Discrepancies and Rates of Publication in Orthopaedic Sports Medicine Abstracts

¹ Columbia University and Massachusetts General Hospital
² New York University/Bellevue Hospital Center
³ Saint John Hospital, Detroit, Michigan
⁴ Columbia University

^* To whom correspondence should be addressed. E-mail: csa4{at}columbia.edu.

	Abstract

Background: Presentations of clinically relevant data at AOSSM national meetings are presented yearly and may influence clinical decision making.

Hypothesis: The incidence of presentations that do not subsequently get published is high, and the numbers of major and minor inconsistencies, once published, are also high.

Study Design: Systematic review.

Methods: A database was created of all abstracts presented at AOSSM meetings from 1999 to 2001 from official program books. To assess whether each abstract had been followed by publication in a peer-reviewed journal, a PubMed search was conducted to include a 5-year follow-up for each conference. Minor inconsistencies included differences in title, authors, presentation of all outcomes, and authors’ interpretation of data. Major inconsistencies included discrepancies in study objective and/or hypothesis, study design, primary and secondary outcome measures, sample size, statistical analysis, results, and standard deviations/confidence intervals.

Results: Overall, 98 of the 165 abstracts presented at AOSSM national meetings from 1999 to 2001 were published in a peer-reviewed journal within 5 years, a publication rate of 59.4%. The median time to publication for all articles was 21 (range, 1-60) months. The majority of articles (61) were published in the American Journal of Sports Medicine (62.2%). The median number of major and minor inconsistencies from abstract to publication was 1 (range, 0-5) and 1 (range, 0-4), respectively. Sixty-two of the 98 published abstracts (63%) had at least 1 major inconsistency, while 79 (81%) had at least 1 minor inconsistency. In 5 manuscripts (5%), the authors’ interpretation of the data had changed, and in 2 (2%), the change essentially invalidated the abstract.

Conclusion: A large number of scientific presentations do not get published in a peer-reviewed journal. In addition, those published have a significant number of changes that, in a small percentage of cases, alter the validity of the original presentation.

Clinical Relevance: Orthopaedic surgeons and other attendees as well as nonattendees who reference conference abstracts need to exercise good judgment when considering the implications of oral presentations of unpublished materials. When reviewing meeting presentation abstracts, readers should remember that the material being presented is often not in its definitive or ultimate form.