We examined factors associated with a reduction of smoking behavior among adolescent daily smokers from a longitudinal school-based cohort study of New Hampshire 8-10th grade students. Cessation behavior was identified by a response on at least one subsequent survey indicating occasional or nonsmoking. We determined associations between cessation behavior and baseline measurements of level of addiction (daily cigarette consumption), years since smoking initiation, opinion of adults smoking more than one pack of cigarettes per day, prior experience with quitting, intent to quit, school performance and sociodemographics. Of 160 adolescents who qualified as daily smokers at baseline, 15 (9.4%) subsequently described themselves as nonsmokers, and 9(5.6%) as occasional smokers. The smoking cessation rate (including only subsequent nonsmokers) was 0.063 per person year. Compared with light smokers(1-9 cigarettes per day), heavy smokers (>10 cigarettes per day) had a significantly longer smoking history, a higher likelihood of past failed quit attempts (23 vs. 58% respectively), a higher intent to continue smoking (61 vs. 80% respectively), and were less likely to disapprove of adults smoking more than one pack of cigarettes per day (82 vs. 61% respectively).

Cessation behavior was associated only with lower frequency of smoking at baseline (10% of heavy smokers vs. 21% of light smokers, p <0.05) and disapproval of adult smoking (26% for disapprovers vs. 11% for those who did not, p < 0.05). Intent to quit measures, including prior cessation experience, the desire to stop smoking now, and intent to be a nonsmoker within one year exhibited no association with subsequent cessation behavior.

Conclusions: Adolescents quit smoking at rates comparable with adults. As with adults, a higher level of addiction is associated with less cessation behavior and a greater likelihood of past failed quit attempts. In contrast with adults, adolescent smokers in this sample were unable to anticipate cessation behavior. This finding casts doubt on the predictive validity of the Transtheoretical Model of Change (Prochaska and DeClemente) regarding the smoking cessation process in adolescents.