Older driver crash statistics: myths, facts and insurance

Drivers in their 70s now crash less often per licensed driver than middle-aged drivers do, but they are more likely to die when they crash, and that fragility is why fatality rates per mile climb from about age 70. Neither "older drivers are dangerous" nor "older drivers are just as safe as everyone else" matches the research. Insurers price older drivers on claims, not on fatality headlines, and the claims data are kinder to people in their late 60s and 70s than most people expect.

Myths and facts at a glance

The sections below explain each line and where the numbers come from.

Three ways to count, and why they disagree

Every statistic about older drivers answers a different question depending on its denominator.

• Per capita divides deaths by the number of people in an age group. It reflects how often a person of that age dies in traffic, whether as driver, passenger, pedestrian or cyclist.
• Per licensed driver divides crash involvement by the number of people holding licenses. It describes the risk that comes with having a license, regardless of how much someone drives.
• Per mile traveled divides crashes by the distance driven. It measures risk in each mile on the road, but it depends on mileage estimates from travel surveys, which are less precise than license counts.

Older people drive fewer miles, so per-licensed-driver and per-mile figures can point in different directions. The Insurance Institute for Highway Safety (IIHS), in its older drivers fact page updated July 2026, reports that fatal crash involvement rates per licensed driver for drivers in their 70s fell below those of drivers aged 35 to 54 for the first time in 2015, while per-mile fatal crash rates start to rise at about 70 (IIHS). Both statements are true because the measures differ. A person who drives 3,000 miles a year and a person who drives 15,000 can have the same per-mile risk and very different per-license risk.

A worked example with invented numbers

These numbers are hypothetical and only illustrate the arithmetic.

Say town A has 1,000 licensed drivers aged 35 to 54 who together drive 13 million miles a year, and 1,000 licensed drivers aged 75 to 79 who together drive 6 million miles a year. Suppose 4 of the middle-aged drivers and 3 of the older drivers are involved in a fatal crash.

• Per licensed driver: 4 per 1,000 for the middle-aged group and 3 per 1,000 for the older group. The older group looks safer.
• Per million miles: about 0.31 for the middle-aged group (4 divided by 13) and 0.50 for the older group (3 divided by 6). The older group looks riskier.

Same town, same crashes, opposite headlines. Neither is wrong, and each answers a different question. The first is the better guide to how likely a person with a license is to be in a fatal crash. The second is the better guide to the risk of an additional trip.

Fragility, not recklessness

IIHS research found that the elevated per-mile fatal crash rates of older drivers in 2005 to 2008 were driven mainly by fragility, the risk of dying in a crash, rather than by over-involvement in crashes, and that similar results held in more recent data covering 2018 to 2024. Fragility rises from about middle age and keeps rising with age (IIHS).

An October 2020 IIHS release makes the same point from the other side. Drivers 70 and over are still more likely to die in a crash than younger drivers because they are physically more vulnerable; the number of deaths per 1,000 police-reported crashes fell about 25% for drivers 70 to 79 from 2009 to 2017, compared with about 15% for drivers 35 to 54 (IIHS news release, October 2020).

This matters for insurance. An insurer pays claims for property damage and injuries, and fragility changes the severity of injury to the older driver, not necessarily the frequency of claims. Greater injury severity to the older driver is mainly a matter for their own health insurance and any medical payments or personal injury coverage; it does not by itself make the driver more likely to hit someone else, and it does not change the cost of repairing a car under collision coverage.

What changed for people in their 70s

The same IIHS release reports that fatal crash rates per licensed driver fell 43% for drivers 70 and over from 1997 to 2018, compared with 21% for drivers 35 to 54, and that drivers 70 and over had fewer police-reported crashes per mile than middle-aged drivers for the first time in 2017. Most of the decline in the older group came in the first half of the period; since then the rate has held steady for older drivers and risen for middle-aged ones.

The release lists several explanations researchers offered:

• Better health that delays age-related declines in eyesight and cognition and makes crash survival more likely.
• Safer vehicles, including side airbags and better crash-test ratings.
• Road changes such as easier-to-see signs and roundabouts, partly spurred by federal training and a 1998 handbook.
• Licensing policy changes, cited as a possible contributor.
• An economic effect: crashes fall in recessions, which may have affected middle-aged drivers more.
• Older drivers speed and drive impaired less often.

The IIHS vice president for research noted that survival in crashes should keep improving as newer safety features reach more older people, because older adults tend to keep vehicles longer and the gains arrive slowly. The upshot for anyone comparing generations is that a 72-year-old today is a different risk than a 72-year-old in 1997.

