A multilevel analysis examining the association between school-based smoking policies, prevention programs and youth smoking behavior: evaluating a provincial tobacco control strategy

This paper examined how smoking policies and programs are associated with smoking behavior among Grade 10 students (n = 4709) between 1999 and 2001. Data from the Tobacco Module from the School Health Action Planning and Evaluation System were examined using multilevel logistic regression analyses. We identified that (i) attending a school with smoking prevention programs only was associated with a substantial risk of occasional smoking among students with two or more close smoking friends and (ii) attending a school with both smoking prevention programs and policies was associated with substantial risk of occasional smoking among students who did not believe there were clear smoking rules present. Students attending schools where year of enrollment in high school starts in Grade 9 were more likely to be regular and occasional smokers. Each 1% increase in Grade 12 smoking rates increased the odds that a Grade 10 student was an occasional smoker. It appears that grade of enrollment, senior student smoking behavior, close friend's smoking behavior and clear rules about smoking at school can impact school-based tobacco control programming. These preliminary study findings suggest the need for further research targeting occasional smoking behavior and the transition stage into high school.     

A randomized intervention of smoking for adolescents in urban Wuhan, China

BACKGROUND: Tobacco use is a significant public health problem in China. Culturally specific smoking prevention programs are needed for Chinese adolescents. This study evaluated a school-based smoking prevention curriculum with a social normative approach developed in the United States for adolescents in urban Wuhan, China. METHODS: As a randomized trial, the intervention was implemented in 1998 with 7th grade students in seven schools with seven matched control schools. Multilevel logistic regression models were used to compare ever and recent (past-month) smoking behaviors for the control and program conditions. RESULTS: At the 1-year follow-up, smoking had increased more rapidly in the control schools than in the program schools. The odds of baseline nonsmokers initiating smoking did not differ between the program and control groups (OR=1.08 with 95% CI=0.71, 1.64). The program prevented progression to recent smoking among boys who were baseline ever smokers. Among boys who were recent smokers at baseline, the prevention program significantly reduced risk of remaining recent smokers at follow-up (OR=0.45 with 95% CI=0.23, 0.88). CONCLUSIONS: This social normative smoking prevention curriculum did not demonstrate a significant primary prevention effect but showed potential for secondary prevention. Culturally specific smoking prevention programs are needed for Chinese adolescents.     

Tobacco smoking among Portuguese high-school students.

The aim of this study was to evaluate the prevalence, behavioural patterns, and determinants of smoking among a large sample of high-school students from Porto, the second largest city in Portugal, information on sociodemographic characteristics and personal history of tobacco, alcohol, coffee, and illicit drug use was obtained from 2974 students, aged 12-19 years (48.7% female, 51.3% male), using an anonymous self-administered questionnaire. Crude and adjusted odds ratios (OR) were calculated by logistic regression analysis to estimate the association between smoking and the characteristics evaluated. Overall, 35.8% students had never smoked, 39.4% had tried it ("experimental" smokers) but were not smokers, 3.3% were former smokers, 6.6% occasional smokers, and 14.9% regular smokers. The mean age for starting smoking was 13.4 +/- 2.1 years for males and 13.4 +/- 1.6 years for females. The prevalence of current smoking was higher among males than females, but the difference was not significant. Male students were significantly more likely to smoke more cigarettes per day than were females. The prevalence of smoking was significantly associated with the following variables: being aged > 12 years; having parents who had attended school for < 4 years; having a mother (OR = 1.88), siblings (OR = 1.96) or friends (OR = 1.75) who smoked; low academic performance (OR = 1.74 for one or two failures and OR = 2.27 for more than two failures at school); and consumption of coffee (OR = 2.90), alcohol (OR = 3.53), or illicit drugs (OR = 6.69). The prevalence of smoking among adolescents increased with age. There is therefore a need for school-based tobacco prevention programmes which also deal with family influences on smoking.     

A multilevel analysis examining the relationship between social influences for smoking and smoking onset

OBJECTIVES: To examine how older smoking peers at school and the smoking behavior of friends and family members are related to youth smoking. METHODS: The School Smoking Profile was used to collect data on tobacco use and determinants of tobacco use from 22,091 students from 29 secondary schools in Ontario, Canada. Correlates of occasional and regular smoking were examined using multilevel logistic regression analyses. RESULTS: Students are at increased risk for smoking if they (a) have smoking friends, (b) have smoking family members, and (c) attend a school with a relatively high senior-student smoking rate. CONCLUSION: These findings suggest that prevention programs should target both at-risk schools and at-risk students.     

