Influential articles about nutritional research are often funded by industry and are four to eight times more likely to reach conclusions that find in favour of its products, says a study in PLoS Medicine (2007;4:e5).
Daniel Ludwig, an author and director of the Optimal Weight for Life programme at Children's Hospital, a Harvard medical institution in Boston, told the BMJ, “It's widely recognised that when a drug company pays for research, the results are likely to be favourable to the company. The question hasn't been widely examined in funding for nutritional research. We don't all take drugs, but we all eat.”
Industry support for nutritional studies is high, which could lead to bias in published research, he said.
The study says that nutritional research influences government and professional dietary recommendations, public health interventions, and regulation by the Food and Drug Administration of health claims by the manufacturers of food products in the United States. The findings on nutritional research are also widely publicised and may affect consumers' choices.
Dr Ludwig's group looked at 206 interventional and observational studies and at scientific reviews relating to milk, soft drinks, and juices, all of which were published between 1999 and 2003.
These drinks were chosen because they are widely consumed by children and adolescents, with possible long term health implications. “The health risks and benefits of these three beverages have been the subject of much recent controversy, and the beverage industry is large and highly profitable, arguably creating an environment in which scientific bias might occur,” say the authors.
Only 111 of the studies included the source of financial sponsorship. Of these, about one in five (22%) had industry funding, almost half (47%) did not, and a third had mixed funding.
None of the interventional studies supported by industry reached a conclusion unfavourable to the industry. “When the food company pays [for the research], the results are four to eight times more likely to be favourable to the company's product than studies with independent financing”, said Dr Ludwig, quoting findings in the paper.
Dr Ludwig said that his group had not looked at whether any of these studies had influenced public policy, but he said, “There is the potential to cause public health harm.”
During the study period, the proportion of researchers who declared sources of funding or conflicts of interest increased from about half to almost 80%, which Dr Ludwig attributed to increasing awareness of the matter and more stringent journal policies.
Increased government funding of nutritional research through a peer review process such as that at the National Institutes for Health was needed, he said.