Study claims obesity research is distorted

By Caroline Scott-Thomas

- Last updated on GMT

Related tags Nutrition Scientific method

Study claims obesity research is distorted
Obesity research is skewed by bias, particularly when reporting the effects of nutritively-sweetened drink consumption, claims a study published in the International Journal of Obesity.

The authors claim that many researchers report findings that are skewed because of bias against the food industry and preconceptions of weight and nutrition topics. They focused on evidence of prejudice in the scientific reporting of soft drink consumption as a possible risk factor for obesity. They found similar results regarding breastfeeding as a possible protective factor for obesity.

Dr Allison and Dr Cope, from the University of Birmingham at Alabama (UAB) School of Public Health, said that their paper was a call for truthfulness in research reporting. They referred to a phenomenon that they called ‘white hat bias’ (WHB), which they explained is a reference to the do-good characters in Hollywood Western movies, who tended to wear white hats. They defined WHB as “bias leading to the distortion of information in the service of what may be perceived to be righteous ends”.

“Although many papers point out what seem to be biases resulting from industry funding, we have identified here, perhaps for the first time, clear evidence that WHBs can also exist in opposition to industry interests,”​ they wrote.

Both authors noted conflicts of interest in their study, having received “grants, honoraria, donations and consulting fees from numerous food, beverage, dietary supplement, pharmaceutical companies, litigators and other commercial, government and nonprofit entities with interests in obesity and nutrition, including interests in breastfeeding and NSBs ​[nutritively sweetened beverages].”

Dr Cope has also recently accepted a position with soy ingredients company Solae.

Dr Allison told that the research was not prompted by any of his contacts in the industry.

Study methodology

The authors reviewed 165 papers in English that cited a 2004 study by James et al,​ and 41 that cited a 2006 study by Ebbeling et al​, both of which focused on the effects of nutritively sweetened soft drink consumption on obesity. Neither study reported a statistically significant relationship between weight and soft drink consumption, they said.

Nevertheless, when Allison and Cope analyzed the studies for how researchers cited them, they found that 84.3 percent of those citing James et al​, and 66.7 percent of those citing Ebbeling et al“described results in a misleadingly positive manner to varying degrees.”

In addition, they highlighted anecdotal examples of the miscommunication of scientific results in press releases, and the inclusion of inappropriate information in scientific reviews, as potentially misleading. They also claimed that there is a publication bias toward those studies that show statistically significant associations between nutritively sweetened beverages and obesity.

“Sound evaluations critically depend on evidence being presented in non-misleading ways,” ​they wrote.

Dr Allison told this website: “I’ve seen many changes in the obesity field. Some of those changes are quite positive. The public is taking much more interest in obesity…What sometimes happens is that scientists, advocates, politicians and so on have their own policies…The take-home point is please be skeptical. That’s what science is about.”

UPDATE: This article has been updated from the originally published version to add Dr Allison's comments.

Seeking your comments is seeking your comments on the issues raised in this article.

Please send your comments of no more than 100 words to caroline.scott-thomas 'at' by Monday, December 7th, putting ‘Obesity research’ in the subject line, and including your name, location and affiliated company or organization.

We will publish a selection of the best responses, covering all angles of the debate, on Tuesday, December 8th.

Please note that comments will be taken as 'on the record', and the sender's name and company/organization will be published.

Source: International Journal of Obesity

(2009) doi:10.1038/ijo.2009.239

“White hat bias: examples of its presence in obesity research and a call for renewed commitment to faithfulness in research reporting”

Authors: MB Cope and DB Allison

Related topics R&D

Related news

Show more

Follow us


View more