Cognitive factors associated with smoking initiation in adolescents

OBJECTIVE: To study the association between cognitive factors of the behavioral change model "Attitude Self Efficacy" (ASE) at different phases of smoking initiation among adolescents. METHODS: We carried out a cross-sectional survey among students in the second grade of Compulsory Secondary Education (13-14 years old) from Cornellà de Llobregat (Barcelona, Spain) in 2000 to obtain information on cognitive factors and smoking. Logistic regression analysis was used to investigate the variables associated with smoking (odds ratio [OR] of experimenters vs. non-smokers and of smokers vs. experimenters). RESULTS: The prevalence of daily smoking was 22.9% (95% CI, 16.5%-29.3%) among boys and 36.2% (95% CI, 29.7%-42.6%) among girls. Factors associated with experimenting (vs. non-smoking) were: attitudes to smoking (disagreement with smoke-free areas [OR = 3.46; 95% CI, 1.65-7.24], agreement with smoking promotion [OR = 3.42; 95% CI, 1.42-8.28]), and subjective norms (perceiving friends as smokers [OR = 2.50; 95% CI, 1.17-5.35]). The variables associated with regular smoking (vs experimenting) belong to: self-efficacy and attitudes to smoking. CONCLUSIONS: Focussing on subjective norms and smoking attitudes with programs targetted younger ages seems appropriate, since these factors are more closely associated with the experimenting phase. Encouraging skills to refuse cigarettes offered by friends is appropriate at a more advanced age, since this determinant is associated with the change from experimenting to regular smoking.     

Youth smoking risk and community patterns of alcohol availability and control: a national multilevel study

STUDY OBJECTIVE: To test whether college youth smoking risks are independently associated with community patterns of alcohol availability and control. DESIGN: Hierarchical multilevel multivariable modelling of cross sectional survey data. Outcomes included self reported current (past 30 day) cigarette smoking and heavy episodic (binge) drinking. SETTING: 120 nationally representative US colleges. PARTICIPANTS: 10 924 randomly selected students. MAIN RESULTS: Individual risks for smoking and binge drinking are independently associated with community patterns of alcohol availability, policy enforcement and control over and above individual perceptions about these factors, student and college characteristics, and school binge drinking rates. Youth exposed to high levels of alcohol availability are at higher risk of smoking (OR 3.61, 95% CI 1.75, 7.44) and binge drinking (OR 4.22, 95% CI 2.25, 7.93) than youth not so exposed; youth exposed to strongly enforced alcohol policy environments are at lower risk for smoking (OR 0.30, 95% CI 0.16, 0.57) and binge drinking (OR 0.17, 95% CI 0.10, 0.31) than youth not so exposed; youth exposed to communities with strong parental controls are at lower risk for smoking (OR 0.05, 95% CI 0.01, 0.23) and binge drinking (OR 0.06, 95% CI 0.01, 0.21) than youth not so exposed. Individual risks related to environmental exposures differ for youth with varying perceptions about alcohol availability and policy control. CONCLUSIONS: Drinking environments in US college communities comprise strong independent risks for smoking. Smoking prevention models should be tested that include environmental drinking prevention strategies tailored to underlying perceptions and experiences of college youth.     

The influence of friends, family, and older peers on smoking among elementary school students: low-risk students in high-risk schools

BACKGROUND: This study examined how older smoking peers at school and the smoking behaviour of friends and family members are related to youth smoking. METHODS: Multi-level logistic regression analysis was used to examine correlates of ever smoking in a sample of 4286 grade 6 and 7 students from 57 elementary schools in Ontario, Canada (2001). RESULTS: Each 1% increase in the smoking rate among grade 8 students increased the odds that a student in grades 6 or 7 was an ever smoker versus never smoker [OR 1.05, 95% CI 1.02 to 1.08]. A low-risk student (no family or friends who smoke) was almost three times more likely to try smoking if he/she attended an elementary school with a relatively high prevalence of senior students who smoke than if he/she attended a school with a low prevalence of senior students who smoke. CONCLUSION: Low-risk grade 6 and 7 students are at significantly greater risk of smoking if they attend an elementary school with a relatively high prevalence of smoking among senior students. Prevention programs should target both at-risk schools and at-risk students.     

