Another community-based prevention program, The Midwestern Prevention Project, was designed to decrease drug use among high-risk adolescents (Chou, Montgomery, Pentz, Rohrbach Anderson, Johnson, Flay & MacKinnon, 1998). The data for this study came from students at schools in Kansas City and Indianapolis. Sixth and seventh graders (n = 3,412) were split into control (n = 1,508) and experimental (n = 1,904) groups. However, only those who reported smoking cigarettes (control = 188, experimental = 212), using alcohol (control = 290, experimental = 323) or using marijuana (control = 38, experimental = 22) were included in the analysis. The experimental group was given a 10-session program that emphasized social skills and drug use resistance skills in 6th or 7th grade. The parents of the experimental group children were taught positive parent-child interaction strategies; community leaders were instructed on how to institute a community-based prevention strategy; and a mass media campaign was launched. For the control group, only community leaders were instructed on how to institute a community-based prevention plan and a mass media campaign was launched. The analysis demonstrated that the prevention strategy reduced cigarette smoking and alcohol use at the 2.5 year follow-up, but by the 3.5 year follow-up there were no significant effects on reducing the prevalence of use, compared to the control group. This result could point to the fact that many adolescents go through a developmental trajectory in which they exhibit some deviant behaviors in early to mid-adolescence, but the deviant behaviors naturally dissipate by the end of adolescence or that alcohol becomes normative by late adolescence. A separate analysis of these data was conducted by Johnson, Pentz, Weber, Dwyer, Baer, MacKinnon, Hansen, and Flay (1990). The results suggest that the prevention strategy reduced the prevalence rates of marijuana and cigarette use in 9th and 10th grade (i.e., at the three year follow-up). However, across both the experimental and control groups prevalence rates increased over the three year period, but not as drastically among the experimental group participants. Having a more intensive strategy for alcohol use might result in significant long-term effects. The results suggest that this community-based program has medium term effects.