Predictors of publication in Dental Clinical Trials

Objectives: To determine predictors of and time to publication of clinical trials presented at 6 dental conferences from 2005–2007. Method: Conference reports from the annual meetings of the International Association for Dental Research, European Orthodontic Society, American Association of Orthodontists, Australian Society of Orthodontists Congress and European Organisation for Caries Research, 2005-07, were hand-searched to identify abstracts of clinical trials. 4025 abstracts were identified from which 229 eligible clinical trials were found. A cross-referenced Medline search was performed to determine whether the trials had been published as a full paper and if so the date of publication was recorded. Factors, thought to influence publication, were assessed. The impact of these on publication rate and time to publication was investigated using univariate analyses, stepwise logistic regression and multivariate Cox hazards model. Results: The overall publication rate was 33.5% and mean time to publication was 20.4 months (95%CI 17.0, 23.8). Stepwise logistic regression revealed that the factors predicting publication rate were the principal author's grade (OR 4.3; 95%CI 2.1, 8.9) and gender (OR 2.3; 95%CI 1.2, 4.6) and disclosure of funding source (OR 2.07; 95%CI 1.1, 4.0). Multivariate Cox hazards analysis showed that the factors predicting time to publication were principal author's grade (HR 2.6; 95%CI 1.57, 4.3) and gender (HR 2.2; 95%CI 1.20, 3.94). The mode of presentation, study design, sample size, submission email address were not significantly associated with the publication rate or time to publication of abstracts presented at key dental conferences. Conclusions: Predictors of the publication rate and time to publication of abstracts presented at key dental conferences were found to be the grade and gender of the principal author. The disclosure of a funding source was also found to be a predictor for the publication rate.