Associations between individual and contextual factors and smoking in 13,293 Mexican students

OBJECTIVE: Factors correlated with cigarette smoking in young people have yet to be documented in most developing countries. This study assesses the correlates of smoking in Mexican young people. METHODS: School-based, cross-sectional study in the central Mexican state of Morelos during the 1998-1999 school year of 13,293 public school students aged 11 to 24 years. Multinomial logistic regression models were constructed with smoking as the dependent variable. RESULTS: Regular smoking (one or more cigarettes daily) prevalence was 13.1% (95% confidence interval [CI]=12.2-13.9) in males, and 6.1% (95% CI=5.6-6.6) in females. Frequent alcohol intoxication was strongly associated with regular smoking (females, odds ratio [OR]=68.5, 95% CI=37.6-125.2; males, OR=34.5, 95% CI=22.6-52.7). Regular smoking was associated with illegal drug use and smoking by both parents in females, and with illegal drug use in males (males, OR=4.9, 95% CI=3.7-6.5). Also associated with tobacco smoking were high socioeconomic status, low academic achievement, illegal drug use by peers, marijuana use by parents, and depression in adolescents. CONCLUSIONS: This study documents a strong correlation between tobacco smoking and other health risk behaviors, especially alcohol and drug abuse. In young women especially, the risk of tobacco use increased with alcohol abuse and higher socioeconomic status. School-based interventions are needed that focus on preventing smoking and also take into account other unhealthy behaviors.     

A Multi-Level Analysis Examining how Smoking Friends, Parents, and Older Students in the School Environment are Risk Factors for Susceptibility to Smoking Among Non-Smoking Elementary School Youth

The purpose of this study was to examine how social models for smoking are related to smoking susceptibility among a sample of non-smoking elementary school students. The Tobacco Module of the School Health Action, Planning and Evaluation System (SHAPES) was administered to 6,431 students (grades 6 to 8) in 57 elementary schools in the province of Ontario, Canada. Multi-level logistic regression analysis was used to examine how smoking friends, parents, and the prevalence of smoking among grade 8 students at a school were related to smoking susceptibility among the 2,478 non-smoking grade 6 and 7 students. Findings indicate that non-smoking grade 6 and 7 students are more likely to be susceptible to smoking if they have (a) smoking friends, (b) a mother who smokes, or (c) two or more close friends who smoke and attend a school with a relatively high smoking rate among the grade 8 students. Sub-populations of non-smoking youth may be at increased risk for smoking because of the elementary school they attend. Future school-based smoking prevention programs might benefit from targeting prevention programming activities to the schools that are putting students at the greatest risk for smoking.     

A period of increased susceptibility to cigarette smoking among high school students

Cross-sectional surveys of randomly selected high school students were conducted in October 1999 and in May 2000 to measure awareness of youth-led tobacco prevention efforts. A secondary goal of the surveys was to learn about tobacco use and attitudes. Respondents who never smoked cigarettes were asked if they had made a firm commitment to not smoke, and respondents who had not made such a commitment were considered susceptible. Among ninth graders, susceptibility increased from 31% in October to 47% in May, while susceptibility decreased from October to May for students in more advanced grades (p = 0.03, interaction of linear trends). Susceptibility was more common among students who felt smoking produced social benefits (OR 1.59, 95% CI 1.22-2.08) or who were tolerant of tobacco company behavior (OR 1.65, 95% CI 1.23-2.21), and less common among students who felt short-term adverse effects of cigarette smoking on health or hygiene were important to them (OR 0.74, 95% CI 0.56-0.97). Further studies should confirm the increase in susceptibility and its underlying mechanisms. Whether associated with ninth grade or with the first year in a new school, this previously unrecognized period of heightened susceptibility might represent an important opportunity for prevention.     

