Driving Statistics That Show Where U.S. Road Risk Still Sits
Driving statistics are often treated as a generic safety topic, but the numbers show a much more specific story: the U.S. has made long-term progress, yet the risk profile shifts sharply by year, behavior, age group, and crash type.
This article pulls together the most useful patterns in the dataset so you can see where driving risk fell, where it rose, and which behaviors still account for too many deaths and injuries.
Table of Contents
- Fast facts
- Long-run fatality trends
- Recent year-to-year changes
- What the major risk factors show
- Drivers by age group
- Seat belts and restraint use
- What the 2024 estimates suggest
- How to read driving statistics responsibly
Fast Facts
- U.S. motor vehicle traffic fatalities fell from 41,059 in 2007 to 32,744 in 2014, then climbed again to 42,939 in 2021 (FHWA FI-200).
- The U.S. recorded 39,254 motor vehicle traffic fatalities in 2024, with 1.19 deaths per 100 million VMT (FHWA FI-200).
- In 2024, NHTSA early estimates said the fatality rate fell to 1.20 per 100 million VMT, the lowest since 2019, while VMT rose 1% (NHTSA 2024 early estimates).
- Speeding killed 11,775 people in 2023 and was involved in 29% of all fatal crashes (NHTSA speeding page).
- Distracted driving killed 3,208 people in 2024 and injured 315,167 people in 2024 (NHTSA distracted-driving page).
- Drunk-driving crashes killed 11,904 people in 2024 and accounted for about 30% of all U.S. traffic crash fatalities (NHTSA drunk-driving page).
- Teen drivers ages 16-19 have a fatal crash rate almost three times higher than drivers ages 20+ per mile driven (CDC teen drivers page).
- Seat belts reduce serious crash-related injuries and deaths by about half (CDC seat-belt facts).
Long-Run Fatality Trends
The clearest long-run signal in the dataset is that U.S. traffic deaths improved through the late 2000s and early 2010s, then worsened again in the 2020s.
In 2007, the U.S. recorded 41,059 motor vehicle traffic fatalities and 1.35 deaths per 100 million VMT (FHWA FI-200). By 2014, that figure had fallen to 32,744 fatalities and 1.08 deaths per 100 million VMT (FHWA FI-200). That is a substantial decline across seven years, and it shows that safety improvements can compound when conditions, enforcement, vehicle technology, and behavior move in the same direction.
The pattern does not stay flat after that. In 2015, fatalities rose to 35,484, then 37,806 in 2016, and 37,473 in 2017 (FHWA FI-200). The series eased slightly in 2018 and 2019, with 36,560 and 36,355 fatalities respectively (NHTSA 2018 traffic deaths release; CDC MMWR high-income countries). Then the number surged again during and after the pandemic period, reaching 38,824 in 2020 and 42,939 in 2021 (FHWA FI-200).
That 2021 figure is the highest in the dataset and a reminder that broad averages can hide sudden reversals. The 2022 count came in at 42,721, then 2023 fell to 40,901, and 2024 fell again to 39,254 (FHWA FI-200; NHTSA 2023 traffic deaths release). The latest two years point in the right direction, but they still sit above the earlier low-risk years of 2013 and 2014.
Year-by-Year View
| Year | U.S. motor vehicle traffic fatalities | Deaths per 100 million VMT | Source label |
|---|---|---|---|
| 2007 | 41,059 | 1.35 | FHWA FI-200 |
| 2010 | 32,999 | 1.11 | FHWA FI-200 |
| 2014 | 32,744 | 1.08 | FHWA FI-200 |
| 2018 | 36,560 | 1.13 | NHTSA 2018 traffic deaths release |
| 2021 | 42,939 | 1.37 | FHWA FI-200 |
| 2023 | 40,901 | 1.26 | NHTSA 2023 traffic deaths release |
| 2024 | 39,254 | 1.19 | FHWA FI-200 |
The table makes the trend easy to see: the U.S. moved from a lower-risk period in the mid-2010s to a much worse period in the early 2020s, then recovered somewhat by 2024.
Recent Year-to-Year Changes
Averages are useful, but the detail that matters most in driving statistics is year-over-year change.
Between 2022 and 2023, U.S. traffic deaths fell by 1,820, from 42,721 to 40,901, which was a 4.3% decline (NHTSA 2023 traffic deaths release). Over the same span, the fatality rate fell 6%, from 1.34 to 1.26 deaths per 100 million VMT (NHTSA 2023 traffic deaths release). That means the risk reduction was not just about fewer deaths in absolute terms; it also improved relative to exposure.
