Retraction Watch readers may recall the case of Don Poldermans, a prominent Dutch cardiology researcher who left a research position in late 2011 amid an investigation into his work. In a letter in the American Journal of Medicine titled “Scientific Fraud or a Rush to Judgement?” Poldermans — three of whose papers are subject to Expressions of Concern — tries to set the record straight, something he has tried to do before.
Poldermans is responding to an editorial by Vineet Chopra and Kim Eagle, “Perioperative Mischief: The Price of Academic Misconduct,” which Chopra and Eagle based on a November 2011 press release:
Let me discuss the press release. I was accused by the Erasmus Medical Center of scientific misconduct in an ongoing study, of which I was the principal investigator. The aim of the study was to assess the predictive value of the hormone N-terminal pro B-type natriuretic peptide (NT-proBNP) for cardiac complications in patients undergoing major vascular surgery. Several assertions were made, including: 1) written informed consent was not obtained from all study patients; 2) some patient data were collected in a negligent way; 3) some of the patient data were fabricated; and 4) fabricated data were used in 2 abstracts submitted to an international scientific conference.
Poldermans acknowledges that “written informed consent for the preoperative measurement of NT-proBNP was not obtained in all patients” in a trial called DECREASE-VI, and notes that he resigned for that violation (although many reports said he was fired). He also admits that “some results of additional cardiac testing were not found in the case report forms,” although he explains why the final analysis did not contain any missing data. And he responds to allegations about data fabrication by saying that information from “routine clinical evaluation and resting echocardiography…was entered into the same database column used for stress test results, but that
there was never any intent on the part of the investigators to represent these data as the results of protocol-driven noninvasive stress testing.
Finally, Poldermans writes that none of the data from DECREASE-VI study were ever published. He concludes:
One of the conclusions of the report is: “no evidence was found for any manipulation of the research results by the researchers in the sense of deliberate steering of results into a particular direction.” I hope this has clarified the press release based on which the unfounded accusations were made in the above-referenced review article. It would have been fair to approach me and just to ask me this before publication. In addition, the investigations are still ongoing, but given this publication, the future outcome will be irrelevant, as the authors have already accused me publicly. In my opinion, scientific integrity has 2 sides; also, the authors should give the accused an opportunity to defend himself and wait until the final report has been published.
Of course, Erasmus has now published that report, as Larry Husten describes here. As Chopra and Eagle note in response to Poldermans’ letter, the report
reiterated that “violation of academic integrity occurred due to the lack of source documents, inaccurate data collection, and the use of fictitious data.”4 It concluded, “The report of the investigative committee on academic integrity dated 8 November 2011 has done considerable harm to the reputation of the research group involved. This 2012 follow-up investigation has not been able to limit this harm.”
Their original paper, say the authors “was not meant to be an indictment of Poldermans.”
Rather, we wished to highlight the caustic nature of scientific misconduct, recognizing that anyone can fall victim to these perils without close attention to warning signs and unending discipline to protocol and procedures. What has befallen Poldermans can happen to any scientific researcher whenever absolute resolve to strict research methods is ignored or allowed to falter. Indeed, Poldermans has stated publicly that he is guilty of these oversights.5 It is our sincere hope that he will emerge from this in such a way that his scientific creativity, once again, shines. However, there is little denying the consequences of these events—both for him, and for us in the perioperative community. Uncertainties about DECREASE IV, DECREASE V-pilot study, and pilot studies for DECREASE VI are soon to be communicated to the respective journals so that further action may be taken.4
Polderman states, ” If I could do it differently, I would cut down the number of fellows,and work very closely with just one or two fellows”. Of course.
David Wright and Susan Titus reported that in a study of closed ORI misconduct cases 62% of mentors had not established procedural standards such as maintaining a notebook and 73% had not reviewed the raw data of their trainees”(1)
II believe it would be helpful if ORI routinely identified mentors who, in some way, contributed to the misconduct of a trainee. They should share responsibility if they co author a publication based on flawed data.
The home institution and the funding agency should be notified.that a mentor’s activity or passivity contributed to that misconduct. Universities are very likely to pay more attention to trainee to mentor ratios and quality of mentorship if they know that these indicators of training quality will be considered when their next training grant application is reviewed. Funding agencies themselves might also consider establishing acceptable trainee to mentor ratios.
The adoption of such policies should help reduce the number of what Ivan refers to as ,” research factories”.
Don Kornfeld
Wright,DE,Titus, SL, Cornelison JB,Mentoring and Research Misconduct,,An Analysis of Research Mentoring in in Closed ORI Cases, Sci Eng Ethics,2008,14:323-336.
This is sad. I agree that in this case the journal has jumped the gun by allowing an article accusing him of scientific fraud for publication before the institutional investigation was completed. In a blog? sure. In the pages of a medical journal is a different story in my opinion.
His explanations seem genuine – could he be playing it down and actually be guilty? Sure, but what he describes sounds like the working environment of pretty much every supervisor or PI I’ve ever known.
The first aspect, that he didn’t collect informed consent does raise a tricky area – for tests done as part of routine medical care there really is no ‘ethical cost’ from the research, as this intervention (a blood test) would happen anyway. People do conduct studies on samples such as these without collecting informed consent from the individuals – is these cases, an ethics committee will grant permission for the researchers to use them, acting on behalf of the individuals (which is well within their mandate of acting on behalf of society as a whole). Because the tests are done anyway, this is really just a question of use of their data, and is similar to how researchers are granted permission to mine databases of various kinds (cancer registries, etc) for additional analyses. I can really see how this kind of ‘oversight’ could happen, and wonder about the implications of him eventually resigning as a result – since as far as I see, there is no ethical cost to those individuals, which pretty much nullifies the reason for obtaining informed consent in the first place. The question would only be if the ethics board had approved his use of this data without obtaining consent (which isn’t entirely clear but seems like probably not). This is certainly a whole different affair to, for example, that anesthesiologist who put patients through differing protocols without consent.
Maybe other aspects of his work is sloppy – i.e. putting data for one test in the same spreadsheet column as an exercise stress test. I also have to note the irony of this post appearing the same day as the other article ‘Scientific Yellow Journalism’ where the author was going on about the rush to publish and the errors introduced into Poldermans’ abstracts as a result of just that. Again, I can see this happening to almost every PI I know.
I’m all for calling scientific fraud out and having the names of people doing it dragged through the mud. But this really seems like the expected end result of a scientific system which demands PIs create big labs, with multiple staff, publish prolifically, and have no time to oversee every detail. If fact, how could such a system ever end differently?
The last paragraph in Chopra and Eagle sums it very well: “Unlike a legal process, once scientific misconduct has been suspected, innocence cannot be presumed unless compelling evidence to the contrary exists. The absence of such evidence, findings of the committee, and Poldermans’ comments enshroud his position in further doubt. We must learn from Poldermans and resolve to move forward with clarity and diligence; only through this approach can further instances be avoided”