BAC Levels
ES

Underage Drinking Statistics 2026: What CDC, NHTSA, and NSDUH Actually Show

Updated April 2026 · 12 min read

About 4,000 young people under 21 die every year from excessive alcohol use, according to the CDC's Alcohol-Related Disease Impact tool. That works out to roughly eleven kids a day, more than a typical American high school's senior class lost every week of the year. The number sits in the same general range it has occupied for most of the past decade, even though teen drinking itself has dropped to record lows — which is one of the more uncomfortable statistical patterns in this data set.

This page lays out the actual numbers — past-month and past-year prevalence, deaths and crashes by year, the abstention trend nobody talks about, and the policy effect that produced the largest single drop in young-driver fatalities in modern American history. Every figure here comes from CDC, NHTSA, NIAAA, SAMHSA's National Survey on Drug Use and Health, or the University of Michigan's Monitoring the Future. Where the most recent number is more than a year old, that is called out so the reader can re-find the source.

How many teens are actually drinking

The 2024 National Survey on Drug Use and Health reported that 1.8 million youth ages 12-17 — 6.9 percent of the cohort — had used alcohol in the past month. About 900,000 in that same age group, or 3.5 percent, engaged in past-month binge drinking. The 2023 NSDUH put past-year alcohol use at 4.4 million teens, or 16.9 percent. One in six is the right mental anchor for past-year use, not the more alarming numbers from older surveys that get recycled through wellness blogs and news graphics.

The structure of those numbers matters. Past-month use of 6.9 percent means that on any given day, the share of teens who actually drink at all is small — somewhere under 7 percent of the age group, much smaller still on a given night. The harm comes from the subset that drinks heavily, not from a teenage population mostly choosing alcohol. That distinction gets lost in the way underage drinking gets covered, but it is what the survey instruments have been showing for two decades.

The decline that doesn't get talked about

The 2025 release of Monitoring the Future, the long-running NIDA-funded survey of 8th, 10th, and 12th graders, reported lifetime alcohol abstention at record highs — 83 percent of 8th graders, 70 percent of 10th graders, and 51 percent of 12th graders had never tried alcohol. Past-30-day binge drinking was effectively zero in 8th and 10th grades and under 10 percent among seniors. Compare that to the late 1990s, when binge-drinking rates among 12th graders ran above 30 percent and lifetime abstention in middle school was a minority position rather than the norm.

This is one of the largest behavioral changes in adolescent health on record, and it gets a fraction of the coverage that smaller, scarier data points receive. The shift is real, durable, and visible in three different national surveys — NSDUH, MTF, and the CDC's Youth Risk Behavior Survey all show the same downward slope. Whatever combination of phones, social media, parental policy, and supply-side enforcement is producing the change, it is producing it consistently, and treating teens like a uniform reckless group does not match the data on the ground.

Where the 4,000 deaths come from

The CDC ARDI estimate of roughly 4,000 underage deaths per year is built from a long list of alcohol-attributable causes. The largest single share is drunk-driving crashes — the NHTSA fatality data covered in the next section. Acute alcohol poisoning, falls, drownings, and other injuries make up the next layer. A meaningful chunk of the total comes from chronic causes that show up later in life but trace back to heavy adolescent drinking — the 4,000 number includes some deaths in the late teens that had a long buildup, not only the headline tragedies.

Parents and educators tend to underestimate alcohol poisoning specifically, because it is the cause that does not generate headlines unless the victim is high-profile. A high schooler who finishes half a bottle of vodka over an hour can hit 0.30% BAC range — territory that involves loss of consciousness and real respiratory-suppression risk — without showing the kind of staggered, slurred presentation people associate with severe intoxication. The BAC level chart covers what each band actually looks like in the body, and the blood alcohol chart by weight shows why a lighter teenager hits the danger band on fewer drinks than an adult drinker doing the same volume.

