Study
Gansky and colleagues (2005) used a cluster-randomized controlled trial to assess the impact of the spit (smokeless) tobacco intervention on collegiate baseball athletes. Eighty-seven colleges were contacted to participate in the study; 59 agreed to participate, but 7 were dropped from the study. Fifty-two California colleges participated in the study, for a total of 883 participants in 27 intervention colleges and 702 participants in 25 control colleges. Schools were stratified on the basis of spit-tobacco use prevalence.
The sample was 70 percent white, 17 percent Latino, 4 percent Asian American, 3 percent African American, and 2 percent each for multiethnic Native American and for “other.” Most athletes were between the ages of 17 and 20.
Spit-tobacco use was assessed over the course of 1 year through self-report. Data was collected on demographic factors, alcohol and lifetime tobacco use, current spit-tobacco use, and type and brand of spit tobacco used. On the follow-up survey, data was collected on tobacco cessation methods tried in the previous year. Analyses to assess group differences were conducted using multivariable logistic regression models for clustered responses. Eighty-one percent of eligible athletes participated in the baseline survey. Seventy-nine percent (1,248 participants) of the original 1,585 athletes recruited completed the 12-month survey. Ninety-two percent (48 of the 52) of the athletic trainers responded to the 1-year follow-up survey. No subgroup analyses were conducted.
Study
Walsh and colleagues (2003) used a cluster-randomized controlled trial to assess the impact of the intervention on spit-tobacco use cessation rates and on initiation rates. The study was conducted in rural areas of California and recruited 516 participants in 22 intervention schools and 568 participants in 22 control schools. Schools were stratified by baseline number and size of the baseball teams, as well as by baseline prevalence of spit-tobacco use. An eligible high school was required to have a baseball team with at least a 20 percent estimated baseline spit-tobacco use prevalence, according to the coach, and at least a 10 percent actual baseline prevalence, as determined by responses to questionnaires administered to team members before randomization.
The control sample was made up of 27.7 percent seniors, 34.3 percent juniors, 28.9 percent sophomores, and 9.0 percent freshman. The intervention sample consisted of 29.1 percent seniors, 40.4 percent juniors, 17.0 percent sophomores, and 13.5 percent freshman. Parental consent was obtained for students to participate in the study.
Self-report was used to measure prevalence of cessation and initiation at baseline and over 1 year. Saliva samples were collected at baseline and 1-year postintervention, and assays were performed on a random subsample of 8 percent of the spit-tobacco nonusers. Analyses used multivariate logistic regression models for clustered responses. Subgroup analyses were conducted to examine the impact grade level and smoking status had on the cessation of spit tobacco use.