OBJECTIVES: The purpose of this study was to identify socio-environmental and personal variables associated with high school students smoking behavior by applying multilevel analyses. METHODS: A cross-sectional survey of the first-year students of five public senior high schools in western Kanagawa Prefecture, Japan was conducted using multilevel logistic regression analyses with students at level 1 and schools at level 2. Self-administered questionnaires were returned by 517 out of 597 enrolled students, and information was collected regarding the prevalence of previous and current smoking and socio-environmental, educational and personal variables. RESULTS: The rates of past and current smoking were found to be 25.2% and 12.6% in males and 16.9% and 5.2% in females, respectively. Maternal smoking, having friends or older school-mates who smoked and lower probabilities of high school academic achievement potential were significantly associated with both past (adjusted ORs of 2.37, 4.28 and 2.98, respectively) and current (adjusted ORs 2.46, 5.57 and 3.02) smoking. CONCLUSIONS: It is recommended that health professionals in charge of school-based educational programs should tailor the teaching methods to fit the students' backgrounds and specific vulnerabilities. Educational programs for smoking prevention focusing on the students' mothers and classmates or students in high schools with reduced academic achievement potential should be developed.
OBJECTIVES: The purpose of this study was to identify socio-environmental and personal variables associated with high school students smoking behavior by applying multilevel analyses. METHODS: A cross-sectional survey of the first-year students of five public senior high schools in western Kanagawa Prefecture, Japan was conducted using multilevel logistic regression analyses with students at level 1 and schools at level 2. Self-administered questionnaires were returned by 517 out of 597 enrolled students, and information was collected regarding the prevalence of previous and current smoking and socio-environmental, educational and personal variables. RESULTS: The rates of past and current smoking were found to be 25.2% and 12.6% in males and 16.9% and 5.2% in females, respectively. Maternal smoking, having friends or older school-mates who smoked and lower probabilities of high school academic achievement potential were significantly associated with both past (adjusted ORs of 2.37, 4.28 and 2.98, respectively) and current (adjusted ORs 2.46, 5.57 and 3.02) smoking. CONCLUSIONS: It is recommended that health professionals in charge of school-based educational programs should tailor the teaching methods to fit the students' backgrounds and specific vulnerabilities. Educational programs for smoking prevention focusing on the students' mothers and classmates or students in high schools with reduced academic achievement potential should be developed.