The effects of an intervention designed to prevent onset of weekly drinking in non drinking students (PAS) were investigated in the group of students that was already drinking at baseline. A cluster randomized trial was used including 3,490 Dutch early adolescents (M age = 12.66, SD = 0.49) randomized over four conditions; 1) parent intervention, 2) student intervention, 3) combined intervention and 4) control group. Outcome measures were amount and growth of weekly alcohol drinking measured 10, 22 and 34 months after baseline. The combined intervention significantly curbed the growth of drinking among both non-drinkers (the target group of the intervention) and drinkers at baseline. Overall, less strong increases of drinking over time are found among non-drinkers compared to drinkers at baseline. Thus, the combined PAS intervention is also effective in curbing adolescents’ drinking behaviour in those who already were drinking at baseline. Broad implementation of the combined parent–student intervention is recommended.