JMIR Publications

ABSTRACT

Background:

E-cigarettes are a significant public health risk and are widely used among adolescents and young adults. Social media is a promising intervention strategy to prevent and control nicotine product use. Recent studies have demonstrated that targeted social media campaigns can reduce risk factors for e-cigarette use. The current study evaluated the effects of 12-month exposure to anti-vaping social media content on subsequent vaping among young adults.

Objective:

The study tested three hypotheses: 1) Higher levels of exposure to anti-vaping social media content will reduce future vaping behavior among never-vapers (prevention); 2) Higher levels of exposure to anti-vaping social media content will reduce future vaping behavior among ever-vapers (cessation); 3) In both strata, higher levels of exposure to anti-vaping social media content will improve secondary outcomes related to vaping.

Methods:

We followed a panel of young adults aged 18-24 years living in the United States to evaluate the effects of AI-enabled anti-vaping social media content on e-cigarette use and related determinants of behavior. The panel was divided into two strata: those who reported at enrollment that they had never used e-cigarettes, and those who reported some e-cigarette use (previous or current) at enrollment. Within each stratum, we conducted a four-arm parallel-group randomized controlled trial. The intervention content consisted of 2 distinct “campaigns” in which anti-vaping video content was delivered to participants through paid advertising on Meta’s Facebook and Instagram. The first campaign ran during the period from baseline to FU1. The second ran from FU1 to FU2, with participants receiving three new videos. We used multiple imputation methods to handle missing data, ordinal logistic regression models for our primary outcome of vaping frequency at FU2, and linear regression models for secondary outcomes.

Results:

We enrolled 3,321 participants to the never-vaper stratum across the four randomized arms and n=5,026 participants in the ever-vaper stratum. In the never-vaper stratum, an increase of 10 impressions per person between waves was associated with a 30% reduction in the odds of being in higher vaping frequency categories at first follow-up, and a 21% reduction in the odds of being in higher vaping frequency categories at second follow-up. For the ever-vaper stratum, higher numbers of impressions per person were associated with small increases in the odds of being in a higher vaping frequency category, but most of these associations are not statistically distinguishable from the null. In the never-vaper stratum higher numbers of impressions per person between waves were associated with higher perceived social unacceptability of vaping scale scores, and with lower levels of intentions to vape in the future.

Conclusions:

This study demonstrated a clear vaping prevention effect among never-vapers. New digital approaches are needed to effectively change behavior among those who use or previously used e-cigarettes. Clinical Trial: This study was registered as a clinical trial at clinicaltrials.gov under identifier NCT04867668.