TY - JOUR T1 - Do “Evidence-Based Recommendations” Need to Reveal the Evidence? Minimal Criteria Supporting an “Evidence Claim” JF - The Journal of Rheumatology JO - J Rheumatol SP - 1737 LP - 1739 DO - 10.3899/jrheum.150846 VL - 42 IS - 10 AU - ROBIN CHRISTENSEN AU - JASVINDER A. SINGH AU - GEORGE A. WELLS AU - PETER S. TUGWELL Y1 - 2015/10/01 UR - http://www.jrheum.org/content/42/10/1737.abstract N2 - “Evidence-based medicine” (EBM) stresses examining evidence from clinical research1 as the preferred method of clinical decision making, de-emphasizing intuition, unsystematic clinical experience, and pathophysiologic rationale. Likewise, evidence-based practice (EBP) holds that evidence should be the basis for particular interventions and management plans that are likely applicable to most patients. Logically, this practice would demand that experts develop evidence-based recommendations founded on valid, reliable, and transparent systematic reviews and/or metaanalyses2,3.In this issue of The Journal, Roubille, et al present recommendations for the management of comorbidities, focusing on 8 areas within rheumatoid arthritis (RA), psoriasis (PsO), and psoriatic arthritis (PsA), based on a review of 407 articles4. Their report summarizes the results of the Canadian Dermatology-Rheumatology (DR) Comorbidity Initiative’s systematic literature searches and consensus-based recommendations from a meeting held in Toronto in 2013, sponsored by the pharmaceutical company AbbVie4. The authors report that they did a thorough, systematic review, followed by data extraction and subsequent metaanalyses (including forest plots summarizing the adjusted relative risk estimates, etc.). However, the authors apparently do not want to reveal their explicit findings (yet!), stating that the details and results of the systematic literature review for each topic will be published separately4. We have no reason to question the integrity or the content of the recommendations that came out of this work, but we worry about a possible trend that could encourage guideline panels, etc., to publish their evidence-synthesis secondarily to their recommendation while claiming in a peer-reviewed journal (like The Journal of Rheumatology) that their work represents evidence-based recommendations.Although realizing that we might be perceived as having intellectual conflicts of interest (i.e., being editors in the Cochrane Collaboration), we would like to encourage systematic and explicit methods of making judgments because they … Address correspondence to Prof. Christensen, Copenhagen University Hospital at Frederiksberg, Musculoskeletal Statistics Unit, The Parker Institute, Nordre Fasanvej 57, Copenhagen F, DK-2000, Denmark. E-mail: robin.christensen{at}regionh.dk ER -