Publication Rates Vary Across Orthopaedic Subspecialties: A Longitudinal Analysis of AAOS Abstracts.

Background: Presentation of research at national orthopaedic meetings and subsequent publication are important for both information exchange among surgeons and individual academic advancement. However, the academic landscape and pressures that researchers face may differ greatly across different subspecialties. This study attempts to explore and quantify differences in research presented at national conferences and its implication on ultimate likelihood of publication in peer-reviewed journals. Methods: All abstracts from the Annual Meetings of the American Academy of Orthopaedic Surgeons (AAOS) from 2016 and 2017 were reviewed and categorized based on subspecialty focus. Resulting publications were identified using a systematic search of PubMed and Google Scholar databases. Multivariate binary logistic regression modelling was used to assess the predictive value of abstract characteristics on eventual publication. Results: A total of 1805 abstracts from the 2016 and 2017 AAOS conferences were reviewed. The overall publication rate of abstracts following the AAOS meetings was 71.6%, with an average time to publication from abstract submission deadline and impact factor of 19.8 months and 2.878, respectively. Statistical differences were observed across subspecialties with respect to publication rate (p<0.001), time to publication (p<0.001), and impact factor (p<0.001). The subspecialty with the highest publication rate, largest impact factor, and shortest average time to publication was Sports Medicine with 83.2%, 3.98, and 17.6 months, respectively; despite lower average sample size (p<0.001) and frequency of multicenter design (p<0.001) compared with other subspecialties. The subspecialty with the lowest publication rate and impact factor was Hand and Wrist with 53.3% and 1.41, respectively. Multivariate logistic regression analysis demonstrates a lower likelihood for internationally authored abstracts (OR: 0.75, p=0.021) and higher likelihood for basic science abstracts (OR: 1.52, p-value=0.023) to reach publication. Conclusion: Differences in publication rate across orthopaedic subspecialties were observed with articles in sports medicine more likely to be published, published quickly, and featured in a higher impact factor journals. Understanding these differences, and how they relate to the publication and promotion of novel research, is important for orthopaedic researchers.