Evidence-based interventions need to be more systematic, not more disruptive.

Comments on the original article, "Disruptive innovations for designing and diffusing evidence-based interventions," by M. J. Rotheram-Borus et al. (see record 2012-10813-001). We concur that psychology has become over-compartmentalized and that research and interventions alike fail to reflect the full complexity of people's lives. More person-centered and family-centered approaches are needed, not only for intervention but also for prevention and research. As Rotheram-Borus et al. suggested, expanding delivery vehicles is one important way forward. Unfortunately, however, our knowledge about what should be delivered by these vehicles is regrettably limited. Psychological interventions are too often developed, evaluated, and disseminated as fully formed entities. As a result, one cannot identify the source of the positive effects when they occur. This creates a problem for Rotheram-Borus and colleagues' suggestion that new interventions could be created by synthesizing information across randomized controlled trials and including the most common elements. "Common" is not equivalent to "robust" or "effective," as illustrated by the historical use of leeches as a common biomedical intervention for a host of ailments. In cases where curricula with common elements have demonstrated positive effects, the research does not indicate which elements produce change and which may have no or even iatrogenic effects (as has been found in more than one violence prevention program).