Best Intentions are Not Enough: Techniques for Using Research and Data to Develop New Evidence-Informed Prevention Programs. Making Choices among Potential Strategies


As potential risk, protective, and promotive factors and strategies to influence them are identified, the list of potential intervention factors may become quite long. Asking some critical questions immediately may serve to shorten the list. For example, is the population that was studied relevant? A correlation found in a sample of delinquent youth might not be relevant for adolescents in foster care. Also, it is valuable to select factors based on multiple studies, though finding multiple studies can be difficult when research on a particular population is scarce, such as immigrant teens.

In addition, it is necessary to identify factors that are malleable. While the education of parents is frequently found to be associated with outcomes for their children (Magnuson, 2007; Duncan, Ludwig & Magnuson, 2007), parent education is not easily malleable and thus may not be a promising factor for an intervention. A malleable factor is an antecedent, reinforcement, physiological or relation frame that can be changed in a timeline that could produce significant outcomes. Often, the most effective strategies in homes, schools, or community settings also have very pronounced immediate (proximal) impacts that can be easily measured behaviorally (not knowledge, per se) that in turn have many positive, long-term (distal) outcomes (Bach, Hayes, and Gallop, 2011; Kellam, Mackenzie, et al., 2011; Prinz, Sanders, et al., 2009). It is rare for there to be long-term benefits without shorter-term measurable changes.

From a practical perspective, we also need to ask whether the association is large enough to warrant the effort and resources required to alter it. The cost and feasibility of including a given program strategy ranges widely, from minimal (sports teams) to very substantial (mental health therapy). This is not an argument for seeking only quick low-cost interventions, however. Rather, we acknowledge that developers of an evidence-informed intervention need to consider whether focusing on a given program element is cost-effective. One way to boost cost-effectiveness is to select intervention options that have multiple, beneficial outcomes (Embry, 2011), instead of a single outcome. This is much easier to hold in mind when one conceptualizes the prevention problem as behavior change rather an awareness or knowledge problem. For example, prevention programs that focus on child-abuse awareness or attitudes have no known impact on the prevention of child maltreatment, but programs that change parent behavior do (Prinz, Sanders, et al., 2009).

Finally, triangulation across varied sources of wisdom (e.g., behavioral psychology, social psychology, medicine or biological sciences, anthropology, epidemiology, and tribal wisdom) increases one's confidence in the potential value of a particular intervention strategy. It is critical to keep in mind, though, that these strategies can only identify candidate factors. It remains to be seen which factors really change outcomes and which can be implemented on the ground to achieve the change envisioned. The next section describes how logic models can be used to illustrate this process.

View full report


"rb_bestintention.pdf" (pdf, 488.63Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®