What Happens To Fake Studies?
Over the past 12 years, anesthesiologist Scott Reuben revolutionized the way physicians provide pain relief to patients undergoing orthopedic surgery for everything from torn ligaments to worn-out hips. Now, the profession is in shambles after an investigation revealed that at least 21 of Reuben's papers were pure fiction, and that the pain drugs he touted in them may have slowed postoperative healing.
A doc on various Pharma payrolls authors studies promoting Neurontin, Celebrex et al, and they turn out to be fakes. There are plenty of people who will/want to make the obvious points, so I'll restrict my comments to the less obvious.
I.
Reuben managed to trick some othopaedic surgeons to be coauthors on his paper. One such orthopod, Evan Ekman, was reviewing one of Reuben's studies.
Ekman agreed to review a Reuben manuscript on surgery on the anterior cruciate ligament (ACL) in the knee. But when [Ekman] asked the anesthesiologist [Reuben] for the name of the orthopedic surgeon on the study, Reuben ceased communication with him.Undoubtedly, everyone is focusing on the identity of the unknown orthopaedic surgeon as a red flag. But here's a question that will become obvious once I ask it: how did Ekman know the identity of the author of the paper he was reviewing? This is "anonymous peer review," right?
Next: even if he was alerted by 9 pixies that it was Reuben who wrote the paper, why would he think he was further entitled to know the rest of the authors' identities?
I'm not saying he should/should not know; I'm saying that this had become so commonplace to him that he doesn't even think twice about asking; or telling Scientific American that he asked. He actually forgot to pretend peer review is anonymous.
You may also ask why Ekman was chosen as a peer reviewer in the first place, since Ekman had previously coauthored other papers with Reuben. Oh, yeah: when you submit a paper to a journal, you choose people you'd like to review it. This way there is no chance whatsoever of institutional bias.
II.
Then, last year, Ekman was invited by Pfizer to give a talk. While there, he was handed a version of the very manuscript Reuben had asked him to review, which had subsequently been published in Anesthesia & Analgesia. To his surprise, and horror, he was listed as a co-author: Reuben had forged his signature on the submission form, Ekman says.If a tree falls in the forest, and no one hears it, does that mean everyone is deaf?
Ekman is an academic doctor who also speaks for Pfizer-- in other words, he's a "thought leader." Does he not read the journals? Do none of his friends read the journals, to send him an email, "hey, saw your study..." If he's not reading them, is anyone else?
What is the impact of a fake study that no one reads? Think about it for a moment, here's my answer: volume. "There is some evidence that..." It doesn't matter if the study is wrong, or silly, or fake; just as long as it exists, to be able to get away with that sentence.
III.
You think I'm kidding.
The faked article that Ekman reviewed and then "coauthored," purported to discover that Celebrex helped reduce pain, and consequently promoted more vigorous participation in rehab.
It also found that patients used a lot less oxycodone:
Casual readers of this blog will perhaps be thinking, "hmm, 20mg less oxycodone, does that seem possible?" But astute readers of this blog will ask the larger question: where the hell did you find the graph? It's not still on PubMed, is it?
Oh, yes. It's also free direct from the journal. Here's a follow-up study, also fake, also free.
"But surely the article is labeled as fake, or something?" Something being like a big red watermark that says, "RETRACTED"? No. The only hint is at the bottom, under "This article has been cited by other articles:" and one of the 12 citing articles is the retraction notice by the editor.
It's hard to explain why the article isn't simply deleted; or, better, loudly labeled as a fake so that we can learn even from fakes.
But it is very easy, very very easy, to explain what will happen to a large number of doctors who never read journals, let alone Scientific American, who may perhaps be motivated just to see how many articles on Celebrex there are in the world-- "wow, that's a lot"-- or read a review article that cites these articles-- reviews that will never find a forum for retraction or explanation.
It's even easier-- sit down for this one-- to explain what will happen to new research on Celebrex-- "does this thing work, or not?"-- now that Pfizer has absolutely no incentive to pay for it.
