Background: Medical guidelines claim to be evidence based, to report newest scientific information and to be transparent in respect to authorship. They want to optimize treatment, improve patient outcome, and reduce unnecessary costs. It must empirically be tested in randomized controlled trials to which degree these claims are fullfilled in day-to-day practice, and whether “guideline exposed physicians” treat their patients differently and, if so, have better or worse treatment results in comparison to “guideline naive physicians”.
Method: Systematic review in the PubMed database of controlled clinical trials comparing “guideline exposed physicians” with “guideline naive physicians” in the area of depression, hypertension, and diabetes mellitus.
Results: Two studies for hypertension, three for diabetes and nine for depression were found. In two studies there was an improvement in treatment outcome in the guideline exposed-group, in one a worsening in treatment outcome and in eleven no differences
Conclusions: Effects of treatment guidelines can be tested in controlled clinical trials. Results of such studies for depression, diabetes and hypertension suggest, similar to conclusions from other authors, that guidelines do often not reach their goal but can even have negative consequences. This can be explained in reference to theoretical knowledge on medical decision making. The term “evidence based” should not be used for guidelines which are literature reviews only, but for guidelines which have been empirically tested by controlled trials.