Fig. 4

Use of causality statements/caveats (error bars are 95%CIs). a ITT was insensitive to differences in news content; AT showed that 20% of news contained causality statements or caveats if the press release did, and almost never otherwise. b ITT shows no reduction of news uptake and AT shows an increase in news for press releases containing causality statements/caveats (see also Additional file 1: Figure S4 for average number of news per press release). c Feasibility is indicated by the increase in spontaneous caveats for observational research since the baseline period (2014/2015). Final press releases showed a further increase following suggestions in relevant trial conditions. For each bar, n reports total number of news (a) or press releases (b, c) in that analysis group (i.e. the denominator of the proportion that the bar displays)