Retraction Watch

Tracking retractions as a window into the scientific process

“Know how to recognize pseudoscience:” Reader reveals how fish oil paper came to be retracted

with 28 comments

Ian Garber

Ian Garber

After our post yesterday on a fishy retraction from author Brian Peskin, a reader who alerted the journal to problems got in touch to give us the lowdown.

Ian Garber is in the last year of medical residency at the University of British Columbia. Here’s the story he told us via email:

A few months ago, I saw an article reposted on Facebook called “Why Fish Oil Fails: A Comprehensive 21st Century Lipids-Based Physiologic Analysis.” I was quite interested, since I had done some work in a lipid disorders clinic not too long ago, and omega-3 fatty acids were something still being recommended by the specialists there. (I’m an M.D., currently completing my final year of specialty training in medical biochemistry at the University of British Columbia).

The article, published in Journal of Lipids, discussed at great length how the initially proposed health benefits of omega-3 fats had not panned out, and in fact polyunsaturated fats could be harmful to your health because they were prone to oxidation into toxic substances. The article was filled with vast numbers of references, and highly detailed mechanistic explanations, but surprisingly little empirical evidence to back these up. This attempt to impress the reader with overwhelming and unnecessary detail made my pseudoscience sense tingle, but it seemed to be in a legitimate journal (the parent company Hindawi publishes over 400 different academic journals), so I read on.

Having taken in the rest of the article, my impression was that there was some interesting information scattered throughout, but overall this was a very poorly written and heavily biased paper. The author made numerous logical errors and wild extrapolations. He repeatedly drew conclusions that didn’t follow from his premises. He used appeal to authority, toward the vague entity of “21st century lipid science”… This is the kind of language I would expect in marketing, not biochemistry. Hmmm… I checked the author affiliations and conflict of interest statement. No conflict of interest declared. Stated affiliation was “The International PEO Society”. When I looked up this organization, the only mention I could find of them was their website: http://www.peo-society.org/  No content, not even an explanation of what PEO stands for, just a logo and another vague statement about “state-of-the-art 21st century medical science”. At this point, alarm bells are screaming: looks like astroturfing to me. Basically a placeholder to give the superficial appearance of a legitimate organization.

I moved on from the affiliations, and did some more digging on the author himself. I was shocked to see that one of the first search hits was “Peskin Pharmaceuticals” (http://www.peskinpharma.com/) with none other than the author himself at the helm. Didn’t the article say no conflicts of interest?? I checked again. Yup. Browsing the website, it became clear that Peskin Pharmaceuticals was selling a supplement stated to be “The first and only organically certified drug to prevent/reverse both cardiovascular disease and cancer utilizing a unique patent-pending formulation — maximizing intracellular oxygen transfer. All active ingredients are natural/organically grown and processed. You and your patients can feel good about taking them.” In other words, a product that would be in direct competition with fish oil!

Further down the search results came the following damning report: http://www.quackwatch.org/11Ind/Peskin/peskin.html This revealed that Brian Peskin does not have any medical credentials, only a degree in a completely unrelated field (electrical engineering), and that he has previously faced criminal charges for misrepresenting his credentials and making fraudulent medical claims!

I emailed the editor of Journal of Lipids, detailing my concerns, and asking that at bare minimum they amend a conflict of interest statement to the article. I received a cursory response, thanking me for the information, and promising that they would investigate further. I didn’t seriously expect that this would go anywhere, and when I didn’t hear back after a few weeks, I assumed the journal had decided not to act. Then last week I received the following email:

“Dear Dr. Garber,

I am writing to you in regard to the investigation that we have conducted into the publication of the manuscript “Why Fish Oil Fails: A Comprehensive 21st Century Lipids-Based Physiologic Analysis.” First, I would like to thank you for your input into this investigation. We have now had a chance to look into a number of concerning points regarding this manuscript, and as a result of our findings we have retracted this article on the grounds that its author had a competing interest that he did not declare. Please check the retraction statement of the article on http://www.hindawi.com/journals/jl/2014/495761/.

If you have any questions regarding this case please let me know, and once again I would like to thank you for your input into our investigation.”

Lessons:

  • Scientific journals are too caught up in the business of publishing to do even cursory checks of their authors’ credentials. This is not the sort of thing peer review can pick up, because the author’s identity is not disclosed the reviewer. So check yourself – especially if it is a single-author paper.
  • This particular journal may warrant further scrutiny, given its acceptance of this highly questionable paper and its “pay-to-publish” model. The references cited in the article include sources like old textbooks and Scientific American, which no serious review process would have allowed to slip through.
  • Know how to recognize pseudoscience, and be vigilant no matter what the source of information. This may seem daunting if you don’t have any formal scientific training, but there are lots of great resources online that can give you the tools you need.

