A paper whose expression of concern we covered in November 2014 has been retracted and republished “because of the extent of the changes necessary,” according to the Lancet Respiratory Medicine.
This isn’t the first time we’ve seen retraction/republications at the Lancet, accompanied by thorough breakdowns of the problem.
This study was a meta-analysis of research on how the timing of tracheostomies — placing a breathing tube directly into the windpipe — affects patients’ mortality rate. The original paper found that critically ill patients who received a tracheostomy earlier fared better than those for whom the procedure was delayed for weeks after intubation, the recommended practice.
However, when the authors calculated how many patients died, they assumed that any patient who wasn’t discharged from the intensive care unit (ICU) had died there, instead of looking for other explanations. This made their estimates unreliable.
The publisher convened a panel, which ultimately decided retraction and republication was the most appropriate course of action.
The original expression of concern contained a quote from the authors about the “data discrepancy” that would later sink the paper:
We have made every effort subsequently to acquire the correct data on ICU mortality for both trials, but have been unable to obtain data for ICU mortality…We (the authors) cannot be sure that our finding regarding ICU mortality is not misleading.
Here’s the retraction notice for “Effect of early versus late or no tracheostomy on mortality of critically ill patients receiving mechanical ventilation: a systematic review and meta-analysis”:
On June 27, 2014, The Lancet Respiratory Medicine published online a systematic review and meta-analysis of early versus late tracheostomy. Following publication, our attention was drawn to some possible data discrepancies affecting the findings for intensive-care-unit mortality and we issued an expression of concern highlighting the details on Oct 14, 2014. The editors have discussed the corrections that are necessary in the paper, and the findings of a panel that we convened, and decided that because of the extent of the changes necessary, the previous version of the Article should be retracted and a corrected version republished after re-review.
The paper has only been cited once, according to Thomson Scientific’s Web of Knowledge. We’ve emailed the author and journal, and will update if we hear back.
Like Retraction Watch? Consider supporting our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, and sign up on our homepage for an email every time there’s a new post.
Having followed RW closely for about 2 years now, and having seen, and witnessed, paper after paper, multiple problems, in the worst case examples, where there are just simply too many errors to merit an erratum, I think that this is the most logical and correct procedure going forward. It reflects careful consideration of the issues, and it reflects also cognizance by all parties involved (authors, editors, publisher), that the problem needs to be fixed. So, retraction, followed by fixing and re-reviewing, although a slow and painful process, seems to be the most viable model moving forward. Do others feel that there is a better model of dealing with the problems?