[Implications of publication bias on guideline development and appraisal]

Abstract

The issue of selective publishing of research results is gaining more and more scientific, public and political awareness. For guideline authors, in particular, it leads to uncertainty about the interpretability of the methodological quality and clinical relevance of the available evidence and the risk of bias where their conclusions and thus guideline recommendations are concerned. The actual impact of publication bias on guideline contents appears to be low if a systematic and methodically sound approach is followed in the process of guideline development. However, the quality of the evidence on this topic is poor. Different strategies to deal with publication bias have been proposed for authors of systematic reviews but they are of limited use for guideline authors. The goal must therefore be to implement appropriate measures in order to avoid the problem ex ante. The first step would be the systematic registration of study results in publicly accessible registers.