• By Dartington SRU
  • Posted on Wednesday 20th June, 2012

Trial and error in Zurich

It is one thing to demonstrate the impact of a program with a small group of participants, specially selected for a particular difficulty; it is entirely another to provide an intervention to a community as universal provision and expect similar effects. This is the experience of a team of researchers working in Zurich, Switzerland, where they undertook the large-scale implementation of two proven prevention programs.The rationale for adopting a universal approach is that it is possible to protect both the high-risk children and families while at the same time enhancing health and development for the general population. Universal approaches are also known to reduce the stigma associated with targeted services. While there are a handful of these kinds of studies available, the findings thus far are mixed and tend to be isolated to one type of intervention (for example, school-based or family-based). There are few studies that examine the individual and combined effects of universal services across several contexts. In addition, even fewer evaluations are conducted independently of the program developer.These studies were the starting point for a team of researchers funded by a group of independent foundations and federal offices in Zurich. The z-proso study (the Zurich Project on the Social Development of Children), recruited 56 Swiss elementary (primary) schools across the city of Zurich, which were randomly assigned to receive either a school-based intervention focused on fostering social competence, a parent training program, both interventions or neither (control). The project selected the two evidence-based interventions carefully. They were “conceptually rooted in a developmental and risk-and-resiliency framework” and shown to be effective elsewhere in preventing aggression. The Triple-P Positive Parenting Program and Promoting Alternative Thinking Strategies (PATHS) were chosen. A feasibility study, conducted before the main trial, revealed that the intervention would be appropriate for the needs of the community.The story here is less about the programs but rather the curious finding that, despite offering two well-evidenced and clearly appropriate interventions to a broad population, the effects on social competence and child behavior problems were minimal. The study was designed to compare children in the control schools with those in schools where the PATHS program was delivered alone or Triple-P was delivered alone or the two programs were offered in combination. Not surprisingly, it was expected the combination would be most potent in reducing aggressive behavior and promoting social competence. These dimensions were rated by three sources: the children themselves, their parents and their teachers. Notwithstanding the few expected significant main effects for the PATHS intervention (but only on problem behavior), the study produced a number of divergent results. Many of the analyses revealed no treatment effects at all and there was no added benefit of the combined interventions as compared to when providing individual interventions. In addition, there were no effects observed directly after treatment (typically found in evaluation studies); rather the effects that did appear for those in PATHS-classrooms became evident only in the longer-term follow-up. The research found a significant effect for children’s baseline scores that say something potentially important about targeting preventative interventions. Children with high levels of problems at the outset of the project appeared to benefit most from either or both interventions compared with those with low levels. Thus, while the rationale for the study was to test for impacts with a general population, it may make more sense to target valuable resources.The authors are quick to point out that large-sample field trials typically produce weaker effects than small, developer-led studies. One difficulty in adopting a universal approach is that participation rates tend to be lower than in targeted studies. Who takes up the intervention, and why, and does this have consequences for understanding the results? [Coming up on Prevention Action tomorrow: When replications don’t replicate]. We continue to need to understanding how evidence-based practices can be effectively applied at a population level. ********Source: Malti, T., Ribeaud, D. and Eisner, M. (2011). The effectiveness of two universal preventive interventions in reducing children’s externalizing behavior: A cluster randomized controlled trial. Journal of Clinical Child & Adolescent Psychology, 40(5), 677-692.

Back to Archives