5.8 The smoking behaviour of peers, and peer attitudes and norms

Last updated October 2014

Smoking during adolescence is primarily a social activity,1 and research has consistently identified peer group influences as a significant factor in uptake of smoking.2–7 Peer influence has been identified as important in future smoking intentions among adolescent smokers and non-smokers,8,9 as well as in the development of nicotine dependence in adolescent smokers.10 Peer groups may variously be defined as best friendships, romantic attachments, small social 'networks' and larger social 'crowds'.i 11 Each of these types of peer group may influence the decision to smoke or not to smoke.11

A recent review of peer group influences on adolescent smoking based on longitudinal studies published from 1999 explored the extent to which peer smoking predicts adolescent smoking.12 The authors identified 40 prospective studies linking peer group smoking or measures of substance use, including smoking, to future adolescent use. All but one of the studies reported positive associations between peer use at baseline and adolescent smoking at follow-up. The review also highlighted the variation of peer influence on adolescent smoking by socio-demographic characteristics, including gender differences (said to be well established, with girls more strongly influenced by peer smoking than boys; see Section 5.8.1), and less clear effects of age group and ethnicity. Social bonds (e.g. with school and family) and individual characteristics such as genetics and personal attributes were also found to moderate peer smoking.12

Peer pressure is one aspect of peer socialisation, in which adolescents are influenced by their friends:12 just how influential it is remains a matter of some debate. Some commentators have argued that the importance of peer influence has been overestimated, and that the clustering of smoking behaviour within peer groups could be because adolescents seek out friendships with individuals who share similar interests, of which smoking may be just one signifier,13,14,15 (i.e. 'selection').12 In a review of peer influence and smoking behaviour, Michell16 concluded that the effect of peer pressure as an influence on adolescent health behaviour is not proven, and is in practice very complex to decipher. It is important to understand that young people are not a homogeneous group, and that there are distinct peer clusters who smoke and do so for different reasons.16 It is probable that peer influences both interact with and are compounded by a host of other predictive factors:17 selection and socialisation processes can operate independently, but may also have reciprocal effects. Studies have found support for an important role of either and/or both processes, but their relative importance is strongly debated.12 Based on a review of 13 studies, Simons-Morton and colleagues concluded that while both socialisation and selection processes contribute to peer group homogeneity in terms of smoking behaviour, evidence from studies using more advanced research designs suggested the influence of selection was somewhat greater.12

It is likely that the nature of peer influences on smoking changes over time and across social and cultural groupings.16 While some research suggests that peer influences may vary in importance at differing points along the adolescent continuum, with the influence of close friends' smoking having most impact in earlier adolescence,1 there is evidence (2009) suggesting that parental and sibling smoking behaviour may be more important earlier.18 Recent research among New Zealand high school students found that the influence of smoking by parents and best friend varied with stage of adolescent tobacco smoking, progressively increasing with smoking frequency.19 There is also some evidence that peer influence on smoking behaviour or intentions may vary with adolescents' previous smoking experience.9

Some features of adolescent social networks may have long-lasting associations with longer term smoking behaviour.20 A recent longitudinal analysis based on a nationally representative sample of US adolescents followed for eight years from high school into adulthood attempted to empirically quantify the role of peer social networks in explaining smoking behaviour.21 Investigators controlled for potential biases in the data: parental characteristics (such as parent smoking status and education) and other socio-demographic factors (e.g. age, grade level, gender, access to cigarettes at home). They estimated that a 10% increase in the proportion of classmates who smoke will increase the likelihood of smoking by more than 3%, while an increase in smoking rates of 10% among an individual's close friends will increase the likelihood of smoking by 5%. They concluded that peer effects are significant determinants of smoking, which persist into adulthood.21 Similarly, recent Australian analyses examined peer effects in smoking for 75 000 adolescents aged 15–16 years across 26 European countries.22 Based on international school survey data–which is highly comparable across countries due to common sampling frames, methods and questionnaires–statistically significant and large peer effects were observed in almost all cases. It was estimated that, on average across countries, a one percentage point increase in the proportion of smoking classmates is associated with an increase of 0.31 to 0.38% in the probability that a 'typical' adolescent smokes, a range that varied between countries.22

