Earlier this month, the U.S. Office for Human Research Protections (OHRP) told a group of researchers led by a Harvard Medical School professor that they hadn’t been forthcoming enough about the risks elderly subjects faced in their trial.
As the Boston Globe reported:
The letter from the Department of Health and Human Services instructed the institutions to contact nursing home residents in Boston, St. Louis, and Baltimore who participated in the completed study and tell them that they should have been alerted to information about risks the scientists discovered during the research.
The scientists, led by Harvard Medical School gerontologist Dr. Douglas P. Kiel, were studying how to reduce often lethal hip fractures.
Federal regulators also ordered the research centers to develop corrective action plans to ensure that their scientists follow rules designed to protect patients and to promptly alert officials monitoring their work when problems arise.
The February 12 letter from OHRP followed a letter last June. From the newer one:
Specifically, individuals who participated in the study, which required them to wear an undergarment with padding on one hip, may have had an increased likelihood of falling to the side of the body that was meant to be protected.
We — along with Elizabeth Woeckner, who flagged the story for us — wondered what might become of the paper the authors published in 2007 in JAMA based on the results.
JAMA tells Retraction Watch there have been no requests to correct or retract the paper, and that the journal is waiting OHRP’s final determination. Kiel referred us to a Hebrew SeniorLife spokesperson, who said “it seems premature to make plans regarding the paper as published in JAMA.”
This issue seems more related to the difficulties of doing clinical research and the ethical problems that can accrue to a project after the fact.
I’d be leery of projecting this blog into a consideration of papers that have not been retracted. We don’t want to seem just science-scolds and calling for a retraction is a very slippery slope indeed….
I’m not that clear on the temporal ordering. As written, the article implies that the potential risks were known prior to conducting the clinical trial and that the participants were insufficiently forewarned as to the possible risks prior to giving informed consent. The article is vague enough that the possible risks were discovered as a result of the clinical trial and the participants were insufficiently debriefed.
The standards here seem very strict, unreasonably so. I read both memos, and the first memo presents a picture of a trial in which an unexpected development occurred, in that the intervention seemed to be related to a side effect of more falls on that side. The numbers are small (11 falls, 8 on the protected side; 40 falls, 28 on the protected side). The difference is statistically significant, but the numbers are small.
The February OHRP memo says the investigators were aware of the effect by October, 2004. That was the month the study was terminated, according to the JAMA article. As the article reports, the participants’ average age was 85 in 2004 and many had some significant degree of cognitive impairment (with consent by legal representative). So, 8 years after the fact, OHRP demands correction of an ex post failure to inform, by demanding notification of hundreds of people — almost all of whom will either be deceased or unable to understand the problem.
In all honesty, I’m not sure I understand the problem. It isn’t as though the total incidence of hip fractures increased in any group. What the &%&* does OHRP think it’s doing, other than perhaps subsidizing a bunch of nursing home lawyers?
If any increased risk was detected at the end of the study, then I also don’t see why not disclosing the (small) additional risk the participants were subjected to during the study should lead to a retraction of the results.
Any clinical study can have results showing that the intervention was unfavourable for the participants. Every clinical study can harm the participants. Understanding this and agreeing to take this risk is part of informed consent.
I do agree that every participant should be informed of the results at the end of a clinical trial, especially if the results are unfavourable for the intervention tested. In this specific case however, the ending of the trial definitely ends any additional risk, unlike some other medical interverntions where longer term negative effects might be possible. So I don’t really see any additional harm here.
Undisclosed financial ties, as discussed by marilynmann below, are more critical in this specific case I would think.
This study has been dogged with controversy. After the study came out, a hip protector manufacturer sued Dr. Kiel in Massachusetts state court for commercial disparagement.
In addition, a journalist dug up the information that some of the authors had ties to bisphosphonate manufacturers and said that should have been disclosed.
http://blogs.wsj.com/health/2008/02/29/hip-protector-maker-sues-harvard-doc-over-jama-study/
http://www.thecrimson.com/article.aspx?ref=522145
Marilyn
Financial ties not revealed in hip study
http://articles.latimes.com/2007/jul/30/health/he-hip30
As they say: Those who can — do. Those who can’t — criticize.
There is some truth to it. Let’s make sure it does not apply to this blog!
Contrary to what JAMA has told retraction watch, I formally requested of Editor, Dr Buchner, this past October (by email and Fedx) that the article be retracted. For some above, who seem not well acquainted with this project there were 3 phases: the pilot study (used to secure the main study) that revealed statistically significant falling to the single pad side, the main study published in JAMA which revealed the same lopsided falls, and a new study beginning in September 2004 (not published)for which informed consent did not inform re-enrollees and new participants about this severe risk.
The operating hypothesis of the single padded design was that participants would fall 50/50 right/left
but the single pad underwear caused them to fall to the single pocketed side thereby negating the operative hypothesis. The investigators supressed this evidence to get the study published. Since the operating
hypothesis of the study design was proven false (and supressed) the JAMA publication is false and should be retracted. OHRP is to be commended for their important work on this project. Some “scientists” like to make this story more complicated, but its really as simple as I have described.—Ed Goodwin, President,
HipSaver Inc.
Correct me if I’m wrong, but journals do not usually grant such requests by third parties, do they? I’ve seen investigators’ home institutions retract articles after a full investigation, but I’d be surprised if JAMA would consider a retraction request by a company that was not involved in the research and is or was engaged in a lawsuit with the authors.
A Journal can retract any study that they deem to be a falsification based upon
valid evidence from any source. In this, case HipSaver provided documents to OHRP.
OHRP gave these institutions two years to reply to OHRP’s findings and they came up with no credible defense. Hence the most recent determination letter from OHRP.
HipSaver’s request for a retraction included a full narative of the rationale for doing so. Although HipSaver is a commercial entity, it has functioned as a wistle blower or private attorney general. Of course if JAMA does not retract the article
it would be counter to their mission of providing valid medical reports.
OHRP found the evidence valid and so should JAMA. My post above was only to correct the absolute statement that JAMA has received NO requests. In the past they have
considered people questioning their journal as “nobodies” so maybe HipSaver is a nobody.