The drunk-driving angle: NHTSA FARS 2023

NHTSA's Fatality Analysis Reporting System recorded 1,030 young drivers ages 15-20 involved in fatal crashes in 2023 who tested positive for alcohol at a BAC of 0.01 g/dL or higher. About 26 percent of young drivers killed in crashes that year were alcohol-impaired — meaning a measured BAC of 0.08 g/dL or higher. The 26 percent figure means that roughly one in four fatal teen-driver crashes involved a legally-drunk teen behind the wheel, not a sober young driver hit by an impaired adult.

The crash-risk math is brutal. NHTSA's Crash Risk of Alcohol Impaired Driving (DOT HS 812 355) put relative crash risk at about 2.07x at 0.05% BAC and 3.98x at 0.08% compared to a sober driver — and that is for the general adult population. For drivers under 21, the same BAC produces materially higher crash risk because of inexperience layered on top of impairment. A teenager at 0.05% is not behaving like a 35-year-old at 0.05%, and the FARS data has shown this differential consistently for two decades. The drunk driving statistics page covers the broader DUI numbers across all ages, and the legal BAC limits by state page lays out the zero-tolerance laws that govern under-21 drivers in every state.

Why the 21 minimum drinking age holds

NHTSA estimates that minimum-drinking-age laws have saved 31,959 lives between 1975 and 2017 — the period during which states moved to MLDA 21 and the data on the policy effect accumulated. Among persons under 21 specifically, drunk-driving fatalities have dropped about 73 percent since 1982, the year the federal push toward MLDA 21 began in earnest. Both numbers come from NHTSA and Responsibility.org's analysis of the same FARS data set.

Whatever the politics on the drinking age, the body count moved when the law moved, and that is rare. Most public-health interventions produce small, contested effect sizes that researchers have to defend at conferences for years. MLDA 21 produced a step-function decline in young-driver fatalities that shows up in raw FARS counts without any need for statistical adjustment — the line bends in 1984-1987 across most states and stays bent. That outcome is a large part of why the policy has held against periodic attempts to roll it back, including the 2008 Amethyst Initiative push for a return to 18.

What hasn't moved: 12th-grade binge drinking and the college transition

The under-10-percent past-month binge-drinking rate among 12th graders is a record low for that group, but the absolute number means roughly 300,000 American high school seniors are still binge drinking in any given month. That is a pool large enough to keep the alcohol-poisoning and crash totals in their current range even as overall drinking declines. The risk is concentrated in a smaller subset of the age cohort, but the subset has not shrunk as fast as the overall numbers suggest at first glance.

The transition from high school to college is the most dangerous single window in the entire age curve. Past-30-day binge drinking among college students roughly triples from late high school to freshman year — NIAAA's college-drinking landing tracks that jump in detail. The first six weeks of the freshman year specifically, what NIAAA calls the "red zone," concentrate a disproportionate share of the worst alcohol harms in young adulthood. If the underage-drinking story has a single most-fixable failure point, it is that transition. The honest answer about sobering up fast covers what does and does not work when a freshman has already over-drunk and friends need to make a real decision about what to do.

The economic cost

CDC's most recent published estimate of the annual cost of underage drinking is 24 billion dollars, from 2010 data. The number covers medical care, lost productivity, criminal-justice and corrections, and motor-vehicle crash losses. CDC has not refreshed the figure in over a decade. Inflation alone would put the equivalent above 33 billion dollars in 2026 dollars, and healthcare costs in the same period have outpaced general inflation by enough that the real underage-drinking cost is almost certainly above the CDC headline number — though no agency has published an official refreshed total. Treat 24 billion dollars as a floor, not a current estimate.

Resources and helplines

If you or someone you know needs help with alcohol use, SAMHSA's national helpline at 1-800-662-HELP (4357) is free, confidential, and available 24/7 in English and Spanish. The CDC's underage drinking page (cdc.gov/alcohol/underage-drinking) is the simplest read for parents and educators looking for a single overview. NIAAA's underage-drinking factsheet at niaaa.nih.gov goes deeper into the science and is the source the CDC's own page often points back to. For the actual math behind a specific drinking situation, the BAC calculator on this site and the sober calculator let you run the numbers without registration or tracking.

