| Sign In to gain access to subscriptions and/or personal tools. |
DOI: 10.1016/J.JHSB.2004.04.005
Duplicate Publication in the Journal of Hand SurgeryFrom the Department of Plastic and Reconstructive Surgery, Radcliffe Infirmary, Oxford OX2 6HE, UK Correspondence: Mr Henk Giele, Department of Plastic and Reconstructive Surgery, Radcliffe Infirmary, Woodstock Road, Oxford OX2 6HE, UK. Tel.: +44-1865-224-793; fax: +44-1865-311-673; E-mail: henk.giele{at}nds.ox.ac.uk
This study evaluates the extent of duplicate publication in the Hand Surgery literature. A retrospective review of original articles published in the American and the British & European editions of Journal of Hand Surgery during the years 1999 and 2000 was performed using MEDLINE (PUBMED) search engine. Index articles suspected of dual publication were identified by using key words in the title and the names of the first, second and last authors. The full initial text was carefully studied and suspected duplicate articles were classified as dual, potentially dual or fragmented. Six hundred articles were evaluated, of which 25 (4%) index articles were identified with 33 "suspects". Eleven "index" and 15 "suspected" articles were cleared on closer scrutiny. Thus 14 "index" articles (2%) were found to be associated with 18 duplicated articles. Of these, four were classified as dual, five as potentially dual and nine as fragmented. We conclude that although duplicate publication of articles in the Journals of Hand Surgery (American and British/European Volumes) does occur, the incidence in the sample studied is lower than some other surgical journals.
Key Words: duplicate publication • hand surgery journals
For many centuries, members of the medical profession have dissipated the art and science of medical practice by sharing their experience. This has crystallized over the last two centuries in the form of published work in medical journals. In the present era of evidence-based medicine, the significance of articles and systematic reviews published in peer-reviewed journals cannot be overemphasized. They form the basis for changes in clinical practice and have profound effects on the quality of patient care and management of resources. This is also the era of "publish or perish" and unfortunately, by mistake or on purpose, sometimes the results of some studies find their way into journals on more than one occasion, either in part or in whole. This is referred to as duplicate publication which is said to occur when the information in an article overlaps substantially with another manuscript. It is also referred to as dual, redundant or repetitive publication. Duplicate publications may also occur when the results of a large study are split and published in more than one article. This is known as salami slicing (salami–science or fragmented publication) (Huston and Moher, 1996; Tobin, 2002). Dual publication is the bane of editors of journals worldwide and has major adverse implications. The incidence varies from 8% to 32% (Bailey, 2002; Rosenthal et al., 2003; Schein and Paladagu, 2001). This study attempts to highlight the incidence of duplicate publication of articles published in the British & European and American editions of the Journal of Hand Surgery.
Original articles published in the British and American editions of the Journal of Hand Surgery during 1999 to 2000 were scrutinized. Review articles, letters, abstracts and editorials were excluded. After exclusions, there were 600 remaining original (index) articles. A MEDLINE search then was performed using key words from the title and the names of the first, second and last authors. Any article appearing to convey similar information was identified as "suspected dual publication". The full version of the index and suspect articles were then read and the contents, methods, subjects and results compared separately by three different reviewers. The results were then collated. In the small number where differences of opinion were found, these were debated until consensus was achieved. Dual publication of the information was graded as suggested by Schein and Paladugu (2001) (Table 1). The full article texts will be made available to the editors of the journals on request.
The MEDLINE search showed that, 25 of the 600 index articles were suspicious of duplicate publication linked to 33 suspect articles. After thorough review of the full texts, 11 of the index articles were found to have been unfairly suspected. However, 14 (2%) index articles were associated with 18 duplicate articles. Of the 18 duplicate articles, four were dual publication (Grade A), five were potentially dual (Grade B) and nine were fragmented publications (Grade C). The duplicated versions were published in non- English language journals in three case, but with no reference to the original article in the English version. Most articles were duplicated in some form only once. However, two articles were duplicated two times and one article three times. Multiple duplication was in the form of fragmentation, dual or a mixture of both. Eleven duplicated articles were published in other speciality journals which probably had a different readership to hand surgery, but three were published in Journals of Hand Surgery. Nine duplicate articles were published after the index article, six duplicate articles were published prior to the index article and three were published in the same month. Table 2 shows the details of duplication in the study sample.
Duplicate publication may be an increasing problem, and a substantial increase was observed in the British Journal of Industrial Medicine between 1986 and 1990 (Waldron, 1992). In the surgical literature, the incidence of duplicate publication varies from 8% to 32% (Table 3). Schein and Paladugu (2001) noted that 92 (14%) of 660 index articles represented some form of redundant publication (Schein and Paladugu, 2001). Of these 20 (14%) articles were categorized as dual publication, 50 articles (34%) as potentially dual and 77 (52%) as salami-sliced (fragmented). Our findings of 14 out of 600 (2%), of which, 1% were dual, 1% were potentially dual and 1.5% were fragmented shows that the incidence of redundant publication is less in the Journals of Hand Surgery (American and British/European volumes). The variation in incidence could have been due to variance in study design, sampling error and strictness and rigidity in the definition of duplicate publication. However, we ensured comparability, by using the study design of Schein and Paladugu (2001) and increased the robustness of the study by utilizing a 2-year period and using three separate reviewers. Where discrepancy or doubt was observed by the reviewers, and no consensus achieved, the suspected articles were not included. Duplicate articles not yet published will obviously not have been detected, and thus, our findings are likely to be a slight underestimate of the true incidence. It is also entirely possible that findings in 1999 to 2000 may not be the same as in other years.
