Retraction Watch

Tracking retractions as a window into the scientific process

Social sciences paper retracted for lack of ethical approval

with 13 comments

social science and medicineA University of Minnesota social scientist who studies health disparities has retracted a study that apparently lacked ethical approval.

Here’s the notice for “Deservingness to state health services for South – South migrants: A preliminary study of Costa Rican providers’ views,” a paper in Social Science & Medicine by Kate Goldade and a colleague:

This article has been retracted: please see Elsevier Policy on Article Withdrawal (http://www.elsevier.com/locate/withdrawalpolicy).

This article has been retracted at the request of the Editors-in-Chief.

The article is based on work that was undertaken without obtaining prior informed consent to conduct human subjects research from the Author’s institution. The scientific community takes a very strong view on any ethical infringement in the conduct of research and apologies are offered to readers of the journal that this was not brought to our attention prior to publication.

Goldade — who won the an award in 2009 from the Society for Medical Anthropology for “Best Published Paper” for another study — did the work while at the University of Arizona, where she earned her PhD. The notice suggests that this line from the original paper was incorrect:

Ethical approval for the research was obtained from the Institutional Review Board at the University of Arizona.

The editors of the journal tell us that while the notice says the retraction was their request, Goldade — whom we have not been able to reach for comment — asked for it first. We’ve asked the University of Arizona how the lack of IRB approval became clear, and will update with anything we learn.

Goldade’s co-author, Kola Okuyemie, who leads the University of Minnesota’s program in health disparities, said through the university that he only interpreted the data and developed the manuscript, so did not have anything to do with the work related to the IRB approval.

The work was funded by “a Fulbright-Hays Doctoral Dissertation Research Abroad grant and a PEO Dissertation Completion Fellowship.”

Update, 3/8/13: Goldade, who has been traveling in Kenya doing fieldwork, tells us that no other papers will be affected:

…IRB approval was obtained for some of the interviews included in the paper but not all.  IRB approval was not sought because I was between institutions (after PhD was conferred but before being hired as an assistant professor) which at the time of submission the manuscript did not occur to me.  Moving forward it did not feel right to include any interview data collected without IRB approval and considering all of my options with mentorship at the medical school I decided to retract the paper entirely.

Written by Ivan Oransky

February 25th, 2013 at 2:01 pm

Comments
  • stpnrazr February 25, 2013 at 2:57 pm

    Quite the blitz of papers from Goldade, Okuyemie and colleagues in the past few years. Since both seem to be based in MN, it would seem like there might have been more communication between them as they sent the manuscripts flying out the door.

    • Catherine Scott February 25, 2013 at 9:38 pm

      I am always astounded that professionals accept that their behaviour has to be policed, that they can’t be trusted to do the right thing without getting a note from someone or other. ‘Doing the right thing’ has turned into ‘getting permission to do it’.

      There was never any evidence that IRBs were necessary for social science research, as in no evidence of widespread misbehaviour and harm to research participants. This is unlike medical research, which can cause death or grievous harm to participants. Asking for evidence for the necessity for ethical clearance in social science research, however, usually results in references to medical experiments from the 1950s or the actions of the Nazi death camp doctors, who I am absolutely convinced would have submitted their research for ethics approval had IRBs existed then. No. Wait. Not.

      I have even served on an IRB and witnessed the utter conviction that something had to be wrong with every proposal and the routine sending back for revision of every single one, but usually for (extremely) minor problems with form completion that had nothing whatsoever to do with ethics.

      Most issues that do arise with research proposals in the social sciences come from graduate students and in my day, sonny, it was considered that their supervisors could be trusted to spot and correct the usually very minor issues.

      And don’t even get me started on the ‘human factor’ issues, like people getting onto the IRB so they could cause their rivals and their rivals’ grad students difficulties and delays. Yep. Saw that.

      • jimgthornton February 26, 2013 at 4:50 am

        Well said.

      • Real Science February 26, 2013 at 6:35 am

        It is widely acknowledged that there has been mission creep by IRBs. People on the ethics boards feel “powerful” and lose sight of the reason why IRBs exist. What completely puzzles me is that reality TV shows routinely put people in situations that no IRB would possibly approve of. So, in our society, it is OK to do things for entertainment purposes that would not be allowed for scientific purposes that might lead to useful knowledge. Something is messed up and needs to change.

  • Dave Langers February 25, 2013 at 5:29 pm

    Hmm, normally you get “informed consent” from your participating subjects, right? The university’s institutional review board gives approval. So what was missing: Were the authors right that they had approval, but did they not seek consent; or was the study unapproved?
    Which brings me to a question I’ve wondered about before: should research that is executed based on proper informed consent, but not approved by an IRB, be considered “unethical”? I suppose that institutes, journals, or countries may impose additional requirement regarding approval, but the Declaration of Helsinki doesn’t require it, right? In any case, a researcher will always first have to decide for him/herself whether approval is required in any particular study setup, I would say.

    • StrongDreams February 25, 2013 at 5:58 pm

      Here, they interviewed Costa Rican health care workers about their attitudes toward providing care for undocumented Nicaraguan migrant workers. So there are potential issues around confidentiality (of the workers and the patients). When studying the “third-world” there are always concerns about exploitation. Normally the IRB will want to see the survey instrument, make sure it is appropriately designed, blinded (if appropriate), that names and identifying info of the subjects is protected, etc.

