Health Education Research, Vol. 19, No. 3, 217-226,
June 1, 2004
© 2004 Oxford University Press
National and school policies on restrictions of teacher smoking: a multilevel analysis of student exposure to teacher smoking in seven European countries
1 Research Centre for Health Promotion, University of Bergen, 5015 Bergen, Norway, 2 Child and Adolescent Health Research Unit, University of Edinburgh, Edinburgh EH8 8AQ, UK and 3 Health Promotion Division, Welsh Assembly Government, Cardiff CF10 3NQ, UKThe views expressed in this paper are those of the authors and not the Welsh Assembly Government 4 Correspondence to: B. Wold; e-mail: bente.wold{at}psyhp.uib.no
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionConclusionReferences |
|---|
The paper examines the association between restrictions on teacher tobacco smoking at school and student exposure to teachers who smoke during school hours. The data are taken from a European Commission-funded study ‘Control of Adolescent Smoking’ (the CAS study) in seven European countries. Multilevel modelling analyses were applied to investigate associations between national legislation concerning tobacco smoking in school, local school smoking restrictions and students’ exposure to teachers who smoke at school. The analyses integrate data from three levels: national (types of policy and how restrictive they are), school (survey among teachers, n = 455) and student (survey among 15-year-old students, n = 10 890). The results suggest that both national- and school-level policies on restriction of smoking among teachers are associated with a decreased probability of students reporting that they are exposed to teachers who smoke indoors, but an increased probability of being exposed to teachers smoking outdoors.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionReferences |
|---|
In accordance with Social Cognitive Theory (Bandura, 1986
), adolescents are more likely to take up smoking if they observe significant others’ smoking. This modelling of smoking is thought to set an example, stimulate imitation and provide social reinforcement for the behaviour. In their capacity as important primary socializing agents (Oetting and Donnermeyer, 1998), teachers are regarded as significant role models. Teachers who smoke may therefore influence adolescents to adopt smoking through direct modelling.
Several empirical studies also suggest that the major factors predicting onset of smoking are socio-environmental, including exposure to smoker role models in family, peer and school settings, and the perception that tobacco use is the norm (Moore et al., 1994; Surgeon General, 1994; Reid et al., 1995; Tyas and Pederson, 1998; NHS Centre for Reviews and Dissemination, 1999; Wakefield et al., 2000).
Thus, restricting adolescents’ exposure to models of adult smokers may prove an effective means of smoking prevention, especially when supported by other preventive measures in the community. Through various public health initiatives in European countries, such as mass media campaigns, adults have been encouraged not to smoke, but few efforts seem to have been made to influence the view of parents and teachers as significant smoker role models (Reid et al., 1995). More effective strategies may therefore require restrictions on adult smoking at places where young people spend time, such as in private homes, schools and other public places. As the regulation of smoking in private homes is difficult to implement in practice, interventions have to rely on voluntary efforts from smokers (Ashley and Ferrence, 1998). In addition, the types of advocacy and educational campaigns which could contribute to achieving smoke-free homes are likely to be very expensive and, probably, not very effective (Reid et al., 1995).
However, achieving smoke-free schools is a more feasible alternative, particularly given the function of school as a formal arena for development and learning. Adults at school may be expected to be motivated to act as exemplars for the students and studies have shown that teachers, including teachers who smoke, agree that school staff should set a good example by not smoking (Galaif et al., 1996). Ideally, it should be possible to prevent student exposure to teachers who smoke through national and local tobacco control policies, developing legislation aimed at restricting the use of tobacco in schools. Brownson et al. (Brownson et al., 1997) found in their review of policies to reduce exposure to environmental tobacco smoke (ETS) that efforts to restrict public smoking have proliferated over the past decade and that bans on public smoking are effective in reducing non-smokers’ exposure to ETS. However, few, if any, previous studies report effects on exposure to smoker role models.
