Health Education Research Advance Access published online on January 31, 2008
Health Education Research, doi:10.1093/her/cym085
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Health risks information reaches secondary school smokers
1 Saving Faces—The Facial Surgery Research Foundation, London EC1A 7BE, UK
2 School of Epidemiology and Health Sciences, University of Manchester, Manchester M13 9PT, UK
3 Department of Oral and Maxillofacial Surgery, Barts and the London NHS Trust, London EC1A 7BE, UK
Correspondence to: * Correspondence to: F. Ridout. E-mail: fran.ridout{at}milfordhouse.net
 |
Abstract |
|---|
 Top Abstract IntroductionMethodResultsDiscussionConclusionFundingConflict of interest statementReferences |
|---|
This cross-sectional study aimed to assess smoking prevention and cessation education delivered as part of the UK National Curriculum and to evaluate the relative effectiveness of health, social influence and other/non-health components. In all, 1789 students aged 11–15 from 12 secondary schools completed online surveys assessing smoking status, factors known to be related to smoking and experience of smoking education. A total of 1421 of 1722 (83%) students remembered some school-based education. Of these, 803 (57%) said that the lessons changed their ideas about smoking. Multinomial logistic regression was used to assess whether lesson recall was associated with smoking status in a model adjusting for age, gender, ethnicity, family and best friend smoking status, socioeconomic status, and school. Quitters were more likely than smokers to report having changed their ideas about smoking as a result of a lesson (OR 5.78, 95% CI 2.44–13.72). The relative effectiveness of 16 lesson themes was assessed. Significantly more students changed their ideas about smoking as a result of ‘health’ compared with ‘social influence’ (
2 (1) 124.0, P < 0.001) or ‘other/non-health’ (2 (1) 63.16, P < 0.001) topics. Mouth cancer was the most effective health topic and may provide a suitable model for both smoking and risky drinking prevention. |
Introduction |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionFundingConflict of interest statementReferences |
|---|
The harmful effects of smoking tobacco are well known. Early age intervention strategies for smoking prevention were based on the ‘rational or information deficit model’ which assumed that by enhancing awareness of the health consequences of smoking, prevalence could be reduced. Programs often employed scare tactics [1]. In 1967 Janis posited his ‘inverted-U’ theory, suggesting that very low or very high levels of fear inhibit both attitude and behavioral modification [2]. The use of scare tactics in health education subsequently declined and the focus of school-based interventions shifted to skill building. Drawing on psychosocial theories such as Bandura's Social Learning Theory [3] and Evans’ seminal research on peer pressure [4], the social influence approach encourages students to recognize and resist the negative peer and familial and cultural factors that influence smoking initiation [5].
Evidence supporting the use of school-based social influence programs is sparse [6]. The largest and most rigorous test of the model, the Hutchison Project, focused on helping pupils to identify and resist social influences, including advertising and peer pressure, while promoting tobacco-free social norms. Motivational and self-help cessation materials were also available and staff were trained to deliver the intervention and to encourage cessation efforts. Control schools followed their usual tobacco education programs. This study provided no support for social influences as a lasting deterrent against youth smoking [7]. A rigorously designed European Union program, which included lessons relating to social influence processes and training in refusal skills, proved to be counter productive in England, the only country that did not address health consequences as part of the program [8]. Other recent approaches to smoking prevention have fared no better in the UK. For example, a rigorously designed trial of a program based on the transtheoretical model of behavior change, which employed an interactive computer program and class lessons, revealed no benefit for smoking prevention or cessation [9].
Although the effects of information-only (health risk) approaches to smoking prevention have not been evaluated in rigorously designed trials, [6] there is some evidence to support the use of shock tactics. For example, a study of the impact of mass media antitobacco advertisements in the United States concluded that young adult quitters were helped more by advertisements that depicted serious harm in an emotional or graphic way than by conventional antismoking aids [10]. The Australian National Tobacco Campaign, which focused on health outcomes, was also effective in promoting antismoking attitudes among adolescents, helping them to quit or resolve to remain non-smoking [11].
