To parents and youth workers, a ground-breaking research article published in 1999 simply provided scientific backing for what they see all the time: Kids who act up are sometimes a bad influence on other kids.
But seven years later, Dr. Yifrah Kaminer says the theory of “deviancy training” is impeding his efforts to raise funds to study adolescent group therapy. Increasingly, he and other researchers are fighting back, publishing challenges to the claim that serving troubled youth in groups often causes more harm than good.
Other researchers, meanwhile, are publishing reports to further support the deviancy training theory, including a book.
This is one academic debate with real implications for much of the country’s youth work, for it involves the benefits and drawbacks of serving youth in groups. A recent report by the theory’s backers, for instance, warns about the dangers of numerous common approaches, including 21st Century Community Learning Center after-school programs, vocational training, group foster care, group homes for mentally ill youth, and community programs and group counseling for deviant youth.
No one says all such programs are bad. At issue is how and when they pose a high risk for deviancy training.
“One bad apple can put bruises in the barrel – no doubt about that,” says Dan Daly, vice president of Girls and Boys Town. But “it’s a bit naïve to say that mixing kids with other deviant kids is an absolute formula for manufacturing a delinquent.”
“I think some folks are getting a bit excited about it because they feel like this issue is undermining their own efforts to create programs for kids,” says Thomas Dishion, whose 1999 article was essentially the opening shot in the debate.
At the root of the dispute: The researchers’ theory of deviancy training, or “contagion,” in youth groups seems to have been stretched beyond what they intended.
An Obvious Problem
The “iatrogenic effect” – a treatment complication caused by the treatment – has been discussed for years, but the controversy in the youth field was ignited with the publication of “When Interventions Harm: Peer Groups and Problem Behavior” (American Psychologist, Sept. 1999). Researchers Dishion, Joan McCord and Francois Poulin reviewed two controlled intervention studies that produced negative effects on high-risk adolescents treated in groups. Coupling those with studies of adolescent social development, they concluded that grouping youth in early adolescence, “under some circumstances, inadvertently reinforces problem behavior.” They said segregating and delivering services to high-risk youth in groups “may inadvertently lead to unintentional harm.”
On its face, that hardly seems controversial. Youth and adults influence each other anywhere they gather, be it recreation centers, backyards, schools or treatment programs. “The best place to get drugs is an NA [Narcotics Anonymous] meeting,” notes Gerry Leventhal, vice president at Graham Windham, a large youth residential agency in New York.
Jennifer Lansford, a research scientist at Duke University’s Center for Child and Family Policy and a colleague of Dishion’s, says contagion typically happens in youth groups like this: Youth “are talking with each other before or after a treatment session and telling each other some trouble that they’ve gotten in to, and one kid might laugh at what another kid is saying and chime in with another story about, ‘Oh, yeah, I did this kind of thing, too.’ ”
That can have “a reinforcing effect on the behavior, and might actually increase the likelihood that those youth would engage in more problem behaviors.”
Youth workers see it all the time: kids relaying tales of their exploits and challenging youth workers in front of other youths. It’s an attitude of “look at how bad I am,” says Mike Garcia, director of residential support services at Children’s Village in Dobbs Ferry, N.Y., which serves youth in child welfare.
Residential programs face a particular challenge, as they often house youth with varying degrees of delinquency and slightly different issues – but can’t keep them apart around the clock. Dennis Smithe faces the challenge at Lutheran Social Services (LSS) in Iowa, where the clients include youth from juvenile justice and child welfare. Smithe, coordinator of quality improvement for the residential sites, says the agency tries to keep youth with severe conduct disorders separate from those with “mental health issues,” because the first group “can really have a dramatic impact, sometimes negatively. It makes it much more difficult to treat the kids with mental health issues.”
Some have taken the contagion theory to mean that grouping together youth who share a problem behavior – such as substance abuse – even in nonresidential sessions is so risky that it shouldn’t be done. That has led many researchers and practitioners to say “hold on.”
Going Too Far?
A research paper soon to be published, “Absence of Iatrogenic or Contagion Effects in Adolescent Group Therapy,” says that over the years, “many have generalized Dishion et al.’s assertion concerning group settings to mean all group settings, disregarding his emphasis on ‘under some circumstances.’ ”
That “has led to multiple instances of federal, state, local and private funders questioning whether they should be funding group therapy” at all, researchers Kaminer, Joseph Burleson and Michael Dennis write in the report, scheduled to appear late this year in the American Journal on Addictions. They say some grant review groups have questioned the ethics of doing research on group therapy for adolescents.
Kaminer says at least two grant proposals he made to the National Institutes of Health (NIH) for research on adolescent substance abuse treatment in groups were returned with comments in the “human subjects protection” section, such as “you can’t do that” to “this is an unacceptable risk to your patients.”
“Meaning, basically that … I’m causing harm,” by placing youth in group settings for treatment, Kaminer says. In making that assumption, NIH’s review board “basically relied on Dishion’s publications.”
“Researchers are always complaining and blaming other people for not getting their grants funded,” Dishion says.
Kaminer is not the only researcher claiming that the contagion theory has been overapplied.
