AIM Measure the impact of brief, academic-led, evidence-based social media videos on vaping harm perceptions among young adults.

DESIGN On-line between-subjects experimental study. Participants were randomized to one of two conditions: experimental (exposed to one of eight brief videos, designed for social media, with academic experts addressing vaping harms) or control. Before and after exposure to the videos, all participants answered questions about their perceptions of vaping and smoking and socio-demographics.

SETTING Qualtrics on-line survey platform.

PARTICIPANTS 593 young adults aged 18-30 years who resided in the UK (49.7% female, 49.2% male; 8.9% exclusively smoked, 32% exclusively vaped, 28.7% did both and 30.4% did neither). Participants were randomly assigned to intervention (n = 279) or control (n = 314) groups.

MEASUREMENTS The primary outcome was the perception that vaping is less harmful than smoking. Secondary outcomes were perceptions that vaping is harmful, vaping is addictive and responses (true, false) to statements that were matched to the videos (e.g. vaping causes cancer, vaping causes lung injuries).

FINDINGS After exposure to an expert video, compared with those in the control group, participants in the intervention group had over three times the odds of perceiving vaping as less harmful than smoking [82.1% vs 57.6%; adjusted odds ratio (AOR) = 3.69; 95% confidence interval (95% CI) = 2.49-5.47; P < 0.001]. Perceptions that the following statements are false were also higher after viewing expert videos than control videos: vaping causes lung injury, vaping leads to cancer, nicotine is harmful when used in ways that does not involve smoking tobacco, pregnant women should not vape, vaping will not help you quit smoking, vaping has no place on the NHS (all P < 0.05). Participants exposed to the 'vaping is as harmful as smoking' misconception video had the highest odds of accurately perceiving vaping as less harmful than smoking (AOR = 13.92; 95% CI = 3.26-59.37; P < 0.001). Videos specifically targeting individual misconceptions (e.g. 'vaping causes lung injury' or 'vaping causes cancer') were particularly effective in improving related perceptions, indicating that the videos functioned as designed. There was little evidence of associations between condition and perceiving that vaping is not harmful (AOR = 2.57; 95% CI = 0.78-8.52; P = 0.122) or not addictive (AOR = 0.49; 95% CI = 0.04-6.67; P = 0.594). Findings were similar among young adults regardless of vaping and smoking status.

CONCLUSION Brief, academic-led, vaping facts films appear to be effective in correcting vaping misperceptions and dispelling common misconceptions.