Full extent of bias into food industry research remains unknown
A new review into bias in industry-sponsored food research has confirmed researchers know little about the full influence of corporate sponsorship or conflict of interest on the results of nutrition studies.
A new review into bias in industry-sponsored food research has confirmed researchers know little about the full influence of corporate sponsorship or conflict of interest on the results of nutrition studies.
Published today in JAMA Internal Medicine, the University of Sydney study concluded there was insufficient evidence to assess the quantitative effect of industry sponsorship on the results and quality of nutrition research.
Researchers from the University's Charles Perkins Centre reviewed 775 reports in the medical literature, narrowing the scope to 12 relevant reports published between 2003 and 2014, to determine whether nutrition studies funded by the food industry were associated with outcomes favourable to the sponsor.
The review and meta-analysis was led by Professor Lisa Bero from the Charles Perkins Centre and Faculty of Pharmacy, and co-authors Dr Nicholas Chartres and Dr Alice Fabbri.
The results suggest – but cannot establish – that industry-sponsored studies may be more likely to have conclusions favourable to the industry than non-industry studies, but the difference was not statistically significant.
"The scrutiny of the funding practices of large transnational food companies has threatened the credibility of nutrition research and researchers. However, without empirical work examining the association of industry sponsorship with the results of nutrition research, researchers, policymakers and the public have no way of quantifying and understanding the extent of industry influence on the data," said Professor Lisa Bero, who heads the Charles Perkins Centre bias node.
"Similar to pharmaceutical or tobacco industry sponsored research, corporate interests could affect the outcomes by setting the research agenda, framing the questions or influencing publication."
We need more rigorous empirical investigation of the effects of sponsorship on the full research cycle and the differences in study results – from asking the questions, to conducting the studies, to publishing the full and accurate results.
Of the 12 reports that met the researchers' inclusion criteria, very few assessed the association of industry sponsorship with the actual numerical results or analyses of the nutrition studies. Only two of the reports assessed the association of food industry sponsorship and the statistical significance of research results, and neither found an association. Only one study found that studies sponsored by the food industry reported significantly smaller harmful effects related to soft drink consumption than those not sponsored by the food industry.
More of the studies (eight of the 12) examined the association of food industry sponsorship with authors' conclusions or interpretations of their studies. A meta-analysis of these eight studies showed that industry sponsored studies had a 31 percent increase in risk of having a conclusion that favoured the sponsor compared to non-industry sponsored studies.
"Conclusions do not always agree with results, but can be 'spun' to make readers' interpretations more favourable. This spin on conclusions has been identified as a tactic used in other industries and can influence how research is understood by the lay community," said Professor Bero.
"But it is the results - the research data and analyses - that really matter. From the standpoint of developing systematic reviews and evidence-based recommendations, the results are more important than the conclusions because only the results are considered.
"Our findings also suggest that there is insufficient evidence to assess the quantitative effect of industry sponsorship on the results of nutrition research and, thus, account for this bias in systematic reviews."
