It won't work if you just think bigger

Past efforts to reduce population-wide rates of antisocial behavior have tended to involve “scaling up” treatments that work for the individuals involved in clinical trials. But a new paper from the Center for Child and Family Policy at Duke University is fueling the argument against simple magnification and multiplication.

It doesn’t work, Center director Ken Dodge writes in Child Psychology and Psychiatry. No individual-level intervention has yet been proven to have a public health effect. Combating risk factors for antisocial behavior at community level calls for community-wide interventions.

Theoretical models of conduct disorder include risk factors such as neighborhood violence, cultural endorsement of aggression and class divide. Individual-level interventions skirt around these risk factors, Dodge maintains. Ignoring cultural and community causes may lead “to the perpetual replication of new cases with little net impact on community rates of problems”.

In some instances it is simply a matter of mathematics. Dodge takes his own Fast Track preventive intervention as an example. The program was proven to reduce conduct disorder by 50% in high-risk groups. But when rates of conduct disorder in the general population and the challenges of engaging people in the program are taken into account the figure plummets to 2%. [See: Long march shows Fast Track’s the way [1]]

And this is a best-case scenario. Usually programs are not delivered to the letter: adaptations are made, things missed out and timings altered. As the report explains, “most interventions lose fidelity when implemented in replication or in communities, and the loss of fidelity is associated with reduced impact”.

The report also cautions that interventions tend to deal with the “easiest cases”. Delivery on a larger scale means treating more challenging individuals.

Dodge also warns that making assumptions that conditions will remain the same when programs are brought to scale can be disastrous – offering as an example the collapse of the Tennessee Student/Teacher Ratio (STAR) project.

The research team had solid evidence to show that children in smaller classes performed better. But when the intervention was rolled out across California it turned out that there were not enough qualified teachers to go round. Standards dropped significantly.

But it is not all doom and gloom. Dodge identifies some considerable benefits of using individual interventions at community-level if the scaling is done right. He points to the “mutual influence that high-risk youth have on each other”. The more individuals involved in a program, the more they can support one another.

He acknowledges the economies of scale that can be achieved. “Given the high fixed costs of bringing an evidence-based program to a single case, a critical mass of interventionists could be trained and supervised at lower per case costs.”

See Dodge K A (2009), “Community intervention and public policy in the prevention of antisocial behaviour” in Child Psychology and Psychiatry, 50, 1-2, pp 194-200