Harkin et al. (Harkin et al., 1997) concluded in their study of smoking policies in the WHO European region that most countries do have some form of legislation covering the restriction of smoking at public places such as schools. In previous analyses of national policy data from the seven countries included in the present study (Wold et al., 2000a; Wold and Currie, 2001), the main differences between countries could be characterized by whether or not national policies restricting teacher smoking existed and how these policies were developed, e.g. by national or local government. Thus, the countries fall into two main categories: (1) those with no national policies related to teacher smoking in schools per se (Denmark, Germany, Scotland and Wales) and (2) those with national laws that explicitly restricted teacher smoking during school hours (Austria, Finland and Norway).
In these latter countries, legislation concerning teacher smoking differed. In Norway, teachers were not permitted to smoke indoors, but were able to smoke outdoors on the school premises. In Finland, smoking among teachers was banned outdoors on the school premises and restricted indoors (i.e. teachers could smoke in designated rooms). In Austria, the law restricted smoking only within the school building, with teachers allowed to smoke in specific rooms. There were no restrictions on smoking outside the building on the school premises.
As suggested by Reid et al. (Reid et al., 1995), the costs to health departments of encouraging restrictions on smoking in schools are minimal. However, there are many obstacles to full implementation unless school administrators are willing to make no-smoking mandatory. There is an abundance of evaluations of specific health education programmes at schools (Surgeon General, 1994; Reid et al., 1995; VanDyke and Riesenberg, 2002; Tingle et al., 2003), but very little is known about the precise role and relative significance of the process of implementing policies on restricting smoking at school (Villalbi and Ballestin, 1994; Chollat-Traquet, 1996).
An indicator of the success of implementation of national and school policies on restrictions of smoking could be the extent to which students are exposed to teachers who smoke at school. The present study seeks to examine whether legislative policies concerning teacher smoking at national and school levels are actually associated with students’ perceptions of exposure to teachers who smoke during school hours.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionReferences |
|---|
The data presented here are taken from a European Commission funded study ‘Control of Adolescent Smoking’ (the CAS study) in seven European countries: Austria, Denmark, Finland, Germany (North Rhine Westphalia only), Norway, Scotland and Wales (Wold and Currie, 2001
). There were three levels of data collection in the study. At the student level, data were collected via an existing transnational survey on health behaviours among children and adolescents, Health Behaviour in School-aged Children (HBSC): A WHO Cross-national Survey (Smith et al., 1992; Currie, 1998; Currie, et al., 2000). The HBSC study represents a behavioural epidemiological approach to research on health behaviours and lifestyles among young people in Europe. Seven of the 29 countries participating in the HBSC joined as partners in the present study, undertaking a series of interlinked special surveys on tobacco control at the national, school and student level.
The student data were collected in 1997/98, with representative samples of students aged 15 years (n = 10 890) from the seven CAS countries. Two-stage cluster sampling was employed, with school or school class the primary sampling unit. Further details of the sample are provided in Table I. A detailed description of the samples from each country is given in Wold et al. (Wold et al., 2000b) and the sampling methods are outlined in more detail in Currie et al. (Currie et al., 1998, 2000).
|
The teachers’ survey was administered as an integral part of the fieldwork with students. In each of the participating countries, questionnaires were sent to two key individuals in each school: the head teacher (or other senior manager) and a teacher with responsibility for health education. Additional members of teaching staff were also involved in Austria, Finland and Germany. A detailed description of the samples from each country and fieldwork dates is given in Roberts (Roberts, 2000
). As the relevant teachers variable included in the present analyses was about factual information regarding smoking policies (whether or not teacher smoking was allowed at their school) and not about the respondents’ subjective views on smoking policies, the same response were expected from all teachers from the same school. Therefore, the responses of one representative from each school (n = 455) were selected on an arbitrary basis (Table I) in the present analyses. Through the use of school identifiers, the school-level data collected could be matched to responses from students in that school.
Instruments
The student questionnaire included selected demographic questions, and questions on school smoking policy and practices, including perceptions of adult and student smoking at school. The specific question on exposure to teacher smoking was: ‘During school hours, how often do you see or know about teachers smoking?’. The students were asked to tick a box for each line on a list of various places at school (staff rooms, canteen/cafeteria, corridors, other parts of the school building, outdoors on the school premises and outdoors off the school premises), indicating that this applied ‘about every day’, ‘sometimes’, ‘never’ or ‘don’t know’.
