Okay. I'm not a scientist. If an officer stops 200 people for a variety of reasons, in 100 of them he suspects impairment; and in 50 of those 100 he suspects alcohol or alcohol and drug impairment, and in the other 50 he suspects drug impairment. He then asks those 100 drivers to undergo the roadside sobriety test.
What you explained to Mr. Ménard was that in the 50 in which he suspected alcohol alone or alcohol and drug impairment, the roadside sobriety test confirmed an impairment in 72% of the cases, meaning that at that level—just at that level—the officer was right in 72% of the cases. He then goes on to do the breathalyzer, etc., and the percentages, from what you have said, go up.
In the case of suspicion of drug impairment, with 50 drivers, the officer conducts the roadside sobriety test. Do you have any studies about what percentage the officer then goes on to require be further tested, through the DRE? Is it 25 out of the 50? Is it 45 out of the 50? Is it 10? Has any study been done on that?