JAMA’s first-ever Expression of Concern appears for hip protector study

JAMA has issued its first-ever Expression of Concern over a 2007 study of hip protectors in the elderly that came under scrutiny from Federal regulators.

As the Boston Globe was first to report yesterday, the journal’s editor and executive deputy editor wrote in a notice published online:

After review of the reports from [the Office of Human Research Protection] dated June 23, 2011,2 February 17, 2012,3 and July 5, 2012,4 and evaluation of additional information requested by JAMA and provided to us by the participating institutions (Hebrew SeniorLife [an affiliate of Harvard Medical School], Washington University School of Medicine, and the University of Maryland School of Medicine), we concur with the OHRP determination that the failure to notify research participants about potential risks they may have experienced by participating in the study represents serious concerns regarding their protection as research participants in this study. Therefore, we are issuing this Expression of Concern regarding the ethical conduct of this study. When the study was first published in 2007, the authors acknowledged1 and an editorial in JAMA5 anticipated that having only 1 hip protected could have altered the gait of participants and the propensity to fall to the protected side. However, at that time, the authors were not forthcoming with providing information JAMA requested that would have been informative in understanding this risk.2 (p9)

In February, JAMA told us that there hadn’t been any requests to correct or retract the paper, and that they were awaiting OHRP’s final determination. In yesterday’s letter, the editors write:

Based on information reported to JAMA by Hebrew SeniorLife, Washington University School of Medicine, and the University of Maryland School of Medicine in response to our requests, there was no evidence provided that raised concerns about the scientific integrity of the data and the veracity of the study conclusions; therefore, we are taking no further action at this time. If additional information should become available about the ethical conduct of this study, about harms to patients that occurred by virtue of failure of notification of possible risk, or about the scientific integrity of the research, we will evaluate that information carefully and determine whether additional notifications will be necessary.

Along with yesterday’s Expression of Concern, the journal published the text of letters that the OHRP required the institutions to send study participants.

Hebrew SeniorLife officials, according to the Globe:

said in a statement Thursday that after federal investigators alerted the organization to problems more than a year ago, it immediately conducted its own probe and made sweeping changes to the research program.

“We know that when an issue like this one arises, it is rarely the responsibility of one person, and so we used our internal investigation of the study as an opportunity to examine and improve all of [our] systems for human subject protection,” Hebrew SeniorLife chief executive Len Fishman said.

Expressions of Concern are, as Retraction Watch readers probably know, far more rare than retractions.

The study by Harvard and Hebrew SeniorLife’s Douglas Kiel and colleagues has been cited 62 times, according to Thomson Scientific’s Web of Knowledge.

Hat tip: Mike Delaney

9 thoughts on “JAMA’s first-ever Expression of Concern appears for hip protector study”

  1. As the “whistleblower” in this case, I provided OHRP with documents resulting in their determination of non-compliance. Since OHRP’s function concerns only harm to subjects I presented only documents related
    to safety. I have written to JAMA editors informing them that I have assisted OHRP and have many other
    documents relating to falsification. They have not responded to me. So when they say that they will evaluate any “additional” information what they mean is that they will ignore any additional information.

    1. Good sense of humor, Ed. What they mean by “we will evaluate” is “we will ignore”. Now, I’m wondering if perhaps they mean “If additional information should become available via spontaneous, voluntary confessions by the authors of the study and their institutions) about etcetera, etcetera, we will evaluate.” Mere evidentiary documents about wrongdoing from an outsider are insufficient to spur the editors to action, like maybe asking the authors to explain what the documents are all about. This is “don’t ask, don’t tell” for scientists.

      If you can say, how did you become the whistleblower in this case? How did you obtain documentation of falsification? What are these documents?

      1. The documents were obtained in court ordered discovery in HipSaver v kiel.
        They outline conflict of interest, supression of evidence, harm to subjects,
        and ethics violations. Hebrew Rehab is in possession of this information so JAMA could get it from them. That would be the additional information they need, but apparently don’t want.

  2. In looking at the links posted, I noticed that Ed Goodwin is the President of HipSaver (comments to the Wall Street Journal Article). I find the unstated conflict of interest troubling. It also seems that Ed Goodwin might have supplied the lawyers.

    I have followed this blog for a while and have become aware that I have been too trusting of work published by my peers. Further, I thought Ed Goodwin made very sensible contributions on other topics. Now, I wonder if I have been too trusting of “whistelblowers”. It appears to be a good thing I do not study human behaviour.

    1. Rob, Human behavior is simple in this case. We have research code of federal regulations and you have motives. In this case publish or perish, grantsmanship, and fraud that are not consistent with the CFRs.
      Its really that simple!

  3. Well spotted vhedwig and ROB. This might be a case of business defending itself against research. In a case like this I give the authors of the paper the benefit of the doubt and am highly suspicious of the company behind the attacks on the scientists.

  4. Indeed, the unstated potential conflict of interest is troubling. Also, it seems possible that nobody has valid scientific data upon which to base a conclusion. If the total number of participants in HipSaver’s own study is truly 38, as explained in the scientificmisconduct blog (linked above), then the result proves nothing. But if the Kiel study tested only one kind of device, then its conclusion is valid only for that kind of device. A random sample of various kinds of hip protectors would be needed if the researcher wants to make conclusions about hip protectors in general.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.