Predictors of publication in dental research

• Main Author: Williams, Gareth
• Other Authors: Harrison, Jayne E
• Format: Text
• Language: English
• Subjects: Orca
• Online Access: http://livrepository.liverpool.ac.uk/9013/; http://livrepository.liverpool.ac.uk/9013/1/Thesis_Submission_following_corrections.doc; http://livrepository.liverpool.ac.uk/9013/3/WilliamsGar_Sept2012_9013.pdf

Aims: 2005-2007 abstracts This study aimed to identify the: • number of clinical trials that were presented from 2005-2007 at the conferences of the: o American Association of Orthodontists (AAO), o European Orthodontic Society (EOS), o International Association for Dental Research (IADR), o European Organisation for Caries Research (ORCA) o Australian Society of Orthodontists Congress (ASO) • abstracts that went on to be published as a full paper in a peer reviewed journal. • time to publication for those abstracts that were subsequently published as a full paper in a peer reviewed journal. • following characteristics of the abstract and determine their influence on the rate of and time to publication: o Result significance: (Significant, Non-significant, or Unclear) o Mode of presentation (Oral or Poster) o Study design (RCT / CCT) o Sample Size: (Absolute number) o Funding disclosure: (Yes / No) o Continent of origin: (North America, South America, Europe, UK, Asia, Africa, Australasia). o Primary author: • Gender (Male / Female / Unclear) • Professional status: (Professor / Non-professor / Unclear) • Identify reasons why abstracts did not achieve publication. University Teachers Group (UTG) abstracts This study aimed to identify the: • number of abstracts presented at the University Teachers Group session, from 1999-2010, at the British Orthodontic Conference. • following characteristics of the abstract and determine their influence on rate of and time to publication: o Funding disclosure: (Yes / No) o Dental School of origin • abstracts that went on to be published as a full paper in a peer reviewed journal. Design: Retrospective, observational study. Subject and Setting: The sample frame included dental clinical trials presented at the conferences of the International Association of Dental Research (IADR), European Orthodontic Society (EOS), European Organisation for Caries Research (ORCA), The American Association of Orthodontists (AAO) and The Australian Society of Orthodontists (ASO) from January 2005 to December 2007. The sample frame for the University Teachers Group (UTG) abstracts, included abstracts presented at the UTG session of the British Orthodontic Conference (BOC) 1999-2010. Sample size Spencer found a publication rate of 38% from abstracts of clinical trials presented at EOS, IADR, ORCA and a 50% increase would be give a publication rate of 57%. Using data from Spencer in Pocock’s formula, 210 abstracts would be required to give 80% power, at the 5% level, and enable me to detect a 50% rise in the proportion of clinical trial abstracts published. Method: Clinical trials presented at above conferences were identified from the associated journals or conference proceedings. Inter-examiner and intra-examiner reliability were assessed using a random 10% sample of abstracts. A MEDLINE search was undertaken to determine whether the abstract had been published in full. The date of publication was recorded. Authors of abstracts that did not reach publication were contacted to determine the reasons. Results: Seven thousand and sixty-nine abstracts presented from 2005-2007 were identified, including 215 clinical trials. 142 abstracts were identified from the UTG session from 1998 – 2008, and all were included. The publication rate for the 2005-2007 sample was 32.6% and the UTG sample 34.5%. There were no predictors of publication in either group studied. The median time to publication of the 2005 – 2007 group was 16.00 months, IQR (10, 26) and the mean time to publication for the UTG group was 18.3 months (95% CI 14.38, 22.19). For the unpublished 2005-2007 group, reasons given for failure to publish were lack of time (8.3%), language, culture, lack of teaching (1.4%), rejection (0.7%), motivation (0.7%), perceived editorial bias (0.7%) and length of review process (0.7%). For the UTG group, reasons given included lack of time (19.4%), lack of interest from SpR (9.7%) or in press (7.5%). Conclusions: No predictors of publication were found for either group studied. For the unpublished 2005-2007 group, main reasons given were lack of time (8.3%), language, culture and lack of teaching (1.4%). For the UTG group, reasons given included lack of time (19.4%), lack of interest from SpR (9.7%) or in press (7.5%). The qualitative results should be viewed with caution due to the low response rate (12.4% for the 2005 – 2007 sample, and 68.8% for UTG).