The association between exaggeration in health-related science news and academic press releases: a replication study.

Exaggerations in health news were previously found to strongly associate with similar exaggerations in press releases. Moreover, such press release exaggerations did not appear to attract more news. Here we tested the replicability of these findings in a new cohort of news and press releases based on research in UK universities in 2014 and 2015. Press releases and news were compared to their associated peer-reviewed articles to define exaggeration in advice, causal claims and human inference from non-human studies. We found that the association between news and press releases did not replicate for advice exaggeration, while this association did replicate for causal claims and human inference from non-human studies. There was no evidence for higher news uptake for exaggerated press releases, consistent with previous results. Base exaggeration rates were lower for human inference from non-human studies, possibly reflecting the Concordat on Openness on Animal Research in the UK. Overall, the picture remains that the strength of news statements is normally associated with the strength of press release statements, and without evidence that exaggerated statements get significantly more news.