A critical review exploring science communication of nutrition and dietetic research: a case‐based approach exploring methodologies.

Background: Nutrition is an area of apparent disagreement among the public and experts. It is also an area which has seen a rapid increase in the number of publications in the past 40 years. With the advent of online media and social media platforms, the volume of news has also increased. This review considered five types of nutrition research and how press releases linked to publications might be reported by the media. Methods: Examples were taken from nutrition‐related articles published in the areas of in vitro work, animal data, epidemiology, clinical trials and data modelling publications which had press releases deposited in online repositories (EurekAlert! and AlphaGalileo). A critical narrative of the source of the media message, estimates of its reach and any potential exaggeration or source of confusion were identified. Results: It was clear that research has been reported by funders, journals and researchers' institutions in ways that claim extended findings of the data beyond that reported in the manuscript. This included inferences of health benefits in humans from laboratory studies, splitting outcome data for the same exposure in epidemiological studies based on perceived public interest, using clinical trials to make media claims that would not be permitted in advertisements and claiming modelled data for cases were actual changes in numbers of cases. Conclusions: It is essential that funding bodies and institutions along with academic journals apply pressure to discourage exaggeration of research. This is necessary to maintain public trust in science and ultimately improve public health. Recommendations: Develop advice on how language linked to causality and certainty is used building on the initial steps made by the Academies of Medical Sciences to assess both press releases and subsequent reporting.Encourage journals to release press releases to external bodies such as the Science Media Centre to seek external comment. This perhaps is preferable to commentary in accompanying journal editorial as this could risk bias, as it is also likely to support journal readership.Consider how results could be extended, and avoid institutional press releases which contain health claims and look to make a health claim.Editors and reviewers should look to moderate how claims can potentially grow through types of research without the evidence being in place, including careful reporting of the nature of the methodologies and models used.Journals should seek to publish more complete data sets rather than apparently splitting outcomes for the same exposure or intervention.Encourage the development of 'lab‐to‐person' research and reviews to develop theories and evidence from mechanism to intervention data. This perhaps would result in fewer publications, but they are likely to provide a more complete view of the evidence.Professional, regulator and statutory bodies, including the Association for Nutrition register (United Kingdom Voluntary Register of Nutritionists) and Health & Care Professions Council (Registered Dietitians), along with learned bodies such as the Nutrition Society should adopt requirements for their members and journals to report only their work in a responsible manner (perhaps aligned to these recommendations). [ABSTRACT FROM AUTHOR]