The crashes where older drivers are over-represented

Older drivers are not equally likely to be in every kind of crash. The IIHS page says that multiple-vehicle intersection crashes accounted for 43% of fatal crashes for drivers 80 and older in 2024, compared with 24% for drivers aged 16 to 59. It adds that older drivers are more likely than younger ones to be involved in angle, overtaking, merging and especially intersection crashes, that failure to yield is the most common error they make in crashes, and that inadequate surveillance was the most frequent error for drivers 70 and older in serious crashes.

An often-cited study of intersection crashes shows how the mechanics differ by age. Researchers at IIHS compared 227 drivers who were at fault in nonfatal injury intersection crashes: 78 aged 70 to 79, 76 aged 80 and older and 73 aged 35 to 54. Failure-to-yield crashes rose with age and mostly happened at stop-sign intersections, usually while the driver was turning left. The 70-to-79 group most often saw the other vehicle but misjudged whether there was time to go; the 80-and-older group mostly failed to see the other vehicle; and the middle-aged group more often made search errors from distraction. Drivers 80 and older had fewer rear-end crashes than the other groups, and both older groups had fewer run-off-road crashes than the 35-to-54 group (IIHS bibliography entry, Braitman and colleagues, 2007).

The study is small and dates from 2007, so treat its percentages with caution, but its pattern matches the later fact page: left turns across traffic and stop-sign gaps are the hard part, and fixes like protected left-turn signals and roundabouts help. The IIHS page reports that roundabout conversions cut injury crashes by 76% in an early study, and that a survey of drivers living near six roundabouts found 65% of those 65 and older favored them.

Who dies in crashes involving older drivers

A headline can make older drivers sound like a threat to others. The data point mostly the other way. IIHS reports that of the people killed in 2024 crashes involving drivers 70 and older, 57% were the older drivers and 13% were their passengers, which leaves 30% who were other road users or occupants of other vehicles (IIHS).

The federal numbers show the burden on older people as road users. The National Highway Traffic Safety Administration's 2023 fact sheet on the older population (age 65 and older) reports 7,891 deaths in that group in 2023, 19% of all traffic fatalities, a rate of 13.32 per 100,000 population, down 5% from 2022. Over 2014 to 2023, deaths in the group rose 38% while the group's population rose 28%, and the per-100,000 rate rose from 12.40 to 13.32, while the rate for people under 65 rose from 9.90 to 11.92 over the same years. Older drivers made up 22% of licensed drivers in 2023, up from 18% in 2014. NHTSA also reports that in crashes involving older drivers, fatalities were more likely to be during the daytime (70%), on weekdays (68%) and in multi-vehicle crashes (68%) than all fatalities (46%, 58% and 46% respectively), and that pedestrian deaths among people 65 and older rose 56% between 2014 and 2023 (NHTSA, Older Population: 2023 Data, May 2025).

The takeaway is that the rise in counts partly reflects a larger older population with more licensed members, and the rates of older and younger people both rose over the decade, so the trend is not unique to older people.

What insurers actually price on

Insurers do not buy their rate from fatality tables. They use their own claims experience, grouped by rating factors such as age band, driving record, mileage, vehicle, location and coverage. how car insurance rates are calculated explains the process in full. For older drivers, the relevant evidence is claim frequency and cost.

IIHS reports, using Highway Loss Data Institute data for calendar years 2021 to 2025, that drivers aged 65 to 69 have the lowest property damage liability and collision claim rates per insured vehicle year, that the rates rise after about age 70, and that they remain far below those of the youngest drivers (IIHS). So in many plans an older driver's premium reflects some increase for age, but the base is a low one, and a clean record, moderate mileage and a modern vehicle can matter as much as the birthday. How the increase unfolds from 65 to 85 depends on the insurer and state, and how car insurance rates change from age 65 to 85 explains it.

Three further points:

• Age rules are state-specific. Some states restrict or prohibit age as a rating factor, so a pattern seen in one state does not apply everywhere. Check your state's insurance department or the car insurance for seniors guide.
• Fragility and claims are different things. Injuries to the older driver would be paid through their own medical or personal-injury coverages, which are rated separately from the liability that pays for harm to others.
• Mileage can offset age. An older driver who drives little can look better to a mileage-sensitive rating plan than a younger commuter. See annual mileage for how it is measured.

Whether an insurer can cancel or decline to renew because of age is a separate question; can an insurer drop you for being old? covers it.