The teacher questionnaire concerned policies and practices of smoking restriction at school, with one question asking ‘Are teachers allowed to smoke on the school premises?’. Respondents were asked to select from ‘no, not at all’, ‘yes, in restricted areas’, ‘yes, anywhere on the school premises’ and ‘don’t know’.
The national policy-monitoring instrument was developed as a standard ‘interview scheme’ for use in interrogating key informants and official documentation. It is presented in full in Wold et al. (Wold et al., 2000a). The instrument was based on Crosswaite and McQueen’s (Crosswaite and McQueen, 1993) model of policy implementation, and guided the collection of both quantitative and qualitative data on national tobacco policy development, implementation and enforcement.
Variables
Students’ responses were dichotomized into whether or not they saw or knew that teachers were smoking indoors at school about every day, and, correspondingly, outdoors on and off school premises during school hours (where 0 = ‘no’ and 1 = ‘yes’ in each case).
From the teacher survey, a school variable was constructed, indicating whether smoking was banned for teachers on the school premises (where 0 = ‘no’ and 1 = ‘yes’). Given that student smoking was not permitted where staff bans were in place, this variable is effectively an indicator of smoke-free schools.
For the multilevel analyses a national-level variable consisting of two categories was constructed, differentiating between those countries with national policies restricting smoking among teachers at school and those without such a policy (where 0 = ‘no’ policy and 1 = ‘yes’).
Data analysis
Relationships between national policies, local school policy and individual exposure to teacher smoking was assessed using the statistical package MlwiN version 1.10 (Goldstein, 1995; Kreft and de Leeuw, 1998). A series of multilevel logistic regression models with binomial variance at the individual level and random intercept variance at the school level was estimated. The random intercept models allow students’ exposure to teacher smoking to differ across schools and nations. Due to the small number of countries, a random intercept variance for country was not modelled. Where feasible, fixed country effects were modelled. Second-order Penalized Quasi-likelihood (PQL) estimation was used as this method tends to give less biased estimates than conventional (MQL) when there are few higher-level units. The estimates in logistic random intercept models are on a logit scale. In the result section, to ease interpretability, the logits were transformed to logistic odds ratios (ORs) and to predicted probabilities.
The multilevel analysis was conducted in two stages. For descriptive purposes, the first stage of analysis set out to establish the magnitude of school-level and country-level variation in exposure to teacher smoking. The observed magnitude of variation provides important information about the scope for incorporating predictors at multiple levels. If exposure to teacher smoking were to show no significant school or country variation, there would be little point in models including national and school-level policies as predictors.
In the second stage of analysis, based on information from the descriptive analyses, multilevel models including national- and school-level policies as predictors of smoking exposure were tested.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionReferences |
|---|
The teacher survey data indicate large variation both between and within countries in terms of smoke-free schools according to data gathered from school staff members, ranging from as little as 1% of schools with a smoke-free policy in Denmark to 65% in Norway.
School-level variation in smoking exposure
A series of logistic random intercept regression models were produced for the following forms of exposure to teacher smoking: (1) exposure to teachers smoking indoors, (2) exposure to teachers smoking outside in the schoolyard, (3) exposure to teachers smoking outside the school premises and (4) a summary measure indicating exposure in any of the three locations.
Table II shows the results from running a ‘null’ model for each exposure variable, where the only explanatory variable included is the school identifier. Compositional effects due to differences in age, gender and student smoking status were partialled out from the variance components. Confounding national level variation was removed from the school variance, by modelling the fixed effect of nation.
|
It can be seen that a substantial number of adolescents reported to be exposed to teacher smoking during school hours on a daily basis (27.2%). Seeing teachers smoke inside was the most frequently reported type of exposure (14%). However, as indicated by the 95% confidence intervals (CIs) of the school variation, the probability of being exposed to teachers who smoke differed substantially across schools, for all three forms of exposure. For example, among the whole population of schools, the expected percentage of adolescents being exposed to teacher smoking indoors would fall within the 95% CI of 2.2–54.1%. The wide confidence interval corresponds to an intraclass correlation (ICC) of 0.236. The intraclass correlation provides an estimate of the relative magnitude of school-level variation in exposure, where an ICC of 0.0 describes a situation with no school-level variation relative to the individual variation. For all types of exposure the 95% CI and the ICC suggests that schools were highly different with regard to smoking exposure.
