> I was guessing it was any amount of positivity, and may be close to the population level, but it's actually impairment levels of THC:

A lot of people are trying to debate the impairment threshold or argue about mean vs median, but 40% of deceased drivers having this much THC in their blood would be a notable result for basically any sample of people for anything other than a music festival or something.

The number of people age 12 or older who report any THC use at all, even once, in the past year is around 20% (or less depending on the survey). Having 40% of a group register levels this high is a very eye opening result.