I thought that as well, but this study appears not to have found that

> and there were no effects of the fall-back transition to Standard Time (ST) on MVA risk

EDIT: It seems our assumption was correct, at least for the UK:

> There is substantial evidence that fewer people would be killed and seriously injured on Great Britain’s roads if this country were to put the clocks forward by one hour throughout the year. The Department should take the lead in re- examining the practice of changing clocks at the end of British Summer Time with other central Government departments.

I wonder why the OP study (using US data) found something different?

Source: https://publications.parliament.uk/pa/cm200809/cmselect/cmpu...

If there's anything to learn from reports like this, it would be that people should be allowed to live on more flexible schedules. Not just around DST changes, but in general. The obsession modern society has with clocks, schedules and calendars bothers me anyway, it assumes every person is equally capable of adapting to what is considered a 'normal' (ie: majority) schedule, which is provably false. But no report will ever investigate e.g. how many traffic accidents would be prevented by not forcing everyone to do the school run at the exact same time and be in the office before 9:00.

Restricting the topic to just DST changes: considering modern society sadly is obsessed with clocks, DST is IMO the best way to maximize how much of the waking hours of people are during sunlight. DST itself isn't the problem, the problem is that apparently people don't get enough time to adjust. So why not introduce a national holiday twice a year, call it 'DST day', so everyone gets an extra day to adjust. I guess that would go a long way to alleviate the negative effects. Obviously then all of a sudden it's not about fatal traffic accidents anymore, but about economic effects and money... :-(