NPC Archive Item: Improving the reporting of conference/journal abstracts: new CONSORT guidance

NOTE – This is an archive post from the NPC and has not been updated since first publication. Therefore, some hyperlinks may no longer be working.
merec_stop_press NPC Logo

Hopewell S, et al. CONSORT for reporting randomized controlled trials in journal and conference abstracts: explanation and elaboration. PLoS Med 2008 5: e20. doi:10.1371/journal.pmed.0050020

What is the background to this?
Abstracts are often the only publicly available information on randomised controlled trials (RCTs) reported at conferences. Even for fully published journal articles, because of difficulties in access or cost, or time available, it may be the only information that is read by many people. Because of word limit restrictions often not all the information needed to allow a robust assessment of a study can be obtained from the abstract; however, it is important that the abstract should not present biased or misleading information, and should accurately reflect the results of the designed study to which it refers.

The Consolidated Standards of Reporting Trials (CONSORT) Group has extended the CONSORT Statement to develop a checklist of essential items that authors should consider when reporting the main results of RCTs in any journal or conference abstract. The authors encourage journals and conference organisers to endorse the use of CONSORT for Abstracts by modifying their “Instructions to Authors”.  The checklist is reported and discussed in this article in PLoS medicine.

What is recommended for inclusion in abstracts?

  • Title: Identification of the study as randomised
  • Authors: contact details of the corresponding author (for conference abstracts)
  • Trial design: description of the trial design (e.g. parallel group, cluster randomised, non-inferiority study)
  • Methods:
    • Eligibility criteria for participants and the settings where data were collected
    • Interventions intended for  each treatment group
    • Specific objectives or hypothesis
    • Clearly defined primary outcome
    • How the participants were randomised (including if and how allocation was concealed)
    • Blinding or masking, i.e. whether or not participant, those administering the intervention and those assessing the outcomes were blinded to group allocation.  Authors should avoid using terms such as “single” or “double” blind.
  • Results:
    • Numbers randomised to each group
    • Trial status e.g. ongoing, closed to recruitment, closed to follow-up
    • Numbers of participants analysed in each group
    • Results for primary outcome i.e. a result for each group and the estimated effect size and confidence intervals for the contrast between groups as a measure of precision
    • Harms i.e. important adverse events or side effects.  If no adverse effects have occurred this should be stated explicitly.
  • Conclusions: general interpretation of results including clinical application and balance of benefits/harms.
  • Registration number and name of trial register
  • Funding

So what?
The CONSORT for Abstracts Statement checklist summarised above, with reference to the full article for explanation and elaboration where necessary, should be an essential companion for people writing an abstract of a RCT, whether for a journal article or a conference presentation. Compliance with the CONSORT for Abstracts Statement will enable a more robust evaluation of conference abstracts, which are often the only published information available for new medicines early in their development. Reviewers of the evidence are encouraged not just to rely on abstracts alone for interpreting the results of RCTs, but refer to the full article where available. However, if the abstract conforms to the CONSORT for Abstracts checklist, it should provide a reasonably balanced summary of the study results that is consistent with the study objectives, and will be valuable for those who cannot, do not wish to, or just don’t have the time to read the whole article.

Action. Authors of abstracts should, where possible, comply with the CONSORT for Abstracts Statement for reporting abstracts of RCTs. Journal editors and conference organisers should endorse the use of CONSORT for Abstracts by modifying their “Instructions to Authors” and drawing their readers’ attention to this reporting guidance. Reviewers of abstracts will find the statement valuable for assessing a trials’ validity and the applicability of reported results.

Please comment on this blog in the NPC discussion rooms, or using our feedback form.