Preventing smoking and other drug use: let the buyers beware and the interventions be apt.

Those concerned with smoking prevention programs in the schools may be misled by a recent report that "we know what works; now let's make it happen." The advocated "social influences" programs appear to have no reliable effects on regular (greater than or equal to weekly) smoking and only short-term and small effects (5-8 percentage points) on "experimental" (less than weekly) smoking. To prevent smoking and other drug abuse among the children most at risk, it may be crucial to use broad-based, multidisciplinary interventions that go well beyond the health education classroom.