[Study Designs for Effect Evaluation in Disease Prevention and Health Promotion - Meaning of Internal and External Validity in Intervention Studies].

BACKGROUND

Health promotion and disease prevention in Germany have become more important as a result of the Prevention Act that gives special attention to the setting. So far, however, there is neither a common understanding of the terms of this approach, nor are its potentials empirically derived. Therefore, a discussion about suitable evaluation and study concepts is required.

AIMS AND METHODS

The aims of this study were to address the meaning of internal and external validity of intervention studies in health promotion and disease prevention. We provide an overview of the achievable bias control for different study designs and provide examples for the evaluation of setting-based approaches.

RESULTS

Interventions in settings are often characterized by a multitude of measures, actors and intervention contexts. Methods of analysis from evidence-based medicine are viewed critically for health promotion and disease prevention in Germany. Such studies are considered to provide a high degree of internal validity, but the extrapolation of the results to "reality" is viewed as low. In contrast, the extrapolation of study results is not more of a challenge for setting projects than for any other research area. It is not limited by different contexts, but rather by different causal relationships. Impact assessment aims at causally attributing an observed outcome to the intervention. Thus, the epistemological requirements do not differ between studies that are designed for internal or external validity. The international discussion focuses on the refinement of (quasi)-experimental study designs. Examples for the evaluation of setting projects, mainly from Germany, illustrate that those alternative methods have already been used in evaluation practice.

CONCLUSION

A challenge for health promotion and disease prevention in settings is to systematically assess the different needs for evidence-basing. At present, there is a wealth of findings in a large number of intervention fields. These findings must be compiled and analyzed to determine whether and to what extent further evaluations need to be initiated and by which methods.