The authors of a paper on websites that sell cholesterol-lowering statin medications without prescriptions have retracted it. The move followed the threat of a lawsuit against the journal by a company included in the study that says it never dispenses sans a script.
Here’s the notice:
The following article from Pharmacoepidemiology and Drug Safety, “Direct to consumer Internet advertising of statins: an assessment of safety”, by Bethan Williams and David Brown, published online on 2nd February 2012 on Wiley Online Library (wileyonlinelibrary.com) and in Volume 21, Issue 4, pages 352–365, April 2012, has been retracted by agreement between the authors, the journal Editor-in-Chief, and John Wiley & Sons Ltd. The retraction has been agreed due to the inability to verify the accuracy of the data in Appendix 1.
Appendix 1 is a list of 184 websites that allegedly sell statins without a prescription. But the journal’s editor, Brian Strom, of the University of Pennsylvania, tells Retraction Watch:
A lawyer for a company specified in the appendix threatened a suit, claiming they never dispensed meds without a prescription. We told the author we would back them, but we needed to see proof the website ever said that. The authors had no evidence to back up their statement.
Corresponding author David Brown, of the UK’s University of Portsmouth, tells us:
We conducted our study in good faith; however, on publication, it was brought to our attention by one of the editors that the producers of one of the websites listed in the Appendix to the paper had indicated that the site was listed in error as offering statins for sale without a prescription.
We did not keep the original screen-dumps of websites at the time of our study (19th November – 23rd December 2010) and as we could not verify the inclusion of the website, we immediately agreed to retract the paper in accordance with the editor’s wishes.
I must stress that this was an honest error on our part; there never was any intention to mislead the journal or its readership. This stance was accepted by the journal editor.
Hat tip: Clare Francis
Okay, sloppy. But was that one website so vital for the conclusions of the paper that a correction would not have done a good enough job?
That is what I was thinking! One site out of 184?!
“By clicking submit you confirm that you possess a valid prescription for statins from a registered medical practitioner”
See? We never distribute statins to people without prescriptions.
Ah, so that’s how it’s done. Clever.
Yes, indeed. If it was only one site of the 184 sites and it was the result of a typo, then a correction and apology should have been enough. But since the researchers didn’t take screen shots of the sites, they have no proof for any of them. Although the Wayback Machine idea is intriguing. Would that stand up in court????
I’m surprised that the researchers didn’t think to document the evidence at the time and that the reviewers didn’t flag this. Did nobody run this by the journal’s lawyer? When you accuse people of breaking the law, it’s a good idea to have proof. When you report that the police have arrested somebody for breaking the law, you are protected by media libel laws as long as you are truthfully reporting what the police said. But when you make the allegation yourself, you’d better have some evidence — receipts, credit card statements, a screen shot of the on-line offer, something besides just your word that it happened.
Rolling on the floor laughing…
My guess is they didn’t have proof for any of the 184 websites, and, rather than risk an onslaught of lawsuit threats, they retracted.
Surely the Internet Archives (Wayback Machine) could have provided the necessary evidence, i.e., sites as they appeared at the time of the research?
The Wayback Machine can sometimes be a lifesaver in these cases, but it does not provide 100% coverage of every site on the internet, and frequently fails to index all or parts of some websites. Having screenshots or printouts is always a good idea.
Shouldn’t the criteria they used to decide whether a website was selling statins without a prescription been made clear in the Materials and Methods? And if it wasn’t, how on earth did they get this published? It seems that unless the authors actively lied that the journal and reviewers are at least a little bit culpable.
The journal is legally liable for what it prints, regardless of what the reviewers and the authors have or have not done in the way of responsible behavior. If an offended party threatens to sue, the journal editor must have the evidence (or be able to get the evidence quickly from the authors) to support what has been printed, or the journal’s lawyer will insist on a speedy retraction and an abject apology. Photocopies of incriminating documents or videotapes of incriminating statements are what the editor needs to support the journal’s published allegations. Or in this modern day, a substitute such as a screen shot of the online offer. Investigative journalists, who do this kind of thing all the time, routinely collect documentation for their allegations. The authors of this manuscript may be new to the game.