Systematic Analysis Underlying the Quality of the Scientific Evidence and Conflicts of Interest in Gastroenterology Practice Guidelines

OBJECTIVES: The practice guidelines published by the American Gastroenterological Association (AGA) and the American College of Gastroenterology (ACG) are used to establish standards of care and improve patient outcomes. We examined the guidelines for quality of evidence, methods of grading evidence, and conflicts of interest (COIs). METHODS: All 81 (AGA and ACG) guidelines available online on 26 July 2012 were reviewed for the presence of grading of evidence and COIs. In total, 570 recommendations were evaluated for level of evidence and methods used to grade the evidence. The data were evaluated in aggregate and by society. RESULTS: Only 31% (n=25) of the guidelines graded the levels of evidence. A total of 12 systems were used to grade the quality of evidence in these 25 guidelines. Of the 570 recommendations reviewed, only 29% (n=165) were supported by the highest quality of evidence, level A; 37% (n=210) level B, 29% (n=165) level C, and 5% (n=30) level D. Since 2007, 87% (n=13/15) of the ACG guidelines graded the evidence compared with only 33% of the AGA guidelines (n=4/12). Furthermore, 70% (n=57/81) of the guidelines failed to disclose any information regarding COIs. Of the 24 articles commenting on COIs, 67% reported COIs. CONCLUSIONS: Although the majority of the gastroenterology guidelines fail to grade the quality of evidence, more recent ACG guidelines grade majority of their recommendations. When the evidence is graded, most of the supporting evidence is based on lower-quality evidence. In addition, most of the guidelines fail to comment on COIs, and when disclosed, numerous COIs were present. This study highlights the critical need to revise the guideline development process. Future guidelines should clearly state the quality of evidence for their recommendations, utilize a standard grading system, and be transparent regarding all COIs.