COVID happened in the year of the discontinuity and caused major changes to commuting behavior as a result of remote work, people afraid of infection avoided mass transit, many people moved out of cities or lost their jobs, people bought cars who didn't used to drive and now there are more new/inexperienced drivers with cars (and it's easier to get a license in the US than Europe), etc.

Also, the numbers for at least the US are apparently just wrong:

https://en.wikipedia.org/wiki/Motor_vehicle_fatality_rate_in...

1.27 fatalities per 100M VMT in 2023 (the latest year with data), 1.11 in 2010, that's a difference of 14%, not 30%. Even the peak during COVID was only 24% above 2010. The only way I can see to get 30% is to use the during-COVID number for only the total number of motor vehicle fatalities without accounting for population growth or vehicle miles traveled, which is not a great metric for making comparisons.

The 30% figure is "correct" if you look at the absolute number of deaths instead of deaths per VMT. But I basically agree with you; that clearly the wrong stat to cite if you are attributing the change to vehicle safety regulations.

Even that is still wrong because you'd have to use the high water mark during COVID and not the more recent numbers which are starting to come back down.

Because a lot of people stopped driving and leaving their home so much during that time.