%0 Journal Article %A Kathrin Lauber %A Darragh McGee %A Anna B Gilmore %T Commercial use of evidence in public health policy: a critical assessment of food industry submissions to global-level consultations on non-communicable disease prevention %D 2021 %R 10.1136/bmjgh-2021-006176 %J BMJ Global Health %P e006176 %V 6 %N 8 %X Background Ultra-processed food industry (UPFI) actors have consistently opposed statutory regulation in health policy debates, including at the WHO. They do so most commonly with claims that regulatory policies do not work, will have negative consequences or that alternatives such as self-regulation work well or better. Underlying this are often assertions that industry is aligned with principles of evidence-based policymaking. In this study, we interrogate if this holds true by exploring the extent and quality of the evidence UPFI respondents employed to support claims around regulatory policy, and how they did this.Methods First, we identified all submissions from organisations who overtly represent UPFI companies to consultations held by the WHO on non-communicable disease policy between 2016 and 2018. Second, we extracted all relevant factual claims made in these submissions and noted if any evidence was referenced in support. Third, we assessed the quality of evidence using independence from UPFI, nature, and publication route as indicators. Lastly, where peer-reviewed research was cited, we examined if the claims made could be justified by the source cited.Results Across 26 included consultation responses, factual claims around regulation were made in 18, although only 10 referenced any evidence at all. Of all 114 claims made, 39 pieces of identifiable evidence were cited in support of 56 claims. Of the 39 distinct pieces of evidence, two-thirds were industry-funded or industry-linked, with only 16 externally peer-reviewed. Over half of industry-funded or industry-linked academic articles failed to declare a conflict of interest (COI). Overall, of only six claims which drew on peer-reviewed and independent research, none appropriately represented the source.Discussion UPFI respondents made far-reaching claims which were rarely supported by high-quality, independent evidence. This indicates that there may be few, if any, benefits from consulting actors with such a clear COI.Data are available in a public, open access repository. All data analysed for this study are in the public domain. %U https://gh.bmj.com/content/bmjgh/6/8/e006176.full.pdf