Clinical trial registration
In an editorial by Drazen and Zarin in this week's NEJM, "Salvation by Registration," there's a brief tale about an author who submitted a paper to the journal, and the reported trial was incorrectly entered into clinicaltrials.gov - the paper was initially refused due to this issue, and it turned out that the trial's sponser was responsible for the incomplete record in the database. One of the co-authors inadvertently worked around this issue by creating a new, complete and accurate record, and the paper is included in this week's issue.
The editorial points out the importance of a complete trial protocol submission to a public registry in compliance with the ICMJE initiative; they also point to the ineffectiveness of duplicate efforts on the part of the study sponsor and the study investigators, like that noted above.
Details on registering trials in clinicaltrials.gov in compliance with these recommendations is available here -- looking at the protocol questions/issues that need to be addressed during clinical trial registration is also a really good refresher course on the key methodological details of such studies.