Abstract
Summary Statement
With increasing emphasis on evidence-based medicine in healthcare, there is global increase in proportion of Level-1 and -2 articles in PUBMED. This study shows the trend of orthopaedic publications from different countries in comparison to other specialties.
Introduction
New medical knowledge is expected to improve health through change in existing practices. Articles need to convince readers of the validity of conclusions in order to bring about a change in practice. The last few decades have witnessed an increasing interest in critical appraisal of research aimed at assessing the ‘quality’ of evidence, a trend towards ‘Evidence Based Medicine’. Whether orthopaedic publications are also becoming more evidence-based has hitherto not been reported. This study aimed to compare the trend of publications originating from orthopaedic services versus other specialties, across different countries, with respect to major categories of levels of evidence.
Methods
PUBMED was used to collect data on yearly publication of articles, categorised by country, specialty (from author address) and article category (Guideline, Review, Case Report, Comparative Study, Clinical Trial and Meta-Analysis).
Results
PUBMED shows 20,572,125 articles published between 1900 and 2010. Of these, 64% were published within the quarter century 1986–2010. Publication type was specified in 25% of articles, of which 8% were review articles, 7% each were case reports and comparative studies while 3% were clinical trials. The top ten countries from which articles originated were US, UK, Japan, Germany, France, Canada, Italy, Australia, Netherlands and Sweden. Comparing the decade of 1990s with 2000s, the number of meta-analyses showed greatest increase from Australia, Japan, Germany, UK and Italy (9.5x-6.1x). When comparing types of articles between the beginning and end of the quarter century, the proportion of clinical trials increased by 2x, guidelines by 5x and meta-analyses/systematic reviews by 28x. There were 121,859 articles from orthopaedic departments/institutes (0.6% of all articles). There was a consistent increase in the proportion of orthopaedic publications from 0.5% to 1.1% over the quarter century; this 231% increase was lower than vascular, cardiothoracic and paediatric surgery but greater than urology, neurosurgery, otolaryngology and obstetrics/gynecology. Among the different types of publications, in orthopaedics the greatest increase between the decades of 1990s and 2000s was in meta-analyses (8.5x) followed by guidelines (5.4x). This increase in number of meta-analyses was higher than other surgical specialties except cardiothoracic and vascular surgery, both of which showed a 13x increase. Among the meta-analyses from surgical specialties, the proportion from orthopaedics and vascular surgery was increasing consistently, with the former now contributing to 13% of all meta-analyses. In comparison, increase in number of clinical trials has been relatively modest for orthopaedics (2.4x), being lower than vascular surgery (3.3x), plastic surgery (2.8x) and neurosurgery (2.6x).
Dicussion/Conclusion
The proportion of clinical trials and guidelines was increasing in a linear fashion in the last 25 years, while the proportion of meta-analyses/systematic reviews was increasing logarithmically. Among European countries, Germany, UK and Italy contributed to the highest number of meta-analyses. Publications from orthopaedic departments showed an overall intermediate rate of increase, but meta-analyses/systematic reviews showed a consistently higher increase compared to other surgical specialties. This analysis shows the trend of orthopaedic evidence in published literature and may be used as a guide to future country-specific research directions in the discipline of orthopaedics.