Factors associated with smoking progression among Spanish adolescents

Smoking onset takes place in the early years of adolescence, and can be seen as a progression through stages of preparation, trying, experimentation, regular smoking and nicotine dependence or addiction. The objective of this study is to identify 1-year predictors of smoking consolidation among occasional smokers (experimenters) and of cessation among all smokers. A questionnaire developed to monitor trends in smoking and other health-related behaviors was administered in the spring of 1992 and again 1 year later (1993). The survey was carried out in six large, private schools in Barcelona and Raimat, in the Lleida province, in Catalonia (Spain). In total, 1460 schoolchildren aged 12-19 responded to both the baseline and follow-up questionnaires; 1236 questionnaires (84.7%) were matched through a personal code and were used for the longitudinal study. At the baseline survey 28% of the schoolchildren were regular smokers and 17.1 % were occasional smokers, with no differences by gender. At the follow-up survey, progression from occasional to regular smoking was observed in 42% of the girls and 22.7% of the boys. Other predictors of consolidation included drinking alcohol (OR = 3.1; 95% CI = 1.1-10.1) and reporting the intention to smoke in the future (OR = 2.5; 95% CI = 1.3-4.9). Among all the smokers at the baseline, predictors of cessation were smoking occasionally (versus regularly) (OR = 4.9; 95% CI = 2.8-8.6), negative attitudes regarding smoking (OR = 3.3; 95% CI = 1.9-5.4), reporting no intention to smoke in the future (OR = 2.2; 95% CI = 1.3-3.8), gender (OR for boys = 2.1; 95% CI = 1.2-3.6), receiving weekly pocket less than 1000 ptas (OR = 1.7; 95% CI = 1.1-5.3) and age (OR for being 15 and younger = 1.6; 95% CI = 1.1-4.9). We conclude that consolidation of smoking is twice as likely among girls than among boys. Intention to smoke seems to be a good predictor of future behavior, while attitudes may predict cessation, but not consolidation. Available pocket money seems to be a strong risk factor for smoking at this age.     

Exposure to, and perceived usefulness of, school-based tobacco prevention programs: associations with susceptibility to smoking among adolescents.

PURPOSE: To examine the relationships of exposure to school-based tobacco prevention programs and perceived usefulness of such programs to smoking susceptibility. METHODS: An ethnically representative sample of 6929 Grade 10 students participated in a cross-sectional survey of tobacco-related activities and behaviors. Perceived usefulness of prevention programs was investigated according to two dimensions: perceived helpfulness of information received at school with regard to making decisions about tobacco use and perceived availability of social resources that might help students cope with tobacco-related issues. RESULTS: The extent of exposure to school prevention programs and both dimensions of perceived usefulness of these programs were inversely associated with susceptibility to smoking. In addition, perceived information helpfulness mediated the effect of program exposure on susceptibility status, whereas perceived availability of social resources moderated that effect. The study also presented evidence suggesting ethnic/cultural variability in the status of susceptibility to smoking at this age. CONCLUSIONS: The study indicates a need to examine the perceptions of students receiving school prevention programs. Perceived usefulness of school prevention programs may be used to identify youth at high risk for smoking susceptibility. Future studies, however, should investigate other dimensions of perceived usefulness as well as intrapersonal, interpersonal, and environmental correlates of this construct.     

青春期性发育对青少年吸烟行为的影响

目的 研究青春期性发育早晚对青少年吸烟行为的影响。方法 2000年11月对随机整群抽取的1453名上海市高中学生进行横断面问卷调查，获取青少年吸烟，男孩首次遗精年龄，女孩月经初潮年龄及可能影响青少年吸烟的社会心理因素等信息。结果 男生的尝试吸烟率在性发育较晚组中最高（28．6％），其次为发育较早（21．3％）和中等组（21．7％）；现在吸烟率在性发育较早组最高（16．4％），然后依次为发育较晚（7．1％）和中等组（4．8％）；女生的曾经吸烟率在性发育早，中，晚各组中依次为24．2％，12．0％，6．3％。在控制影响青少年吸烟的一系列社会心理因素后，性发育较早是导致男生现在吸烟的独立危险因素（OR＝3.68,95%,CI:1.23-10.99)，也是女生吸烟的危险因素（OR＝2.23,95%,CI:0.89-5.60)；而性发育较晚可能导致男生尝试吸烟（OR＝1．48，95％CI：0．72－3．06），对女性却可能有一定的保护作用（OR＝0.64,95%CI:0.22-1.86)。结论 学校，家庭和社会应该关心，引导性发育时间偏离正常的男女青少年，加强吸烟危害健康教育。     

Nine-year prospective association between older siblings' smoking and children's daily smoking.

PURPOSE: To explore the hypothesis that older siblings' smoking, after controlling for parents' smoking, influences children's smoking. METHODS: Study participants were 2981 students in the control cohort of a school-based smoking prevention randomized trial for whom parents' smoking and older siblings' smoking data were collected at 3rd grade and daily smoking data was assessed 9 years later through a self-report questionnaire at the 12th grade. Data were analyzed using conditional logistic regression and likelihood ratio tests. RESULTS: For families in which no parent smoked, the 12th grade prevalence of daily smoking was 31% when at least one older sibling smoked compared to 18% when no older sibling smoked. For families in which at least one parent smoked, the 12th grade prevalence of daily smoking was 41% when at least one older sibling smoked compared to 29% when no older sibling smoked. There was a substantial increase in the odds (OR = 1.60, p =.004) of children's daily smoking at 12th grade when their older siblings smoked, even after adjusting for parents' smoking. CONCLUSIONS: Older siblings' smoking is associated with increased risk of children's smoking after adjusting for parents smoking. Furthermore, there was no evidence to suggest that the influence of older siblings' smoking was different in families where no parent smoked compared to those where at least one parent smoked. Also, there was no evidence that the influence of older siblings' smoking was different in boys vs. girls.     

