Developing effective prevention services for the real world: a Prevention Service Development Model.

There is a growing awareness that our current models for prevention research are not working as intended. The dominant current model (Mrazek & Haggerty, 1994) describes a progression of scientific studies in which research on the development of problems leads to the development and evaluation of interventions, which in turn lead to research on the widespread dissemination and implementation of effective interventions. However, despite considerable progress in developing scientifically validated preventive interventions (Durlak, 1997; Greenberg, Domitrovich, & Bumbarger, 1999), there is little evidence that these interventions have been widely implemented or that they have led to a significant reduction in the rates of behavioral health problems in the population (e.g., Biglan & Taylor, 2000). The purpose of this paper is to present a new research model for more rapid development and widespread implementation of effective prevention programs. The model integrates concepts and methods from business with those traditionally used in prevention research. Business models have been very successful in guiding the development of a broad range of products and services that are widely used by the public (Cooper & Edgett, 1999), but these models are not commonly used in behavioral health contexts where the primary aim is well-being of the public rather than profit. Prevention research models have been very effective in developing programs that have demonstrated a potential to benefit the public, but have not facilitated their wide-scale implementation. Integration of these two models is needed for rapid development and wide-scale implementation of effective prevention programs.

The paper will first present a version of the current dominant model of the Prevention Research Cycle (PRC), review sources of dissatisfaction with this model and describe proposals to increase its utility. We will then describe a model of service development in the business literature, the New Service Development Process (NSDP). A new model, the Prevention Services Development Model (PSDM), that integrates concepts from business and prevention research will then be presented. Studies from our research with children from divorced families and bereaved children will be used to illustrate activities proposed by the integrated PSDM.

[FIGURE 1 OMITTED]

Stakeholder Dissatisfactions With Prevention Research Cycle and Some Proposed Solutions

A five-phase version of the PRC, derived from alternative versions described previously (Greenberg & Cullen, 1984; Mrazek & Haggerty, 1994; Price, 1983; Roosa, Wolchik, & Sandler, 1997) is presented in Fig. 1. The phases describe a logical series of studies (with multiple feedback-loops) that lead from the identification of potentially modifiable risk and protective factors, design of interventions to change these factors, well-controlled trials of the efficacy of the intervention under optimal conditions, evaluation of program effectiveness when delivered under more naturalistic conditions, and finally widespread dissemination and ongoing evaluation. Multiple stakeholders have expressed concerns about the PRC model. Prevention scientists express concern that prevention programs that have been demonstrated to be efficacious in well-controlled experimental trials often do not get delivered on a large scale to the public (Rotheram-Borus & Duan, 2003). In contrast, programs that have little or no research support are sometimes widely disseminated. For example, a national survey of school-based prevention programs (Gottfredson & Gottfredson, 2001) found that DARE is the most widely disseminated substance abuse prevention program, despite the fact that evaluations of this program have failed to demonstrate positive effects (Clayton Cattarello, & Johnstone, 1996; Lynam et al., 1999). Even when evidence-based prevention programs are adopted, they are often not implemented with fidelity by community agencies, and thus are not likely to be effective (Gottfredson et al., 2002).

Community agency stakeholders express concern about the lack of fit between research-based prevention programs and their organizational capabilities as well as key stakeholders' (e.g., parents, mental health advocates, providers) preferences or values. Furthermore, there is concern about the applicability of the findings on program effectiveness to any particular community given that the clients, providers, and organizational context are likely to differ from those in the original evaluation of the program (Green, 2001; Green & Mercer, 2001).

Multiple approaches have been proposed to bridge the gap between research-based prevention programs and prevention services in the community. One approach has been to identify effective research-based programs using strong scientific criteria, to publicly certify their efficacy, and to provide funding for training and implementation of these programs in the community. A second approach involves building community or agency capability to identify and adopt programs that meet the community's needs (Hawkins, Catalano, & Associates, 1992; Morissey et al., 1997). For example Wandersman, Imm, Chinman, and Kaftarian (2000) developed a 10-step process (empowerment evaluation) in which communities identify their needs and research-based programs that meet these needs, assess how well these programs are implemented and obtain feedback to improve implementation of these programs in the local community. A third approach is to build stronger community-university collaborations to develop and evaluate prevention programs (Jensen, Hoagwood, & Trickett, 1999; Nelson, Pancer, Hayward, & Kelly, 2004).