Good luck.
Digg | 
del.icio.us | 
StumbleUpon | 
Reddit | 
Yahoo! | 
My favourites
March 21, 2009 11:42 AM | Posted by : | Reply
From context, I interpreted "asked to review" as something other than anonymous peer review by a publisher prior to publication.
Reuben and Ekman had collaborated on a prior publication. So I suspected that Reuben was asking for Ekman's input on the paper, in exchange for making him a co-author.
Score: 1 (1 votes cast)
March 21, 2009 1:57 PM | Posted by : | Reply
i agree with dr. henway: when you go to pub a paper, you want a team of authors to cover all bases. if it is about pain in acl reconstruction, you get an orthopod to join the authorship group. not too difficult to achieve legitimately: the orthopod might simply need to contribute some specialty-focused talk to th eintro and discussion, regarding the rason d'etre for the study, and implications, and the orthopod now has contributed substantially.
wait - i forgot latin was banned on Last Psy: for the latin-impaired: ...regarding the whole main reason for even conducting the study, and imlpications, and the orthopod now has contributed substantially.
Score: 0 (2 votes cast)
March 21, 2009 1:59 PM | Posted by : | Reply
LP: this is too funny: if your fake study has a retraction published, it improves your citation rating, indicating even better impact! This set of studies will have many comments published in peer-reviewed journals for a long time.
Score: 5 (5 votes cast)
March 21, 2009 2:15 PM | Posted by : | Reply
On a side note: I took Vioxx for pain from torn ACL, and got a bunch of samples. It worked really well. I also used them for post-op pain (along with oxycontin) after reconstruction of ACL. Man-o-man that vioxx worked good. And so does the oxycontin. I would never suspect a study supporting Vioxx for post-op pain as being fake. Especially the way they faked that jaggedy line across the assessment timepoints.
Wait: that jaggedy line is fairly smooth, and the confidence intervals are quite stable. But oxycontin data points are not like pain assessment points - oxycontin comes in 10, 20, 40, 80, and 160 mg tabs. Post-op pts are probably only gonna take one or two dosages the entire recovrey timeframe - so people should really be dropping step-like in dosage, which should make it less likely to have such a consistent std deviation across recovery.
Basically, if the std deviations are calc the normal (no pun intented) way, then 2/3 of the pts should have data at or within those hash marks. the hash mark spread around the vioxx group consistently spans about 7 mg, as dose evenly drops from a 20mg tab-plus-10mg tab to a 20-mg-tab (or two 10mg tabs).
Does that seem odd? I will have to keep my eye out for other post-op pain studies to see how avg and how std deviation goes across a discreet recovery period.
Score: 0 (0 votes cast)
March 21, 2009 2:30 PM | Posted by : | Reply
It's pretty obvious to me that medicine, like everything else involving money, is a gigantic scam. It's unbelievable until you do the research and find out about the wide eyes open deception on behalf of pharma companies and their shill docs, and politicians who don't care because they're cashing in too.
A doctor said I had "bipolar spectrum", and I absolutely needed medicine. She vehemently refused the possibility that what I experienced may be normal, or could be helped by therapy. I was buying into it, until I realized the following:
1) I function pretty well, albeit not perfectly: where is evidence of absolute necessity of brain-destroying antipsychotics then? I leave the house. I have a job. I can control the majority of my symptoms with lifestyle changes, and the rest I can't control I can ride out. So, clearly, I need to be on meds stat, right.
2) This clicked into place a few weeks ago: BIPOLAR ISNT REAL.
The concept of the "bipolar spectrum" was made up by pharma shill doctors to sell sedative medicine to reasonably normal people, with little to no benefit to the patient. Taking medicine would only impair my functioning, not enhance it.
I'm not saying manic depression doesn't exist... I'm saying bipolar is BS, it went from a valid disease featuring major depression alternated with psychosis, to a mockery.
