Nutrition data follow sponsorship

Industry-supported nutritional research on milk, juice, and soft drinks is as much as seven times more likely to support its sponsors than research paid for by disinterested parties, according to a study published this week in the Public Library of Science.

While research has demonstrated sponsorship bias in pharmaceutical investigations, the relatively limited study by a team at Children's Hospital Boston marks the first systematic effort to measure such bias in nutritional research. The authors hope that the findings will help open a discussion of bias and funding in nutrition research, and whether the source of sponsorship is subtly affecting investigations or preventing negative results from being published.

"One of the key concerns is, if independent funding for nutrition research is inadequate, then industry funding is going to look especially attractive to researchers," possibly opening the door for more bias, senior author David Ludwig, director of the Optimal Weight for Life program at Children's Hospital, told The Scientist. "This is a well-documented phenomenon, and every researcher is going to think that they're not susceptible to the effect."

Ludwig and his team restricted the literature search to scientific reviews, observational studies, and interventional studies on the human health effects of milk, juice, and soft drinks conducted between the beginning of 1999 and the end of 2003.

Two separate investigators classified each of the resulting 206 articles as having a favorable or unfavorable conclusion about a beverage's health effects, while other members of the team documented each article's funding sources and determined whether results benefited sponsors. Only a small subset of 111 articles reported funding sources. The team omitted from their analysis sources of "mixed" industry and independent funding, including the U.S. Department of Agriculture, which contains agencies that promote industry.

According to the team's analysis, industry-funded studies were 4.17 times more likely to support a sponsor's commercial interests than studies paid for by disinterested parties. When the researchers controlled for beverage type, potential author conflicts of interest, and publication year, industry-funded studies became 7.61 times more likely to favor sponsors.

Ludwig and colleagues note that the influence of sponsors is likely subtle, and "could occur at many levels," Ludwig suggested. These influences may be apparent in "how questions are posed in the hypothesis, how a study is designed, what data are collected and are not collected, how data are analyzed, and how conclusions are derived," said Ludwig.

Researchers may also only seek industry sponsorship when they suspect favorable findings, said David Klurfeld, national program leader for human nutrition at the U.S. Department of Agriculture, who wasn't involved in the research. "You're not going to send Coca-Cola a study saying 'We hypothesize that Coca-Cola kills you,' but you might send them a study that says, 'We hypothesize that Coca-Cola is not bad for you,'" he said.

And industry is not the sole culprit here, argued Greg Miller, executive vice president of science and research for the National Dairy Council, who noted that sponsors in general -- including government agencies and activist groups -- also tend to fund research that is aligned with their interests. Furthermore, the PloS paper does not take the quality of research among the two types of studies into account, noted Miller, a factor that may influence the results.

Klurfeld also noted that leaving USDA-funded work out of the final analysis may have skewed the findings. "The research components of the department do not promote consumption of any specific food or commodity," while USDA grants to outside researchers share the same evaluation process used by the US National Institutes of Health, he said.

Chris Womack
mail@the-scientist.com

Links within this article:

LI Lesser, et al. "Relationship between Funding Source and Conclusion among Nutrition-Related Scientific Articles." PloS. 4: e5. January 9, 2007.
http://www.plos.org

JE Bekelman, et al. "Scope and impact of financial conflicts of interest in biomedical research: A systematic review." JAMA. 289: 454-465. January 22, 2003.
http://www.the-scientist.com/pubmed/12533125

K Fodor, "Trials of the Pharmaceutical Industry," The Scientist, November 22, 2004.
http://www.the-scientist.com/article/display/15076/

David Ludwig
http://www.childrenshospital.org/cfapps/research/data_admin/Site114/mainpageS114P0.html

KY Kreeger, "Studies call attention to ethics of industry support," The Scientist, March 31, 1997.
http://www.the-scientist.com/article/display/17475/

David Klurfeld
http://www.ars.usda.gov/pandp/people/people.htm?personid=36443

Greg Miller
http://www.nationaldairycouncil.org/