Written by Cat Ferguson

November 12th, 2014 at 9:30 am

Comments
  • emesax November 12, 2014 at 9:44 am

    Great guy!!! And with a good “pseudoscience sense”. Just a warning: do not jugde scientific legitimacy of a journal by the number of journals published by the parent company.

    • Neuroskeptic November 12, 2014 at 9:53 am

      Very true. Anyone can start a junk journal. And anyone who can start one junk journal can start lots of them.

  • hvergelmir November 12, 2014 at 10:30 am

    Isn’t anything from Hindawi pretty dodgy anyway?

    • yc November 12, 2014 at 11:08 am

      Well, Hindawi, indeed started as one of those predatory open access publishers that like mushrooms – here today (to take your money) – gone tomorrow. Yet, Hindawi, refused to go away and being backed by serious money started to clean its act and get it together. They are getting much better. I am not at the point where I would publish there myself but I would not dismiss them out of hand either. They are also going through a process of careful journal acquisition and pruning. In summary, while jury is still out to the overall quality of Hindawi they are trying to move in a right direction. I would be watching them carefully and may be in 2-5 years they will become one of the good alternatives – they do have several interesting (at least in paper) titles in my field. They also tend to specialize as opposed to publishing World Journals of Everything Remotely Sounding Scientific.

    • Rob Siebers November 12, 2014 at 12:43 pm

      Hindawi is not dodgy. Jeffrey Beall initially listed it in his predatory open access publishers list, but after further investigation of the company removed it from that list. A number of Hindawi journals are ISI listed. I have published in one of their journals and in my opinion, their review process was professional.

      • hvergelmir November 13, 2014 at 11:45 am

        Thanks for that. Good to see that they’re cleaning up their act. I stand tentatively corrected!

        • DocMartyn November 16, 2014 at 3:10 pm

          I have published with them. Their low prices for color images is important if you have lots of images.

          • JATdS November 16, 2014 at 3:21 pm

            Surely, an irrelevant issue for an open access journal, as are journals published by Hindawi. PDF files cost no more money for colour figures, or for lengthy manuscripts.

    • Charles Hoogstraten November 13, 2014 at 10:36 am

      They seem to have done the right thing in this case. You’d never see that from one of the truly predatory places.

  • Michael Carnovale November 12, 2014 at 10:37 am

    “if you don’t have any formal scientific training, but there are lots of great resources online that can give you the tools you need.”

    May someone point me towards these resources? I’m fairly new to critical appraisal.

    Thanks

  • Lo Mein November 12, 2014 at 10:54 am

    There will always be those trying to cheat the system. My concerns are: where were the reviewers? Where was the editor?

  • Other Ian November 12, 2014 at 12:43 pm

    Is it common for some journals not to disclose the author’s identity? This is not usual in the journals I publish in. I have reviewed a couple articles myself (I’m a new post-doc) and I see the authors names. If I have never heard of them I always look them up to see if the research is in their realm of expertise. At the very least the editor should have their identity and should look them up. I’ve never heard of journals that don’t disclose the authors to the reviewers.

    • Marco November 12, 2014 at 12:54 pm

      Double-blind peer review is not uncommon in medical journals, nor is it in some social sciences, if I remember correctly.

  • JATdS November 12, 2014 at 12:55 pm

    Other Ian. Blind or double-blind peer review = not disclosing the identities of the authors/reviewers and/or vice versa. In some ways its good, because it eliminates bias, but it’s not too effective if the person has published extensively since in the “blind” copy of the article that is sent out for peer review, there are still many hints or vestigial traces as to who the research group is, e.g., in the methodology that might refer to previous methods used, or the Introduction or Discussion that might state “Our previous study …”. So, currently, there is very little choice as to how anonymity can be implemented in the traditional peer review system, which is why the F1000Research system and even that by Frontiers is good (despite other criticisms of these publishers), because all names (authors, editors, reviewers) are on the table from the word go, and nothing can be hidden. However, what if there are potential rivalries or COIs between the authors and the editors? Surely, then the authors will be disadvantaged from step 1, and will most likely never get a fair, honest, or professional peer review? The risks become larger as the field of study becomes narrower simply because the selection of target journals also becomes narrow.

  • rose water November 12, 2014 at 2:41 pm

    I get emails from Hindawi just about every day asking me to submit a paper for publication, often from journals far outside my area of expertise. Their model is one of payment for publication. Ideally that would also come with real peer review.

