Authors lied about ethics approval for study on obesity, depression

Obesity has retracted a study that suggested overweight people may be less depressed than their slimmer counterparts in cultures where fat isn’t stigmatized, after realizing the authors lied about having ethical approval to conduct the research.

The authors claimed their research protocol had been approved by Norwegian and Bangladeshi ethical committees, but, according to the retraction note, part of the study “was conducted without the required approval of the university ethics board.” The journal’s managing editor told us that there is no evidence that there was harm to the study subjects.

Here’s more from the retraction note for “In Bangladesh, overweight individuals have fewer symptoms of depression than nonoverweight individuals:”

The above article, published online on 1 October 2009 in Wiley Online Library (wileyonlinelibrary.com), and in volume 18, pp. 1143-1145, has been retracted by agreement between the authors, the journal Editors-in-Chief, Eric Ravussin and Donna Ryan, the Obesity Society and Wiley Periodicals, Inc. Although there is no implication that the data collection was carried out unethically, the retraction has been agreed to because the subsequent data analysis was conducted without the required approval of the university ethics board while the published paper incorrectly stated that Norwegian ethical approval had been obtained.

The purpose of the study was to look at mood in a place where fat isn’t stigmatized the way it is in Western cultures, where depressed people are more likely to be obese; it concluded that cultural context could contribute to the correlation between depression and body mass index. The authors surveyed and measured people in Bangladesh where, due to the high rate of illiteracy, they could only obtain “verbal consent,” according to the paper.

They also claim that all necessary ethical approvals were in place:

The protocol was approved by both Norwegian and Bangladeshi ethical committees for medical research.

Corresponding author Andres Magnusson, a psychiatrist at National University Hospital in Iceland, told us he was not involved in data collection, and wouldn’t be the one to ask about the ethics approval:

I came to the study after the data had been collected, the last author would be the one to ask.

We have reached out to the last author on the paper, Hussain Akhtar at the University of Oslo, but haven’t heard back.

Magnusson did say why he and his co-authors chose to obtain verbal consent from participants:

As explained in the paper, consent was obtained verbally after informing all the participants carefully, and they were informed that they could withdraw at any time…if only those who could read and sign would be included, this would not have been a proper epidemiological study, you would only include those who can read and write, probably richer, better educated, having a different diet, life style and another risk for diabetes. As stated in the paper, if we would have relied on written consent, one could not have obtained a non-biased sample of the population. It would be difficult to perform epidemiological studies in poor countries were literacy is low if one has to rely on written consent, not oral.

Magnusson also noted that none of the patients received any treatment as part of the research:

Note that this was not an intervention study, i.e. participants were not subjected to any treatment, only answered questions and gave blood samples.

The managing editor of Obesity, Allison Templet, told us how the journal learned about the issues with ethics approval:

An author of the paper notified our publisher, Wiley, and the Editors-in-Chief that the paper was inaccurate in reporting approval by 2 human ethics committees. He reported that the paper had received approval from an ethics board in Bangladesh but that they did not actually apply to the National Committee for Medical and Health Research Ethics (REC) of Norway for approval.

Meanwhile we received a separate letter from the University of Oslo (with which the authors were affiliated) indicating the letter from the author was sent at the university’s request. The university confirmed that the research in question had not been approved by the university’s ethics board before it was conducted using the population in Bangladesh.

Templet told us why approval from Norway was required, even though the authors weren’t doing their research there:

We corresponded with the author and the Norwegian university to confirm that the authors were indeed affiliated with the University of Oslo, that they required ethics board approval of research in this case, and that it was not obtained. The Dean and Faculty Director of the Faculty of Medicine told us that approval of the Norwegian REC was required by the Health Research Act of 2009. A letter from them stated: “Approval from Norway is required when a Norwegian institution is responsible for the research in a developing country or financed the research. Before the Health Research Act of 2009, this was based on a regulation made by the Norwegian Ministry of Education and Research, implemented around 1990.”

 Based on these facts, we retracted the paper.  It should be noted that the retraction was based on failure of authors affiliated with a Norwegian university to obtain  ethics board approval from their institution prior to their participation in human research, even though the research was conducted on a population in  Bangladesh. There was no evidence that the population in Bangladesh was subject to unethical treatments or assessments.

The article has not been cited, according to Thomson Scientific’s Web of Knowledge.

We also reached out to the University of Oslo for more information and will update with anything else we learn.

Update 12/11/15 9:38 a.m. eastern: Helpful reader “Morty” tipped us off to the fact that Hussein has issued two other corrections, also over ethics approvals.

Here’s one for a 2012 paper published in the Journal of Nutrition and Metabolism:

With regard to the paper titled “Effect of Intervention in Subjects with High Risk of Diabetes Mellitus in Pakistan” by Hydrie et al., it was stated that the protocol was approved by the Norwegian Research Council. But the statement was made by mistake.

All the subjects were from Pakistan and the project was approved by the appropriate Ethics Committee of Baqai Medical University (Baqai Institute of Diabetology and Endocrinology). All participants provided informed consent prior to the collection of any data and the project was conducted in accordance with the Declaration of Helsinki.

And here’s the correction to “Increasing Prevalence of Type 2 Diabetes in a Rural Bangladeshi Population: A Population Based Study for 10 Years,” published by the Diabetes & Metabolism Journal:

With regard to the paper by Bhowmik et al. (Diabetes Metab J 2013;37:46-53), the statement that the protocol was approved by the Norwegian Ethical Committee is incorrect. The authors deeply apologize for the imprecise statement. All the subjects were from Bangladesh and ethical clearance was approved only by the Ethical Committee from Bangladesh. All necessary information was collected from the participants following participant consent. The research was conducted following Declaration of Helsinki.

Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, and sign up on our homepage for an email every time there’s a new post. Click here to review our Comments Policy.