      Since these were face-to-face interviews it seems unlikely that the participants/subjects did not know what was going on. Still, IRB review is required, and I fully support retracting studies that don’t get it, even if the study is perfectly good. It’s a lesson to others.

      And yes, doing a proper study without IRB approval is unethical even if the study protocol itself is not unethical, because (1) you promised the funder that you would get IRB approval as a condition of receiving the money, and (2) you promised the journal that you got approval as a condition of publishing.

      • Dave Langers February 26, 2013 at 5:58 pm

        I agree that that is the case if you made these promises to journals and funders, and like I said they can impose extra requirements. But suppose that in my make-belief world I have my own bag of money and there is some vague journal that doesn’t have that requirement, and suppose I obtain informed consent but have no access to an IRB, could I not do research for the good of mankind? Or, more realistically, there are fields where “informed consent” is not an issue (like physics), so suppose I’m on the borderline between fields; say, I want to study the functioning of a hygrometer that operates by means of a blond human hair, which I got from my wife’s haircomb with her informed consent, but I didn’t get prior approval from an IRB. I submit to a physics journal that doesn’t worry about the ethical issues surrounding the use of human tissue, and the research came about as my own hobby project, would that be unethical?

        • StrongDreams February 26, 2013 at 11:53 pm

          IRB approval assures that a study is ethical by reviewing it from a theoretically neutral point of view (not invested in the hypothesis, for example), but it doesn’t *make* the study ethical. So I expect it is possible to carry out a study in an ethical manner even it is not approved by an IRB–presuming of course that IRB approval is not required by jurisdictional law, the funding agency, the journal, etc.

  • StrongDreams February 25, 2013 at 5:50 pm

    Here is the abstract of the original paper, for reference:

    Health services for undocumented migrants highlight the complex politics of the “right to health”. South-South migrants, an emerging focus of migration scholarship, compose an estimated 40 percent of the world’s 200 million international migrants. In Costa Rica, internationally renowned for its public health achievements, undocumented Nicaraguan migrants number between 8 and 16 percent of the population. In spite of historical, linguistic, and ethnic congruencies between peoples of the sending and receiving countries at the ends of this migratory path, access to health services is limited for migrants experiencing illegality in the global economic South. Costa Rican health providers articulated concepts of deservingness to health services for undocumented Nicaraguan migrants. This article is based on a preliminary study with a purposive sample of 22 Costa Rican health services providers. Interviewed over two field research periods (June 2005-July 2006; July, 2008), providers addressed four types of health services for undocumented migrants. Overall their views on the deservingness of health services for undocumented migrants reflected a utilitarian approach. Specifically, their talk reflected: 1) the limits to state responsibility for ensuring the health of individuals not pertaining to the nation; and 2) a concern for the threats posed to the health of Costa Rican nationals. Costa Rican providers’ perceptions on health services for migrants offer partial insight for the development of future migrant health policies in receiving countries of the global economic South.

    • StrongDreams February 25, 2013 at 5:53 pm

      Note that there is no data in the paper. They interviewed 22 health care workers, and performed 40 hours of “focused observation” in two hospitals, to observe how migrants receive health care. There are no data tables, all results are subjective. There is no supplement giving the questionnaire or other details.

  • Ressci Integrity February 25, 2013 at 7:43 pm

    Here is a situation. I had mentioned this yesteday, I think. Social Science publications are of this type. In this case there are multiple authors. However, if they interview people (say students, health care workers, policy makers and may be others) and publish a paper based on their interviews, does this need ethical approval? Usually it does not need? Theoretically, this involves human subjects as well and a survey (or an interview). If you just interview and survey helath care providers, do you need an IRB approval?

    • StrongDreams February 25, 2013 at 10:04 pm

      From http://www.niaid.nih.gov/researchfunding/sci/human/Pages/hshandbook.aspx#A2

      HHS’ Protection of Human Subjects in 45 CFR Part 46 defines a human subject as a living person about whom an investigator obtains either 1) data through intervening or interacting with the person or 2) identifiable, private information…In general if you’re using coded private information, data, or specimens, NIH will consider your research to involve human subjects unless it meets both of the following conditions:
      *You are not collecting samples by interacting or intervening with living people.
      *None of the investigators or collaborators listed in the application can identify the subjects through coded private information or specimens (e.g., an investigator’s access to identity is prohibited by a written agreement).

      For research funded by the federal government, all human subjects research requires IRB approval unless it meets one of the specific exemptions, in which case you have to justify the exemption in writing as part of your grant application.

      Research involving personal interviews, where the subject could get in trouble based on the views expressed, is not exempt.
      http://www.hhs.gov/ohrp/policy/checklists/decisioncharts.html#c1

  • DrugMonkey March 10, 2013 at 10:44 am

    Pet owners and farmers also do all kinds of horrible stuff to their animals that would never fly with an IACUC. That’s no reason to argue that science shouldn’t have the oversight though. Nor is reality TV proof of IRB overreach.

  • Post a comment

    Threaded commenting powered by interconnect/it code.