risk of 0.97, confidence interval 0.86 to 1.13).
The difference is worth noting: CTC appears to be more effective in preventing serious behaviour than milder forms. One possible explanation is that preventing violence requires structural change at community level, which is exactly what CTC aims at.
Urban implementation: the Chicago example
The Bronzeville study from Chicago (Gorman-Smith et al., 2024) deserves particular attention. Unlike the other studies, it was carried out in a large-city setting with high social deprivation. The research team adapted the CTC approach to local conditions and placed particular emphasis on building the organisational capacity of local partners. The result was a risk reduction of nine per cent in police-recorded offences, including property crime, robbery, aggravated assault, shootings and homicide.
What the effect means in practice
A ten per cent reduction in the risk of violence sounds abstract. What would it mean for a community?
A purely illustrative example: take a town with 10,000 young people between twelve and eighteen in which, say, 400 of them come to police attention for violent offences in a year. A reduction of ten per cent means 40 fewer young people appearing as perpetrators, and correspondingly fewer victims. Over five years, that adds up to some 200 incidents prevented.
Behind those numbers stand young people who do not acquire a criminal record; victims spared harm; families who do not have to live with the consequences of a violent act; and a community in which young people grow up more safely.
The review’s recommendations
From the findings, the authors derive concrete recommendations.
On choosing where to implement
Choose an environment open to prevention: CTC thrives where politicians and administrators are receptive to preventive approaches and provide the resources to match.
Build on existing structures: successful implementations benefit from existing cooperation structures and from a supply of effective programmes. Where these are missing, more lead time is needed.
Choose sensible area boundaries: CTC areas should coincide with administrative structures. That makes funding and cooperation easier.
On implementation
Secure funding for coordination: the funding of the coordinator post is especially critical. When it falls away, CTC coalitions frequently dissolve, even where the substantive work was going well.
Invest in training: the participation of key leaders and coalition members in CTC training proved a central active ingredient. It fosters the adoption of a science-based understanding of prevention, which the studies identify as the most important mediator of CTC’s effects.
Provide continuous support: sustained professional support over a longer period is decisive for implementation fidelity.
Allow adaptation: measured adaptation to local conditions makes implementation easier, as long as the essential components are preserved.
Avoid the known pitfalls
The studies identified recurring obstacles: poor acceptance of the long-term CTC process, difficulty in winning important partners (schools in particular), competition with existing initiatives, and, above all outside the United States, a lack of effective programmes for the target group in question.
The verdict, and how to read it
On the basis of this review, the Fund’s Toolkit rates CTC as low impact, very low confidence and high cost, and advises commissioners in England and Wales: do not commission or scale new CtC programmes until further research demonstrates that the approach can be delivered effectively in the UK context.
Two things are worth knowing about that verdict. First, none of the studies pooled came from the United Kingdom; the concern is about deliverability in a setting where CTC has not run since 2009 and was last evaluated in the mid-2000s, not about the model. Second, the pooled estimate points in the same direction as every other study on the evidence page; what the review could not establish, on the studies available, was that the effect is reliably different from zero.
This synthesis is the first to examine systematically whether CTC reduces youth violence and offending. The results suggest an effect, particularly on violence. The evidence base rests on four independent studies and needs further research, ideally in European contexts, where several of the pitfalls the review names – the shortage of tested programmes above all – have been addressed in recent years through national evidence registers. We think the review is a fair challenge, and the right answer to it is European evidence rather than dismissal. Germany is generating it with CTC-EFF, Sweden, Estonia and Ireland are building the survey base for it, and Croatia contributes survey and implementation evidence; the country pages record where.
References
Chilenski, S. M., Frank, J. L. & Fagan, A. A. (2019). Pennsylvania’s Communities That Care prevention system: Planning, initial implementation, and student outcomes. Community Development, 50(1), 24–45.
Gorman-Smith, D., Cosey-Gay, F., Henry, D. B., Wright, E. M., Quinn, C. & Tolan, P. (2024). Adapting Communities That Care for urban violence prevention: The Bronzeville experience. Journal of Urban Health, 101(5), 864–878.
Hawkins, J. D., Catalano, R. F., Arthur, M. W., Egan, E., Brown, E. C., Abbott, R. D. & Murray, D. M. (2008). Testing Communities That Care: The rationale, design and behavioral baseline equivalence of the Community Youth Development Study. Prevention Science, 9(3), 178–190.
Toumbourou, J. W., Rowland, B., Williams, J., Smith, R. & Patton, G. C. (2019). Community intervention to prevent adolescent health behavior problems: Evaluation of Communities That Care in Australia. Health Psychology, 38(6), 536–544.
Valdebenito, S., Baidawi, S., Rankin, S., Jiang, J., Smith, S., Lewis, J. & Shlonsky, A. (2025). Communities That Care: Toolkit technical report. Youth Endowment Fund / National Children’s Bureau. youthendowmentfund.org.uk/toolkit/communities-that-care