This page is informational and not medical or legal advice. Underage drinking laws in the United States are zero-tolerance — any measurable BAC for a driver under 21 is illegal in every state. If you are responsible for a young person who has been drinking heavily and is unresponsive, irregularly breathing, or showing signs of hypothermia, treat it as a medical emergency and call 911. Most states have medical-amnesty laws that protect callers from prosecution for the underlying underage drinking when they are seeking help for an alcohol-poisoning emergency.

Sources

Frequently Asked Questions

How many teens drink alcohol in the United States?
About 1.8 million youth ages 12-17 — 6.9 percent of that age group — reported drinking alcohol in the past month, according to the 2024 National Survey on Drug Use and Health. Past-year use is roughly 16.9 percent, or 4.4 million teens, based on the 2023 NSDUH. Both figures have trended downward over the past decade rather than upward, and the long arc is the most under-reported part of the underage-drinking story.
How many young people die from underage drinking each year?
Roughly 4,000 people under age 21 die each year from excessive alcohol use, based on the CDC's Alcohol-Related Disease Impact tool using 2020-2021 data. That figure includes alcohol poisoning, drunk-driving deaths, and chronic causes — not just the headline crash fatalities. Drunk driving accounts for the largest single share of the total.
Has underage drinking gone up or down?
Down, and by a wide margin. The 2025 Monitoring the Future survey reported lifetime alcohol abstention at record highs — 83 percent of 8th graders, 70 percent of 10th graders, and 51 percent of 12th graders had never tried alcohol. Past-30-day binge drinking was near non-existent in 8th and 10th grades and under 10 percent in 12th. The decline has been steady for 20 years and has accelerated since the early 2010s.
How many young drivers cause fatal alcohol crashes?
In 2023, NHTSA's Fatality Analysis Reporting System recorded 1,030 young drivers ages 15-20 involved in fatal crashes who tested positive for alcohol at BAC of 0.01 g/dL or higher. About 26 percent of young drivers killed in crashes that year were alcohol-impaired at BAC of 0.08 g/dL or higher. Those numbers are down from 2010s peaks but have been roughly flat for the past three years.
How much does underage drinking cost the US economy?
The most recent CDC estimate, from 2010, put the annual cost at 24 billion dollars. The figure covers medical care, lost productivity, criminal justice, and motor-vehicle crash losses. CDC has not published a refreshed national total since, so the real 2026 number is almost certainly higher when you account for inflation and growth in healthcare costs alone — somewhere in the 30-40 billion range is a reasonable mental ballpark, though no agency has published that estimate officially.
How effective was raising the drinking age to 21?
NHTSA estimates that minimum-drinking-age laws saved 31,959 lives between 1975 and 2017. Among people under 21, drunk-driving fatalities have decreased 73 percent since 1982, when most states moved to MLDA 21. Few American policies have a body-count case this clean — the lives-saved estimate is one of the most-cited numbers in traffic-safety research, and the decline tracks almost exactly to when the law changed.
What percent of high schoolers binge drink?
About 9 percent of US high school students reported past-month binge drinking on the 2023 Youth Risk Behavior Survey, defined as five or more drinks in a row for males or four or more for females within a couple of hours. Binge drinking peaks late in high school and rises sharply in the first year of college — the transition jump is the single most dangerous moment in the age curve.
Why does underage drinking still get so much attention if it is declining?
Because the consequences per drinking episode are worse for teens than for adults. Brain development through age 25 means alcohol does more lasting damage in adolescents, the early-driver inexperience and high BAC combination produces fatal-crash risk that is multiples higher than for adult drivers at the same BAC, and teen alcohol poisoning is more likely to be missed by adults who do not realize how quickly a binge can land in the danger zone. The numbers are dropping, the harm-per-incident is not.

Need help understanding what a BAC number means in real terms? See the BAC chart and legal limits by state. For more BAC-related reading: drunk driving statistics, blood alcohol chart by weight, how to sober up fast, and the rest of the blog.