Retraction Watch

Tracking retractions as a window into the scientific process

Nature fixes highly cited paper suggesting food additives hurt the gut

with one comment

Nature_latest coverA 2015 study about dietary emulsifiers has been corrected by Nature after another researcher pointed out a few ambiguities.

When it first appeared, the study — which showed emulsifiers cause inflammation in the guts of mice — received a fair amount of media attention, including from Nature’s own news department. But since publication, a researcher noted some imprecision around the ages of mice used in the sample, affecting the paper’s calculations of weight gain over time. Andrew Gewirtz, co-author of the study from Georgia State University, told us the change did not affect the conclusions of the paper.

Here’s the corrigendum for “Dietary emulsifiers impact the mouse gut microbiota promoting colitis and metabolic syndrome”:

Some clarifications are provided to this Letter; these do not alter any of the central conclusions but, rather, are provided in the interests of transparency and reproducibility. Our Letter indicated that experiments were performed on 4-week-old mice (unless stated otherwise). In fact, for several experiments, mice ranged from 5 to 7weeks as follows: Fig. 4a–h, Extended Data Fig. 9g–w, b′ –t′ , z: 5 weeksold; Figs 1, 2, 3a–d, 4i–o, Extended Data Figs 1a–d, 2, 4, 5s–v, 6, 7h–k, 8l–s, 9a–f, x, y, 10: 6 weeks old; Fig. 3e–l, Extended Data Fig. 1e–l, 5q, r, 7a–g, l–h′ , 8a–k, t–o′ , 9a′ : 7weeks old. The weight gain versus time curves are affected by mouse age and hence explain why the kinetics of weight gain differ among control mice when comparing between different experiments.For each experiment, we listed the average n value for all the conditions within each panel, which differed from the exact n for each experimental condition within each figure, as shown in Supplementary Table 1 of this Corrigendum.

Furthermore, our Letter reported relative changes in body mass and absolute mass of fat pads, thus not permitting assessment of absolute weight changes nor fat pad mass relative to total body mass. Hence, we provide measures of absolute and relative body and fat pad mass in a side-by-side manner in the Supplementary Data to this Corrigendum.

The 2015 study has been cited 40 times since publication, making it a highly cited paper, according to Thomson Reuters Web of Science.

At the time of publication, Gewirtz told “Not Exactly Rocket Science,” a blog by Ed Yong on National Geographic, that one implication of the study is that:

…current methods for testing food additives for safety are not adequate.

Gewirtz told us the corrigendum was

issued at the request of the journal following inspection of our raw data by a scientist who requested access to it. Briefly, the researcher inspecting it noticed a couple of area of imprecision re the ages of mice and that we provided average n per condition instead of the precise n for each condition…

When asked about the identity of the scientist, Gewirtz said the researcher has requested anonymity. Gewirtz, nevertheless, noted that the paper originally had that level of “precision,” but this was taken out for brevity in the final editing process after peer review. He added: 

The corrigendum does not alter the conclusions of our study in ANY way whatsoever re the effects of the emulsifiers that were studied…

He added that journal policies around brevity often require papers to be “considerably shortened,” which, especially after peer review, increase the chances of important information being omitted.

Gewirtz also corrected a 2008 paper published in The Journal of Immunology where a figure was accidentally published twice.

Hat tip: Jonathan Eisen

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.

Comments
  • Salve May 22, 2016 at 12:19 pm

    This paper had serious credibility problems, which raises many questions regarding the quality of work published in NATURE magazine. I have never heard that reporting the number of replications (n values) as “averages” of several experimental groups is even allowed in a scientific publication. This is not “lack of precision”, clearly, Nature published an article with misleading information.

    The statements made by Dr. Gewirtz are rather astonishing (to say the least). In particular, the comments where he states that “the paper originally had that level of “precision,” but this was taken out for brevity in the final editing process after peer review” and further that “journal policies around brevity often require papers to be “considerably shortened,” which, especially after peer review, increase the chances of important information being omitted” seem astonishing.

    If indeed Nature’s “policies around brevity” induce authors to publish misleading information, then there is something tragically wrong.

    It would be of interest to see Nature’s editor response to this.

  • Post a comment

    Threaded commenting powered by interconnect/it code.