117 The Full Publication Rate of Abstracts Presented at the British Cardiovascular Society Annual Conference Remains High, Comparing Favourably with other Cardiology Annual Scientific Sessions Worldwide

Abstract

BackgroundIn recent years, there has been significant translation of scientific knowledge, from discovery to clinical practice in cardiology. Although a critical stage of this process is the dissemination of findings, it remains uncertain what proportion of research presented at scientific meetings is subsequently published in peer-reviewed journals. The aim of this study was to evaluate the publication rate of abstracts presented at the British Cardiovascular Society (BCS) Annual Conference over a 10-year period, and identify factors associated with full publication.MethodsAll abstracts presented at the BCS between 2002–2012 were retrieved and assessed. Key details, including author, institution, study type and subject matter were extracted. Abstracts were excluded if subject to multiple presentation, prior publication, or if the primary institution was non-UK-based (Figure 1). Manual, structured database searches (Medline, Embase, Web of Science) were performed on 1st September 2014, cross-referencing primary and senior author, abstract title and keywords. Manuscripts were retrieved and verified in tandem with the presented abstract; validated by 2 independent observers. Publication metrics (including impact factor; IF) were ascertained. Data are presented: median (interquartile range).Abstract 117 Figure 1A flow diagram of abstracts accepted for presentation, 2002–2012ResultsFrom 2002–2012, n = 2,392 abstracts were accepted for presentation at BCS. The mean number of abstracts presented annually was 173 (46). The overall publication rate was 62%. Lag time to full publication was 1.4 (1.6) years; >95% occurred by 4 years (Figure 2). Average IF was 3.8 (3.8); 47.6% were published in high IF journals (IF ≥4). No significant differences in overall publication rate, impact factors or time to publication were observed between 2002 and 2012. To date, publications presented at the BCS were cited 41,600 times; with an average citation rate of 4.4 citations y-1. Basic science abstracts were twice as likely to achieve full publication as epidemiological and clinical abstracts (odds ratio 2.35; 95% confidence intervals 1.73–3.19; p < 0.0001). Abstract subject matter (i.e. disease focus) was not a predictor of full publication. Inter-observer agreement for abstract and publication matching was high (κ = 0.96; p < 0.001).Abstract 117 Figure 2(a) Secular trends in abstract presentation and publication; (b) Lag time between presentation and subsequent publicationConclusionCardiovascular research within the UK remains robust. Presentation at the BCS suggests a two-thirds chance of subsequent full publication, comparing favourably with top-level annual scientific meetings of other clinical specialities within the UK, and cardiology societies worldwide.