Symptoms of nicotine dependence and other predictors of student smoking at school: implications for school smoking policy

Students who violate school smoking policies present a difficult health and disciplinary issue for school officials. Professionals know little about the characteristics of students who smoke at school. In a prospective study of 679 students in two cities in central Massachusetts, researchers examined how nicotine dependence contributes to the problem of smoking at school. After three years of follow up, smoking at school was reported by 10.3% of students. Among subjects who admitted to smoking at school, 63% reported that symptoms of nicotine dependence preceded their smoking at school. After adjusting for other variables, student smokers with symptoms of nicotine dependence were nine times more likely to report smoking in school (OR 9.1, 95% CI 2.9-28.5) than were student smokers without symptoms. Smoking at school was more common among daily smokers and those who paid for their own cigarettes. Age, gender, race, and parental smoking status were not significantly associated with students' reports of smoking at school. These data suggest nicotine dependence as an important contributor to the problem of smoking at school, but not the only reason why students violate school smoking policies. Disciplinary action against students caught violating school smoking policies should be supplemented with an offer of treatment for nicotine dependence.     

Peers, schools, and adolescent cigarette smoking.

PURPOSE: To investigate the effects of popularity, best friend smoking, and cigarette smoking within the peer networks on current smoking of seventh- through 12th-grade students. These factors were examined for adolescents attending schools with varying rates of student cigarette smoking. METHODS: This study used data from the saturated school sample of National Longitudinal Study of Adolescent Health (Add Health), a nationally representative school-based sample. The sample for the present study was 2525 adolescents in Grades 7-12 who completed an in-school questionnaire and an in-home interview. Information from the in-school questionnaire was used to construct measures of school smoking prevalence and popularity. Using peer nominations from the in-home interview, best friend smoking, and peer network smoking exposure were constructed using the peers' own reports of their cigarette smoking. Multiple regression techniques were used to estimate the risk of current cigarette smoking as a function of popularity, best friend smoking, peer network smoking, and school smoking prevalence, and all first-order interactions between measures of peer and school smoking prevalence. RESULTS: Adjusting for age, gender, race/ethnicity, parent education, school, and availability of cigarettes in the home, the risk of current smoking was significantly associated with peer networks in which at least half of the members smoked [odds ratio (OR) = 1.91], one or two best friends smoked (OR = 2.00), and with increasing rates of school smoking prevalence (OR = 1.73). In addition, there was a significant interaction of popularity and school smoking prevalence such that risk of current smoking was somewhat greater among popular students in schools with high smoking prevalence than among popular students in schools with low smoking prevalence. CONCLUSION: Findings suggest that school environments are important contexts for understanding peer group influences on adolescent cigarette smoking.     

Senior student smoking at school, student characteristics, and smoking onset among junior students: a multilevel analysis

BACKGROUND: Current research on the etiology of tobacco use has largely focused on identifying the influential psychosocial characteristics of individual students; the influences of characteristics in the school environment are generally ignored. The purpose of this study was to simultaneously examine how school and individual student characteristics were related to smoking onset. METHOD: Multilevel logistic regression analysis was used to examine how the senior student smoking rate at a school and the psychosocial characteristics of students were able to differentiate tried-once smokers from experimental smokers in a sample of 4850 grade 9, 10, and 11 students from the School Smoking Profile (SSP) project. RESULTS: Each 1% increase in smoking rate among high school seniors increased the odds that a junior student was an experimental smoker vs. a tried-once smoker (OR 1.07, 95% CI 1.03-1.12). A significant contextual interaction was identified where the senior student smoking rate at a school moderates the negative influence of having close friends who smoke. Influential student characteristics were also identified. CONCLUSIONS: The smoking prevalence of older students at a school is directly related to smoking onset among younger students at that school. Prevention programs should target schools that put students at-risk.     