Cross-national variation in smoking exposure
Cross-national variation in exposure to smoking was estimated in a fixed-effect model. Being the most restrictive policy environment, Finland was chosen as the reference for cross-national comparison. It can be seen from Table III that the probability of exposure to teacher smoking varied strongly across nations. The cross-national variation was particularly strong for indoor smoking. According to Table III, the probabilities of Finnish and Norwegian adolescents reporting seeing their teachers smoke indoors was very low (2.2 and 4.6%, respectively). In contrast, there was a 39% probability of the Danish students reporting to see their teachers smoke indoors on a daily basis. Expressed in ORs, Danish adolescents were 28 times more likely to report exposure to indoor teacher smoking than Finnish adolescents.
|
Students from different countries also differed in their reports of teacher smoking in the schoolyard and teacher smoking outside school premises. Norwegian adolescents were comparatively more likely to report such outdoor exposure. Among Norwegian adolescents, the probability for reporting seeing their teacher smoke in the schoolyard was 16.9% and for seeing their teachers smoke outside school premises was 17.4%.
When exposure to teacher smoking was collapsed across the different kinds of exposure, cross-national differences remained strong. Four in 10 Danish adolescents reported seeing teachers smoke, whereas only one in 10 Finnish adolescents reported such exposure.
Overall, the magnitude of cross-national differences indicated scope for multilevel modelling of national influences on exposure to teacher smoking. Drawing on prior knowledge about the smoking policies, the observed cross-national differences in students reported exposure appear to have high face validity. Finland has been known for restrictive policies, whereas Denmark has been known for a non-restrictive policy.
Smoking policies and exposure to teacher smoking
The descriptive analysis of school and country variation in exposure to teacher smoking suggested a potentially significant role for both national and local factors in shaping such exposure. To study the particular role for national and local smoking policies, and the interaction between such policies, adolescents were divided into four groups. It can be seen from Table IV that there was a clear relationship between a restrictive national policy and number of smoke-free schools. In the three restrictive countries (Finland, Norway and Austria), 101 from 214 schools reported to be smoke-free (47%), whereas in the less-restrictive countries (Denmark, Germany, Scotland and Wales), only 42 from 231 schools were smoke-free (18%).
|
This analysis suggests the existence of four subgroups. First, those living in countries without a national policy regarding teacher smoking and without any school-level policies banning such smoking. Second, those in a country with polices restricting teacher smoking, but attending a school without such policies. Third, those in a country without national policies, but attending a school where teacher smoking is banned. Fourth, those students living in a country with both national and school policies restricting teacher smoking in schools.
In a series of logistic regression models with random intercept variance at the school level, exposure to teacher smoking was regressed on policy group membership. In these models exposure was allowed to vary by school. The ‘no policy’ group served as the reference for comparisons. To control for compositional effects, all estimates were adjusted for age, gender and smoking status. It can be seen from Table V that the four groups of smoking policies were significantly related to all forms of exposure. The group differences were most pronounced for teacher smoking indoors. When both national- and school-level policies banned teachers smoking, less than 3% of adolescents would be expected to see their teachers smoke indoors. However, when no policies restrict teacher smoking, the model predicts that 29% of students would see their teachers smoke indoors daily.
|
A totally different pattern emerged for exposure for teacher smoking outdoors. Having a policy banning teachers smoking, either at the national or school level, was associated with a higher expected probability of seeing teachers smoke outdoors. In particular, when teacher smoking was banned both at the school and national level, adolescents were twice as likely to see their teacher smoke outdoors compared to when no such policies were in place (OR = 2.18), suggesting that teachers smoke outside when they are not allowed to smoke inside the school.