In the United Kingdom, the government remains committed to reducing teenage smoking prevalence [12]. Smoking is emphasized as a personal health issue in the National Curriculum (a statutory framework for the education of 5- to 16-year olds), within which schools are free to plan and organize teaching and learning in the way that best meets the needs of their pupils. There is some evidence from national surveys that pupils who remember lessons on smoking are less likely to be regular smokers than those who do not [13] and that health education works cumulatively by repeated exposure to the same message presented in different ways [14]. Identifying the most effective components of this education could therefore help in the development of future programs. The current study uses data collected as part of an assessment of an online survey instrument, to assess the relative effectiveness of health information, social influences and non-health-related themes.
|
Method |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionFundingConflict of interest statementReferences |
|---|
There are two main components of the National Curriculum that relate to smoking. The statutory science order requires pupils to be taught about the effects of tobacco and how these relate to their personal health. The Personal, Social and Health Education (PSHE) curriculum is non-statutory. Guidelines are provided, but the content, quantity and quality of lessons delivered varies considerably between schools and teachers may receive no subject-specific training. The guidelines suggest that in contrast to the science (knowledge based) curriculum, PSHE should focus on developing transferable skills and attitudes that will help young people recognize and manage risk and make safer choices about healthy lifestyles. Suggested activities for pupils aged 7–11 include discussions on why people smoke or the harmful effects on families when individual members smoke or suffer illness as a result. Pupils may be asked to investigate the reasons why people do things, which they know are harmful and discuss ways to resist such pressures. Secondary school pupils, aged 11–16, should be encouraged to recognize when pressure from others threatens their personal safety and well-being and to develop effective ways of resisting these pressures. Participatory teaching styles are recommended including role-play, case studies and debates. Many interactive Web-based resources are available via the government's ‘Wired for Health’ Web site, which supports the PSHE curriculum [15].
The cross-sectional survey was conducted using a proprietary online survey tool (Survey Monkey). Twenty-two schools in Surrey, Hampshire and central London were approached. Ten schools declined to participate giving the lack of computer availability (four), participation in other studies (two) or full timetables (four) as reasons. The 12 participating schools provided a broad range of school and pupil characteristics. Schools were asked to randomly select one or more mixed-ability tutor groups from years 7–10 to take part in the survey. One thousand seven hundred and eighty nine pupils aged 11–15 from these selected groups attended a timetabled supervised computer session and completed the anonymous, self-administered, Web-based questionnaire. Absentees were not followed up.
Procedure
Ethical approval was provided by the relevant local research ethics committees. Consent was obtained from all respondents and a passive consent process was used to secure the consent of participants’ parents. Skip logic was built into the survey to ensure that students were not asked to respond to inapplicable questions. For example, those who were self-categorized as ‘never smokers’ were not asked further questions about their smoking habits. Individual items in listed response choices were presented in random order to reduce potential order effects. Data collection took place during PSHE lessons. A researcher attended the first session at each school to ensure that there were no technical problems and to train staff how to administer the survey. Teachers were provided with a detailed lesson plan. The survey took approximately 20 min to complete.
Measures
Standard questions from the national survey in England [13] were used to measure factors that are known to be associated with smoking including ethnicity, gender, age, parental, sibling and best friend smoking status. Smoking status was measured using six standard options: I have never smoked; I have only tried smoking once; I used to smoke, but I never smoke now; I sometimes smoke, but not as many as one a week; I smoke between 1 and 6 cigarettes a week, and I usually smoke more than 6 cigarettes a week. In accordance with the national survey, regular smoking was defined as the consumption of at least one cigarette per week and the last two items were therefore treated as one category. Since this measure cannot be assumed to represent an interval scale, the last four categories were converted into dichotomous dummy variables.
Eight options for ethnicity were presented in the survey: white, Black Caribbean/West Indian, Black African, Indian, Pakistani, Bangladeshi, Chinese and other. As several of these options had fewer than 50 responses, they were recoded as white, black (Black Caribbean/West Indian and Black African), Pakistani, Bangladeshi and other (Chinese, Indian and other). The questions relating to students’ experiences of smoking prevention and cessation lessons were developed in collaboration with a panel of experts who had extensive experience of school-based studies and piloted them with a class of 11- to 12-year olds prior to the start of the study.
