An independent study by US researchers has suggested that a significant proportion of research on the health impact of certain types of beverage contains sponsorship bias.
The study is published in the online journal PLoS Medicine.
Sponsorship bias is said to exist when the results of the research favour the financial interests of whoever pays for it. There is some evidence that this happens in biomedical research, but until this study, there was little evidence of this happening in food and drink research.
The investigation was conducted by a team of scientists mostly from the Department of Medicine at the Children’s Hospital Boston, Massachusetts. They looked at all the research literature from January 1999 to December 2003 on three types of non-alcoholic drinks, including soft drinks, juices and milk. They reviewed the papers that described research carried out on humans that had conclusions about impact on health or disease.
They classified the conclusions as favourable, unfavourable or neutral in terms of the health outcomes of the products. They classified the funding into all industry, no industry, or mixed.
Altogether they found and analysed 206 articles. Of these only 54 per cent declared funding.
They found that articles that were sponsored entirely by industry (as opposed to partly) were 4 to 8 times more likely to be biased in favour of the sponsors compared to articles that received no funding from industry. By industry they meant food and drinks companies.
These findings are similar to those found in drugs research sponsored by drug companies, the researchers said.
An optimistic, possibly reasonable reaction to this news could be not to throw the baby out with the bathwater – but consider who paid for the study, how this might affect the conclusions, and use a number of sources before making up your mind.
However, as the author of an accompanying article in the same journal, Martijn B. Katan, Royal Netherlands Academy of Sciences Professor of Nutrition at Vrije Universiteit, Amsterdam reminds us, when manufacturers plan their sponsorship they naturally pick a product for which there is already some favourable evidence. While this is understandable, it is also the start of a very slippery slope.
“When an industry is the major sponsor of research on its own product, unfavorable effects of that product are less likely to be investigated, ” says Katan. The slope becomes even slipperier when the sponsor steers product dosage so as to increase the likelihood of beneficial outcomes, or so that “adverse effects will not reach statistical significance,” or unwelcome results are omitted from the abstract or the press release. He suggests there are also times when publication is cancelled or delayed when the results are unfavourable to the sponsor.
Anyone interested in keeping themselves informed on the health properties of what they eat and drink or recommend to their patients and clients would do well to heed the old cliché “he who pays the piper calls the tune” and the tricky but thoughtful nugget “absence of evidence does not mean evidence of absence.”
“Relationship between Funding Source and Conclusion among Nutrition-Related Scientific Articles.”
Lesser LI, Ebbeling CB, Goozner M, Wypij D, Ludwig DS.
PLoS Medicine Vol. 4, No. 1, e5 doi:10.1371/journal.pmed.0040005
Written by: Catharine Paddock
Writer: Medical News Today