VOL. 86-B, No. 5, JULY 2004 743 General One in 13 ‘original’ articles in the Journal of Bone and Joint Surgery are duplicate or fragmented publications S. E. Gwilym, M. C. Swan, H. Giele From Nuffield Orthopaedic Centre, Oxford, UK  S. E. Gwilym, MRCS, Senior House Officer in Orthopaedic Surgery Northampton General Hospital, Cliftonville, Northampton NN1 5BD, UK.  M. C. Swan, MRCS, Senior House Officer in Plastic Surgery Radcliffe Infirmary, Woodstock Road, Oxford OX3 7LD, UK.  H. Giele, FRCS, Consultant in Hand Surgery, Nuffield Orthopaedic Centre, Windmill Road, Headington, Oxford, UK. Correspondence should be sent to Mr S. E. Gwilym at 29 Marriott Close, Oxford OX2 8NT, UK. ©2004 British Editorial Society of Bone and Joint Surgery doi:10.1302/0301-620X.86B5. 14725 $2.00 J Bone Joint Surg [Br] 2004;86-B:743-5. Received 24 June 2003; Accepted 21 August 2003 Duplicate publication in orthopaedic journals may further an author’s academic career but this is at the cost of both scientific integrity and knowledge. Multiple publications of the same work increase the workload of editorial boards, misguide the reader and affect the process of meta-analysis. We found that of 343 ‘original’ articles published in the Journal of Bone and Joint Surgery in 1999, 26 (7.6%) had some degree of redundancy. The prevalence of duplicate publications in the orthopaedic literature appears to be less than that in other surgical specialties but it is still a matter of concern. It is the author’s responsibility to notify the editor of any duality when submitting a paper for publication. Publication of identical data in different jour- nals has been strongly criticised by editorial boards as being unethical. In addition, there is the cost of peer-reviewing when the informa- tion is essentially ‘redundant’ and the con- founding effect on meta-analysis. The level of this practice has now been measured in some surgical specialties. The edi- torial boards of Surgery and Otolaryngology – head and neck surgery have both produced consensus statements on duplicate publica- tions in their respective fields. To date no such examination of the orthopaedic literature has been undertaken. Our aim therefore was both to quantify and grade the severity of duplicate publication in the orthopaedic literature and to compare our findings with those in other specialties. The grade of duplication was based on a scale previously described by Schein and Paladugu 1 (Table I). Materials and Methods Original articles published in both the British and American issues of the Journal of Bone and Joint Surgery (JBJS) in 1999 were identi- fied. The on-line search engine MEDLINE (WinSpirs 4.0) was used to search all indexed publications between January 1996 and July 2002 to identify those which may have repre- sented a duplication of the ‘original’ JBJS arti- cle. MEDLINE was searched using the surname of the first, second and last authors (if applicable) and two or three key-words from the title. From the abstracts generated we were able to confirm the JBJS reference articles and identify their related ‘suspect’ articles. Full-text versions of all reference and sus- pect articles were obtained. These were col- lected and each author independently reviewed the papers in full and documented the presence of duplication. When duplication was thought to be present, it was graded as either ‘dual’, ‘potentially dual’ or ‘salami-sliced’. When there was inconsistency in the grading of the papers this was discussed until a consen- sus was reached. Results Details of the findings are given in Tables II to V. Of the 84 papers which were identified as ‘suspicious’ from their MEDLINE abstract, 50 were found to be non-duplicate when the full text versions were reviewed by the authors and 34 were confirmed as representing some degree of duality (Table III). These related to 26 ‘orig- inal’ articles published in JBJS. Of the 34 duplicate articles, 22 (65%) were published in English-language orthopaedic Table I. Details of the grading of duplication based on the scale of Schein and Paladugu 1 Grade Description Dual publica- tion Identical material, methods and conclu- sions Potentially dual Almost identical materials, methods and conclusions Salami-slicing (fragmented) Suspected article represents a part of, continuation of, or partial repetition of the index article