Send email for updates
Hawkins J.D., Oesterle S., Brown E.C. et al.
Archives of Pediatric and Adolescent Medicine: 2009, 163(9), p. 789–798.
Unable to obtain a copy by clicking title? Try asking the author for a reprint by adapting this prepared e-mail or by writing to Dr Hawkins at firstname.lastname@example.org. You could also try this alternative source.
With its appealing mix of science and community empowerment, the US Communities That Care prevention process has spread to the UK and other countries. This first randomised trial confirmed that given promising towns and rigorous execution, it can curb adolescent smoking and drinking.
Summary Developed by the University of Washington, Communities That Care is a prevention system provided in the USA by the Substance Abuse and Mental Health Services Administration, and promoted in Britain by Rainer, a national charity for under-supported young people. Rather than a specific intervention, CTC offers a process aiming to generate science-based, effective prevention initiatives led by community coalitions. First step is to construct the coalition, which then uses 'diagnostic' surveys and local knowledge to assess the community's strengths and vulnerabilities (protective and risk factors) in relation to preventing substance use problems and delinquency among its children. Next steps are to formulate and implement an action plan to address these, drawing on a menu of proven interventions tackling for example drinking, drug use, smoking, violence, family conflict, life skills, HIV/AIDS risk, dating safety and anger management, depending on community need.
The Community Youth Development Study was the first to randomise communities to implement this process or to act as controls against which the results could be benchmarked. A preceding study had compared communities which according to state authorities were trying to mount prevention initiatives based (like CTC) on protective and risk factors, against matched In respect of size, poverty, diversity, and crime indices. communities not pursuing this strategy. In the event, in 13 of the 20 matched sets, none of the communities had actually implemented such an approach. Of these 13, 12 pairs of communities agreed to join the featured study. One from each pair was randomly allocated to the CTC process. Each was a relatively self-contained, geographically distinct small town.
A previous report had established that CTC and control communities rarely differed in pre-study trends and prevalences of youth substance use and delinquency. Another found equivalent progress in implementing research-based prevention activities, in cross-sector collaboration generally, and in collaboration over prevention initiatives. From this common baseline, on all three implementation measures CTC communities had made greater progress by the year after the CTC process had started. During this time certified CTC trainers had held six training sessions, and community leaders had identified or created coalitions which had selected priority risk factors and made plans to target these with on average three prevention policies/programmes The 12 intervention communities selected 13 different prevention programmes to implement during the 2004–2005 school year, 16 during the 2005–2006 year, and 14 during the 2006–2007 school year. These included school-based programmes (All-Stars, Life Skills Training, Lion's Quest Skills for Adolescence, Project Alert, Olweus Bullying Prevention Program, and Program Development Evaluation Training), community-based youth-focused programmes (Participate and Learn Skills, Big Brothers Big Sisters, Stay Smart, and academic tutoring), and family-focused programmes (Strengthening Families 10–14, Guiding Good Choices, Parents Who Care, Family Matters, and Parenting Wisely). each year over the next three school years aimed at 10–14-year-olds and their families.
The featured study tested whether this activity had made a difference to youth behaviour. Just before the activities had started, it recruited grade five (10–11-year-old) pupils, and then followed them up annually until they were aged 13–14. Nearly three years after the activities had first been implemented. The 4407 pupils While most were recruited from the start of the study, about a quarter could not be recruited until the following year, when some of the activities would already have been underway. Beyond this year, follow-up rates were around 96%. comprised just over three quarters of the relevant classes. By age 12–13, growth in delinquency had already been significantly curbed in CTC communities (and continued to be so), but only over the next year was there a significant impact on substance use. Between ages 12–13 and 13–14, fewer children in CTC towns who had not previously tried these substances had tried drinking (17% v. 25%), smoking (8% v. 12%), or smokeless tobacco (4% v. 6%). There were no such impacts on trying cannabis or inhaling solvents. Combining all these substances, there was a statistically significant reduction in substance use initiation across all the years of the study.
Also, in the final year of the study children in CTC towns were less likely to currently (past month) be using substances. Breaking this down, statistically significant impacts were seen for drinking (16% v. 21%), binge Drinking five or more US drinks at a sitting during past fortnight. drinking (6% v. 9%), and use of smokeless tobacco, but not for cannabis, smoking, solvent abuse or use of other drugs, though in all cases the proportions using these substances were lower in CTC towns.
The authors concluded that within four years of adopting the CTC system, community coalitions can curb the numbers of children starting to use alcohol, tobacco, and smokeless tobacco and committing delinquent acts. By age 14, the result is fewer children drinking or binge drinking or using smokeless tobacco, and fewer delinquent acts, with possible long-term public health benefits.
commentary Among CTC's attractions are the empowerment of local communities to select their own priorities and responses, the possibility that this can underpin wider and lasting improvements, the alliance between localism and centrally determined scientific 'diagnostic' tools and response options, and the way it targets a range of risk factors potentially affecting several social problems. The findings of this first randomised trial should help sustain expansion from its US base to the UK and other countries.
