Do recorded abstracts from scientific meetings concur with the research presented?

Research abstracts for scientific meetings are usually submitted several months in advance of the meeting. Authors may therefore be tempted to submit an abstract on the basis of the research that is ongoing or not yet fully analysed. This study aims to determine the extent to which submitted abstracts, often disseminated in printed form or online, differ from the research ultimately presented. The risk taken by clinicians considering changes in practice on the basis of presented research who refer back to the printed abstract can be assessed.

All posters presented at the Royal College of Ophthalmologists Annual Congress 2007 were compared with abstracts in the 'Final Programme and Abstracts'. Discrepancies were recorded for authorship, title, methodology, number of cases, results and conclusions.

A total of 171 posters were examined. The title changed in 21% (36/171) and authorship in 25%. The number of cases differed in 22% (number of cases in the poster ranging from less than one quarter to more than triple the number in the abstract). Differences between abstract and poster were found in the methodology of 4%, the results of 11% and conclusions of 5% of studies.

Scientific meetings provide an opportunity for timely dissemination of new research presented directly to clinicians who may then consider change of practice in response. Caution is advised when referring back to printed records of abstracts, as substantial discrepancies are frequently seen between the published abstract and the final research presented, which, in a minority of cases, may even alter the conclusions of the research.