Has the quality of abstracts for randomised controlled trials improved since the release of Consolidated Standards of Reporting Trial guideline for abstract reporting? A survey of four high-profile anaesthesia journals.

BACKGROUND AND OBJECTIVE

Randomised controlled trial (RCT) abstracts published in journal articles have traditionally been deficient of crucial information. To improve the quality of RCT abstracts, in January 2008, the Consolidated Standards of Reporting Trial (CONSORT) group published a checklist of essential information for inclusion. The current study assessed whether there has been an improvement in the quality of RCT abstracts published in main anaesthesia journals since this new guideline was introduced.

METHODS

Articles involving human RCTs published in four high-profile anaesthesia journals (Anaesthesia, Anesthesia & Analgesia, Anesthesiology and the European Journal of Anaesthesiology) were reviewed, comparing those published from October 2005 to September 2006 (pre-CONSORT abstracts) with those published from October 2008 to September 2009 (post-CONSORT abstracts). Trials involving healthy volunteers or cadavers, cost-effectiveness studies, meta-analyses and letters were excluded. Abstracts from remaining RCTs were randomly assigned to four reviewers in a blinded fashion and reviewed for content using the new CONSORT checklist.

RESULTS

In total, 527 RCT abstracts (pre-CONSORT RCTs, n = 275 and post-CONSORT RCTs, n = 252) were analysed. The majority of abstracts in both groups provided an appropriate description of study interventions (73.1 and 73.8%, pre-CONSORT abstracts versus post-CONSORT abstracts, respectively), objective (91.3 and 90.1%) and conclusions (72.4 and 66.3%). From pre-CONSORT to post-CONSORT guidelines for abstract reporting, there were significant improvements in correctly identifying blinding (18.2-29%) and harmful effects (31.6-42.1%). The improvement in reporting the nature of the trial in abstract titles (20.1-29%) and primary outcome measure in the methods section (22.9-30.6%) did not reach significance. There was no clear improvement in the already poor reporting of trial design, participants, randomisation, recruitment, outcomes, trial registration and funding sources.

CONCLUSIONS

Despite some promising improvements and inter-journal differences, the overall quality of RCT abstracts and adherence to the CONSORT checklist for abstracts remains poor.