Vehicle technology and what it changes

IIHS summarizes several findings about safety features and older people. Side airbags were estimated to reduce nearside fatalities by 45% for front-seat occupants 70 and older, compared with 30% for ages 13 to 49, a 2013 estimate; older drivers tend to drive vehicles less likely to have them. Seat belts and frontal airbags work about equally well at all ages. IIHS researchers estimated in 2023 that front crash prevention, blind spot detection and lane departure warning together relate to up to a fifth of older drivers' crash involvements and a quarter of their fatalities, and a 2017 IIHS study found rearview cameras and parking sensors especially effective for drivers 70 and older in preventing reported backing crashes. The Highway Loss Data Institute (HLDI) found that all age groups see lower claim rates with front crash prevention and lane departure warning, but drivers under 65 benefit more than those 65 and older (IIHS).

The practical reading is mixed. Features are valuable, but they do not guarantee savings or safety, and older drivers often have not had access to the newest ones. Whether they lower a premium depends on the insurer's filing; see vehicle safety ratings and insurance costs and the article driver assistance features for older drivers: what helps, what pays.

Licensing policy and what it does and does not do

States set different renewal rules for older drivers, and the evidence on them is mixed. IIHS says that in-person renewal or vision testing is associated with lower fatality rates among drivers 85 and older (a 2014 study), while road testing, knowledge testing or shorter renewal periods were not associated with lower fatality rates per licensed driver for drivers 55 and older. A 2025 study associated longer renewal intervals for drivers 75 and older with higher injury rates and some remote renewals for drivers 65 to 74 with increased crash rates. IIHS also reports that Illinois' road-test requirement was associated with lower claim rates for drivers 75 and older in a 2016 HLDI study, but with fewer older people being insured, suggesting some stopped driving, and that Illinois raised the starting age for road tests from 75 to 87 effective July 1, 2026 (IIHS). For the state-by-state rules, see older driver license renewal rules by state.

The takeaway is not that testing is pointless. It is that age alone is a blunt instrument, and the best evidence supports in-person contact and vision screening more than blanket road tests.

How to read an older-driver headline

Use this sequence on any article or report.

1. What is the denominator? Count, per 100,000 population, per licensed driver or per mile. If the article does not say, the claim is not yet interpretable.
2. Which ages? 65 and older, 70 and older and 80 and older are very different groups. A statistic for "seniors" often blends a healthy 66-year-old with a frail 88-year-old.
3. Who was killed? Deaths of older drivers and their passengers are not the same as danger to others.
4. Which years? Trends from the 1990s may describe people who are no longer the typical older driver. IIHS's 1997 to 2018 comparison is a good illustration.
5. Is it fatal crashes or all crashes? Fragility inflates fatal statistics; claim rates and police-reported crashes tell a different story.
6. Who is the source? IIHS, NHTSA and the Highway Loss Data Institute publish methods. A press release from a company selling a service rarely does.

Common questions

Are older drivers more dangerous than teenagers?

No. Teen drivers have far higher claim rates, and IIHS reports that older drivers' claim rates stay far below those of the youngest drivers. Older drivers face higher risk of dying in a crash, which is a different thing.

Why do premiums often rise in the late 70s and 80s if claim rates are low?

Claim rates in the IIHS data begin to rise after about age 70, and rating plans pass part of that rise on in price. The pattern varies by insurer and state. a first at-fault accident after decades of clean driving explains how a first at-fault accident is treated.

Do these statistics apply to me personally?

No statistic describes an individual. Group averages say little about a particular driver's eyesight, reaction time or health. If a family member is worried about someone's driving, a clinical or driving-rehabilitation evaluation is a better guide than any chart.

Where can I find the original data?

IIHS maintains fact pages on older drivers and fatalities, and NHTSA publishes annual fact sheets on the older population. Both state their methods and data years. The figures here are as published in 2025 and 2026, and later editions may differ.

Key takeaways

• Per licensed driver, drivers in their 70s now have lower fatal crash rates than middle-aged drivers; per mile, fatal rates rise from about age 70, mostly because of fragility.
• Most people killed in crashes involving a driver 70 or older are that driver or the driver's passengers, not other road users.
• Older drivers are over-represented in intersection crashes, especially failure-to-yield left turns, which is why design and driver-assistance fixes target them.
• Insurers price on claims data, where drivers 65 to 69 have the lowest claim rates and rates stay far below those of young drivers after 70.
• When you read a headline, check the denominator, the age range, who was killed and the years covered.