Wherever possible, free text responses to ‘other—please specify’ list items were recoded independently by two researchers into existing categories. Students’ answers to the question ‘Can you suggest any ways of improving the lessons, so that more students would take notice of the anti-smoking message’ were coded by two researchers into recurrent themes. Where there were discrepancies between researchers in the coding, responses were recoded by mutual agreement.
Analysis
SPSS 13 for Windows was used for the analysis. The intraclass coefficient was calculated. As the number of participating tutor groups, and therefore students, varied between schools, an estimate of mean cluster size was employed in the calculation [16].
Chi-square was used to test for differences in familial and best friend smoking prevalence between never smokers, triers, quitters and occasional and regular smokers, and to determine whether there were between-group differences in students’ experiences of lessons about smoking. The relative effectiveness of 16 lesson themes was assessed. For the purpose of analysis these were divided into three categories: ‘health’ (smoking and lung cancer, smoking and mouth cancer, health risks of smoking, smoking and heart disease, smoking is addictive, diseases caused by smoking and patients suffering from diseases caused by smoking), ‘social influences’ (ways to refuse cigarettes and pressure from others to smoke) and ‘other/non-health’ (reasons why people smoke, reasons why people stop smoking, cost of smoking, substances in tobacco, tobacco laws, economic effects of tobacco and smoking and environmental effects of tobacco and smoking). Chi-square was used to test for differences in proportions of students changing their ideas about smoking across the three categories. Data were analyzed with and without the general health topics ‘health risks of smoking’ and ‘diseases caused by smoking’, as a disproportionate number of students might be expected to choose these very general options.
Multinomial logistic regression analysis was used to determine whether pupils’ recollections of school-based smoking prevention lessons were associated with smoking status, in a model that adjusted for age, gender, ethnicity, family and best friend smoking status, percentage of pupils known to be eligible for free school meals (used as a proxy for socioeconomic status) and school.
|
Results |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionFundingConflict of interest statementReferences |
|---|
Our data revealed an intraclass correlation coefficient (ICC) of 0.013 for regular smoking, which is relatively small when compared with published data for smoking prevalence in the age group [17]. This is to be expected since large clusters generally have small ICCs [18] and the clusters in our study were quite large (an average of 149 pupils per school). Other research confirms our observation that weekly smoking incidence exhibits little clustering in schools [19]. Additionally, the observed odds ratios (ORs) for our main outcomes are large with highly statistically significant P-values (<0.001), which would not change materially after adjusting for clustering using an ICC of 0.013. With these considerations in mind, the presence of intact social groups within schools will therefore have had a negligible impact on our results. The number of respondents included in each analysis varied due to missing data and skip logic. For example, only 1753 of 1789 pupils provided information on their smoking status. No parents withdrew their children from this study. Fifty-four students did not give consent. Seven of the 12 schools provided information on the number of eligible students in participating classes. The mean attendance rate in these schools was 83%.
Demographics
Regular smoking prevalence was 1%, 3%, 6%, 16% and 21% for ages 11, 12, 13, 14 and 15, respectively. This compares with 1%, 2%, 6%, 12% and 21% in national survey results collected during the same year [13]. Table I shows the demographic characteristics of the sample and their self-reported smoking status.
|
Reported effects of lessons
Of the 1722 respondents who answered a question about school-based antismoking education, 83% remembered some lessons. Eleven- and fifteen-year olds were less likely to remember lessons than other age groups (
2 (4) 169.0, P = 0.002). Of the 1421 pupils who remembered lessons, 803 (56%) said that the lessons had changed their ideas about smoking. Four hundred and forty never smokers had changed their ideas and of these 206 (47%) reported that they decided not to start smoking as a result of this education. Similarly, 123 of 285 (43%) ‘former smokers’ (quitters and triers) reported that they had stopped smoking as a result of the lessons. Table II shows responses to questions relating to recall of lessons about smoking.
