America Got Safer in 2024. Unless You're Over 65.

NHTSA dropped its 2024 full-year estimates last month, and the headline number was genuinely good: 39,345 traffic deaths, marking the first time since 2020 that the annual count fell below 40,000, capping eleven consecutive quarterly declines and pushing the fatality rate down to 1.20 per 100 million VMT.^[3] By any honest reading of the data, that qualifies as progress.

So I ran the subcategory breakdowns by age group, expecting the decline to distribute more or less evenly across demographics. Every adult cohort between 15 and 64 tracked the overall trend downward: the 25-to-34 bracket dropped 8 percent, teens fell 4 percent, and middle-aged drivers saw decreases of 6 to 7 percent across the 45-to-64 range.

Then one row in the table broke the pattern entirely.

Americans 65 and older: 8,081 estimated deaths, up 2 percent from 7,923 in 2023, which is not a rounding error or a statistical wobble within the confidence interval but a directional divergence from a national trend that moved every other demographic group in the opposite direction. Seniors are now the single largest age-group death toll in the NHTSA data, accounting for 20.5 percent of all traffic fatalities, up from 19.4 percent just one year prior.^[1]

Roughly twenty-two seniors die on American roads every day, and while the rest of the country gets incrementally safer, this group is drifting away from the curve in a direction that the structural data makes disturbingly easy to explain.

Aging Metal

NHTSA's own subcategory data splits passenger vehicle occupant deaths by vehicle age, and the split is damning. In 2024, vehicles ten years or older accounted for 62.4 percent of all passenger vehicle occupant deaths: 13,964 out of 22,395, with deaths in newer vehicles dropping 8 percent year over year while deaths in older vehicles dropped only 5 percent.^[1] Safety dividends from modern crash structures, automatic emergency braking, and electronic stability control are demonstrably real, but they only protect you if you are sitting inside a vehicle that was built with those features rather than one that predates them by a decade.

Cross-referencing FARS fatal crash data from 2014 through 2023 sharpens the picture considerably. Vehicles with the highest occupant-death-per-fatal-crash ratios read like a retirement community parking lot: the Buick LeSabre at 82.3 percent, Mercury Grand Marquis at 76.1 percent, Buick Century at 76.8 percent, Cadillac Seville at 69.4 percent.^[2] When a LeSabre appears in a fatal crash, the person inside the LeSabre dies more than four out of five times, a ratio that turns the vehicle from transportation into an actuarial liability. Compare that to the Ram 2500 at 20.5 percent or the Ford F-250 at 36.4 percent, where the structural mass of the vehicle absorbs the energy that would otherwise kill the occupant and redistributes it, often lethally, to whoever they hit instead.

Why the Gap is Widening

Three forces are compounding simultaneously, and none of them are slowing down. The 65-plus population is growing: the Census Bureau projects 77 million Americans will be 65 or older by 2034, up from 58 million in 2022, which means more elderly drivers, more elderly passengers, and more elderly pedestrians sharing road infrastructure that was engineered around the reaction times and visual acuity of younger demographics. Older Americans also retain their licenses longer and drive more miles than previous generations did at the same age, a cultural shift that puts more vulnerable bodies behind more steering wheels for more hours per year. And they are disproportionately driving older vehicles with pre-modern safety equipment, vehicles like a 2014 Buick LeSabre that sits on the road in 2024 as a ten-year-old car without lane departure warning, without automatic emergency braking, without pedestrian detection, and possibly without electronic stability control depending on trim, carrying a five-star NHTSA rating from a decade ago that was tested under a protocol indifferent to the crash types that actually kill its driver demographic.

Meanwhile, the 25-to-34 cohort that saw the steepest decline of any age group in 2024 (down 8 percent, from 7,613 to 6,989) is exactly the demographic most likely to be driving vehicles under five years old with the full suite of active safety technology, which means the gains are not distributing equally but concentrating in the part of the population that can afford to buy them.

Counter-Evidence

A 2 percent increase on a base of 7,923 is 158 additional deaths, and given that NHTSA's subcategory estimates use inflation-adjusted FARS projections, the confidence interval could plausibly contain zero change. NHTSA themselves note that "estimated fatalities by subcategories may vary due to the continuous updating of 2024 FARS data,"^[1] so it is possible that when the final 2024 FARS annual report file publishes, the 65-plus number will flatten to roughly even with 2023, which would still represent a divergence from the 4 to 8 percent declines seen in every other adult age group but would weaken the claim that senior fatalities are actively rising rather than merely stagnating.

An aging population also mechanically inflates raw counts: a flat per-capita rate with a growing population produces more deaths without any worsening of individual risk, and without age-adjusted fatality rates per VMT for the 65-plus cohort specifically, the raw count increase could reflect nothing more than demographic growth rather than deteriorating safety outcomes for any given senior on the road.

Limitations

This analysis has genuine blind spots that should temper the conclusions. FARS only captures fatal crashes, so we cannot assess whether the injury rate among seniors is improving even as fatalities increase, a distinction that matters because improved emergency medicine could be saving more lives without reducing the crash rate itself. The death-per-crash ratios from FARS reflect all occupant ages rather than exclusively elderly occupants, so attributing the LeSabre's 82.3 percent ratio to senior vulnerability specifically requires an inference about who drives LeSabres rather than direct age-linked crash data from the same vehicle. The vehicle age split in NHTSA's subcategory report does not break down by driver age, which means we cannot confirm from this dataset alone that the 65-plus deaths are disproportionately occurring in older vehicles, even though the demographic correlation between vehicle age and driver age is well-documented elsewhere. Our FARS data covers 2014 to 2023 while the NHTSA estimates are for 2024, creating a one-year gap in the cross-tabulation that the final FARS release will eventually close.

What to Do

If you have a parent or grandparent driving a vehicle manufactured before 2018, check the IIHS ratings at iihs.org/ratings for their specific model and year. Look for vehicles without ESC (pre-2012 mandate for cars, pre-2012 for SUVs), without AEB (not standard until the 2022 voluntary commitment), and without side-curtain airbags. Run their VIN through nhtsa.gov/recalls for open recalls. A 2014 sedan with an unfixed Takata airbag inflator is not a "perfectly good car." It is a vehicle whose primary safety system may fragment and kill its occupant in the crash it was designed to survive.

A newer vehicle is the cheapest intervention. A 2020 anything with ESC, side-curtain airbags, and AEB is structurally a different class of protection than a 2010 anything without them. Median transaction price for a used 2020 sedan is roughly $18,000 to $22,000. That is not trivial. But it buys the occupant-protection technology that appears to be driving the 8 percent fatality decline in the under-ten-year vehicle cohort. Whether the person who can least afford the upgrade is also the person who needs it most is a question this data can't answer, but it's the one that should keep transportation policymakers awake.

Sources & References

1. NHTSA, Early Estimates of Motor Vehicle Traffic Fatalities and Fatality Rate by Sub-Categories in 2024 (DOT HS 813 729), May 2025. crashstats.nhtsa.dot.gov
2. NHTSA, Fatality Analysis Reporting System (FARS), 2014–2023. nhtsa.gov. Death-per-crash ratios computed from FARS fatal crash involvement counts and occupant fatality counts.
3. NHTSA, Early Estimate of Motor Vehicle Traffic Fatalities in 2024 (DOT HS 813 710). nhtsa.gov
4. IIHS, Vehicle Size and Weight. iihs.org
5. IIHS, Vehicle Ratings. iihs.org
6. NHTSA, Recalls Database. nhtsa.gov/recalls