Use of systematic reviews in a clinical practice guideline on colorectal cancer

Article type
Authors
Kuijpers T, Deurenberg R, Rosenbrand K, Moret M, Burgers J
Abstract
Background: Ideally, recommendations in guidelines are based on strong evidence as summarised in systematic (Cochrane) reviews. However, some case studies have shown that the use of reviews in guidelines is quite low and could be improved.
Objective: To identify the number of systematic reviews, including Cochrane reviews, used to answer the clinical questions in a guideline on colorectal cancer.
Methods: A working group of clinical experts formulated 12 key questions on diagnosis and therapy for a guideline on colorectal cancer. An information specialist, assisted by clinical experts, searched for systematic reviews and additional studies that could potentially answer the clinical questions. Sources used were existing high quality guidelines, MEDLINE and EMBASE. Search filters were used to identify systematic reviews on the specific topic. After drafting the guideline, we analysed the type of literature on which the final statements and recommendations in the guidelines were based. Within this set, the number and proportion of systematic reviews was calculated.
Results: The search retrieved 456 systematic reviews for the 12 key questions of the guideline. Five (1%) of the reviews were produced by the Cochrane Collaboration. The final guideline contains 71 evidence statements supported with 89 different citations. 16 (18%) of these were systematic reviews, supporting 29 (40%) statements. Only two Cochrane reviews were used for supporting the conclusions.
Conclusions: The number of systematic reviews used in formulating the evidence statements was low, although the guideline was developed according to principles of evidence-based medicine. The low number may be due to the fact that the reviews available do not address the main questions raised in practice and covered by guidelines. Collaboration on agenda setting between Cochrane review groups and guideline developers could enhance the effective production and use of systematic reviews in clinical guidelines.