Facebook Twitter LinkedIn

Home Back Issues No. 5: Research Ethics: Is misleading data fraud?

Nov 30

No. 5: Research Ethics: Is misleading data fraud?

Posted by: PIA in

Tagged in: Untagged 

Sign Up to receive free weekly articles like these

Research Ethics:

Is misleading data "fraud"?

Reader Question: As PIs, we understand of course that research fraud is intolerable. However, at times misleading data could be reported to a journal or sponsor merely because of a serious, though fully accidental, mistake in recordkeeping, faulty statistical calculations etc. Does such a mixup constitute "fraud"? What is the difference in the eyes of granting agencies and legal doctrine? When and how does "accident" cross over to PI "crime"?

Expert Comments:

The U.S. Public Health Service Policies on Research Misconduct, 42 CFR Part 93, state that research misconduct does not include honest error or differences of opinion (§ 93.103(d)). The misconduct must be a significant departure from accepted practices, and it must be committed intentionally, knowingly or recklessly (§ 93.104).

However, the respondent, i.e., the person accused of the misconduct, has the burden of proving "affirmative defenses" (§ 93.516) to the conduct alleged. In other words, the respondent has to prove that, even if the factual allegations are admitted or proven, there is some other justification or excuse for the action that would limit or eliminate his or her culpability. For example, he or she made an honest and reasonable error or mistake.

The level of intent needed for culpability under these regulations gives guidance as to whether accidents, mistakes, faulty calculations, or other “mixups” should constitute research misconduct. Intentional acts are those where the person has a conscious objective to engage in the conduct that caused the offensive result. Knowing acts are acts in which the alleged offender is aware of the nature of the conduct and the fact that it will almost certainly cause the offensive result. Reckless acts are those in which the person acts with a conscious disregard of a substantial and unjustifiable risk that the offense will result from the conduct. Reckless acts exhibit a gross disregard for the standard of conduct that a reasonable person would observe.

From the preceeding common interpretations, it could certainly be argued that a mere act of negligence or inadvertent mistake, without circumstances that would point to a significant departure from and disregard of accepted practices in the profession, should not constitute the level of intent needed to prove research misconduct under the PHS regulations..

Comments by Kendra Dimond, JD and Leslie Platt, JD. Ms Dimond is a Director at Daylight Forensic and Advisory LLC, Washington, DC. Mr. Platt is a Managing Director at the firm, and leader of its Healthcare Practice. Ms. Dimond is a former Counsel to the NIH Office of Scientific Integrity, and former Acting Director of NIH's Office of Legislative Policy and Analysis. Mr. Platt is a former Deputy General Counsel-Legal Counsel in HHS and former Executive Assistant to the Director and Chief of Operations, Office of the Director, NIH.

Enjoy this article? Sign Up to receive these free every week

Comments (10)
written by Sherlock, November 24, 2009
Why do these "mistakes" become "innocent" only upon start of the investigation? Did nobody vet the item beforehand?
written by Brendan G., November 24, 2009
Most fakery gets but a "slap on the wrist". Ergo: research fraud pays! Tragic but true.
written by Golden Flece, November 24, 2009
Raid your editorial budget and offer $100,000 for every new instance of research fraud which can be proven. I predict your postbox will overflow and your bank account will be emptied. Sadly, it will take money to pry loose the cheaters.
written by Anon, December 02, 2009
Upon finding a false entry in an income tax return, how often does the IRS accept that it was just "an innocent mistake"?
written by Laughable, December 02, 2009
Go out and take an anonymous poll of investigators as to how much of the literature they believe and you will soon realize how bad this problem really is.
written by Dr. Deek, December 02, 2009
It is fraud. Of this, you can be sure. Denis English, Ph.D.
written by Erich von Holder, December 09, 2009
as long as the studies support global warming, fudging the data is ok
written by Anonymous, February 16, 2010
Science is about correcting mistakes. If you are a scientist, you will make all possible efforts to correct the mistake. Admitting to a mistake can cause some minor embarrassment, but it does make your other results more trustworthy because it shows you're willing to double check.
Only insiders can tell
written by Unemployed, August 09, 2010
The only way to uncover fraud is from the inside. My boss did a gene chip experiment and, over my objections, removed two vital control samples because "they cost too much". Then, when the data turned out to be completely useless (due to lack of the aforementioned controls), she proceeded to fabricate data to fit her hypothesis. Attempt 1: when calculating the fold difference between group A and group B, take the average of group A and call that the difference. Attempt 2: log transform the data multiple times to amplify differences. Attempt 3: Divide the average of group A by the standard deviation of group A. She attempted to include this fabricated data in my manuscript. When I refused to allow her to do this, she harassed me to the point where I became physically ill and I had to quit. Honest, hardworking people are forced out of science while cheaters get promoted.
written by Fraud aware, August 09, 2010
There's the old saying that at a scientific presentation, the audience will accept that data, but may debate the conclusion reached. On the other hand the speaker is sure the conclusion is correct, but knows that he had fudged the data a little to make his point.

That aside, if erroneous data or interpretation and conclusions get published, an honest scientist must somehow make it known, and if the conclusions are of some consequence, the retraction must be as conspicuous as the original. So, perhaps another paper in which the purpose of the experiment is summarized, the error (made by the computer, of course) is explained, and the correct interpretation provided. Additional data confirming the correct interpretation would be helpful.

Of course, if the data were fabricated and the originator doesn't own up to it, it is up to someone more senior than the authors (the department chair, the Dean, the editor, ...) to point out the fradulent nature of the published article. Subsequent manuscripts by that author will be suspect and probably never published. Such is the just reward for intentional fraud by a non-repentant scientist. Unfortunately, sometimes the institution wants to be rid of the discovered fraudulent scientist and provides a high recommendation to another institution, andnever reveals the fraud. There is double shame.

Write comment
smaller | bigger

Write the displayed characters