|
Lesson themes
Health topics including mouth cancer, which encouraged more than half the students to change their ideas about smoking, were the most effective (Table III). A significantly greater proportion of students changed their ideas as a result of health compared with social influence (
2 (1) 124.0, P < 0.001) and other/non-health (2 (1) 63.16, P < 0.001) topics. When the analysis was repeated with the general health topics removed, the results were similar and are not therefore reported here.
|
Factors associated with smoking status
Multinomial logistic regression revealed that ‘never smoked’ status was associated with having a non-smoking best friend, father, mother or brother and to being younger than current smokers (Table IV). Quitting was related to having a non-smoking best friend and to having changed ideas as a result of lessons about smoking. Having ‘tried’ smoking was associated with having a non-smoking best friend, father or brother and to having changed ideas about smoking as a result of lessons. There was no association between remembering one or more lessons and current smoking status (data not shown). However, both triers (OR 3.69, 95% CI 2.36–5.78) and quitters (OR 4.02, 95% CI 2.02–7.99) were more likely to report changing their ideas about smoking as a result of lessons received than pupils who had never smoked. These former smokers were also more likely to change their ideas than current smokers (Table IV). There was no significant difference in the likelihood of quitters compared with triers changing their ideas. Similarly, there was no significant difference between never smokers and current smokers.
|
Reasons for quitting
Table V shows the responses of students who had tried to stop smoking to the question ‘Why did you decide to stop or not continue smoking?’ In all, 38% of quitters reported that lessons about heart disease and lung cancer had influenced their decision to stop smoking.
|
Students’ suggestions for making lessons more effective
A total of 1032 students suggested at least one way of making lessons more effective. The most frequent suggestions were the inclusion of information about health risks and/or ex-smokers’ real-life stories and the use of graphic pictures or videos.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionFundingConflict of interest statementReferences |
|---|
The results of this study suggest that most 11- to 15-year olds in the United Kingdom have experienced some National Curriculum smoking education. It is encouraging that 15% of those who remembered one or more lessons reported that they had decided not to start smoking as a result. Historically, interventions have tended to concentrate on prevention rather than cessation. However, three-quarters of the former smokers (quitters and triers) in this study claimed to have changed their ideas about smoking as a result of lessons. These ex-smokers were five times more likely to have changed their ideas than current smokers. They were also more likely to have changed their ideas than never smokers and while it is not possible to determine whether lessons prompted or reinforced cessation, our results suggest that they may have played a role.
Compared with all other groups, fewer regular smokers remembered lessons and fewer than one in three said that lessons had changed their ideas. It is not clear whether these findings reflect a genuine association between regular smoking and lack of exposure to antismoking education. Perhaps regular smokers really did attend fewer lessons, as adolescent smokers may be twice as likely to be absent from lessons as their non-smoking colleagues [20]. However, it is also possible that there were few or no between-group differences in exposure to tobacco related education. It is generally accepted that people selectively attend to information that reinforces their current behaviors and opinions so regular smokers may have had less recollection of lesson content. Additionally, some young smokers may have used dissonance-reducing strategies to escape the personal relevance of the messages [21].
Health themes were the most frequently remembered topics, although it is not clear whether this accurately reflects the main focus of the lessons delivered. They were also more effective in changing students’ self-reported ideas about smoking than other topics. Mouth cancer was less frequently remembered than many other topics. This may reflect the general lack of awareness of this disease [22], but our results suggest that this disfiguring disease could be an effective deterrent in an age group which is very much concerned about its physical appearance [23–25]. Both smoking and heavy drinking are risk factors for mouth cancer [26]. Further research is needed to confirm the usefulness of this disease as a model to discourage smoking and to determine whether the threat of possible disfigurement can be a sufficient deterrent when current concerns about appearance, evident in the increasing use of cigarettes amongst girls as an appetite suppressant [27], may be perceived as being more relevant.