The common perception that 'peer group pressure' equates to open coercion is not necessarily the case: most of the evidence indicates that socialisation is mainly a normative process and not one of overt peer pressure.12 Initiating smoking may arise as a response to more subtle influences, such as being a means of facilitating acceptance and bonding, and avoiding exclusion from peer groups.7,11 Qualitative research from Western Australia found that some adolescent experimenters and smokers saw trying a cigarette in the spirit of 'joining in' or 'giving it a go'. However, the same research found that young people of Indigenous or lower socio-economic status background were much more likely to describe overt peer pressure or inducement to try smoking.23,24

Socialising processes that facilitate smoking can also discourage use.12 For example, being 'cool' is important to teenagers,25 although what is deemed to be cool also changes across time, peer groups and social contexts. Smoking has traditionally been viewed as one of the badges of 'coolness' among teenagers.26 While 'coolness' is still identified by young people as one of the reasons why some of their peers smoke,27 research undertaken in Western Australia for the Smarter than Smoking project suggests that the inverse is increasingly true, with those who smoke often regarded as 'losers' or 'trying too hard to be cool'.24 Refusing an offer of cigarettes or declaring that 'I don't smoke' is increasingly socially acceptable and normative among many youth cohorts.24 Among groups with a negative prevailing attitude to smoking, peer influence may of course deter uptake of smoking.1,11,12

British research has found that forming romantic attachments ('dating') at an earlier age is a predictor for becoming a smoker later, independent of other possible confounding factors. The authors speculate that dating and smoking behaviours may be connected by a desire to appear to be more grown up,28 which is consistent with tobacco industry advertising linking its products with sex appeal and popularity.29 Research also suggests that young people who are lesbian, gay or bisexual (LGB) have a higher risk of taking up smoking.30,31 Reasons for this could be that LGB teenagers may be at greater risk of experiencing loneliness, isolation, harassment and depression. The LGB population has also been specifically targeted in tobacco advertising.30

Social norms need only to be perceived to influence behaviour:12 young smokers tend to congregate together, and also to overestimate the extent of smoking in their own age group, giving them a distorted sense of what is normal behaviour.6,12,32 The National Drug Strategy Household Survey for 200433 found that about two-thirds (64%) of recent smokers aged 12–15 said that 'all or most' of their friends and acquaintances were smokers as well. A further 20% said that 'about half' of their peer group were smokers. Only 1% of smokers in this age group said that none of their friends and acquaintances smoked. The effect declined with increasing age, with 28% of smokers aged 20 years or older reporting that 'all or most' of their friends and acquaintances were smokers, 32% saying 'around half were', and 39% saying only a few of their peer group smoked. Adolescents who think that smoking is normative and that most of their peers smoke are more likely to start smoking.1,34

5.8.1 Influence of gender

A large number of studies have examined whether boys and girls are similarly affected by the various factors that influence smoking behaviour. In a major review of the literature, the US Surgeon General's report for 2001 (Women and Smoking) concluded that 'Most risk factors for smoking initiation appear to be similar among girls and boys' (p477.32) However the review did find that girls may be more likely to be influenced by positive images of smoking, perceptions about smoking and weight control, and improvement of mood. There was also some evidence that girls are more likely to smoke than boys out of rebelliousness, rejection of conventional values, lack of religious conviction, poor self-esteem and emotional distress. Recent Dutch research based on a national cross-sectional survey among young people aged 9–13 years found gender differences in the association between pre-adolescent smoking initiation and emotional or behavioural problems, with relationships between smoking and problem behaviour (attention problems, thought problems and delinquent behaviour) observed in girls, but not in boys.35

