This project contains open data for our randomised controlled trial looking at the influence of press releases on news. We have subsequently used subsets of the data in this database for two further analyses, explained below. For more information on the trial see [here][1], and the registered protocol can be found [here][2]. All raw data is available and has also been processed into a database in .sqlite3 format. Details of how to access and use the database can be found in the file 'Insciout_Documentation.html'. Any issues or queries can be sent to us at insciout@cardiff.ac.uk The trial is published as: Adams, R. C., Challenger, A., Bratton, L., Boivin, J., Bott, L., Powell, G., et al. (2019). Claims of causality in health news: a randomised trial. BMC Medicine, 17(1), 91. http://doi.org/10.1186/s12916-019-1324-7 Further analyses: 1. We used data from the baseline period (2014 and 2015) in this database for a direct replication of the analyses in Sumner et al. (2014). The association between exaggeration in health related science news and academic press releases: retrospective observational study. Bmj, 349(dec09 7), g7015–g7015. http://doi.org/10.1136/bmj.g7015 An additional folder has been added in 'processed data' containing the extracted data for this replication. It comes from the 'before and after' section of the raw data files and the database (reason for label explained in 2. below; see Insciout_Documentation.html for access to database). The replication is limited to the universities analysed in Sumner et al. 2014. The replication is published in Bratton et al. (2019) *Wellcome Open Research*. 2. We also used data from the baseline period to compare press release exaggeration rates between 2014 and 2015, i.e. 'before' and 'after' the publication of Sumner et al. (2014). This was in order to seek evidence of any impact associated with that publication. An additional folder has been added in 'processed data' containing the extracted data for this impact analysis. The results are submitted to *Wellcome Open Research*. 3. We also analysed how often news and press releases provide information on funding and conflicts of interest. We used data from the baseline period (2015) in combination with the database provided by Sumner et al. (2014) at (http://dx.doi.org/10.6084/m9.figshare.903704), which contains news and press releases from 2011. An additional folder has been added in 'processed data' containing the extracted data from both datasets for this funding analysis. The results are published as [to be updated]. [1]: http://sites.cardiff.ac.uk/insciout/rct/ [2]: http://www.isrctn.com/ISRCTN10492618