Our findings provide little support for the use of a social influence approach to reducing adolescent smoking. The proportion of students recalling lessons about peer pressure and ways to refuse cigarettes was similar to the number who recalled mouth cancer. However, in spite of a very strong association between smoking behavior and best friend's smoking status, these lessons were relatively ineffective in changing students’ ideas about smoking. Few respondents gave peer pressure as a reason for trying to quit. One possible explanation for this is that smoking status plays a symbolic role in the establishment of a young person's self-identity, so that he or she is reluctant to recognize the extent of peer influence on behavior. It is also possible that cross-sectional studies tend to overemphasize the peer group's role in smoking initiation. Recent longitudinal research suggests that adolescent smokers may simply choose like-minded friends making the focus on peer pressure less relevant to young people's behavior [28].
Some potential limitations of this study need to be considered. As the participating schools and pupils were self-selecting, there is potential for selection bias that could limit the generalizability of the results. With a cross-sectional design, results can only be suggestive of an association between antismoking education and smoking status. It is possible that a disproportionate number of smokers missed lessons and that the non-follow-up of absentees therefore led to undercounting the highest risk youths. While our sample of schools was chosen to provide a broad range of demographic and socioeconomic characteristics, our results may not be generalizable to the United Kingdom as a whole, but our smoking status results are in accordance with national data collected during the year of the study. The list of lesson topics presented in this survey covers the themes most commonly taught in UK schools. Some important prevention approaches may therefore have been omitted. For example, several US studies have found that strong anti-industry advertisements, which highlight the predatory marketing practices of the tobacco industry and the misinformation they broadcast can be effective prevention tools [29,30].
Although our measures were based on students’ self-reports, it is generally accepted that young people's accounts of their behavior tend to be reliable if, as was the case in this study, confidentiality can be assured [31]. The national survey, from which our questions on smoking prevalence were derived, has been validated with biochemical tests. Pupils were found to be largely honest in their reported smoking and no significant differences in reported prevalence between test and non-test halves of the sample were found [13]. This suggests that the omission of biochemical testing would have minimal impact on prevalence estimates. The online survey included questions that have been validated for use only in ‘pen and paper’ questionnaires. A recent large-scale lifestyle survey that included standard questions about smoking was offered in parallel paper and Web-based versions. The results of this survey, which was administered to 8- to 15-year olds, suggest that if questions are presented in a similar format there are very few significant differences in responses [32]. Pupils’ self-reported responses suggest that health risks, social influences, and more general topics all featured in lessons. The non-statutory PSHE guidelines for this curriculum emphasize that teaching young people about the risks of smoking is not sufficient and that social influences and skills should be the main focus of these lessons. However, it was not possible to collect accurate data about actual lesson content due to the inconsistency of approach to smoking education, both between and within schools. Additionally, many pupils indicated that they were able to recall lessons taught in primary schools that did not participate in this study.
This study corroborates the results of previous research, which found that adolescents recalled health-based messages more frequently than other topics and considered them to be highly effective smoking deterrents [29]. The use of scare tactics to graphically depict the health effects of smoking has been found to be popular with young people [33]. If it is indeed the case that health information is having a positive effect on this age group, the increasing availability of visually shocking interactive Web-based resources might be a contributory factor. It is possible that young people find scare tactics more acceptable and believable in this modality compared with traditional classroom teaching. Additionally, the images viewed by youth today are generally more horrific and explicit than those employed in earlier scare tactic programs and this may explain why their effect may differ from that documented in earlier studies.
The importance of theory-based education programs has been stressed by many authors [34], but this should not deter researchers from assessing the scare tactic approach, as it is well grounded in health behavior change theories such as the Health Belief Model [35] and Protection Motivation Theory. This theory was originally proposed to give conceptual clarity to the understanding of fear appeals and is widely used in health promotion [36–39]. It could therefore provide a sound basis for the development and testing in randomized controlled trials of innovative interventions such as the use of mouth cancer as a visually shocking smoking and binge drinking deterrent.
|
Conclusion |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionFundingConflict of interest statementReferences |
|---|
Messages about the negative health effects of smoking may change adolescents’ ideas about smoking. With mounting evidence, a reevaluation of scare tactic approaches to smoking prevention and cessation is indicated. The use of scare tactics appears to have considerable student support.
|
Funding |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionFundingConflict of interest statementReferences |
|---|
Facial Surgery Research Foundation—Saving Faces.
|
Conflict of interest statement |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionFundingConflict of interest statementReferences |
|---|
None declared.
|
References |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionFundingConflict of interest statementReferences |
|---|
1. Lantz PM, Jacobson PD, Warner KE, et al. Investing in youth tobacco control: a review of smoking prevention and control strategies. Tob Control (2000) 9:47–63.