Other research has proposed that males are more likely to smoke as a result of 'psychosocial' factors (such as risk taking, rebelliousness, self-esteem and coping ability), whereas girls tended more to be influenced by 'environmental' factors such as parental smoking habits, peer group attitudes and behaviours.36 A social network analysis of adolescent smoking in Finland, however, found that both male and female smokers were influenced by parental smoking behaviour and tended to select other smokers as friends, while only females were influenced to smoke by their peer group.37 The strength of the relative influence of psychosocial and environmental factors is likely to change during teenage years.36 Koval and colleagues found that psychosocial factors are more closely associated with smoking in young teenage girls than in older girls, who are more influenced by attitude variables (including beliefs about smoking), while younger boys are less susceptible to the influences of both psychosocial factors and environmental variables than older males.

Australian studies have reported that uptake of smoking in adolescent girls is strongly related to a desire to adopt and reinforce their reputation among a specific peer group,38 and that strength of self-concept in girls (defined as how an individual perceives her physical presentation and appearance to others) is more closely connected with increased likelihood of smoking than in boys.39 The authors of this study speculate that environmental pressures on adolescent girls to be self-confident, socially aggressive and sexually precocious may lead to cigarette smoking, in an effort to boost physical self-concept.39 Do concerns about body weight influence the uptake of smoking?

The perception that smoking depresses appetite, hence assisting with weight control, has long been considered a possible enticement for smoking, especially among females.ii Over many decades the tobacco industry has overtly targeted the female market with brands and imagery connecting cigarettes with a slim and shapely female form.32,40,41

A large number of studies have investigated the relationship between adolescent smoking and body weight. A review by Potter and colleagues42 analyses 55 studies published between 1980–2003. Taking into account the wide variation in study methodologies, this review concluded that there was some evidence that:

  • young smokers were more likely to perceive that they were overweight
  • some adolescents smoked because they thought it would help with weight control
  • adolescent smokers were more likely to have engaged in dieting, the evidence being strongest for girls.

Research from Western Australia24 has shown that young people (both smokers and non-smokers) cite 'to be thin' as a reason for smoking. Recent evidence links smoking among young females with body shape concerns and eating disorder symptoms,43,44 including extreme dieting (e.g. fasting or use of laxatives or diet pills). 45 There are several recently published US studies linking increased smoking over time with weight issues among young females. For example, research found increases in smoking prevalence over four years of high school among adolescent girls to be associated with self-perceptions of overweight,46 while a twins study found both underweight and overweight/obesity to be positively associated with the transition from experimentation to regular smoking among females aged 18–29 years.47 Analyses of data from a large longitudinal US adolescent survey suggest that for females aged 11–23 years, being overweight–based on measured body mass index (BMI) in 1996–was associated with frequent/heavy smoking five years later.48

Other research links body weight, dietary behaviours and smoking among young males and females. For example, a Canadian study found that girls who perceived themselves to be overweight and boys with higher self-reported BMI at two time points during high school were more likely to be smokers as young adults.49

Extreme dieting among nationally representative samples of US adolescents in grades 9–12 between 1999 and 2007 has been linked with current smoking regardless of gender and weight status:45 extreme dieters showed a significantly higher current smoking rate than their peers, and higher variability of current smoking rate. Trends in this data also suggest that the belief in the link between smoking and weight control may be becoming more pervasive: while some of the extreme dieting behaviours were not associated with current smoking in the years 1999–2003, all of the extreme behaviours were associated with adolescents' current smoking from 2005, regardless of gender and weight.45 While the magnitude of the association between smoking and extreme dieting became smaller among adolescents in later survey years, it remained unchanged among non-overweight girls over that same time period. The authors speculate that this might reflect the stubborn nature of smoking behaviour in this group and therefore that it would be more difficult to implement smoking cessation intervention for female extreme dieters than for other groups.45 They also suggest that as extreme dieters showed greater variability in smoking rates than non-extreme dieters, extreme dieters are more heterogeneous in smoking behaviour, and therefore interventions against adolescent smoking among this group may need to be more varied than among non-extreme dieters.45

