Indeed. But the party line is that the problem lies in the substance, not the people.
I've never seen it stated that way.
But it just makes sense that if a certain percentage of people who use Substance A go on to try Substance B, reducing the number of people who use Substance A would result in fewer people going on to try Substance B.
(Granted, there are people who are using Substances X or Y or Z (or no substances) that go on to try Substance B, but that's not the subject of this article.)