[Selective use of systematic reviews in decision making]

When formulating recommendations, guideline developers have opportunities to use systematic reviews (SRs), which are widely considered to be the best available evidence. However, in practice the use of SRs by guideline developers is limited. Recent research has shown that only 22% (60/279) of the recommendations on screening of newborns for rare diseases were based on SRs. Moreover, the non-use of SRs was correlated with positive recommendations. This commentary discusses the reasons for limited SR use and the influence of wishful thinking on decision making. Although standards for guideline development are evolving, differing methods can still be used without compromising quality. Expert opinion is always needed to weigh the benefits and harms of the available management options. As long as potential conflicts of interest and considerations as well as choices made during the process are explicitly described, guidelines and policy decisions can be considered to be trustworthy.