As expected, when collapsing the different kinds of exposure, the ‘no policy’ group had a higher likelihood of seeing teachers smoke daily (33.5%), whereas the ‘national and school policy’ group had the lowest exposure to teacher smoking (18%).
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionReferences |
|---|
The results of the analyses presented here suggest that students’ exposure to teachers who smoke at school are related to national and school policies on restrictions on smoking at school. Both national- and school-level policies restricting smoking among teachers are associated with a reduced probability of students reporting that they are exposed to teachers who smoke indoors, but an increased probability of being exposed to teachers smoking outdoors. Thus, as suggested by other studies [e.g. (Brownson et al., 1997
)], this research indicates that restrictive tobacco control policies at national and local levels seem to be effective in reducing non-smokers’ exposure to ETS in school. However, the present study also suggests that such policies are less effective in reducing adolescents’ exposure to people who smoke tobacco in their environment, the potential for increased exposure to smoker role models outdoors being clearly demonstrated.
One of the implications of the present study is that policies to restrict teacher smoking may need to be more sensitive to the social ecology of smoking. Restricting smoking in one social arena may in fact increase social modelling in other arenas that are not sanctioned by such restrictions. For example, although local and national policies towards teacher smoking appear to prevent exposure to teacher modelling of smoking indoors, such policies tend to lead to an increase in modelling of smoking outside.
The present study highlights the interactive effects between smoking policies at the local school level and at the national level. For indoor smoking, the level of teacher modelling of smoking was low when either the national or the local policy was restrictive, but very high when neither of these policies were restrictive.
Moreover, local school policies predicted exposure to smokers in addition to national policies. In general, local school policies tended to be more restrictive than national policies, thus adding to the effect of national policies. As may have been expected, school policies seem more effective in countries lacking national policies regarding school smoking restrictions. This implies that when national policies on restriction of smoking at school are absent, restrictive local policies may be quite effective because teachers seem to comply with the regulations in their school. This study, however, indicates that a combination of national policies and restrictive local policies is the most effective way of reducing student exposure to teachers who smoke indoors at school during school hours.
The results suggest that although there were consistent and statistically significant associations between policies restricting teacher smoking and student exposure to such smoking, the effect sizes tended to be moderate to low. A substantial proportion of students in smoke-free policy schools reported that teachers smoke during school hours, indicating that one of the reasons for the weak associations may be that teachers who smoke find it difficult to comply with the restrictions. In Norway and Austria, earlier proposals on national laws to ban smoking in schools altogether met with strong opposition by teachers’ unions and resulted in compromises to allow teachers to smoke (Wold et al., 2000a). Other studies have also shown that as a group, staff smokers tend to be unfavourable toward a no-smoking policy at school (De Moor et al., 1992; Galaif et al., 1996) and several studies suggest that compliance with bans on smoking in school is low (Galaif et al., 1996; Hartland et al., 1998; Wakefield and Chaloupka, 2000).
As in all research, the results of this study must be viewed in light of the strengths and weaknesses of the research design. A particular problem that the study has had to deal with is the difficulty of collecting truly comparable cross-national data on tobacco policy, the reasons being that data were not available on the same issues in every country and the form that the data took varied from country to country. Nevertheless, we were able to use the data to make broad comparisons between countries and confidently dichotomize national smoking policies in terms of restrictions on teacher smoking.
