Mo1101 Systematic Analysis of the Quality Underlying the Scientific Evidence and Conflicts of Interest in Interventional Subspecialty Practice Guidelines

Abstract

Background: The Institute of Medicine defines guidelines as statements to assist physician and patient decisions regarding appropriate health care in clinical situations. Guidelines are meant to create a succinct roadmap for diagnosis and treatment by analyzing and summarizing the abundance of medical studies. This is predicated on guidelines being unbiased and based on strong quality of evidence. Currently, though, no studies have reviewed the quality of the interventional medical subspecialty guidelines Methods: We reviewed the interventional medicine subspecialty society's websites for interventional guidelines. The American Society of Gastroenterology Endoscopy (ASGE), American Association of Diagnostic and Interventional Nephrology (ASDIN), American Association for Bronchology and Interventional Pulmonology, American Thoracic Society, CHEST, and Society for Cardiovascular Angiography and Interventions (SCAI) were reviewed for guidelines that graded the level of evidence, methods used to grade the evidence, and disclosures of conflicts of interest (COI). The data were evaluated in aggregate and by society. Results: 153 interventional guidelines were evaluated of which two were duplicates. 46% (n=70/151) of the interventional guidelines graded the quality of evidence. ASGE graded 72% (n=48/67) compared to 28% (n=22/78) by SCAI, and 0 (n=4) by ASDIN and 0 (n=2) by the pulmonary societies. Of the 3460 recommendations reviewed 11% (n=374) were supported by the highest quality of evidence, level A (multiple randomized controlled trials), 41% (n=1414) were level B (single or nonrandomized controlled trials), and 48% ((n=1645) were level C (expert opinion). See table 1 for breakdown by society. While SCAI used two systems to grade the level of evidence, ASGE used 5 different systems to grade its evidence. Furthermore, 62% (n=94/ 151) of the guidelines failed to comment on COI. Of the 57 articles that disclosed COI, 91% (n=52) reported potential conflicts. In total 1855 COI were reported of which 967 were non-research based COI. Authors with conflicts (n=320/705) averaged 5.8 conflicts/ authors. While the SCAI disclosed potential COI in 52% of their guidelines, ASGE did so in only 24% of their guidelines, and the ASDIN and pulmonary societies did not disclose COI. Conclusions: Majority of the medicine subspecialty interventional guidelines fail to grade the quality of evidence, and when present, most used lower quality evidence. Furthermore, most guidelines fail to comment on potential COI. When assessed, numerous COI were present. Interventional procedures carry unique risks to the patients and guidelines should ideally be based on strong quality of evidence. Future interventional guidelines should clearly state the quality of evidence for their recommendations, utilize a standard grading system, and be transparent regarding all potential biases. Table 1. Total number of practice guidelines with graded evidence and quality of evidence for recommendations