Rigorous Development Does Not Ensure That Guidelines Are Acceptable to a Panel of Knowledgeable Providers
Published in: Journal of General Internal Medicine, v. 23, No. 1, Jan. 2008, p. 37-44
Posted on RAND.org on January 01, 2008
BACKGROUND: Rigorous guideline development methods are designed to produce recommendations that are relevant to common clinical situations and consistent with evidence and expert understanding, thereby promoting guidelinesgas acceptability to providers. No studies have examined whether this technical quality consistently leads to acceptability. OBJECTIVE: To examine the clinical acceptability of guidelines having excellent technical quality. DESIGN AND MEASUREMENTS: The authors selected guidelines covering several musculoskeletal disorders and meeting 5 basic technical quality criteria, then used the widely accepted AGREE Instrument to evaluate technical quality. Adapting an established modified Delphi method, the authors assembled a multidisciplinary panel of providers recommended by their specialty societies as leaders in the field. Panelists rated acceptability, including perceived comprehensiveness (perceived relevance to common clinical situations) and perceived validity (consistency with their understanding of existing evidence and opinions), for ten common condition/therapy pairs pertaining to Surgery, physical therapy, and chiropractic manipulation for lumbar spine, shoulder, and carpal tunnel disorders. RESULTS: Five guidelines met selection criteria. Their AGREE scores were generally high indicating excellent technical quality. However, panelists found 4 guidelines to be only moderately comprehensive and valid, and a fifth guideline to be invalid overall. Of the topics covered by each guideline, panelists rated 50% to 69% as comprehensive and 6% to 50% as valid. CONCLUSION: Despite very rigorous development methods compared with guidelines assessed in prior studies, experts felt that these guidelines omitted common clinical situations and contained much content of uncertain validity. Guideline acceptability should be independently and formally evaluated before dissemination.