Exposure to alcohol-related social media content and young adults’ desire to drink: a randomized controlled experiment

INTRODUCTION

The use of social media (e.g., Instagram) is common among young adults and is characterized by a continuous stream of content from peers and influencers,¹ which can feature alcohol in aspirational or humorous contexts.² Although prior research has linked exposure to pro-alcohol-related content on social media with pro-alcohol-related attitudes and behaviors among young adults,³ most studies have been cross-sectional, unable to establish temporal precedence, and were limited by unmeasured confounding variables. Establishing experimental evidence regarding the role of social media content on pro-alcohol-related attitudes and behaviors is critical for guiding evidence-based policy on social media.⁴ This randomized controlled experiment examined whether exposure to pro-alcohol-related social media posts increased desire to drink among young adults.

METHODS

From August 22, 2025, to September 04, 2025, young adults (ages 18–24 years) living in the United States (U.S.) with internet access were recruited by YouGov, a research panel agency, to complete an online survey (See online supplemental file for details on methodology). Participants were randomly assigned to a treatment or control condition designed to simulate a typical social media browsing experience. In the treatment condition, participants viewed 20 Instagram posts from lifestyle influencers that visibly depicted alcohol consumption or pro-alcohol-related imagery (e.g., holding alcoholic beverages). In the control condition, participants viewed 20 matched posts from the same influencers, but without any alcohol-related imagery (Online supplemental file).

Desire to drink was assessed using a single-item measure adapted from prior research.⁵ Covariates included perceived influencer credibility, sex, age, race, exposure to offline alcohol advertising, exposure to alcohol-related content on social media, daily social media use, and lifetime alcohol use (Online supplemental file).

Ordinal logistic regression was used to test the relationship between the experimental condition (treatment vs. control) and participants’ desire to drink, adjusting for covariates (Online supplemental file). Multiple sensitivity analyses were conducted (Online supplemental file). This study followed the American Association for Public Opinion Research (AAPOR) best practices for survey research.

RESULTS

Participants in the alcohol-related social media exposure condition had significantly greater odds of reporting higher desire to drink compared with those in the control condition (aOR=1.73, 95% CI=1.46–2.07, p<.05) (Table 1). Participants who perceived the influencers as highly credible, compared with those who did not, had greater odds of reporting higher desire to drink (aOR=5.53, 95% CI=3.50–8.76, p<.05). Similar positive associations between prior alcohol use and desire to drink were observed among participants who reported past 30-day alcohol use (aOR=1.40, 95% CI=1.20–1.65, p<.05) and those who reported past 30-day binge drinking (aOR=1.30, 95% CI=1.09–1.55, p<.05). Results were consistent across sensitivity analyses.

Table 1.

Weighted ordinal logistic regression models of desire to drink (N=2000).

Regressors	Desire to Drink (aOR [95%CI])
Main effects model
Pro-alcohol-related influencer content (vs. control)	1.73 (1.46, 2.07)
Interaction effects models
Experimental condition x perceived influencer credibility	5.53 (3.50, 8.76)
Experimental condition x past 30-day alcohol use	1.40 (1.20, 1.65)
Experimental condition x past 30-day binge drinking	1.30 (1.09, 1.55)

Note. All models were weighted using post-stratification weights and adjusted for gender, age, race or ethnicity, prior exposure to offline alcohol advertising, prior exposure to social media alcohol content, daily social media use, and lifetime alcohol use. Results were statistically significant at p < .05. Perceived influencer credibility, past 30-day alcohol use, and past 30-day binge drinking were entered as ordinal predictors; as such, each one-unit increase in these variables reflected a higher category of the predictor, corresponding to the multiplicative change in odds shown in the table.

DISCUSSION

Findings indicated that exposure to pro-alcohol-related social media content may influence desire to drink across varying levels of prior alcohol use and that social media influencers may contribute to the normalization of alcohol consumption among young people. These findings add experimental evidence to a growing body of research showing that exposure to pro-alcohol-related content, particularly on social media, is associated with pro-alcohol-related attitudes and behaviors among young adults.⁶

Limitations

This experiment assessed pro-alcohol-related attitudes following exposure to social media content; however, it remains unclear whether exposure to pro-alcohol-related content subsequently contributes to increased drinking behavior among young adults. Reciprocal processes are also possible; that is, individuals with a greater desire to drink may be more likely to seek out pro-alcohol-related content online. The influencer stimuli used in this study may not be representative of the broader range of alcohol-related imagery found in the online environment. Posts were limited to one social media platform, Instagram, which may not reflect the content popular on other social media platforms.

CONCLUSION

Addressing the role of social media content in shaping pro-alcohol-related attitudes and behaviors may be critical for curbing drinking among young adults. Future research should examine how repeated exposure to alcohol content affects drinking intentions and behaviors over time, and whether these effects differ by content source, such as brand-generated advertising, or peer- or influencer-generated posts.

Data availability statement:

Data and associated study materials can be found at https://github.com/SDonaldsonRutgers/Alcohol-Influencers-Study.