  • JATdS November 12, 2014 at 6:41 pm

    Hindawi: “we have retracted this article on the grounds that its author had a competing interest that he did not declare.” This only covers one aspect of the problems listed by Garber. Sorry to say, the peer review at Hindawi, a claimed peer reviewed journal [1], failed to some extent. Indeed, Peskin has taken the blame, because he did fail to declare COIs, and has suffered a consequence: a retraction. But let Hindawi also be held accountable for an incomplete view of the issue. So, justice was dealt to Hindawi in the form of exposure of the truth here at RW: a public notary of the full issues. I wonder if Peskin paid the Hindawi open access fee of 800 US$ [2]. Perhaps Dr. Garber could submit a letter to the editor that critiques the other academic issues, and try and have it published in the very same journal. That would be the ultimate test of transparency and academic integrity of Hindawi. Those in passionate defense of Hindawi’s apparent improvement in quality are reminded of the following retractions covered by Retraction Watch [3-8], a post-humous review and publication [9]. This is because retractions can sometimes say as much (if not more) about the publisher than they do about the authors, e.g., failed “peer” review that is supposed to be fail-safe and trusted. If there are problems with peer review, should Hindawi be making the profit margin that it does [10]? Admittedly, it does allow for some good post-publication peer review to get published (e.g., [11]).

    [1] http://www.hindawi.com/journals/jl/ (“Journal of Lipids is a peer-reviewed, open access journal”)
    [2] http://www.hindawi.com/journals/jl/apc/
    [3] http://www.hindawi.com/journals/crid/2014/382142/
    http://retractionwatch.com/2014/09/12/wrong-in-the-tooth-faked-data-authorship-issues-force-retraction-of-dental-paper/
    Case Reports in Dentistry
    [4] http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4038915/
    http://retractionwatch.com/2014/08/18/i-know-you-are-but-what-am-i-school-program-paper-pulled-for-duplication/
    The Scientific World Journal
    [5] http://www.hindawi.com/journals/rerp/2014/130543/
    http://retractionwatch.com/2014/06/30/substantial-flaws-trip-up-big-toe-paper/
    Rehabilitation Research and Practice
    [6] http://www.hindawi.com/journals/bmri/2012/158232/
    http://retractionwatch.com/2014/06/19/recombinant-protein-paper-retracted-for-recombining-others-work/
    BioMed Research International
    [7] http://www.hindawi.com/journals/bmri/2014/308263/
    http://retractionwatch.com/2014/06/18/duplication-forces-retraction-of-liver-cancer-paper/
    BioMed Research International
    [8] http://chp.sagepub.com/content/19/3/220.full (was the journal bought by Sage Journals?)
    http://retractionwatch.com/2014/06/05/plagiarism-spells-demise-of-complementary-medicine-paper/
    Journal of Evidence-Based Complementary Medicine
    [9] http://www.hindawi.com/journals/omcl/2014/345105/
    http://retractionwatch.com/2014/04/03/late-resveratrol-researcher-dipak-das-manages-to-revise-and-publish-paper-from-the-grave/
    [10] http://scholarlyoa.com/2013/04/04/hindawis-profits-are-larger-than-elseviers/
    [11] http://www.hindawi.com/journals/bmri/2014/247218/

  • JATdS November 12, 2014 at 7:17 pm

    Actually, a search at Hindawi reveals alot more retractions. A simple search gives search results of 99 and 58 for retraction [1] and retracted [2], respectively, but not all are actual lins to retractions or retraction notices. My best estimates are that we are upwards of 60 retracted papers. This is something worth noticing and perhaps RW could focus on Hindawi in a separate future story* because 60+ retractions (estimate) is not a random or an insignificant number. Especially if we consider two aspects: a) all Hindawi journals are purportedly peer reviewed; b) there is an APC (article processing charge), which varies massively, for example 800 US$ for Journal of Lipids, but as much as 2000 US$ for BioMed Research International [3], home to at least one other retraction [4].
    [1] http://www.hindawi.com/search/all/retraction/
    [2] http://www.hindawi.com/search/all/retracted/
    [3] http://www.hindawi.com/journals/bmri/apc/
    [4] http://www.hindawi.com/journals/bmri/2014/965213/

    * I would even volunteer to compile the full list to assist RW to get a complete and full coverage of the actual scenario related to retractions. It would save RW alot of time, and avoid having to post 60 stories. One mega-centralized story on one publisher could be extremely valuable for RW, and for members of the scientific community that publish at Hindawi.

    • Nils November 13, 2014 at 3:40 am

      JATdS
      * I would even volunteer to compile the full list to assist RW to get a complete and full coverage of the actual scenario related to retractions. It would save RW alot of time, and avoid having to post 60 stories. One mega-centralized story on one publisher could be extremely valuable for RW, and for members of the scientific community that publish at Hindawi.

      Dear Jaime,
      This would be very useful indeed. I seem to recall that some Hindawi journals, among those having a very broad scope, have huge editorial boards with 100+ members. If I’m not mistaken, there are journals where the board members do not even send the manuscript to referees: instead several editors handle the paper, and a majority vote determines its fate. If you do such a survey, it would be interesting to keep track of how the number of retractions correlates with the type of peer review model used.