Every couple of years fad diseases pop up, and these fad diseases are associated with new lucrative drugs to treat them. The drug precedes the disease now. This is because medicine has become a scam and a mockery.
"1 in 250 americans struggle with a disorder called blurry eye syndrome... symptoms include heavy eyes and deficits in attention after 12 hours wakefulness... this disorder has been tied to premature death, accidents, depression, diabetes, cancer, insomnia, hypersomnia... now there is hope, with RESTOREX... some misinformation about BES state it is a simple case of sleep deprivation; while sleep loss can exacerbate BES, the symptoms for BES sufferers are experienced subjectively as more extreme and life-impairing, to the point where they cannot force themselves to stay awake much longer after 12 hours"
SRSLY, it's such a simple formula
1) Make up a disease and popularize it to hell by talking about it in every media outlet, much like an up and coming starlet (i.e. take relatively common unpleasant symptoms and present them as pathological and urgent)
2) Make up a drug with a catchy title which claims to treat aforementioned made up disease
3) Have a drug education section on consumer website/pamphlet which indoctrinates the future patient away from reason (i.e. convince the patient that these normal symptoms are not normal, use wording that encourages the individual to exaggerate their distress/symptoms ... )
4) Use the word hope a lot, always say there is hope, as this gets the patient used to thinking that they are hopeless without the drug
5) Continue this game until the drug goes generic or it is proven to cause horrible diseases which can no longer be covered up by fake studies
6) Start back at point 1
Score: 2 (2 votes cast)
March 21, 2009 9:44 PM | Posted by : | Reply
"raison d'être" is french, not latin
Score: 5 (5 votes cast)
March 22, 2009 6:59 PM | Posted by : | Reply
What is even more troubling about that fake graph is that it does not even indicate that Celebrex reduces the need for opioid medication by 20 mgs. It simply suggests that patients who were assigned to Celebrex group required less oxycodone from Day 1 (e.g. they had less pain). Given the severity of post op pain within the first 24 hours (ask anyone who had any type of operation that involved cutting skin and muscle, no celebrex will touch that initial , you would expect this difference to be smaller initially
Moreover, there was no difference between the rates of reduction in need for opioid medication in these groups. Obviously, post op pain gradually decreases over time, which explains why patients in the control group required less oxycodone on day 14 compared to day 1. The rate of this reduction (can be calculated from the slope of the graph) seems to be a slightly steeper in the control group than the
Score: 0 (0 votes cast)
March 22, 2009 7:10 PM | Posted by : | Reply
CORRECTED COMMENT:
What is even more troubling about that fake graph is that it does not even indicate that Celebrex reduces the need for opioid medication by 20 mgs. It simply suggests that patients who were assigned to Celebrex group required less oxycodone from Day 1 (e.g. they had less pain). Given the severity of post op pain within the first 24 hours you would expect this difference to be smaller initially. Moreover, there was no difference between the rates of reduction in need for opioid medication in these groups. Obviously, post op pain gradually decreases over time, which explains why patients in the control group required less oxycodone on day 14 compared to day 1. The rate of this reduction (can be calculated from the slope of the graph) seems to be a slightly steeper in the control group than the celebrex group, which also shows that celebrex may have slowed down healing.
On a different note: What makes us think that an ortho surgeon is qualified to review a research article of this kind? I am sure he is fully qualified to watch the video of an operation to review the performance of another surgeon, but reviewing an original research article requires a firm background in statistics, which most MDs utterly lack. It is so funny, how our society requires a license for prescribing a medication to a patient, but allows anyone to analyze data that can then be used to prescribe a particular medication to millions of people.
Score: 0 (0 votes cast)
March 23, 2009 1:18 AM | Posted by : | Reply
Re youngturks
1) Pain is actually greatest a little bit after post op day 1, and the reason this is so is that there is often residual anesthetic from surgery so that LOC and sensation of pain is less acute
2) I assume people who review studies do have both the educational background in necessary to interpret study data, as well as relevant training to understand the content of the studies (e.g. surgeon has proper training to understand stuff like post op pain is greatest a little after surgery rather than immediately after).
