One useful point of comparison here would be the percent of the driving population overall who have some THC in their system in the same way as these researchers are measuring it. I wouldn’t guess that 40% of drivers would test positive for recent THC use, but I can’t understand the 40% number here without knowing the percent for the overall population.
The number of people who use any THC at all isn’t even close to 40% thought. The highest survey numbers I found had 20% of people reporting any use, even once, in the past year.
There’s no way to normalize a result of 40% of a population sample having significant THC concentrations. That’s way higher than any conceivable sample of the general population.
There sure is, when the survey question boils down to "have you committed a Federal crime?"