Predictors of the onset of tobacco consumption among secondary school students of Barcelona and Lleida

BACKGROUND: Knowing the factors predisposing the smoking onset of smoking at school age may help to improve the preventive strategies and make the starting age later. This study is set out to describe the natural history of smoking throughout the teen-age years and the main factors which may predict teenagers developing the smoking habit. METHODS: This is a longitudinal study of a survey type, using the school room as the sampling unit at six schools in Barcelona and Lleida. A total of 1,460 students within the 12-19 age range took part. Two surveys were conducted, an initial survey in the spring of 1992 and a follow-up survey one year later. A total of 1,236 questionnaires (84.3%) were paired by means of a personal code used for the longitudinal study. RESULTS: A total of 25% of the students were regular smokers at the start of the study, 17.1% smoking occasionally or having experimented with smoking, there being no differences between sexes for both of these categories. Among those who were non-smokers at the starting survey, 12.7% stated that they smoked occasionally and 6.2% regularly one year later. The females started smoking at an earlier age, the differences as compared to males being statistically significant at 14 years of age. The predictors of the smoking onset include favorable attitudes toward cigarettes (OR = 3.1; CI5% (1.6-5.6), drinking alcoholic beverages (OR = 2.2; CI95% (1.4-3.3), being younger than age 15 (OR = 2.0; CI95% (1.2-3.4), having friends who smoke (OR = 1.9, CI95% (1.2-3.0) and the stated intention of smoking in the future (OR = 1.6; CI95% (1.0-2.5). CONCLUSIONS: The students showing a higher degree of tolerance toward cigarette smoking, entailing favorable attitudes or friends who smoke or who intend to smoke in the future at the start of the study are at a much higher risk of Smoking onset than their classmates who are not of these characteristics. The cigarette smoking prevention programs must focus special attention on the teenagers most highly vulnerable to these factors.     

心理压力及疲劳对大学生不同吸烟行为的影响

目的研究心理压力及疲劳对大学生不同吸烟行为的影响。方法采用整群抽样法抽取某大学1～3年级在校大学生8138名，以自填式结构问卷调查学生的心理压力、自身和同伴吸烟情况及社会经济学指标，Chalder疲劳量表测量其躯体和精神疲劳。以“从未吸烟者”作为对照组，分别以“尝试吸烟者”和“现吸烟者”作为病例组，进行Multinomial logistic回归分析心理压力及疲劳对大学生不同吸烟行为的影响。结果进入大学后仍有学生尝试或形成吸烟行为。调整社会经济学指标和同伴吸烟行为的影响后，躯体疲劳（OR=1．044，95％CI：1．013～1．075）、对学业和未来工作的担心（OR=1．020，95％CI：1．001～1．040）是男生尝试吸烟的危险因素；学习压力（OR=1．063，95％CI：1．012～1．117）是女生尝试吸烟的危险因素；人际与环境压力（OR=1．152，95％CI：1．012～1．312）是女生现吸烟的危险因素。结论心理压力及疲劳和大学生吸烟行为存在一定程度的关联，且对不同性别和不同吸烟行为的影响是不同的。     

Effect of a Smoking Ban and School-Based Prevention and Control Policies on Adolescent Smoking in Spain: A Multilevel Analysis

We evaluated the impact of a smoking ban in schools and of school-based smoking prevention and control policies on adolescent smoking. Annual surveys carried out between 2001 and 2005 that were representative of students in the 4th year of secondary education in the Madrid region, with 203 schools and 9127 students participating. The student questionnaire gathered information about personal and family variables. The contextual factors were: the periods before (years 2001?C2002) and after the law; and through a survey of school management boards: compliance with the law, policy reflected in the school regulations, existence of complaints against smoking, and undertaking of educational activities regarding smoking. Multilevel logistic regression models were constructed with two dependent variables: current smoking and the proportion giving up smoking. Smoking declined in 2003, the first year after the law came into force (Odds ratio: 0.80; CI 95%: 0.66?C0.96), and this decline was maintained in 2005. By contrast, smoking increased in those schools that did not undertake educational programmes regarding smoking (Odds ratio: 1.34; CI 95%: 1.13?C1.59), and in those that received complaints about smoking (Odds ratio: 1.12; CI 95%: 0.96?C1.29). This association is partly due to the effect of the increase in giving up smoking. The inclusion of contextual variables into the model with the individual factors reduces the variability of smoking between schools by 32.6%. In summary, the coming into force of a law banning smoking in schools, and the implementing of educational policies for the prevention and control of smoking are related to a lower risk of adolescent smoking.