2. Janis IL. Effects of fear arousal on attitude change: recent developments in theory and research. In: Advances in Experimental Social Psychology—Berkowitz L, ed. (1967) New York: Academic Press. 166–224.
3. Bandura A. Self-efficacy mechanism in human agency. Am Psychol (1977) 37:122–47.[CrossRef]
4. Evans RI, Dratt LM, Raines BE, et al. Social influences on smoking initiation: importance of distinguishing descriptive versus mediating process variables. J Appl Soc Psychol (1988) 18:925–43.[CrossRef][Web of Science]
5. NHS Centre for Reviews and Dissemination. Preventing the uptake of smoking in young people. Eff Health Care Bull (1999) 5.
6. Thomas R, Perera R. School-based programmes for preventing smoking. In: Cochrane Database of Systematic Reviews 2002 (2003) (Issue 4). Chichester, UK: John Wiley and Sons, Ltd. Art. No: CD001293; doi: 10.1002/14651858.CD001293.pub2.
7. Peterson AV Jr., Kealey KA, Mann SL, et al. Hutchinson Smoking Prevention Project: long-term randomized trial in school-based tobacco use prevention—results on smoking. J Natl Cancer Inst (2000) 92:1979–91.
8. de Vries H, Mudde A, Kremers S, et al. The European Smoking Prevention Framework Approach (ESFA): short-term effects. Health Educ Res (2003) 18:649–63.
9. Aveyard P, Cheng KK, Almond J, et al. Cluster randomised controlled trial of expert system based on the transtheoretical ("stages of change") model for smoking prevention and cessation in schools. Br Med J (1999) 319:948–53.
10. Biener L, Reimer RL, Wakefield M, et al. Impact of smoking cessation aids and mass media among recent quitters. Am J Prev Med (2006) 30:217–24.[CrossRef][Web of Science][Medline]
11. White V, Tan N, Wakefield M, et al. Do adult focused anti-smoking campaigns have an impact on adolescents? The case of the Australian National Tobacco Campaign. Tob Control (2003) 12(Suppl. 2):ii23–29.
12. Department of Health. "Smoking Kills" A White Paper on Tobacco (1998) London: The Stationary Office.
13. Fuller E. National Centre for Social Research and the National Foundation for Educational Research. Smoking, Drinking and Drug Use among Young People in England in 2004. A survey carried out on behalf of the NHS Health and Social Care Information Centre (2005) Leeds: HSCIC.
14. Mason JO, Ogden HG, Berreth DA, et al. Interpreting risks to the public. Am J Prev Med (1986) 2:133–9.[Medline]
15. Department of Health and the Department for Education and Skills. Wired for Health: Helping Schools to be More Effective. Available at: http://www.wiredforhealth.gov.uk. Accessed: 15 June 2007.
16. Armitage P, Berry G. Statistical Methods in Medical Research (1994) Oxford: Blackwell.
17. Siddiqui O, Hedeker D, Flay BR, et al. Intraclass correlation estimates in a school-based smoking prevention study. Outcome and mediating variables, by sex and ethnicity. Am J Epidemiol (1996) 144:425–33.
18. Gulliford MC, Adams G, Ukoumunne OC, et al. Intraclass correlation coefficient and outcome prevalence are associated in clustered binary data 3. J Clin Epidemiol (2005) 58:246–51.[CrossRef][Web of Science][Medline]
19. Murray DM, Rooney BL, Hannan PJ, et al. Intraclass correlation among common measures of adolescent smoking: estimates, correlates, and applications in smoking prevention studies. Am J Epidemiol (1994) 140:1038–50.
