Policy Statement for Intervention Review

The Prevention Research Centers (PRC) Program writes case studies for PRC-tested public health interventions (e.g., research, dissemination, or health policy interventions) found to be adoptable, effective, promising, or notable, according to definitions established by the program. The review procedure and criteria are described below.

When the PRC Program learns of advances in the testing of interventions by any of the individual PRCs, the work is reviewed to determine whether the information about it is sufficient for sharing with the public health community. A panel, comprising members of the PRC Program’s scientific and leadership staff, asks the principal investigator of the research or other work to submit detailed information, which may include scientific articles, data presentations, questionnaires, reports, and other materials related to the work. The panel reads and discusses the material, and consults with outside experts, when needed, who help the panel members interpret the implications of findings that are unusually complex or potentially controversial.

The panel uses the following set of questions, definitions, and values as criteria in deciding the strength of research so that it can be classified and shared through the program’s Web site.

• Have results of the work been published in a peer-reviewed scientific journal?
• Did the research use sound experimental design (such as a randomized study or a quasi-experimental design) and control for potential sources of bias?
• Are the outcome measures of statistical significance (p<0.05)?
• Are the outcome measures of public health significance? (Note: The answer to this question is sometimes a subjective determination by the panel and the experts with whom it consults.)
• How well do the main outcomes satisfy the seven widely accepted criteria for causality: (1) strength of association, (2) dose-response effect, (3) temporality, (4) consistency of the findings, (5) biological and theoretical plausibility of the hypothesis, (6) coherence of the evidence, and (7) specificity of the association?*
• Has the intervention been successfully implemented in populations other than the original study population?

*For detailed definitions, see Kleinbaum DG, Kupper LL, Morgenstern H. Epidemiologic research: principles and quantitative methods. New York: John Wiley and Sons, 1982.

When additional work on a project is completed, the panel reviews any new information to update the case study and determine whether it needs to be reclassified.