  • Rob Warner November 13, 2014 at 1:43 pm

    Question all sources

  • Chad Atkins (@chemchad) November 13, 2014 at 9:32 pm

    Nice sleuthing.

    As previously suggested in the comments, Hindawi has questionable practices as a publisher, regardless of how many journals they publish. For more information on “predatory publishers” : http://scholarlyoa.com/2014/01/02/list-of-predatory-publishers-2014/ The author of this list states: “Hindawi is not on my list of questionable publishers. I do receive complaints about Hindawi, however. They use spam a lot, most of their over 500 journals lack editors in chief, and it seems to be a publisher that focuses just on the authors’ needs and not so much the readers’.”

    I’d also caution against the generalization buried within the following sentence: “Scientific journals are too caught up in the business of publishing to do even cursory checks of their authors’ credentials.” That is certainly true of SOME journals — but on the whole, such journals constitute the minority of the scientific publishing world, and often have obvious signs (such as a publisher with a shady record). We can’t let the bad apples spoil the reputation of the scientific community.

  • Lee Hover, D. Med. Hum. November 13, 2014 at 10:14 pm

    And let us not forget those authors who publish the same article in two different journals.

  • Lucas Gallindo November 15, 2014 at 3:41 pm

    Yet the paper has two citations on Google Scholar…

  • JATdS November 15, 2014 at 4:15 pm

    Lucas, astute observation and good catch. I checked Google myself, and found the following papers had referenced the Peskin retracted paper (2 as you indicated):
    Reference No. 99 in Fezza et al (2014) MDPI’s Molecules:
    http://www.mdpi.com/1420-3049/19/11/17078
    Reference 75 in John A. Loudon’s Herbert Open Access Journals’ Cardiovascular Systems:
    http://www.hoajonline.com/cardiovascsyst/2052-4358/2/6

    These authors, journals, editors and publishers shoudl be alerted immediately about the existence of a retracted paper in the reference lists of these two papers, such that an erratum can be issued.

    Of note, both publishers are listed as “predatory” open access publishers by Jeffrey Beall:
    http://scholarlyoa.com/publishers/
    This could make the correction of the literature tricky, if not impossible. It would be nice to get Peskin on RW to comment.

  • Anthony Brenton December 19, 2014 at 10:34 am

    This is just a “thank you” note to Ian Garber. I read Professor Brian Peskin’s article which to my untutored mind was alarming…to say the least. I also noticed that he was an electrical engineer.

    My own background is in philosophy, but I was married for 24 years to a biochemist who had had post-doctoral fellowships in both cancer research and sexually transmitted diseases.

    I had learned enough for alarm bells to start ringing, especially as the possibility arose of my having prostate cancer…fortunately, discounted following a biopsy.

    The trouble is, I also read an article (I can’t find it at the moment) by some Japanese researchers I think, also stating that Omega-3 fish oil was a likely cause of aggressive prostate cancer.

    Although Peskin appears to be a charlatan, is it possible that he is right, and do you know of any research that either confirms any or all his claims, or refutes them entirely?

    Finally, there is a company called “Yes” marketing nutritional supplements that refers visitors to the PEO site. It seems to be associated in some way with Brian Peskin and states that it’s based in California, although I’ve been unable to find an address or contact details other than the means of sending an email.

    The whole setup is decidedly fishy.

  • Eric August 12, 2015 at 7:04 pm

    Ian. With all due respect, regardless of what you thought when you produced this article the American Heart Association is in fact seriously questioning any benefits from fish oil and also starting to evaluate the potential harm. Yet just as Brian Peskin’s literature suggested years ago The American heart Association is now seeing that high sensitivity C reactive protein levels are the best indicator for cardiac Disease. And their traditional methods of cholesterol analysis have been shown to mean squat. Maybe you should read his books.

  • Marco August 13, 2015 at 6:12 am

    Perhaps of relevance to this thread: Peskin has had two other papers retracted: http://www.scirp.org/journal/PaperInformation.aspx?PaperID=36133
    http://www.scirp.org/journal/PaperInformation.aspx?PaperID=37563
    The notices are a bit vague, so some sleuthing by the RW team may be of interest

    • Eric August 13, 2015 at 4:30 pm

      I agree Marco. Actually all the retractions are quite vague. As a matter of fact this original article that we are all responding to is extremely vague. I’d like somebody to give one example of something Brian Peskin has actually said in his articles that is just not true. Ian talks about pseudoscience and yet half the treatment approaches in modern medicine are just that. Why in the world would a particular question have to have research done over and over dozens of times? Why do clinical guidelines keep changing over and over and even reversing themselves? All Brian is doing is challenging those in the ivory towers and they just don’t like it. Of course I want to be clear that is just my opinion. I have not done the research to prove it.

  • Post a comment

    Threaded commenting powered by interconnect/it code.