The reasons for the occurrence of duplicate publication include naivety on the part of the authors, failure to read the instructions and a desire to reach a different audience (Bailey, 2002; Tobin, 2002). However, we feel in some cases, that duplicate publication occurs deliberately, probably to boost the author’s bibliography. Such pressure to publish for career progression has been noted by others (Bailey, 2002; Benninger, 2002; Tobin, 2002). Yank and Barnes (2003) in a survey of authors and editors reported that redundant publications occurred because authors feel the pressure to publish and journals do not do enough to publicize, criticize and punish cases. The consequences of duplicate publication are considerable. From a scientific point view, it skews the results of meta-analyses, which provide the highest level of evidence of a study (Huston and Moher, 1996; Tramer et al., 1997). This can have a profound effect on good practice guidelines and a reader who comes across the results of the same study in different articles and in different journals may erroneously consider it as reinforcing evidence and make wrong inferences (Huston and Moher, 1996; Tobin, 2002). One case study demonstrated that inclusion of duplicated data in meta-analysis led to a 24% overestimation of anti-emetic efficacy of a drug (Tramer et al., 1997). An average scientific paper in the British Medical Journal is read by eight people before publication, takes up 8 hours of a technical editor’s time, and costs around £1300 to process, print and distribute. Thus duplicate publication is a wasteful expenditure of both money and resources (Lowry and Smith, 1992). The International Committee of Medical Journal Editors has clearly laid out the guidelines for authors. It states that manuscripts must be accompanied by a covering letter signed by all co-authors which must include information on prior or duplicate publication or submission elsewhere of any part of the work (International Committee of Medical Journal Editors, 1991). Secondary publication in another language is one possible justification provided certain conditions are met. The conditions include informing the editors of both journals about the article and providing the editor concerned with secondary publication with a photocopy, reprint or manuscript of the primary version, allowing a publication interval of at least 1 to 2 weeks and including a footnote, which declares that the article is based on a study that has been previously published. It is expected that full reference to the primary publication is supplied (International Committee of Medical Journal Editors, 1991; Special Report NEJM, 1997). In the three foreign language dual publications seen in our study, the original paper was not referenced. This is suspicious of an attempt to deceive readers, reviewers and the editors. Similar, probably deliberate, failure to cross-reference salami sliced papers or dual papers can only be seen as attempts to deceive. If the findings of a study have been presented in a conference or published in press reports, subsequent publication is not considered as dual publication. (International Committee of Medical Journal Editors, 1991; Special Report NEJM, 1997). Producing more than one manuscript from a single dataset can be legitimate, provided the manuscripts are cross-referenced and each addresses distinct and important questions. However, the authors should declare this to be the case (Britton and Knox, 1999; Tobin, 2002; www.wame.org/duplicate.htm). Such papers proved to be the most contentious to assess. However, we felt that if a single dataset was used to produce essentially similar papers and outcomes and were not cross-referenced, then they were duplicate. Many other examples of probable salami slicing were not counted as they could be interpreted as legitimate. However, we suspect that some papers were deliberately split. It is difficult for the editors to detect dual submissions. It has been pointed out rightly on the WAME (World Association of Medical Editors, 2002) website (www.wame.org) that whilst a MEDLINE search might detect prior publication, it is almost impossible to detect dual submission. Therefore, it is up to the authors to ensure that their manuscripts are legitimate. Various measures have been suggested to tackle the problem of dual publication. It has been proposed that clinical trial registries are used to overcome the problem of the results of multi-centre trials being published by different sets of authors in different journals. The editor of the British Journal of Industrial Medicine has suggested that the most effective deterrent may be to require applicants for posts or grants to submit copies of six of their most important papers for the committee to read rather than a whole bibliography list (Waldron, 1992). Exposing dual publication and naming and shaming the authors have also been suggested as deterrent (Yank and Barnes, 2003). We have attempted to determine the magnitude of this problem in the hand surgery literature by studying the American and European volumes of the Journal of Hand Surgery. It is hoped that this will raise the awareness of the readers to notify the editors of any suspicious articles and to warn authors submitting the articles to ensure the legitimacy of their manuscripts. The website of the Council on Publication Ethics in the United Kingdom (www.publicationethics.org.uk) contains more information on publication ethics. Declaration: Henk Giele has submitted, with other authors, similar studies on duplicate publication in orthopaedic and plastic surgery literature.
Received for publication November 13, 2003. Accepted for publication April 14, 2004.
  CiteULike  Connotea  Del.icio.us  Digg  Reddit  Technorati What's this?
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
 |
Sign In to gain access to subscriptions and/or personal tools.
TOP
Abstract
INTRODUCTION
