BAC Levels
ES

Drunk Driving Statistics 2026

Updated October 2026 · 10 min read

Drunk driving still kills about 12,000 people a year in the United States — 11,904 in 2024, according to NHTSA. The agency's latest cost study put the bill at $58 billion a year in direct economic losses, and $296 billion once lost quality of life is counted.

The data here pulls together the latest drunk driving figures from key sources: the National Highway Traffic Safety Administration (NHTSA), the FBI Uniform Crime Report, the Centers for Disease Control and Prevention (CDC), and state-level reporting agencies. Fatality figures come from NHTSA's 2024 data, released in June 2026. Use our BAC Calculator to understand where you stand before getting behind the wheel.

Annual DUI Arrests

CDC's impaired-driving page says hundreds of thousands of people are arrested each year for driving under the influence of alcohol or drugs, citing FBI data. National totals differ by source and year, so we are not printing one precise count. The point is scale: that many people were stopped and charged.

A 1988 Bureau of Justice Statistics report puts the peak at 1,921,000 DUI arrests in 1983. Arrests fell from there as MADD reshaped public attitudes and states tightened their laws. Keep in mind that arrest counts follow police priorities as much as drinking behavior, which makes them a poor stand-in for how much drunk driving actually happens.

Alcohol-Related Crash Fatalities

NHTSA's 2024 data (DOT HS 813 815, released June 2026) counted 11,904 people killed in alcohol-impaired-driving crashes — 30% of all 39,254 traffic fatalities. That is 3.9% below 2023's final count of 12,382, a slightly smaller drop than traffic deaths overall (4.3%).

That works out to one death every 44 minutes, or about 33 people a day.

Most of the people who die are not the drunk driver. Of the 11,904 deaths, 7,247 (61%) were the alcohol-impaired drivers themselves. Another 1,483 were passengers riding with them, 1,858 were occupants of other vehicles, and 1,316 were pedestrians, cyclists or other nonoccupants. Among children 14 and younger, 221 of 1,032 traffic deaths (21%) happened in alcohol-impaired-driving crashes.

The timing is lopsided. NHTSA found that 31% of drivers in nighttime fatal crashes were alcohol-impaired versus 10% in daytime ones, and 28% on weekends versus 15% on weekdays. Weekend nights are the peak: 35% of drivers in fatal crashes on weekend nights (Friday 6 p.m. to Monday 6 a.m., after dark) were alcohol-impaired.

Worth noting: "alcohol-impaired" in NHTSA statistics means a driver or motorcycle operator with a BAC of 0.08% or higher, and it describes the driver's BAC, not proof that alcohol caused the crash. Crashes where the highest driver BAC was 0.01% to 0.07% are counted separately. Add them and 13,932 deaths, or 35% of the 2024 total, involved a driver with any alcohol.

Economic Cost

NHTSA's latest cost study uses 2019 data, the most recent year with cost estimates. It attributes $58 billion in economic losses to alcohol-impaired-driving crashes: medical expenses, lost productivity, workplace costs, legal and court costs, emergency services, insurance administration, congestion and property damage. Value the lost quality of life from serious injury and death as well, and the total societal harm rises to $296 billion.

That broader figure tries to capture what the invoices miss — the devastation to families, the permanent disabilities, the psychological trauma inflicted on survivors and witnesses, and the lost earning potential of people killed in their twenties and thirties.

You will also see $69 billion quoted. That is NHTSA's 2019 figure for all alcohol-involved crashes, including ones where no driver reached 0.08%. Both numbers are in 2019 dollars, so today's real cost is higher.

Trends Over Time: The Long View

NHTSA's long-run table shows 26,173 traffic deaths in 1982 involved alcohol at any level, 60% of that year's 43,945 deaths. In 2024 the equivalent count was 13,932, or 35% of 39,254. The share fell from 60% to 35%. NHTSA has changed how it estimates missing BAC results over the decades, so read the comparison as a direction, not a precise measurement.

MADD's founding in 1980 rewired public attitudes fast enough that within four years Congress passed the National Minimum Drinking Age Act, raising the limit to 21 nationwide. States then spent two decades ratcheting BAC limits down from 0.15% to 0.10% to 0.08%, with federal standardization by 2004. The timing lines up with falling death counts, though the data cannot prove which law did the work.

Then the progress stalled. Alcohol-impaired deaths ran between roughly 10,200 and 11,000 a year from 2015 to 2019, with a low of 10,196 in 2019, then jumped to 11,727 in 2020 and 13,599 in 2021. They have eased to 11,904 in 2024.

The pandemic scrambled the numbers. People drove less in 2020, yet the alcohol-impaired death rate per 100 million vehicle miles rose from 0.31 in 2019 to 0.40 in 2020 and 0.43 in 2021. It is down to 0.36 now, which is still above where it stood before the pandemic.

Age Demographics

Drunk driving hits some age groups harder than others. In 2024, 28% of drivers aged 21-24 in fatal crashes were alcohol-impaired, the highest share of any group. Ages 25-34 followed at 27%, then 35-44 at 23%, 15-20 at 20%, 45-54 at 19%, 55-64 at 16%, 65-74 at 12% and 75 and older at 7%.

These are shares of drivers in each age group who were involved in fatal crashes, not shares of all drivers on the road. The pattern is still clear: the drivers most likely to be over the limit when a crash turns deadly are in their twenties and early thirties.

Drivers 15 to 20 made up 989 of the 11,367 alcohol-impaired drivers in fatal crashes, under 9% of the total.

Older drivers (65+) have the lowest alcohol-impaired shares but face unique risks: reduced alcohol tolerance, medication interactions, and age-related declines in driving ability that amplify the effects of even small amounts of alcohol.