In order to correctly understand a research study you need both, and like I said, I assume any physician involved in research has a "firm background in statistics" as well as being a top notch doc.
Score: 0 (0 votes cast)
March 23, 2009 3:47 AM | Posted, in reply to , by : | Reply
1) That is not necessarily true (Mc Kean et al. 2006)
2) Don't.
--
ps: Below is a list of papers that show what happens to fake studies: They get cited!
1: Citation of retracted articles: a call for vigilance. Drury NE, Karamanou DM. Ann Thorac Surg. 2009 Feb;87(2):670.
2: Research misconduct, retraction, and cleansing the medical literature: lessons from the Poehlman case. Sox HC, Rennie D. Ann Intern Med. 2006 Apr 18;144(8):609-13. Epub 2006 Mar 6.
3: Effects of article retraction on citation and practice in medicine. Budd JM, Sievert M, Schultz TR, Scoville C. Bull Med Libr Assoc. 1999 Oct;87(4):437-43.
4: The continued use of retracted, invalid scientific literature. Pfeifer MP, Snodgrass GL. JAMA. 1990 Mar 9;263(10):1420-3.
Score: 1 (1 votes cast)
March 23, 2009 5:13 PM | Posted by : | Reply
I'll be curious to see if Judicial recourse.
Not sure what exactly he could be convicted with, but something about this is so wrong, yet I doubt he'll see the inside of a jail cell.
Maybe a Fine, but no one even seems to be sure how much he was paid.
Sure his license is revoked, but he can still run around calling himself a doctor. He'd be a good candidate for a school superintendent or such.
Score: 0 (0 votes cast)
March 23, 2009 5:58 PM | Posted by : | Reply
Perhaps a single public database of retracted studies could be created (by whom? Gov't? which one?) that would be a one stop shopping place to check before you cite anything.
Score: 0 (0 votes cast)
March 23, 2009 7:56 PM | Posted by : | Reply
@information addict: Well, pubmed is a public government database. Seems like a good place to start...
Score: 0 (0 votes cast)
March 24, 2009 7:31 AM | Posted by : | Reply
While a majority of peer review is anonymous, there is nothing all that rare about peer review that is not anonymous, and there are plenty of high level advocates of open peer review.
Score: 0 (0 votes cast)
March 24, 2009 9:09 AM | Posted by : | Reply
Anonymous review is an oxymoron: even if you blind the reviewer to the applicant's identity (in psychiatry) there are lots of people -usually the most "successful" ones- who have very specific interests that it would still be relatively easy to identify them. What we need is something like a revolution (I'm only partially joking) that is going to change the rules of the game in a much more radical way.
Score: 0 (0 votes cast)
February 2, 2010 1:15 AM | Posted by : | Reply
Similar to the Ricaurte studies on MDMA and "holes in the brain"/parkinsons. The articles are still posted on Medline even though they've been retracted (he used meth and not MDMA). He was never held accountable, nor were his claims that he was supplied the wrong drug shown to be true. The studies were (not surprisingly) government funded and the US government pushed these studies for propaganda purposes ("this is your brain on ecstasy" cards) but never apologized/made it clear that the study was bogus.
Not trying to justify the use of MDMA, but it is interesting how this belief about "holes in the brain" still persists today. If this is true for the "war on drugs" I can't imagine what other BS government funded scientists fabricate to push an agenda. Ricaurte is at an extremely accredited research university (JHU), which makes all of this even more depressing.
Score: 1 (1 votes cast)
October 13, 2010 4:47 AM | Posted by : | Reply
Find the Best Psychologist in mumbai. Connect with the top Psychiatrist, Counsellor, Mental disease,Mental Health Specialist. Get comprehensive information about the best doctors in mumbai .Make an informed choice, get the right healthcare solution in mumbai
Score: 0 (0 votes cast)
Comments