Article
Lessons learned using AGREE II: Linking methodological rigour of a guideline and guideline content
Search Medline for
Authors
Published: | July 10, 2012 |
---|
Outline
Text
Background: The AGREE-Collaboration suggests a relation between the methodological quality of a guideline as assessed with the AGREE-instrument and the guideline content. It is assumed that high methodological quality implies high quality of the content.
Objectives: To identify the correspondence between the methodological quality of a guideline as assessed with the AGREE II-instrument and the quality of the guideline's content.
Methods: Two widely used guidelines on antenatal care (British NICE and German Mutterschafts-Richtlinien) were appraised by four raters with the AGREE II-instrument. A comparison of both guidelines content followed.
Results: Applying the AGREE II tool, the NICE-guideline scored high in all domains, whereas the German guideline scored extremely low. However, when comparing guideline content, i.e. the individual recommendations, an overlap of 79% was found.
Discussion: Discrepancies in content are associated with inconclusive evidence, which allows recommending a test, or not. The overall methodological quality of guidelines and the quality of their content are not related. It remains unclear whether 'good' guidelines make 'good' recommendations. In contrast to that, 'poor' guidelines do not necessarily give 'poor' recommendations with regard to scientific evidence.
Implications: The quality of guideline content is most likely an independent characteristic from a guideline's overall quality as assessed with the AGREE II-instrument. The AGREE-instrument needs a considerate user, who is aware that it does not appraise guideline content. It is not recommended to be used without expert knowledge in the field to which it is applied.