I note with interest that you neglected to mention the recent increase.
I also note with interest that robertpaulsen is very selective about which data to look at. Two rows higher on the same chart shows drug usage increasing quite steadily from 31.3% in 1979 to 41.7% in 2001. Those stats are for all persons who have ever tried an illicit drug. Or, at least, all persons who have ever tried an illicit drug and actually 'fessed up to the pollster. Not very indicative of a working ban to me...
That's what it is, a recent increase.
Given that in the last 24 years, drug use declined 60% then remained flat for 10 years, do you consider an increase in the last year or two significant? Obviously, I don't.