Nutrition science in the media: you are what you readBMJ 2016; 353 doi: https://doi.org/10.1136/bmj.i1879 (Published 07 April 2016) Cite this as: BMJ 2016;353:i1879
- Navjoyt Ladher, clinical editor, The BMJ
Is a low carbohydrate diet more healthy than a low fat diet? Does eating red meat cause cancer? Will a glass of red wine a day keep the doctor away?
Whether it’s the latest purported health benefit of a so called superfood or a nutrient that has fallen out of favour because of a link to a harmful health outcome, nutritional research is all over the news. Media reporting influences people’s choices about what to eat and drink. But coverage is often incomplete, inaccurate, or sensationalised, leading to confusing, sometimes conflicting, messages around food and diet.1 2
“There seem to be more health stories in the media,” Richard Smith, former editor of The BMJ, told a recent meeting at the Royal Society of Medicine. “But more information has not meant more understanding because information is poorly presented and inaccurate.”
Going behind the headlines
A vast research literature has found associations between all manner of foods and health outcomes.3 Each new study may add incrementally to the evidence, but a news story about a single study might contradict previous research—missing the big picture.
Vivek Muthu of the Economist Intelligence Unit, also at the meeting, explained that the presentation of risk is a perennial problem, with relative risk often trumping absolute risk as the more dramatic looking statistic.
For example, many stories last year about the link between processed meat and cancer cited the relative risk for bowel cancer as 17% more for people who ate processed meat. But few stories mentioned baseline risk and absolute figures. For every 1000 people in the UK about 61 will develop bowel cancer at some point in their lives, compared with about 56 per 1000 people who eat the least processed meat and 66 per 1000 for people who eat the most.4
Failure to explain the context of health claims “is the biggest media error and is powerfully misleading,” said Tim Rycroft of the industry’s Food and Drink Federation.
Other problems include extrapolating from studies that are in vitro, in animals, or have small sample sizes and a relative lack of information on who funded the research.
The hype machine
“Everyone is incentivised to overhype,” said Muthu. Researchers are under pressure to publish. Academic institutions are competing for exposure and funding. Journals are chasing readership and citations. News outlets seek to maximise readers and viewers. And reporters lack the time and expertise to critically appraise original research that often lacks robustness.
“Weak science” is how Smith described the body of research, pointing out an endless stream of observational studies, many of which are underpinned by poor statistical evidence.3
He said that journals were partly to blame for the inflated standing of the nutritional evidence base and urged journal editors to take greater responsibility for health news reporting.
Higher quality nutrition research would help. More robust methods—moving away from self reported food intake questionnaires and observational studies towards large randomised trials—would make the evidence base more reliable.5 We know that food industry funding sets the research agenda and can bias study results.6 More public funding or an “arm’s length” grant model similar to the Wellcome Trust could reduce industry influence.
Press releases give the scientific community an essential opportunity to engage with the media, are key in determining what news outlets cover,7 and help journalists make sense of the science. But evidence suggests they can contribute to hype, with overstatements in journal8 and academic9 press releases leading to exaggeration in subsequent news. The science writer and researcher Ben Goldacre has called for academia to raise its standards.10 He suggests naming authors to “create professional reputational consequences for misrepresenting scientific findings,” publishing releases alongside research papers, and encouraging post-publication debate on press releases. Others, such as The BMJ’s columnist GP Margaret McCartney, think journals and institutions should issue press releases more judiciously.11
Lastly, better public understanding of scientific methods could enable people to judge the quality of stories for themselves. Websites such as NHS Choices’ Behind the Headlines and Health News Review already provide unbiased analysis of health news. Critical appraisal could be taught in health literacy initiatives in schools. These skills could help the public and professionals evaluate media messages and make better informed decisions about what to eat and drink.
Navjoyt Ladher attended “Behind the headlines: analysing confusing media messages and food policy” on 23 March 2016 at the Royal Society of Medicine, London.
Competing interests: I have read and understood the BMJ policy on declaration of interests and declare the following interests: None.
Provenance and peer review: Commissioned; not externally peer reviewed.