PT - JOURNAL ARTICLE AU - Margaret R. Becklake AU - Heberto Ghezzo AU - Pierre Ernst TI - Childhood predictors of smoking in adolescence: a follow-up study of Montréal schoolchildren AID - 10.1503/cmaj.1041428 DP - 2005 Aug 16 TA - Canadian Medical Association Journal PG - 377--379 VI - 173 IP - 4 4099 - http://www.cmaj.ca/content/173/4/377.short 4100 - http://www.cmaj.ca/content/173/4/377.full SO - CMAJ2005 Aug 16; 173 AB - Background: The factors that cause children to become smokers in adolescence remain unclear. Although parental smoking and peer pressure may play a role, physiological factors such as lung volume have also been identified. Methods: To investigate these and other possible childhood predictors of teenage smoking, we gathered follow-up data on 191 Montréal schoolchildren, aged 5–12 years (average 9.2 yr) when first examined. At an average age of 13.0 years, they answered further questions on their health and smoking behaviour and provided a second set of spirometric measurements. Results: At the second survey, 80% of the children had entered high school and 44% had become smokers. Reaching puberty between the surveys was the most significant determinant of becoming a smoker: 56.4% of the 124 children postpubertal at the second survey had taken up smoking, versus 17.9% of the 67 who were still prepubertal (p = 0.001). We found salivary cotinine level, a measure of uptake of environmental tobacco smoke, to be an independent predictor of becoming a teenage smoker; even after adjustment for sex, socioeconomic status of parents, a crowding index, and the numbers at home of siblings, adult smokers and cigarettes smoked, it remained significant for both groups: postpubertal (odds ratio [OR] 1.2, 95% confidence interval [CI] 1.2–3.0) and prepubertal (OR 2.1, 95% CI 1.0– 4.5). The influence of forced vital capacity was marginally significant only in the postpubertal group (OR 5.0, 95% CI 0.88–28.3). Interpretation: The proportion of nicotine absorbed from that available in environmental tobacco smoke during childhood is associated with subsequent smoking in adolescence. The more efficient absorption of nicotine seen in some children may be related to physiological factors such as lung capacity.