Incomplete reporting of recruitment information in breast cancer trials published between 2003 and 2008
1Quality, Safety & Informatics Research Group, University of Dundee, Kirsty Semple Way, Dundee DD2 4BF, UK. streweek@mac.com
Objectives:
To review the reporting of key items of recruitment information in trial reports and estimate the number needed to screen to recruit one additional participant.
Study Design And Setting:
Review of breast cancer trials published in the years 2003-2005, 2007, and 2008.
Results:
The search identified 1,570 potentially eligible studies. After a random selection of 20% from each year and checking against inclusion criteria, a total of 207 studies were included in the review. Some items of information were well reported, such as the number included in the analysis. Sample size calculations were often not presented, but reporting is slowly improving. Who recruits participants and how many individuals were screened are often not reported. The median number needed to screen to recruit one additional participant was two (range, 1-593).
Conclusions:
Without reporting the when, where, by whom, and how many of recruitment, trialists deny readers part of the contextual description they need to judge whether a trial's results are applicable to their own situation. Trialists and journal editors need to be more diligent in following the reporting recommendations of the Consolidated Standards of Reporting Trials statement.