The 2024 early estimates suggest that the downward movement continued. NHTSA estimated that 35 states and Puerto Rico saw traffic fatalities decline in 2024, while 14 states and the District of Columbia saw increases, and one state was unchanged (NHTSA 2024 early estimates). That distribution matters because it shows the national trend was broad, but not universal.
The same data source also noted that VMT increased by 1% in 2024 while the fatality rate fell to 1.20 per 100 million VMT, the lowest since 2019 (NHTSA 2024 early estimates). That combination is important. If travel increases while the fatality rate still drops, the safety system is absorbing more exposure with less harm per mile than before.
What the Major Risk Factors Show
The headline fatality counts only tell part of the story. The dataset also shows which behaviors are still tied to a large share of deaths and injuries.
Speeding
Speed remains one of the most damaging behaviors in road safety. In 2023, 29% of all fatal crashes involved speeding (NHTSA speeding page). That same year, speeding killed 11,775 people in the United States (NHTSA speeding page).
A few related facts help explain why the problem persists:
- In 2023, 51% of passenger vehicle drivers involved in speeding-related fatal crashes were unbuckled (NHTSA speeding page).
- In 2023, 37% of men ages 15-20 involved in fatal crashes were speeding (NHTSA speeding page).
Those facts suggest speeding is not just about road design or enforcement in isolation. It also interacts with restraint use and driver age.
Distracted Driving
Distracted driving is smaller than speeding in fatality count, but it still has a large injury footprint. In 2024, distracted driving killed 3,208 people in the United States and injured 315,167 people (NHTSA distracted-driving page). That is a reminder that some behaviors create a large burden even when they do not always appear in the same way in headline fatality counts.
The earlier data point in the set shows that in 2022, distracted drivers were involved in 3,308 traffic deaths and distracted driving was involved in 8% of fatal crashes (NHTSA distraction page). The same source said that more than 32,000 people had died in distracted-driving crashes over the previous decade as of 2022 (NHTSA distraction page).
Alcohol-Impaired Driving
Alcohol remains one of the most persistent and preventable risk factors. In 2024, drunk-driving crashes killed 11,904 people in the United States, about 30% of all U.S. traffic crash fatalities (NHTSA drunk-driving page). The same source says one person was killed in a drunk-driving crash every 44 minutes in the United States in 2024 (NHTSA drunk-driving page).
The CDC data points give the problem a broader population context. In 2022, alcohol-impaired drivers were involved in 13,524 U.S. crash deaths, or 32% of all traffic-related deaths (CDC impaired-driving facts). That year, 37 people per day were killed in crashes involving an alcohol-impaired driver, and 283 children ages 0-14 were killed in those crashes (CDC impaired-driving facts).
There is also self-reported behavior data. In 2023, 15.5 million U.S. residents ages 16+ reported driving under the influence of alcohol in the past year (CDC impaired-driving facts). The same source says 12.1 million reported driving under the influence of cannabis, and 2.2 million reported driving under the influence of illicit drugs other than cannabis (CDC impaired-driving facts).
A Quick Comparison
| Risk factor | Recent count in dataset | Share or related measure | Source label |
|---|---|---|---|
| Speeding | 11,775 deaths in 2023 | 29% of fatal crashes involved speeding | NHTSA speeding page |
| Distracted driving | 3,208 deaths in 2024 | 315,167 injuries in 2024 | NHTSA distracted-driving page |
| Drunk driving | 11,904 deaths in 2024 | About 30% of traffic crash fatalities | NHTSA drunk-driving page |
| Alcohol-impaired driving | 13,524 crash deaths in 2022 | 32% of traffic-related deaths | CDC impaired-driving facts |
This comparison shows that the largest risks are not isolated edge cases. They are mainstream road-safety problems with large fatal and nonfatal consequences.
Drivers by Age Group
Age is one of the strongest predictors of crash risk, and the dataset shows clear differences between teen, young adult, and older drivers.
Teen drivers ages 16-19 have a fatal crash rate almost three times higher than drivers ages 20+ per mile driven (CDC teen drivers page). That kind of gap is too large to ignore. It is one reason graduated driver licensing exists.
The dataset says all 50 states and the District of Columbia use a three-stage GDL system for teen drivers (NHTSA teen driving page). NHTSA says GDL can reduce teen crash risk by as much as 50% (NHTSA teen driving page). The same source recommends at least 30 to 50 supervised practice hours before full licensure, a learner’s permit age of no younger than 16, an intermediate-license minimum age of 16.5, nighttime driving restrictions from 10 p.m. to 5 a.m., and passenger limits during the intermediate stage (NHTSA teen driving page).