One advantage of the study is that the national data were collected by the national members of the research team. This allowed a high degree of adaptation from the international standard protocol to national circumstances to ensure that the purpose of the study and the data which were collected were compatible, and thereby securing a high degree of validity. Another advantage is that the study of national smoking policies is linked to related research findings derived from a survey on smoking and other health behaviours of school-aged children, and a survey on smoking policies in schools in the participating countries. This particular combination of data from three levels—individuals, schools and nations—provides an excellent basis for further analyses as suggested by Rütten et al. (Rütten et al., 2000).
|
Conclusion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionReferences |
|---|
The present study suggests that national tobacco control policies to a great extent seem to be effectively implemented. and underpin local school-level policy and its implementation. Students were less likely to be exposed to indoor tobacco smoke in schools in countries where restrictive national smoking policies were in place. However, it should be noted that in some countries, very restrictive national policies on indoor smoking at school can push teacher smoking outdoors, resulting in the negative and unforeseen side effect of making smoking (and smoker role models) more visible to students. Thus, policies restricting teacher smoking at school may actually enhance the recruitment of young smokers. These findings emphasize that although the level of restriction is an important issue, the development of effective tobacco control policies also depends on sensible considerations about the nature/type and quality of restrictions, as well as efforts aimed at promoting their enforcement and compliance. In this regard, the need for developing policy in collaboration with important organizations such as teachers’ unions may prove to be necessary. Also, policies to restrict teacher smoking may need to be more sensitive to the social ecology of smoking.
|
Acknowledgements |
|---|
The project was funded by EC BIOMED II grant BMH4-CT98-3721 and a grant from the Norwegian Research Council. Thanks are due to the following partners of the Control of Adolescent (CAS) study for their collaboration and use of their data: Wolfgang Dür (Austria), Lasse Kannas and Jorma Tynjälä (Finland), Bjoern Holstein (Denmark), and Klaus Hurrelmann and Bettina Schmidt, Germany
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionReferences |
|---|
Ashley, M.J. and Ferrence, R. (1998) Reducing children’s exposure to environmental tobacco smoke in homes: issues and strategies. Tobacco Control, 7, 61–65.
Bandura, A. (1986) Social Foundations of Thought and Action. Prentice-Hall, Englewood Cliffs, NJ.
Brownson, R.C., Eriksen, M.P., Davies, R.M. and Warner, K.E. (1997) Environmental tobacco smoke: health effects and policies to reduce exposure. Annual Review of Public Health, 18, 163–185.[CrossRef][Web of Science][Medline]
Chollat-Traquet, C. (1996) Evaluating Tobacco Control Activities. WHO, Geneva.
Crosswaite, C. and McQueen D.V. (1993) The Implementation of Smoking and Nutrition Policies in Scotland and Finland: A Review. Research Unit in Health and Behavioural Change, University of Edinburgh, Edinburgh.
Currie, C. (1998) Health Behaviour in School-Aged Children: A WHO Cross-National Survey (HBSC): Research Protocol for the 1997/98 Survey. Research Unit in Health and Behavioural Change, University of Edinburgh, Edinburgh.
Currie, C., Hurrelmann K., Settertobulte, W., Smith, R. and Todd, J. (eds) (2000) Health and health behaviour among young people: Health Behaviour in School-aged Children: A WHO Cross-National Study (HBSC) International Report. WHO Policy Series Health Policy for Children and Adolescents Issue 1. WHO, Copenhagen. Available at: http://www.hbsc.org.
DeMoor, C., Cookson, K., Elder, J.P., Young, R., Molgaard, C.A. and Wildey, M. (1992) The association between teacher attitudes, behavioral intentions and smoking and the prevalence of smoking among seventh-grade students. Adolescence, 27, 565–578[Web of Science][Medline]
Galaif, E.R., Sussman, S. and Bundek, N. (1996) The relations of school staff smokers’ attitudes about modeling smoking behavior in students and their receptivity to no-smoking policy. Journal of Drug Education, 26, 313–322.[Web of Science][Medline]
Goldstein, H. (1995) Multilevel Statistical Models. Edward Arnold, London.
Harkin, A.M., Anderson, P. and Goos, C. (1997) Smoking, Drinking and Drug Taking in the European Region. WHO, Copenhagen.
Hartland, J., Tudor-Smith, C. and Bowker, S. (1998) Smoke-free policies in schools: a qualitative investigation of the benefits and barriers. Health Education Journal, 57, 51–59.
Kreft, I. and de Leeuw, J. (1998) Introducing Multilevel Modelling. Sage, Newbury Park, CA.
