Design and Setting
Voluntary, self-administered survey of 1265 sixth- through 12th-grade students representing 79% to 95% of all students attending 5 rural New Hampshire and Vermont public schools in October 1996. We examined the association between ownership of a CPI and smoking behavior through regression models and conducted a sensitivity analysis on the findings.

Main Outcome Measures
Adjusted odds of being a smoker (lifetime use of 100 cigarettes) and, among never smokers and experimental smokers, adjusted cumulative odds of having higher levels of smoking uptake given CPI ownership.

Results
One third of students owned a CPI. Prevalence of ownership did not vary by grade or sex, but was higher among poor-to-average school performers (45.0% vs 21.0% for excellent school performers, P<.001) and children whose friends and family members smoked (43.4% vs 13.8% for students with no family members or friends smoking, P<.001). Cigarette promotional items included articles of clothing (T-shirts, hats, backpacks, and jackets), smoking paraphernalia (lighters and ashtrays), camping gear, and electronics. More than half of CPIs (58.2%) bore the Marlboro logo, and almost one third (31.7%) bore the Camel logo. These items were obtained directly from catalogs or vendors 22.4% of the time. Whereas only 4.5% of students reported bringing a CPI to school with them the day of the survey, 44.5% reported seeing such an item at school the day of the survey. After controlling for confounding factors, such as having friends who smoke, students who owned CPIs were 4.1 times more likely to be smokers than those who did not own CPIs (95% confidence interval [CI], 3.1-5.5). Never and experimental smokers (n=1008) who owned CPIs were more likely to be in a higher category on the smoking uptake index in grades 6 (cumulative odds ratio [OR]=5.7,95% CI, 1.9-16.8), 7 (OR=1.8,95% CI, 0.9-3.7), 8 (OR=2.3, 95% CI, 1.1-4.8), and 9 (OR=2.1, 95% CI,1.1-3.9), periods when children are most vulnerable to initiating cigarette use. A sensitivity analysis indicated that an unmeasured confounder of CPI ownership and smoking was unlikely to alter our conclusions.

Conclusions
Cigarette promotional items are owned by one third of students in these rural northern New England schools. These items are highly visible in the public school setting, and their ownership is strongly associated with initiation and maintenance of smoking behavior. These data lend support to a ban on CPIs to be included in US Food and Drug Administration regulations to prevent tobacco use among US youth.

Arch Pediatr Adolesc Med. 1997;151:1189-1196