[Requirements on evidence for system-level decision making regarding the adoption of medical services into the Statutory Health Insurance (SHI) catalogue].

In Germany, recommendations regarding health care (system) decisions dealing with medical services which are to be included in (or excluded from) the SHI benefit catalogue clearly relate to the principles of evidence-based health care which can be defined as "the conscientious, explicit and judicious use of the current best external, scientific evidence in making decisions on a system level". If benefit is understood as a "causally related positive effect" of a medical intervention, we may directly assume that only those trials which allow for causal conclusions are to be included in benefit assessments and regarded as part of the basis of the decision-making process. In general, these will be randomised controlled studies. There may be exceptions to this rule, especially when we are dealing with particularly severe disease calling for immediate action. In these cases, an adequate justification is required which must not be based solely on the absence of conclusive or more conclusive trials. Evidence-based benefit assessment comparing two or more treatment options has both qualitative (bias potential) and quantitative aspects. The quantitative aspect can be subdivided into "precision of the estimate" and "magnitude of the observed difference". On the analogy of statistical hypotheses testing the following can be postulated: if in a comparative trial the (observed) difference is sufficiently large in relation to the bias potential, it will not rest on potentially confounding mechanisms (alone), but at least on the effect of the applied intervention as well. The analogy is limited, though, as the bias potential cannot precisely be quantified. Applying the methods of evidence-based health care requires the right balance between (appropriate) scepticism on the one side and an (inappropriately) over-critical attitude on the other.