The teen fatality numbers show why those rules matter. In 2024, 2,320 people were killed in crashes involving a teen driver ages 15-18, and 752 teen drivers died in crashes (NHTSA teen driving page). Those are not abstract counts; they represent families, passengers, and other road users affected by inexperience and high-risk conditions.
Older adults show a different pattern. Drivers aged 70+ have higher crash death rates per 1,000 crashes than middle-aged drivers aged 35-54 (CDC older adult drivers page). CDC also notes that older adults are more likely to have safer driving behaviors than other age groups, but night driving and poor weather increase the likelihood of crash injuries and deaths for older adults (CDC older adult drivers page).
That means age-related risk is not a simple story about one group being uniformly unsafe. Risk changes with context: time of day, weather, exposure, and driving conditions.
Seat Belts and Restraint Use
Seat belts are one of the clearest examples of a low-cost behavior with a large safety payoff. CDC says seat belts reduce serious crash-related injuries and deaths by about half (CDC seat-belt facts). That is a large effect for a single behavior.
The dataset also gives a look at usage rates. Seat belt use was 92% in states with primary enforcement laws in 2022 and 90% in states with secondary enforcement laws or no seat belt laws in 2022 (CDC seat-belt facts). As of September 2023, 29 states did not have a primary enforcement law covering all seating positions for adults (CDC seat-belt facts).
National usage numbers are high, but still not perfect. In 2024, seat belt use was 91% for drivers nationwide and 91% for right-front seat passengers nationwide (IIHS seat belts page). Rear-seat belt use was 80% nationwide in 2023 (IIHS seat belts page).
That rear-seat number matters. When restraint use is lower in the back seat, the risk can be shifted onto passengers who may mistakenly assume the rear seat is inherently safe without a belt. The statistics do not support that assumption.
The dataset also adds the cost side of the issue. More than 1.8 million drivers and passengers were treated in emergency departments for motor-vehicle-crash injuries in 2021, and non-fatal crash injuries to drivers and passengers caused over $220 billion in lifetime medical, work, and quality-of-life loss costs in 2021 (CDC seat-belt facts). This is why restraint use is not just about surviving a crash. It is also about reducing the long tail of injuries, disability, and economic harm.
What the 2024 Estimates Suggest
The 2024 estimate set is encouraging, but it should be read carefully.
The fatality rate improved to 1.19 deaths per 100 million VMT in FHWA FI-200, and NHTSA early estimates say the rate fell to 1.20, the lowest since 2019 (FHWA FI-200; NHTSA 2024 early estimates). At the same time, 39,254 deaths is still a very high number, and the state-level pattern was mixed, with 35 states and Puerto Rico improving but 14 states and the District of Columbia worsening (NHTSA 2024 early estimates).
That combination suggests three things:
- The U.S. can reduce risk even while people drive more.
- Progress is uneven across states.
- The major behavior risks remain large enough to keep fatalities elevated.
The 2024 number should not be treated as a finished endpoint. It is better understood as a partial recovery from the worst recent years, not a return to the safer levels seen in the mid-2010s.
How to Read Driving Statistics Responsibly
A good driving-statistics read needs more than one number. Here is the practical way to interpret the dataset without over-reading it.
Use rates, not just totals
Fatalities can rise because more people are exposed, more miles are driven, or risk per mile is worse. That is why deaths per 100 million VMT are so useful. The rate makes years with different travel volumes more comparable (FHWA FI-200).
Separate broad trends from behavior-specific risks
A national fatality decline does not mean speeding, alcohol, distraction, or teen driving stopped being dangerous. The dataset shows each of those behaviors still drives substantial harm (NHTSA speeding page; NHTSA distracted-driving page; NHTSA drunk-driving page; NHTSA teen driving page).
Look at injury counts too
Deaths are the most severe outcome, but injuries reveal scale. Distracted driving injuries in 2024, and the 1.8 million crash-related emergency department treatments in 2021, show how large the nonfatal burden remains (NHTSA distracted-driving page; CDC seat-belt facts).
Watch the state split
National averages can hide local gains and local reversals. The 2024 estimate that 35 states and Puerto Rico improved while 14 states and the District of Columbia worsened shows why state-level reporting matters (NHTSA 2024 early estimates).
The overall picture is straightforward: the U.S. has made real progress in some periods, but the road safety system still carries large losses from speeding, distraction, alcohol, teen inexperience, and inconsistent restraint use. Those are the statistics that matter most when the goal is not just to count crashes, but to reduce them.