In a 2009 Dutch study, depressive symptoms were related to smoking among adolescent boys and girls similarly, but the effect for depression became non-significant only for girls when controlling for the effect of weight concerns and dieting. Researchers therefore proposed that smoking prevention programs for depressed girls might be enhanced by challenging the idea of smoking as a diet strategy and incorporating a focus on healthy weight regulation.50

The relationship between smoking uptake and mental health problems such as depressive symptoms is explored further in Section 5.5.2.


i The tobacco industry's use of 'viral' and 'buzz' marketing capitalises on the effectiveness of peer 'crowds'. See Chapter 11, Section 11.7.

ii Although smokers weigh, on average, less than people who have quit smoking and who have never smoked, the effect is modest and accrues over decades of smoking. Taking up smoking does not appear to be associated with weight loss.32

Relevant news and research

For recent news items and research on this topic, click here.(Last updated October 2020)


1. Turner L, Mermelstein R and Flay B. Individual and contextual influences on adolescent smoking. Annals of the New York Academy of Sciences 2004;1021:175–97. Available from: http://www3.interscience.wiley.com/journal/118766022/abstract

2. Conrad K, Flay B and Hill D. Why children start smoking cigarettes: predictors of onset. British Journal of Addiction 1992;87(12):1711–24. Available from: http://www3.interscience.wiley.com/journal/119335396/abstract

3. US Department of Health and Human Services. Preventing tobacco use among young people: A report of the Surgeon General. Atlanta, GA: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health, 2012. Available from: http://www.cdc.gov/tobacco/data_statistics/sgr/2012/

4. US Department of Health and Human Services. Reducing tobacco use: a report of the Surgeon General. Atlanta, Georgia: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health, 2000. Available from: http://www.cdc.gov/tobacco/sgr/sgr_2000/index.htm

5. Wang M, Fitzhugh E, Westerfield R and Eddy J. Family and peer influences on smoking behaviour among American adolescents: an age trend. Journal of Adolescent Health 1995;16:200-3. Available from: http://www.ncbi.nlm.nih.gov/pubmed/7779829

6. Tyas S and Pederson L. Psychosocial factors related to adolescent smoking: a critical review of the literature. Tobacco Control 1999;7(4):409–20. Available from: http://tobaccocontrol.bmj.com/cgi/content/full/7/4/409

7. Walsh R and Tzelepis F. Adolescents and tobacco use: systematic review of qualitative research methodologies and partial synthesis of findings. Substance Use & Misuse 2007;42(8):1269–321. Available from: http://www.informaworld.com/smpp/content~db=all?content=10.1080/10826080701204904

8. Ling PM and Glantz SA. Why and how the tobacco industry sells cigarettes to young adults: evidence from industry documents. American Journal of Public Health 2002;92(6):908–16. Available from: http://www.ajph.org/cgi/content/full/92/6/908

9. Vitória P, Salgueiro M, Silva S and De Vries H. The impact of social influence on adolescent intention to smoke: combining types and referents of influence. British Journal of Health Psychology 2009;14(Pt 4):681–99. Available from: http://www.ncbi.nlm.nih.gov/pubmed/19288976

10. de Leeuw R, Engels R, Vermulst A and Scholte R. Relative risks of exposure to different smoking models on the development of nicotine dependence during adolescence: a five-wave longitudinal study. Journal of Adolescent Health 2009;45(2):171–8. Available from: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=19628144

11. Kobus K. Peers and adolescent smoking. Addiction 2003;98(suppl.1):i37-55.

12. Simons-Morton B and Farhat T. Recent findings on peer group influences on adolescent smoking. Journal of Primary Prevention 2010;31(4):191-208. Available from: http://www.ncbi.nlm.nih.gov/pubmed/20614184

