After publishing a paper about neuropathy in diabetic patients last week, The New England Journal of Medicine (NEJM) immediately corrected it after editors learned of errors and some missing disclosures within the article.
The notice explains that the sole author of the paper, “Diabetic Sensory and Motor Neuropathy,” reported incorrect doses for several medications, and received royalties for the tool to measure quality of life used in the paper. The author told us all the declarations were “discussed in detail” between him and the journal, and both parties agreed to the final decision.
Let’s take a look at the lengthy correction notice — what some of our readers might call a “mega-correction:”
Diabetic Sensory and Motor Neuropathy (Clinical Practice, N Engl J Med 2016;374:1455-1464). In Table 1 (page 1457), in the second entry in the third column, “impaired autonomic dysfunction” should have been “impaired autonomic function.” The third entry in the first column should have been “Recognize clinical implications.” The fifth entry in the first column should have been “Consider differential diagnosis.” In Table 2 (pages 1460-1), the initial dose of topiramate should have been 25 mg/day, and the effective dose 25−100 mg/day. For gabapentin, the upper bound of the confidence interval for the number needed to treat should have been 8.3. The effective dose of amitriptyline should have been 25−150 mg/day. The effective dose of tramadol should have been 100−200 mg/day, and the number needed to treat should have been 4.7. In the final sentence of the second paragraph under Areas of Uncertainty (page 1461), the term “(frank deficiency)” should not have appeared. The disclosure statement at the end of the article (page 1462) should have included the following information, which Dr. Vinik regrets not having provided earlier: Dr. Vinik reports having received consulting fees from Cline Davis Mann and System Analytic. Dr. Vinik reports having received royalties for the use of the Norfolk QOLDN tool, a quality-of-life instrument owned by his medical school, which he codeveloped for use in clinical trials in patients with diabetic neuropathy. Dr. Vinik also reports being the inventor of a dietary supplement that includes a mixture of alpha lipoic acid, methylcobalamin, benfotiamine, dihomo-γ-linoleic acid, cholecalciferol, and ascorbic acid, the rights to which were assigned to his medical school. Dr. Vinik reports that he, his laboratory, and his department have not received and will not receive any income that might derive from this product. The article is correct at NEJM.org.
The paper has not yet been indexed by Thomson Reuters Web of Science.
A NEJM spokesperson told us about the changes that were introduced to effective drug dosages:
The upper end of the effective dose range for amitriptyline was changed from 100 mg/day to 150 mg/day. Initial dose of topiramate changed from 15 mg to 25 mg/day and the lower end of the effective dose range changed from 30 mg to 25 mg/day. The effective dose for tramadol changed from 210 mg/day to 100-200 mg/day.
The dosages listed initially would not have posed a threat to patients, the NEJM spokesperson said. Nevertheless, the spokesperson added:
The author and editors felt it was appropriate to correct them when errors were noted.
Author Aaron Vinik, who is the director of Research & Neuroendocrine Unit at the Eastern Virginia Medical School in Norfolk, Virginia, told us:
It should be clear that I have disclosed all sources of income to Eastern Virginia Medical School. I did not perceive that there would be a conflict of interest related to the items listed and the content of the manuscript. For e.g. use of the Norfolk quality of life instrument is a measure of quality of life used in clinical research trial (but not in clinical practice) which is owned by the medical school that has licensed it to a number of companies and received the royalties which have been disbursed according to the school policies. After discussion with the editors of the journal we jointly resolved that it would be preferable to avoid any misconceptions on what may be conceived of as a conflict of interest.
Vinik added:
The only royalties I have received is for the QOL tool use and this information can be obtained from the medical school.
NEJM has been criticized in recent months after issuing a controversial editorial downplaying the sharing of data, and for being slow to correct their published content.
Certainly no one can criticize the journal for not “doing the right thing” by delaying this latest correction since the original paper is dated April 14, 2016, whereas the correction is dated April 13. The NEJM spokesperson explained the situation to us:
A pharmacist reviews our content prior to publication to check accuracy of drug dosages. In this case, the process was delayed and errors were identified after the issue was printed. The errors were corrected prior to online publication. Omitted disclosure information was brought to our attention by a colleague who received an embargoed copy of the issue. A formal correction was issued because the article had already been printed at the time corrections [were] made, though it had not yet been released on our website.
When questioned about whether the journal has made policy changes to speed up its correction processes, the NEJM spokesperson said:
We are always open to suggestions on how to improve our processes and we take all such suggestions seriously.
We’ve also found three other errata for papers that Vinik has previously co-authored. In one paper, “Biochemical testing for neuroendocrine tumors,” an author’s first name was changed from “Gene” to “Eugene” (see second erratum at this link). Vinik said:
Gene Woltering and I are close friends and he has always been Gene to me.
In another corrected paper, “Adding insulin glargine versus rosiglitazone: health-related quality-of-life impact in type 2 diabetes,” a mislabeled figure was tweaked. According to Vinik, the erratum
…has not impacted the interpretation of the data.
Finally, the erratum to the last paper, “Diabetes-related complications, glycemic control, and falls in older adults,” tweaked three incorrectly reported percentages in the first sentence of the abstract’s results section. Vinik was one of the middle authors on this paper.
Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up on our homepage for an email every time there’s a new post, or subscribe to our new daily digest. Click here to review our Comments Policy. For a sneak peek at what we’re working on, click here.