Effectiveness of Text-Only E-Cigarette Warnings: A Meta-Analysis.

Importance: A single text-only e-cigarette warning about nicotine addiction is required by the US Food and Drug Administration in the US, yet little is known about whether health harms warnings are more effective than the required nicotine addiction warning. Objective: To assess the effectiveness of text-only e-cigarette warnings on message processing, risk belief, and behavioral intention outcomes, as well as the impact of health harm warnings and the required nicotine addiction warning or similar addiction warnings. Data Sources: A comprehensive search was conducted using PubMed, Embase, Scopus, PsycINFO, and Business Source Premier from inception through February 2024. Review articles were also examined for potential studies. Study Selection: Experimental studies that randomized participants to view nicotine addiction or health harms e-cigarette warnings were included. Studies had to report 1 or more message processing, risk perception, or behavioral intention outcomes. Data Extraction and Synthesis: Using Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) reporting guidelines, 2 independent coders (Y.J. and J.S.) screened and coded all studies. Effect sizes characterizing overall warning effects and health harm vs addiction warning effects were calculated using the standardized mean difference (d). Effect sizes were pooled using random-effects models and tested for heterogeneity using the Q statistic and I2. Effect sizes with their 95% CIs, as well as prediction intervals, are reported. Main Outcomes and Measures: The outcomes were attention, negative affect, and effects perceptions (message processing); addiction, risk, and relative risk beliefs (risk perceptions); and intentions to vape and intentions to quit vaping (behavioral intentions). Results: Across 24 studies with 22 549 participants (mean [SD] age, 27.91 [6.36] years; 53.9% female), e-cigarette text-only warnings were associated with improvement of most outcomes. Compared to control, e-cigarette warnings were associated with greater attention (d, 0.52; 95% CI, 0.33-0.70), negative affect (d, 0.65; 95% CI, 0.49-0.81), and effects perceptions (d, 0.95; 95% CI, 0.69-1.21), as well as increased addiction beliefs (d, 0.24; 95% CI, 0.05-0.42) and risk beliefs (d, 0.26; 95% CI, 0.16-0.36) but not relative risk beliefs (d, 0.00; 95% CI, -0.11 to 0.12). E-cigarette warnings also were associated with reduced intentions to vape (d, -0.14; 95% CI, -0.27 to -0.01) and increased intentions to quit vaping (d, 0.34; 95% CI, 0.09-0.58). Health harm warnings outperformed nicotine addiction warnings on many outcomes. Compared to addiction warnings, health harm warnings were associated with more negative affect (d, 0.44; 95% CI, 0.32-0.52), attention (d, 0.37; 95% CI, 0.26-0.49), and effects perceptions (d, 0.36; 95% CI, 0.23-0.49), as well as increased risk beliefs (d, 0.07; 95% CI, 0.00-0.13) and intentions to quit vaping (d, 0.17; 95% CI, 0.06-0.27). There were no differences in addiction beliefs, relative risk beliefs, or intentions to vape for health harms vs addiction warnings. Conclusions and Relevance: In this meta-analysis of experimental studies, text-only e-cigarette warnings were associated with increased beliefs about the harm and addictiveness of e-cigarettes without creating the misperception that e-cigarettes are more harmful than cigarettes. Warnings were also associated with reduced intentions to vape and increased intentions to quit vaping. These findings support adding health harms to e-cigarette warnings and have important implications for warning policy.