Moore, S., Wolfe, S.M., Lindes, D. and Douglas, C.E. (1994) Epidemiology of failed tobacco control legislation. Journal of the American Medical Association, 272, 1171–1175.
NHS Centre for Reviews and Dissemination (1999) Preventing the uptake of smoking in young people. Effective Health Care Bulletin, 5(5).
Oetting, E.R. and Donnermeyer, J.F. (1998) Primary socialization theory: the etiology of drug use and deviance. Substance Use and Misuse, 33, 995–1026.
Reid, D.J., McNeill, A.D. and Glynn, T.J. (1995) Reducing the prevalences of smoking in youth in Western countries: an international review. Tobacco Control, 4, 266–277.
Roberts, C. (2000) Control of Adolescent Smoking (the CAS study) 1997/1998: Technical Report on the Staff Survey in Eight European Countries. Report to European Commission. Research and Evaluation Branch, Health Promotion Division, National Assembly for Wales.
Rütten, A., Lüschen, G., von Lengerke, T., Abel, T., Kannas, L., Rodriguez Diaz, P., Vinck, J. and Van Der Zee, J. (2000) Health Promotion Policy in Europe. Olbenbourg, München.
Smith, C., Wold, B. and Moore, L. (1992) Health behaviour research with adolescents: a perspective from the WHO Cross-National Health Behaviour in School-Aged Children Study. Health Promotion Journal of Australia, 2, 41–44.
Surgeon General (1994) Preventing Tobacco use among Young People. US Department of Health and Human Services, Washington, DC.
Tingle, L.R., DeSimone, M. and Covington, B. (2003) A meta-evaluation of 11 school-based smoking prevention programs. Journal of School Health, 2, 64–67.
Tyas, S.L. and Pederson, L.L. (1998) Psychosocial factors related to adolescent smoking: a critical review of the literature. Tobacco Control, 7, 409–20.
VanDyke, E.M. and Riesenberg, L.A. (2002) Effectiveness of a school-based intervention at changing preadolescents’ tobacco use and attitudes. Journal of School Health, 72, 221–225.
Villalbi, J.R. and Ballestin, M. (1994) Developing smoking control policies in schools. Health Promotion International, 2, 99–104.
Wakefield, M. and Chaloupka, F. (2000) Effectiveness of comprehensive tobacco control programmes in reducing teenage smoking in the USA. Tobacco Control, 9, 177–186.
Wakefield, M.A., Chaloupka, F.J., Kaufman, N.J., Orleans, C.T., Barker, D.C. and Ruel, E.E. (2000) Effect of restrictions on smoking at home, at school and in public places on teenage smoking: cross sectional study. British Medical Journal, 321, 333–337.
Wold, B. and Currie, C. (2001) Control of Adolescent Smoking: Transnational Variation in Prevalence of Adolescent Smoking: The Role of National Tobacco Control Policies and the School and Family Environment. Report to EC BIOMED II. Research Centre for Health Promotion, University of Bergen, Bergen.
Wold, B., Holstein, B., Griesbach, D. and Currie, C. (2000a) Control of Adolescent Smoking. National Policies on Restriction of Smoking at School in Eight European Countries. Report to EC BIOMED II. Child and Adolescent Health Research Unit, University of Edinburgh, Edinburgh.
Wold, B., Currie, C. and Lund M. (2000b) Control of Adolescent Smoking (The CAS study) 1997/1998. Technical Report on Surveys of 15 year-olds in Nine European Countries. Report submitted to the European Commission under contract BMH4-CT98-3721. Research Centre for Health Promotion, University of Bergen, Bergen.
Received on January 3, 2003; accepted on March 31, 2003
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  The GTSS Collaborative Group The Global School Personnel Survey: a cross-country overview. Tob. Control, June 1, 2006; 15(suppl_2): ii20 - ii30. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Wold, C. Currie, C. Roberts, and L. E. Aaroe National legislation on school smoking restrictions in eight European countries Health Promot. Int., December 1, 2004; 19(4): 482 - 488. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||