Do alcohol ads make kids more likely to start drinking?
Yes. Young people with high exposure to alcohol marketing had about a 50% higher risk of drinking than those with low exposure (relative risk 1.50). That comes from 19 long-term studies following 82,498 people under age 25 across 10 high-income countries.
This was a systematic review and meta-analysis published in JAMA Pediatrics. A meta-analysis pools the results of many separate studies so the combined answer is steadier than any single one. Every study included here was longitudinal, meaning researchers measured how much alcohol marketing a young person was exposed to first, then followed them forward in time to see who started drinking. That order matters. It rules out the simplest reverse explanation, where kids who already drink just happen to notice more alcohol ads.
The researchers searched three large medical databases, MEDLINE, Scopus, and PsycInfo, from their beginning through September 2, 2025. They counted alcohol marketing broadly, not just television commercials. Online advertising, alcohol shown in movies and shows, promotional events, owning branded merchandise like a beer T-shirt or hat, liking a particular ad or brand, and combined measures of all of the above all counted as exposure.
What the data show
The headline number is a 50% higher risk of drinking outcomes in young people with high marketing exposure compared with low exposure. The range around that estimate was tight, so the true effect is very likely somewhere between 37% and 63% higher risk (95% CI 1.37 to 1.63). “Drinking outcomes” here means starting to drink in the first place and binge drinking, not casual sips.
That single number rests on a lot of underlying data. The 19 cohorts produced 23 published papers and 94 separate risk estimates. The authors converted odds ratios into relative risks so everything could be compared on one scale, then used a multilevel analysis to combine them.
Among the different types of marketing, television advertising had the strongest link to drinking. The authors also ran a narrative synthesis, which is a structured summary of studies that could not be pooled with numbers, and it pointed to how the effect seems to work: exposure shifts a young person’s attitudes and beliefs about alcohol, and those shifted beliefs are what show up later as drinking. The marketing works on what a young person believes about alcohol, not on a single purchase decision.
One thing this pooled figure cannot tell you is how many extra teenagers start drinking. A relative risk describes how much the risk multiplies, not how many additional kids that works out to in a classroom or a town.
Dr. Kumar’s Take
The alcohol industry has argued for decades that marketing only moves market share between brands and does not create new drinkers. This analysis is hard to square with that. Nineteen separate cohorts, across 10 countries with different ad rules and different drinking cultures, pointing the same direction, is not the kind of pattern you get from noise.
The mechanism finding is the piece I find most useful clinically. If the pathway runs through attitudes and expectations, then the damage is not done at the moment of purchase. It is done years earlier, accumulating in what a kid assumes adulthood looks like. That changes what a reasonable response looks like.
This is observational research, and it cannot prove that the ads caused the drinking. Given that, the design here is close to the best available: exposure measured first, outcomes measured later, in large general populations rather than volunteers.
How strong is the evidence?
The authors graded the evidence using GRADE, a formal system for rating how much confidence a body of research deserves. They landed on moderate certainty, and notably they upgraded it because of dose response. Dose response means more exposure went with more drinking, in a stepwise way. That pattern is one of the classic fingerprints of a real causal relationship rather than a statistical coincidence, which is why it earns an upgrade.
Bias was assessed with a tool called ROBINS-E, built for exactly this kind of non-randomized exposure study. Every single study was judged to be at risk of bias for one specific reason: drinking was self-reported. Young people asked about their own alcohol use may underreport it, overreport it, or answer differently depending on who is asking. That limitation runs through the entire evidence base, and it is the main reason the certainty is moderate rather than high.
Where the evidence is thinnest
All 19 cohorts came from high-income countries. The authors call this out directly and say more research is needed in low- and middle-income countries, where alcohol marketing is expanding. Those are precisely the places where ad regulation is often weakest and where the results would matter most, and applying this number to them is an extrapolation.
The other gap is measurement. Exposure was captured in a patchwork of ways across studies, from television hours to merchandise ownership to brand liking. The authors argue for single measures that capture multiple marketing channels at once, which is a fair point in an era when a teenager encounters alcohol on a phone, a streaming show, and a stadium banner within the same hour.
Practical Takeaways
- Treat alcohol marketing as an exposure worth limiting the same way you would limit any other, since young people with high exposure had a 50% higher risk of drinking outcomes.
- Television advertising had the strongest link of any channel studied, so ad-supported live TV and sports broadcasts deserve more attention than they usually get in conversations about screen time.
- The effect appears to run through attitudes and expectations about alcohol, so talking with kids about what ads are actually selling matters more than a one-time warning about drinking.
- Branded merchandise counted as exposure in these studies, which makes a free hat or shirt with a beer logo on it worth a second thought rather than a shrug.
Related Studies and Research
- Binge drinking once a month triples your risk of serious liver damage
- Circadian light therapy for young people with depression: systematic review
- Heavy pesticide exposure linked to 2 to 8 times higher cancer risk
- PPIs cause low magnesium: meta-analysis reveals 43% higher risk
FAQs
Does this prove alcohol advertising causes teen drinking?
No, and no observational study can. Proving cause would require randomly assigning children to grow up with or without alcohol marketing, which is neither possible nor ethical. What this analysis offers instead is the next best thing: exposure measured before drinking started, across 19 separate cohorts and 10 countries, with a dose-response pattern strong enough that the authors upgraded their confidence rating because of it. That combination is why the evidence was graded moderate certainty rather than low.
Why does television advertising stand out when so much marketing has moved online?
Television advertising showed the strongest association in the subgroup analyses, but that finding needs context about when these cohorts were recruited. Longitudinal studies take years to run, and a body of research searched through September 2025 necessarily includes cohorts that were followed when television held a bigger share of a young person’s attention. The abstract does not break the estimates down by study year. Online advertising was included as an exposure type, so the online channels were studied, they simply did not top television in the pooled comparison.
What counted as “high” alcohol marketing exposure in these studies?
There was no single threshold. Different cohorts measured exposure in different ways, including television advertising, online ads, alcohol portrayals in entertainment media, promotional activities, owning branded merchandise, liking an ad or brand, and combined measures. Where a study measured exposure on a continuous scale, the researchers scaled it to compare low exposure against high exposure within that study. So the 50% figure describes heavily exposed young people versus lightly exposed ones, using each study’s own yardstick, rather than a fixed number of ads per week.
Bottom Line
Pooling 19 longitudinal cohorts and 82,498 young people under 25, heavy alcohol marketing exposure went with about a 50% higher risk of drinking outcomes, including starting to drink and binge drinking, and the true effect is very likely between 37% and 63% higher. Television advertising showed the strongest link, the association appeared to work by shifting attitudes and expectations about alcohol rather than by direct selling, and a dose-response pattern was clear enough that the authors raised their certainty rating to moderate. The evidence comes entirely from high-income countries and relies on self-reported drinking, but the direction and consistency across countries and marketing channels is difficult to explain away.