20. Charlton A, Blair V. Absence from school related to children's and parental smoking habits. Br Med J (1989) 298:90–2.
21. Festinger L. Conflict, Decision and Dissonance (1964) Stanford, CA: Stanford University Press.
22. Canto MT, Kawaguchi Y, Horowitz AM. Coverage and quality of oral cancer information in the popular press: 1987–98. J Public Health Dent (1998) 58:241–7.[Web of Science][Medline]
23. MacFadyen L, Amos A, Hastings G, et al. ‘They look like my kind of people'—perceptions of smoking images in youth magazines. Soc Sci Med (2003) 56:491–9.[CrossRef][Web of Science][Medline]
24. Stice E, Shaw H. Prospective relations of body image, eating, and affective disturbances to smoking onset in adolescent girls: how Virginia slims. J Consult Clin Psychol (2003) 71:129–35.[CrossRef][Web of Science][Medline]
25. Tomeo CA, Field AE, Berkey CS, et al. Weight concerns, weight control behaviors, and smoking initiation. Pediatrics (1999) 104:918–24.
26. Moreno-Lopez LA, Esparza-Gomez GC, Gonzalez-Navarro A, et al. Risk of oral cancer associated with tobacco smoking, alcohol consumption and oral hygiene: a case-control study in Madrid, Spain. Oral Oncol (2000) 36:170–4.[CrossRef][Web of Science][Medline]
27. Cawley J, Markowitz S, Tauras J. Lighting up and slimming down: the effects of body weight and cigarette prices on adolescent smoking initiation. J Health Econ (2004) 23:293–311.[CrossRef][Web of Science][Medline]
28. de Vries H, Engels R, Kremers S, et al. Parents’ and friends’ smoking status as predictors of smoking onset: findings from six European countries. Health Educ Res (2003) 18:627–36.
29. Biener L. Anti-tobacco advertisements by Massachusetts and Philip Morris: what teenagers think. Tob Control (2002) 11(Suppl. 2):ii43–46.
30. Bauer UE, Johnson TM, Hopkins RS, et al. Changes in youth cigarette use and intentions following implementation of a tobacco control program: findings from the Florida Youth Tobacco Survey, 1998–2000. JAMA (2000) 284:723–8.
31. Milton B, Cook PA, Dugdill L, et al. Why do primary school children smoke? A longitudinal analysis of predictors of smoking uptake during pre-adolescence. Public Health (2004) 118:247–55.[CrossRef][Web of Science][Medline]
32. Regis D. Differences between paper and online surveys. Educ Health (2006) 24:56.
33. Wakefield M, Flay B, Nichter M, et al. Effects of anti-smoking advertising on youth smoking: a review. J Health Commun (2003) 8:229–47.[CrossRef][Web of Science][Medline]
34. Thrush D, Fife-Schaw C, Breakwell GM. Evaluation of interventions to reduce smoking. Swiss J Psychol (1999) 58:85–100.[CrossRef]
35. Janz NK, Becker MH. The Health Belief Model: a decade later. Health Educ Q (1984) 11:1–47.[Web of Science][Medline]
36. Ben Ahron V, White D, Phillips K. Encouraging drinking at safe limits on single occasions: the potential contribution of protection motivation theory. Alcohol Alcohol (1995) 30:633–9.
37. Greening L, Stoppelbein L. Young drivers’ health attitudes and intentions to drink and drive. J Adolesc Health (2000) 27:94–101.[CrossRef][Web of Science][Medline]
38. Milne S, Orbell S, Sheeran P. Combining motivational and volitional interventions to promote exercise participation: protection motivation theory and implementation intentions. Br J Health Psychol (2002) 7:163–84.[CrossRef][Web of Science][Medline]
39. Neuwirth K, Dunwoody S, Griffin RJ. Protection motivation and risk communication. Risk Anal (2000) 20:721–34.[CrossRef][Web of Science][Medline]
Received on April 12, 2006; accepted on November 1, 2007
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||