Retraction Watch

Tracking retractions as a window into the scientific process

University of Utah finds former faculty member guilty of misconduct because of “reckless disregard”

with 6 comments

utahSince last May, we’ve been reporting on a case at the University of Utah involving two retractions and two corrections. When the story first broke, the lab blamed a former worker for inappropriately removing data from the premises, and the university has been investigating. Last month, we reported that Ivana De Domenico, the junior faculty member who was first author on those papers, had left the university, and that senior author Jerry Kaplan had retired.

But the problems went beyond the four papers we reported on, as the Salt Lake Tribune reported on Friday. In a 9-page report provided to the Tribune, the panel overseeing the investigation details its examination of 11 papers. This is the first of six findings:

1. Paper 11 contained images that were intentionally manipulated to present false data. This was done by computer copying of information from one scanned image to another or in some instances by splicing of gel images to make one image. While the Panel agrees that this manipulation does appear to have been intentional, we cannot conclude that it was done by Dr. De Domenico as opposed to an unknown third person. Nevertheless, the error appears so obvious that it should have been caught by a responsible lead or senior author.

The panel did not think much of De Domenico’s explanations:

There are multiple instances of errors in 10 of the 11 papers examined. The only one in which no errors were found was the one in which Dr. DeDomenico was the “Senior Author,” meaning that other authors were responsible for the initial preparation of data, charts, and images. As first author on the other papers, she bore responsibility for checking all graphs and figures against the raw data and then maintaining an adequate record from which later researchers could replicate the experiment. Her explanations of these errors all relied on asserting that someone else did key parts of the work, but in almost all instances she was responsible for producing the actual images that were presented to the publisher. For two reasons, her explanations are inadequate:

a. if true, then her explanations show a lack of control over data for which she was responsible as first author or preparer of the figure for publication

b. it strains credulity that so many people could make similar errors without her catching at least some of them

This seems to us to be a very important point that bears repeating in such cases: If you want to take credit for a paper for which others have done some of the work, you also need to take the blame when things go wrong.

The panel summarizes as follows:

…our findings on the overall record are similar to what we found with regard to Paper 1. Although the Panel does not find that intentional irregularities can be attributed to Dr. DeDomenico, the number and severity of errors in the publications where she is first author show reckless disregard for accepted practices. Recklessness consists of proceeding with publication in the face of a very substantial risk of disseminating inaccurate information to the scientific community. The substantiality of risk stems from the number of publications affected and the severity of the errors.

The investigating committee did find intentional falsification or fabrication, although it

could not determine by a preponderance of the evidence that these items were the work of Dr. DeDomenico.

Still, the committee found there was misconduct

based on reckless disregard of accepted practice, a finding heavily influenced by the number of departures from established norms over an extended period. We recognize that the “accepted practices” in this particular laboratory and some of its collaborating laboratories may have countenanced the carelessness and obfuscation described by both prior cOlnmittees, but these irregularities exceeded both accepted practice in the broader research community in which Respondent was educated as well as basic norms of right and wrong.

(Contrast this with the findings in the Rui Curi case, where problems with images led two groups to find “failure to exercise rigor in the conduct and dissemination of results” but no fraud.)

Through her lawyer, De Domenico told the Tribune

that she didn’t purposely falsify or fabricate research and disputed the finding that she had committed accidental misconduct.

Still, the panel

regretfully concurs with the Investigation Committee’s recommendation that Dr. DeDomenico be terminated or allowed to resign from the University.

And this case is not over:

The Panel is concerned that the University take full appropriate action with regard to the institutional difficulties uncovered in this proceeding. The problems outlined by the Investigation Committee are not likely to be resolved with one termination and one retirement. Thus, we recommend further investigation of the laboratory procedures uncovered in this investigation.

The Tribune reports that there have been three corrections and two retractions, although the third correction does not seem to be indexed yet. The university, which redacted some names in the report, will now contact the editors of the journals that published the other six papers in question, according to the newspaper, and leave it to them and the authors to decide how to deal with those studies. The Office of Research Integrity has also been informed, because the work was federally funded.

Written by Ivan Oransky

August 5th, 2013 at 9:30 am

  • Don Kingsley Jr, MD. August 5, 2013 at 11:14 am

    A great service, much appreciated by”the rest of us” Perhaps this will decrease self citation.

  • Junk Science August 5, 2013 at 1:42 pm

    No surprise! I find it a bit peculiar though that the investigation could not link the image files to the computer of DeDomenico.

  • Dan Zabetakis August 5, 2013 at 5:12 pm

    The story here is not clear. De Domenico is said to be “junior faculty” but is clearly not the leader of an independent research group. The UU report seems to indicate broad concern about a wider net of scientific misconduct. But there are many words redacted from the pdf and it is not clear who is responsible for what questionable conduct.

    Also, I think the committee reads too much and too little into authorship. They are making assessments about the duties of a first-author and a last-author. Aren’t we agreed that all authors share equal responsibility in regard to a manuscript. Author order is more of a convention than a strict aspect of scientific ethics.

    • genetics August 6, 2013 at 6:38 am

      Equal responsibility, sure. But in case of fraud or misconduct, that does not mean that every co-author is equally culpable. First and senior author are responsible for knowing both the raw data and for the correct transfer of raw data into correct and non-misleading publication figures. Co-authors that added important data and ideas must decide on their own, if they trust first and senior authors, or if they want to see all the raw data themselves. Especially when from other institutions, co-authors wil rarely insist on seeing the raw data. So yes, they take responsibility, which manifests in having a retraction on their record if something provided by the other authors is fishy. But co-authors remain innocent until proven guilty, as long as it concerns data provided by others. If you want full culpability for every co-author, meaningful cross-institutional cooperation becomes highly impractical and we will quickly be back to single-author papers only.

  • iron-man August 8, 2013 at 6:20 am

    @Dan Zabetakis
    Then there’s the defense De Domenico AND Kaplan had chosen:
    ‘The data were lost when an employee, who was dismissed, discarded lab notebooks without permission.’

  • Post a comment

    Threaded commenting powered by interconnect/it code.