The gender gap is stark. In 2024, 22% of male drivers in fatal crashes were alcohol-impaired versus 16% of female drivers. By count, that was 8,889 men versus 2,209 women, about four male alcohol-impaired drivers for every female one.

By vehicle, the highest shares were motorcycle riders and passenger-car drivers at 25% each, then light trucks at 19% and large trucks at 4%.

BAC Distribution in Fatal Crashes

Most drunk driving deaths don't involve borderline cases. Of the 13,500 drivers in 2024 fatal crashes who had alcohol in their systems, 84% (11,367) were at or above 0.08%. Fifty-six percent of all drinking drivers (7,547) were at 0.15% or higher, which is about two-thirds of the alcohol-impaired group and nearly double the legal limit.

Of the 11,904 alcohol-impaired-driving deaths, 68% (8,097) happened in crashes where at least one driver was at 0.15% or above. The most frequently recorded BAC among drinking drivers in fatal crashes was 0.15%.

The policy takeaway is blunt: the deadliest drunk drivers are not borderline cases who had one extra glass of wine. They are heavily intoxicated. Prior convictions are only part of the picture: 6% of alcohol-impaired drivers in fatal crashes had a DWI conviction in the previous five years, versus 2% of drivers with no alcohol. That is three times as likely, and still a small minority.

At 0.15% BAC, a person experiences significant impairment: major loss of balance, blurred vision, vomiting, substantially impaired reaction time, and severely impaired judgment. At 0.20% and above, confusion and disorientation make driving almost impossible to do safely. Yet thousands of people attempt it every night.

State-by-State Variation

NHTSA itself warns against ranking states too eagerly. The share of drivers in fatal crashes with a known BAC test result runs from 8% in Mississippi to 86% in Montana, and for untested drivers NHTSA fills the gap statistically.

With that caveat, here is what the 2024 numbers show. By share of traffic deaths in alcohol-impaired-driving crashes, Vermont is highest at 41%, followed by South Carolina and Texas at 40% each, and Alaska is lowest at 19%, against 30% nationally. By rate per 100 million vehicle miles, South Carolina is highest at 0.67 and Massachusetts lowest at 0.16, against a national 0.36. By raw count, Texas (1,676), California (1,311) and Florida (769) lead, which largely reflects their size.

Utah, the one state with a 0.05% limit, shows 26%, below the national 30% but not an outlier, and a single year cannot isolate what the lower limit did. Check the legal BAC limits by state for a complete breakdown of each state's thresholds.

The Ride-Sharing Effect

Anderson and Davis used Uber's own tract-level data for their Review of Economics and Statistics paper and found that ridesharing has decreased U.S. traffic fatalities by 5.2% in the areas where it operates, worth about $6.8 billion a year in lives saved. The effect concentrates on nights and weekends — the hours when a cheap ride home is the realistic alternative to driving drunk. The paper covers all traffic deaths, not only alcohol-related ones.

Earlier studies that only looked at when Uber entered a city reached inconsistent answers, which is why the tract-level data matters. Access is the open question: the places with the most drunk driving per person often have the fewest ride options.

What the Numbers Mean for You

Those 12,000 annual deaths are not a statistic — they are parents, children, friends. Every one was preventable. Every one started with a decision that a phone call or a designated driver could have stopped.

Most people underestimate how few drinks it takes to hit 0.08% — two strong beers in an hour can put a 140-pound person over the limit.

Know Your Limit

Use our free BAC calculator to estimate your blood alcohol content. It takes 30 seconds and could save your life — or someone else's.

Open BAC Calculator

Resources

If you or someone you know is struggling with alcohol, help is available 24/7.

Frequently Asked Questions

How many people die from drunk driving each year?
NHTSA's 2024 data (DOT HS 813 815, released June 2026) counted 11,904 people killed in alcohol-impaired-driving crashes — 30% of the 39,254 traffic deaths that year, or one every 44 minutes. That is 3.9% below 2023's 12,382, but still above the 10,196 recorded in 2019, so the country has not returned to its pre-pandemic baseline.
How many DUI arrests are there per year?
Hundreds of thousands. CDC's impaired-driving page describes arrests for driving under the influence of alcohol or drugs in those terms, citing FBI data. For scale, a 1988 Bureau of Justice Statistics report put the 1983 peak at 1,921,000 arrests.
What is the economic cost of drunk driving?
NHTSA's latest cost study (2019 data, the most recent year with cost estimates) put alcohol-impaired-driving crashes at $58 billion in direct economic losses — medical bills, lost productivity, property damage, legal costs. Add the monetized loss of quality of life and the total reaches $296 billion.
What age group is most at risk for drunk driving?
Drivers aged 21-24. In 2024, 28% of drivers that age in fatal crashes were alcohol-impaired, with 27% for ages 25-34 right behind. The share then falls step by step with age, to 7% for drivers 75 and older.
Has drunk driving gotten better or worse over time?
Better than the early 1980s by a wide margin, but not on a straight line. NHTSA counted 26,173 traffic deaths involving alcohol at any level in 1982 (60% of all deaths) versus 13,932 in 2024 (35%). Alcohol-impaired deaths bottomed out at 10,196 in 2019, climbed to 13,599 in 2021, and were 11,904 in 2024.
At what BAC do most fatal crashes occur?
Well past the limit. NHTSA's 2024 data show 7,547 of the 11,367 alcohol-impaired drivers in fatal crashes — about two-thirds — had a BAC of 0.15% or higher, nearly double the legal limit, and 0.15% was the most frequently recorded level.