Child cycling deaths fell 88 percent. Adult deaths rose fivefold.
Since 1975 the American cyclist most likely to be killed changed from a twelve-year-old to a fifty-year-old. Almost nothing about how we talk about cycling safety has caught up.
The Insurance Institute for Highway Safety keeps the long series on this, back to 1975. Two lines run in opposite directions and they are both dramatic.
Deaths among cyclists under 20 have fallen 88 percent since 1975. Deaths among cyclists aged 20 and over have risen almost fivefold. In 2024, 89 percent of the Americans killed riding a bike were adults.
Two different things happened
The child figure is one of the largest sustained safety improvements in American road history, and it deserves to be said out loud before the caveat. Fewer children die on bikes now than at almost any point in the last fifty years.
The caveat is that a good part of it is not a safety achievement. Children stopped riding. The collapse in independent childhood mobility over the same period — the disappearance of the ride to school, the ride to a friend's house — removes deaths by removing exposure. Some of that 88 percent is better helmets, slower residential streets and safer vehicles. Some of it is an empty street.
The adult line has no such ambiguity available. Cycling by adults rose over the same period, so exposure rose too, but a fivefold increase in deaths is steep even against that.
The picture inverted geographically as well
In 1975 cyclist deaths were split roughly evenly between rural and urban areas. In 2024, 81 percent were urban.
Put the two shifts together and the American cyclist killed in traffic went from a child on a residential or rural road to an adult on an urban one. IIHS puts 66 percent of 2024 deaths on major roads other than interstates, with 29 percent on minor roads. The killing happens on the arterial: the four-lane road with a 45 mph limit that somebody has to ride on because it is the only route across town.
Why the mismatch matters
Cycling safety messaging in the United States is still largely inherited from the era when the victims were children. Helmets, visibility, riding predictably, obeying signals: all of it aimed at the rider, most of it developed when the rider in question was twelve.
Those things are not worthless. But an adult commuter on a 45 mph arterial at seven in the evening is not failing to obey a signal. They are in a place where a moment's inattention by a driver is fatal, and no amount of rider education changes the geometry of that road.
IIHS reports that 68 percent of those killed in 2024 were not wearing a helmet, and that figure gets quoted heavily. It is worth reading carefully: 19 percent were recorded as unknown, which is a large uncertainty, and helmets are designed for impact energies well below those of a car at arterial speed. It is a real finding about a real protective device, and it is not an explanation for the fivefold rise.
The measurement gap, again
What would help is knowing which arterials, and how bad they are on an ordinary Tuesday rather than on the day somebody died. A road that carries the only bridge across a river will show up in a fatality table once a decade, and in the lived experience of every rider who uses it every single week.
Sources: IIHS, Fatality Facts: Bicyclists (2024 data); NHTSA, Traffic Safety Facts: Bicyclists and Other Cyclists, 2024 Data (DOT HS 813 822, July 2026).