PRC Home

Policy Statement for Intervention Review

Procedure
When the PRC Program learns of advances in the testing of interventions by any of the 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, and reports. When needed, the panel consults with outside experts who help interpret the implications of findings that are unusually complex or potentially controversial.

When additional research is completed, the panel reviews new information to update the case study and determine whether it needs to be reclassified.

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

Have results of the work been published in a peer-reviewed scientific journal?
Did the researchers use sound intervention research design (such as a randomized study, quasi-experiment, or another design defined by new standards†)?
Did the researchers control for potential sources of bias?
Did the researchers use appropriate analytic methods?
Are the outcomes of interest health conditions, personal behaviors, or community- or system-level (e.g., health system, schools, neighborhoods) changes in practice or policy?
Are the outcome measures of public health significance? (The answer may be a subjective determination by the panel and the experts with whom it consults.)
Are the outcome measures of statistical significance (p<0.05)?
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?

Discussion
Public health interventions, which must be effective in everyday practice, are complex because of many interacting components, and the development of these interventions may not be linear. Intervention research can be expressed by a complex causation model for health promotion or disease prevention. In the health promotion model, the goal is to improve one or more health-related risk behaviors (e.g., tobacco use, sun exposure). In the disease prevention model, the ultimate goal is to reduce the occurrence, complications, or deaths from one or more diseases (e.g., diabetes, asthma). Intervention research may be designed to assess directly the final health promotion or disease prevention outcome of interest. Intervention research may also test the efficacy of one or more steps or strategies that may lead to the final desired outcome. The PRC Program criteria (listed above) applies to both types of intervention research. For example, a case study may describe research outcomes for intermediate measures, such as changes in knowledge, attitudes, environment, or policy. A case study also may describe outcomes directly related to the ultimate goal of the intervention research, such as a demonstrated change in health-related risk behaviors (e.g., smoking), health conditions (e.g., obesity), or diseases (e.g., heart attack).

† Medical Research Council. Developing and evaluating complex interventions: new guidance. London, 2008. http://www.mrc.ac.uk/complexinterventionsguidance; see also Craig P, Dieppe P, Macintyre S, Mitchie S, Nazareth I, Petticrew M. “Developing and evaluating complex interventions: the new Medical Research Council guidance.” British Medical Journal 2008:337;a1655; http://www.bmj.com/cgi/content/full/337/sep29_1/a1655

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