[Emerging concepts in clinical practice guidelines]

The German Society for Digestive and Metabolic Diseases (Deutsche Gesellschaft für Verdauungs- und Stoffwechselkrankheiten, DGVS) has already published an impressive number of clinical practice guidelines, with more topics currently being under development. The guideline method within the DGVS as well as between medical professional societies varies greatly. In particular, the levels of evidence as well as the grades of recommendations are often applied very differently. Most guidelines in Germany use the grades of the US Agency for Health Care Policy and Research (AHCPR), which were first published in 1992. Some newer guidelines, e.g., the ulcerative colitis guideline of the DGVS or the guidelines of the German Diabetes Association, try to circumvent problems of this method. They adopted variations of the very simple AHCPR grades. Sometimes these new grading methods have serious shortcomings too. They may not be reproducible, ignore important parts of the consensus process, can only be applied to therapeutic interventions, or they do not adequately distinguish between the strength of a recommendation and the quality of evidence. Meanwhile, an enormous number of different and partially contradictory methodologies for the grading of guideline recommendations are being used. In the GRADE working group several renowned clinical epidemiologists came together to develop a uniform grading systematic that avoids the problems of earlier guideline methods. With GRADE first the quality of evidence is assessed in a systematic manner. The quality of each study plays an important role in this process. Hence randomised trials of poor quality can end up with a lower overall quality rating than well performed observational trials. The strength of recommendation is based on the balance between the expected benefit and harm of the intervention in the respective population for which the recommendations of the guideline are made. The quality of evidence is only one factor contributing to the strength of recommendation. The GRADE method can be more cumbersome in its application compared to older methods. Nevertheless, more and more organisations are adopting GRADE for their guidelines and recommendations. Possibly the DGVS guidelines could benefit from this new system too.