Last year, a Duke University panel asked Mark Lipsey, director of Vanderbilt University’s Center for Evaluation Research and Methodology, to examine the use of peer groups for delinquency prevention. His paper, “The Effects of Community-Based Group Treatment for Delinquency,” says a meta-analysis of nearly 375 published and unpublished studies found “no evidence that working with anti-social youth in groups in community-based programs has negative effects on their subsequent delinquency.”
Also last year, researchers from Vanderbilt University and Harvard University published a study, “Iatrogenic Effects of Group Treatment for Antisocial Youth,” (Journal of Consulting and Clinical Psychology, December 2005) that looked at 66 studies involving 115 treatment groups. While agreeing with Dishion and others that “more research is warranted,” they wrote, “we find little support in the literature for iatrogenic effects, deviancy training-based or otherwise.”
“We’re not saying that negative effects don’t happen,” says one of the researchers, Bahr Weiss. “Just that if they do happen, they don’t seem to happen very much, because if you look at a large bulk of the studies you don’t see it.”
There’s more. For their paper, to be published soon, Burleson, Kaminer and Dennis found similar results in examining data from 400 youths in 67 therapy groups as part of the Cannabis Youth Treatment Study. “Not only did we find no evidence of a peer contagion,” Dennis says. “We actually found evidence that … kids did slightly better in groups” than in some other types of treatment.
Contagion theorists are producing new research of their own.
Guilford Press has just published the findings of the Duke University Executive Sessions Panel on Deviant Peer Contagion, led by Dishion and Dodge, which included more than 25 academicians, policymakers and practitioners. (Deviant Peer Influences in Programs for Youth: Problems and Solutions.)
In a recent Social Policy Report article that draws heavily from the book, Dishion, Dodge and Lansford list 19 scientific studies that show evidence of the adverse effects of “deviant peer-group aggregation on youth outcomes.”
The article, “Deviant Peer Influences in Intervention and Public Policy for Youth,” lists programs and policies that “aggregate deviant peers and sometimes have harmful effects.” The list includes some specific program names, but also some broad descriptions – like “group homes.” Particularly striking is the inclusion of 21st Century Community Learning Centers, which vary so significantly from each other in their approaches and the youths they serve that it’s difficult to imagine how they could be lumped together as potentially harmful.
“We should be studying these [CLC programs] to see which of those programs produce positive effects and which don’t, and why,” Dishion says.
“All we’re trying to do is shed light on this very important issue and to really look at the data carefully to see to what extent we should be approaching interventions with kids based on some of these findings,” he says.
The authors list more than 40 “effective programs that offer viable alternatives to aggregating deviant peers,” such as Functional Family Therapy, Multisystemic Therapy, the Iowa Strengthening Families Program and Big Brothers Big Sisters.
What to Do?
Where does this leave practitioners? “Whenever possible, Don’t aggregate kids who have behavior problems with each other,” Lansford says.
It’s unlikely that agencies will give up on group settings, given the pervasiveness of group interventions and the evidence that they often work. Those approaches are “easy to teach, easy to disseminate, cost-effective, and the kids like it and stay to complete it,” Dennis says.
The upcoming paper by Burleson, Kaminer and Dennis says grouping youth with shared problems and experiences helps them develop positive socializing techniques in a setting that mirrors their daily social interactions, try out new behaviors in a safe environment that offers immediate peer validation and work on interpersonal learning and trust skills.
Minimizing contagion in such settings, researchers and practitioners say, requires focusing on such factors as the degree of structure, the adult/youth ratio, staff training and the mix of youth within the group.
“You’re talking about the quality of the treatment,” says Daly at Girls and Boys Town. “Putting kids together is not the risk factor.”
One big factor is the amount of structure and supervision of the group. The recent paper by Dishion and his colleagues lists unstructured recreation programs among the group approaches that “sometimes have harmful effects.”
Daly says he knows a dangerous program when he sees one: “You see kids not dressed very well, you see kids who ignore the adults around them, you see kids who are way too informal in their interactions with adults and or you see a lot of attitude.
There’s no doubt that deviancy training’s occurring in those kinds of places.”
Practitioners say it’s important to train staff to recognize opportunities for contagion and to intervene. Daly says Girls and Boys Town staffers try to discourage youth from discussing their past deviant acts with one another. At Graham Windham in New York, Leventhal says staff members know the importance of responding when youths talk in ways that could have a bad influence on others. “If you hear something that violates the norm and you don’t confront it, it means you accept it,” he says.
At LSS in Iowa, Smithe says, the staffers identify kids who are potential deviancy leaders and followers, and try to keep them separate, “making sure they’re not spending time alone together.”
In essence, practitioners and researchers are seeking common ground.
“I don’t think we would want the message to be that all groups are bad,” Lansford says. “There are certain situations where you probably do have to group together kids who are acting out.”
“It’s not that Dodge and Dishion don’t have a point when it comes to how kids interact,” Kaminer says. The problem, he says, is that “the volume was turned up” too far on deviancy training.
“We know that unstructured environment and activities can go bad,” he says. “But you cannot say unconditionally, ‘You cannot do group therapy.’ ”
