Do orthopaedic journals provide high-quality evidence for clinical practice?

Background: In the hierarchy of research designs, randomized controlled trials and meta-analyses are considered to be evidence of the highest grade, and scientific journals are the main source of scientific information. Methods: Using the National Library of Medicine Medline database, all randomized controlled trials and meta-analyses from 1966 to 1999 were retrieved from the journals indexed in the core list of the Science Citation Index in 1999, dedicated primarily to orthopaedics. The abstracts of the articles were reviewed independently by each author and classified by the year, journal name and subject. Results: The total number of articles was 36,293, of which only 671 were randomized controlled trials (1.85%) and 12 were meta-analyses (0.03%). Although there was a progressively increasing trend for randomized controlled trials, more than half of them (81.9%) were published in four journals. Of the randomized controlled trials, 66% was about arthroplasty, and hip and knee arthroplasties covered 90.7%. Conclusion: Although the number of randomized controlled trials and meta-analyses is tending to increase, the conclusion of this study is that the high-quality evidence provided by the major orthopaedic journals is quite low, and more randomized controlled trials and meta-analyses are needed for evidence-based orthopaedic practice