Novel programs, international adoptions, or contextual adaptations? Meta-analytical results from German and Swedish intervention research

One of the major dilemmas in intervention and implementation research is adaptation versus adherence. High fidelity to an intervention protocol is essential for internal validity. At the same time, it has been argued that adaptation is necessary for improving the adoption and use of interventions by, for example, improving the match between an intervention and its cultural context, thus improving external validity. This study explores the origins of intervention programs (i.e., novel programs, programs adopted from other contexts with or without adaptation) in two meta-analytic intervention data sets from two European countries and compares the effect sizes of the outcomes of the interventions evaluated. Results are based on two samples of studies evaluating German child and youth preventative interventions (k = 158), and Swedish evaluations of a variety of psychological and social interventions (k = 139). The studies were categorized as novel programs, international adoption and contextual adaptation, with a total of six subcategories. In the German sample, after statistically controlling for some crucial methodological aspects, novel programs were significantly more effective than adopted programs. In the Swedish sample, a trend was found suggesting that adopted programs were less effective than adapted and novel programs. If these results are generalizable and unbiased, they favor novel and adapted programs over adopted programs with no adaptation and indicate that adoption of transported programs should not be done without considering adaptation.