Here are traffic deaths by age group: https://www.cdc.gov/mmwr/volumes/68/wr/mm6806a8.htm. Let's pad those numbers up by a significant amount and assign 20 per 100,000 from birth, and remove those deaths from the life tables (ie let's say noone dies from traffic accidents in the USA). That still only lifts life expectancy by 1 year (from 78.6 to 79.6)
You can keep adding other parts of accidental deaths but: 1) they don't fit the story of USA dying more from driving 2) the leftovers are already below the accidental death rate from countries with much higher life expectancy 3) a big chunk of the rest is accidental falls which predominantly affects very old people so doesn't have a big impact on life expectancy
I wince when I hear 1 hour commute times. The risk of injury and death from driving 10k miles per year versus 25k miles per year is material in my opinion.
The primary source of injury in young people creating this situation? Car accidents. And remembering how me and my friends were behind the wheel at that age, this is completely unsurprising. Young people in many other countries do not drive nearly as much.
Everyone looks for medical reasons but it is largely fatal accident rates.