Twenty-one of 29 rural counties (72%), 41 of 73 randomly selected high schools (56%; of the 217 eligible schools), and 4,731 of the 9,391 eligible students (50%) participated in the study. At 1-year follow-up, 3,072 students participated for 65% retention. Year in high school (52% retention for senior, 62% for juniors, 68% for sophomores, and 72% for freshman), race/ethnicity (46% retention for Black, 56% “other” race/ethnicity, 72% for Asian/Pacific Islander, 63% for Hispanic/Latino, 66% non-Latino White, and 65% Native American), baseline ST use (52% retention for user and 66% for nonuser), perception of harm (55% for none/low and 67% for moderate/great), baseball (64% retention for nonparticipant and 70% participant), and soccer (64% retention for nonparticipant and 71% participant) were significantly related to retention in a multivariable GEE model. The intervention group had 63% retention, while the no-intervention group had 67% retention (GEE p = .713; multivariable GEE model p = .316). Baseline ST use among retained participants was 7.6% (5.3%–9.8%) in the control group and 8.0% (4.9%–11.2%) in the intervention group, while among those lost to follow-up, baseline ST use was 12.9% (10.5%–15.2%) in the control group and 13.5% (9.9%–17.2%) in the intervention group. A GEE interaction test of Group × Retention as it related to baseline ST use was nonsignificant (p = .941).
Overall, most ST users in our study used dip exclusively or both dip and chewing tobacco (40% each); used ST on at least 5 days in the past week; used Copenhagen, a brand of ST offering a high level of bioavailable nicotine (Hoffmann et al., 1995
); and used ST within 30 min of waking, suggesting nicotine dependence (Boyle, Jensen, Hatsukami, & Severson, 1995
; Ebbert, Patten, & Schroeder, 2006
). The mean age ST users reported first having tried ST was 12.1 years, and the mean age they began regular ST use was 13.5 years (data not shown).
shows the baseline characteristics of the study sample by randomization group. The majority of students were White, followed by Latino and other Hispanic, Asian/Pacific Islander, Native American, and Black. Overall baseline prevalence of ST use, smoking, and combined cigarettes and ST use were 9%, 14%, and 6%, respectively (data for overall not shown in table). Three of the 24 baseline characteristics—Asian/Pacific Islander, year in high school, and playing soccer—differed significantly between the two randomization groups at baseline; although none would be considered statistically significant with a Bonferroni multiple comparison correction (α* = .002), subsequent analyses were also performed adjusting for year in high school.
Baseline characteristics of study sample by group (N = 4,731)
shows 1-year follow-up by randomized group: the overall prevalence of ST use, the percent initiation of ST use among baseline non-ST users, and the percent of ST cessation among baseline ST users. There were no significant differences between the intervention and no-intervention groups overall, in initiation or in quitting (all ORs’ 95% CIs include 1.0).
One-year follow-up prevalence of overall ST use, ST initiation, and ST cessation in prior 30 days by randomization group (complete and imputed data)
Subgroup analysis, however, showed that baseline smoking was a significant intervention effect modifier: Baseline Smoking × Group yielded p = .019 and adding year in high school as a covariate did not change that result. In addition, ST use intensity (low/moderate/heavy), first ST use after waking (≤30 min/>30 min), year in high school, and high school type (continuation/regular) were not significant effect modifiers (all p > .861).
shows 1-year follow-up ST use by baseline smoking status: overall prevalence of ST use, ST initiation, and ST cessation among males who were baseline nonsmokers and baseline smokers. A highly significant intervention effect was seen in reported ST quitting among the baseline nonsmokers (p < .001): 62% of those who used ST but did not smoke at baseline in the intervention group reported quitting compared with 36% of those in the no-intervention group. These results remained when adding year in high school as a covariate.
One-year follow-up prevalence of ST use, ST initiation, and ST cessation in prior 30 days by randomization group and baseline smoking status (complete and imputed data)
Of baseline ST users who smoked, 40% used ST within 30 min of waking compared with 22% of exclusive baseline ST users (p = .011). The number of days per week of ST use was similar between baseline dual users and baseline ST only users (p = .747). Thus, ST users who smoked at baseline appeared to be more addicted than exclusive baseline ST users but did not differ in the number of days of ST use per week. shows the percentage of exclusive baseline smokers (i.e., non-ST users) and exclusive baseline ST users (i.e., nonsmokers) who reported using another form of tobacco at 1-year follow-up. Exclusive baseline ST users (i.e., baseline nonsmokers) reported a significantly higher percentage smoking at follow-up than exclusive baseline smokers (i.e., baseline ST nonusers) reported using ST at follow-up. (Exact binomial 95% CIs do not overlap, demonstrating statistical significance.) Thus, ST use appears to have facilitated initiation of smoking in this adolescent population.
One-year initiated using tobacco in new form follow-up of baseline exclusive ST users and exclusive smokers
No adverse events were reported during the study.