Skip Navigation

Link to  the National Institutes of Health  
The Science of Drug Abuse and Addiction from the National Institute on Drug Abuse Archives of the National Institute on Drug Abuse web site
Go to the Home page

Home > Publications > NIDA Notes > Vol. 17, No. 5 > Research Findings

Grouping High-Risk Youths for Prevention May Harm More Than Help
Research Findings
Vol. 17, No. 5 (January 2003)

By Jill Schlabig Williams, NIDA NOTES Contributing Writer

Grouping high-risk youths in early adolescence may inadvertently reinforce problem behavior, according to a NIDA-funded study by researchers Dr. Thomas Dishion and his colleagues at the University of Oregon Child and Family Center. They found that 11- to 14-year-olds at high risk for drug abuse and increasingly serious delinquency who were grouped together for a 12-week program designed to reduce problem behavior actually increased their levels of self-reported smoking and teacher-reported delinquency over 3 years. These youths exhibited significantly worse behaviors than similarly at-risk youths who were given prevention materials to study by themselves individually or received no intervention at all.

Problem Behavior Increases Following Peer Group Intervention
Problem Behavior Increases Following Peer Group Intervention

Teacher reports of problem behavior among the 11- to 14-year-olds in the prevention study showed that adverse effects of the peer-group interventions extended to 3 years after the end of the study period. A score of 0 would indicate a student is seen as never engaging in problem behavior; the maximum in this study, a score of 12, indicates several problem behaviors reported.

These findings raise important questions about the common practice of grouping high-risk, aggressive, or antisocial adolescents together in interventions. The results are consistent with a number of studies elucidating negative peer dynamics within high-risk youth groups.

"The failure of some very sophisticated interventions to improve behavior has been a mystery in the field for many years," says Dr. Elizabeth Robertson, chief of NIDA's Prevention Research Branch. "Now, this research from Dr. Dishion offers some clues to help us understand what's going on."

The youths originally took part in a 12-week cognitive-behavioral intervention called the Adolescent Transitions Program (ATP), designed by Dr. Dishion and his colleagues. ATP aimed to improve the behavior of young adolescents identified by their parents as having problems in at least four of the following risk areas: closeness to parents, emotional adjustment, academic engagement, involvement in positive activities, experience-seeking behaviors, substance use, having friends who use substances, or family substance use history. The study was designed to compare the effectiveness of intervention approaches focusing on parental influence and on peer influence.

A total of 158 at-risk youths (83 boys and 75 girls) aged 11 to 14, enrolled in grades 6 through 8, were randomly assigned to a parent-only group intervention, a peer-only group, a parent-and-peer group, or control conditions that involved no group meetings. The control group received reading materials and videotapes or else no intervention at all.

All of the group interventions used discussion and role-play to build skills in managing emotions, setting limits, building relationships, and refusing drug offers. Parent-family management practices and communication skills were the focus of the parent-only intervention. Researchers followed study participants at 1, 2, and 3 years postintervention, looking at teacher reports of delinquency and the participants' self-reports of frequency of tobacco use -- confirmed by assessment of expired carbon monoxide levels.

At the end of 12 weeks, researchers found that all three intervention groups improved parent-child interactions and skills acquisition. However, at 1-year followup, participants in the peer-only and the parent-and-peer intervention groups had increased rates of self-reported smoking and teacher-reported problem behavior. After 3 years, adolescents grouped with peers for the 12-week program reported twice as much tobacco use as those who were not grouped with peers. In addition, teacher reports of delinquency were 75 percent higher for study participants grouped with peers than for all other groups. Additional analysis revealed that youth with initially low levels of problem behavior showed the greatest increase in problem behaviors.

"For the ATP program, we hypothesized the opposite of what we found," says Dr. Dishion. "We wanted to use the reinforcing power of peers to promote abstinence and improve behavior. Instead, at 1-year followup we found that kids grouped with peers were actually exhibiting more problem behaviors than those who had not been grouped with peers."

Dr. Dishion describes the method by which peers negatively influence one another as "deviancy training": For example, rule-breaking discussions about drug use receive positive responses from peers, further encouraging such behavior and more discussions. In a previous study of adolescents interacting one on one, Dr. Dishion had found that antisocial peers react positively to rule-breaking discussions, while nondeviant peers react positively to normative discussions. He hypothesizes that this process may also occur among youths grouped for treatment, thwarting the assumed benefits of the group setting.

These interactions are very subtle and can occur even in situations designed to promote positive behavior. In the ATP trial, for example, all sessions were videotaped. Later analysis of the tapes revealed that interactions before and after sessions and during breaks actually negated positive interactions that went on during the intervention.

"This research is most relevant for the youngest adolescents -- from junior high to early high school. We're hypothesizing that negative peer influences are most dangerous at the outset of drug use," he says.

A great deal more research is needed to study the unintended effects of interventions. Not all interventions with peer groups exhibiting problem behavior have had adverse effects. "We're working to understand which situations are the most problematic. We don't want to make a blanket statement that aggregating kids will always have a negative effect," he says. "Additional data also exist," Dr. Dishion notes, "that indicate that peer interventions may result in beneficial effects, especially when the groups include prosocial youth."

Dr. Dishion says also that the more effective interventions for adolescents target parenting practices. "Adult involvement is needed to prevent escalating cycles of risk in young people. Without adult involvement, youngsters are left vulnerable to peer effects," he says.

NIDA's Dr. Robertson agrees that finding better ways to help at-risk children and adolescents is of utmost importance. "A small percentage of troubled kids grow into those adults with the majority of social and health problems, including substance abuse, criminality, and sexually transmitted diseases. It is worthwhile to find better ways to identify at-risk kids and intervene with them, their families, and their schools earlier in their lives--as early as preschool."


Dishion, T.J.; McCord, J.; and Poulin, F. When interventions harm: Peer groups and problem behavior. American Psychologist 54(9):755-764, 1999. [Abstract]

Poulin, F.; Dishion, T.J.; and Burraston, B. 3-year iatrogenic effects associated with aggregating high-risk adolescents in cognitive-behavioral preventive interventions. Applied Development Science 5(4):214-224, 2001.


Volume 17, Number 5 (January 2003)

NIDA Notes graphic

Topic Collections of



For additional information about NIDA Notes, send e-mail to

Archive Home | Accessibility | Privacy | FOIA (NIH) | Current NIDA Home Page
National Institutes of Health logo_Department of Health and Human Services Logo The National Institute on Drug Abuse (NIDA) is part of the National Institutes of Health (NIH) , a component of the U.S. Department of Health and Human Services. Questions? See our Contact Information. . The U.S. government's official web portal