Résumé :
|
[BDSP. Notice produite par INIST-CNRS rR0xr97k. Diffusion soumise à autorisation]. Aims : Grading of evidence of the effectiveness of health promotion interventions remains a priority to the practise of evidence-based health promotion. Several authors propose grading the strength of evidence based on a hierarchy : convincing, probable, possible and insufficient ; or strong, moderate, limited and no evidence. Although these grading hierarchies provide simple and straightforward rankings, the terms that describe the categories in the hierarchies, however, do not explain, in an explicit manner, in what way the strength of the evidence in one category is more, or less, superior than that in another. Methods : To enhance the explanatory power of the hierarchy, we propose that evidence be classified into three grades, each with a short explanatory note on the basis of three criteria : the degree of association between the intervention under study and the outcome factors, the consistency of the findings from different studies, and whether there is a known cause-effect mechanism for the intervention under study and the outcome factors. Conclusion : For more in-depth grading, a three-grade expanded hierarchy is also recommended. Examples are given to illustrate our proposed grading schemes.
|