CTC incorporates many of the lessons of international research on community drug/alcohol interventions. These include: devolve decision-making to the community while supplying research-based knowledge; rapid feedback of results motivates participants and keeps projects on track; recruit influential and respected local leaders; considerable lead-in time is needed to build the social and organisational infrastructure for community action, and projects need a few years to fully deliver; project staff must expect and permit adaptation not just of methods but also aims in response to the community's strengths and self-perceived needs; success comes easier in communities where the project's aims are already high on the agenda; a key element is the surer detection and sanctioning of transgressors brought about by the more intensive use of existing legal powers; however, these legal powers must in the first place have the potential to be effective.
Despite these strengths, there are doubts about whether the diagnostic indicators are strongly enough related to substance use to guide the targeting of interventions, and over whether some of CTC's interventions menu (for example, Project Northland, Project ALERT, Life Skills Training, the Midwestern Prevention Project) really are generally effective. Nevertheless, CTC's process is a big step up from interventions mounted without a needs assessment and chosen without regard to the evidence base.
The study's methodological qualities included a comprehensive process to establish that CTC and control communities were truly comparable, Where they differed, in all three cases trends favoured the control communities. tests of significance which did not inflate the chances of significant findings by assuming that the intervention can only improve things, and the construction of single, combined measures of substance use and delinquency, which avoid capitalising on chance positive findings from multiple tests. The main questions concern the generalisability of the findings to other areas; details below.
Possibly the study's towns were unusually well placed to take advantage of the CTC process. In these small, self-contained communities with relatively socially homogenous populations, it was probably easier to identify opinion leaders and for them to exercise widespread influence than in larger or more diverse conurbations. A commentary on the study pointed out that the populations were 90% white and teenage delinquency was relatively rare. Community norms and availability restrictions also have their greatest impacts in self-contained, stable communities whose residents and businesses cannot easily escape their impact. Within this promising type of community, those selected for the study were particularly promising. They were located in seven states identified as leaders in risk- and protection-focused prevention planning. These states in turn identified towns leading the way On inspection by the researchers they did not seem so, but this does not seem sufficient to dismiss the states' assessments. within their borders, which formed half the pool allocated to the CTC approach or to act as controls. It remains to be seen whether the findings would be replicated in severely disadvantaged communities whose very need to address pressing social problems is indicative of the difficulties they face in doing so.
The importance of the community was demonstrated by an evaluation of the first three In Barnsley, Coventry and Swansea. CTC programmes in Britain. Though drawing on the same framework and resources, in each area the approach and the results were very different. Where the areas started from was the key to whether coherent community action emerged. In one, local people were already involved in community development, and the infrastructure and experience of successful partnership working provided a platform from which the new project quickly moved forward. In the other two, poor infrastructure or tensions between professionals and local people seriously impeded implementation.
As the researchers acknowledged, the featured study was a test of rigorously implemented and monitored versions of the CTC process, conditions which may not be widely replicable. The CTC towns also had resources not commonly available. These included a series of 'diagnostic' surveys from the preceding project on which to base their programmes, and the financial and in-kind resources of the study, whose staff included the developers of the CTC process. In several social research areas (1 2 3), studies led by developers or people with other forms of 'allegiance' to the trialled programme have been found to produce more positive findings than fully independent research.
The impacts recorded in this rigorous trial were (relative to other prevention trials) substantial. It is important though to remember that statistically significant reductions in the uptake of substance use emerged only between grades seven and eight. Focusing just on this final year As the programme's US government distributors did in their media release on the study: “By the eighth grade, students in the CTC communities were 32 percent less likely to begin using alcohol, 33 percent less likely to begin smoking, and 33 percent less likely to begin using smokeless tobacco than their peers in the control communities.” http://www.drugabuse.gov/newsroom/09/NR9-07.html, accessed 12 October 2009. when the proportions of children starting to drink, smoke or use smokeless tobacco were cut by a third, gives a false impression of the impact over the entire study and over the entire cohort of children, including those who had already tried these substances before the study started. From this broader and longer perspective, it can be estimated Calculations by Findings based on figure 2 of the featured study and baseline prevalence of having tried these substances reported in table 2. Available on request from email@example.com. that by grade eight about 14–16% fewer children had tried smoking or drinking in CTC communities, and about 6% fewer smokeless tobacco. For every 100 children in grade five, by grade eight these figures meant that respectively 8, 4 and 1 fewer would have tried these substances. Also there were no significant findings in respect of solvent abuse, cannabis, or other prescription or illicit drugs, though except for solvents, these might be expected to emerge only after age 13–14, the last follow-up in the study.
Last revised 19 October 2009
Comment/query to editor
Give us your feedback on the site (one-minute survey)
Open Effectiveness Bank home page
Add your name to the mailing list to be alerted to new studies and other site updates
REVIEW 2015 Prevention of addictive behaviours