13. Urberg K, Degirmencioglu M and Pilgrim C. Close friend and group influence on adolescent cigarette smoking and alcohol use. Developmental Psychology 1997;33:834-44. Available from: http://www.ncbi.nlm.nih.gov/pubmed/9300216

14. de Vries H, Candel M, Engels R and Mercken L. Challenges to the peer influence paradigm: results for 12-13 year olds from six European countries from the European Smoking Prevention Framework Approach study. Tobacco Control 2006;15(2):83–9. Available from: http://tc.bmjjournals.com/cgi/content/abstract/15/2/83

15. Mercken L, Candel M, van Osch L and de Vries H. No smoke without fire: the impact of future friends on adolescent smoking behaviour. British Journal of Health Psychology 2011;16(1):170-88. Available from: http://www.ingentaconnect.com/content/bpsoc/bjhp/pre-prints/bjhp924

16. Michell L. Loud, sad or bad: young people's perceptions of peer groups and smoking. Health Education Research 1997;12:1-14. Available from: http://www.ncbi.nlm.nih.gov/pubmed/10166897

17. Holliday J, Rothwell H and Moore L. The relative importance of different measures of peer smoking on adolescent smoking behavior: cross-sectional and longitudinal analyses of a large British cohort. Journal of Adolescent Health 2010;47(1):58–66. Available from: http://www.jahonline.org/article/PIIS1054139X09007046/fulltext

18. Mercken L, Candel M, Willems P and de Vries H. Social influence and selection effects in the context of smoking behavior: changes during early and mid adolescence. Health Psychology 2009;28(1):73–82. Available from: http://psycnet.apa.org/index.cfm?fa=main.doiLanding&uid=2009-00026-014

19. Scragg R, Glover M, Paynter J, Wong G and McCool J. Association of parent and best friend smoking with stage of adolescent tobacco smoking. The New Zealand Medical Journal 2010;123(1326) Available from: http://www.nzma.org.nz/journal/123-1326/4440/

20. Pollard MS, Tucker JS, Green HD, Kennedy D and Go M-H. Friendship networks and trajectories of adolescent tobacco use. Addictive Behaviors 2010;35(7):678-85. Available from: http://www.ncbi.nlm.nih.gov/pubmed/20332061

21. Ali M and Dwyer D. Estimating peer effects in adolescent smoking behavior: a longitudinal analysis. Journal of Adolescent Health 2009;45(4):402–8. Available from: http://www.jahonline.org/article/PIIS1054139X09001001/fulltext

22. McVicar D. Estimates of peer effects in adolescent smoking across twenty six European Countries. Social Science & Medicine 2011;73(8):1186-93. Available from: http://www.sciencedirect.com/science/article/pii/S027795361100493X

23. Leavy J, Wood L, Rosenberg M and Phillips F. Try and try again: qualitative insights into adolescent smoking experimentation and notions of addiction. Health Promotion Journal of Australia 2010;21(3):208-14. Available from: http://www.ncbi.nlm.nih.gov/pubmed/21118068

24. Wood L, Lang A and Coase P. Smarter than Smoking Qualitative Research. A research report. West Perth, Australia: TNS Social Research, 2005.

25. Watson N, Clarkson J, Donovan R and Giles-Corti B. Filthy or fashionable? Young people's perceptions of smoking in the media. Health Education Research 2003;18:554-67. Available from: http://her.oxfordjournals.org/content/18/5/554.full.pdf+html

26. Danesi M. Cool: signs and meanings of adolescence. Toronto: University of Toronto Press, 1994.

27. Eureka Strategic Research. Youth tobacco prevention research project. Undertaken for the Australian Government Department of Health and Ageing. Canberra: Department of Health and Ageing, 2005. Available from: http://www.health.gov.au/internet/main/publishing.nsf/Content/phd-pub-tobacco-literature-cnt.htm

28. Fidler J, West R, Jarvis M and Wardle J. Early dating predicts smoking during adolescence: a prospective study. Addiction 2006;101:1805-13. Available from: http://www.ncbi.nlm.nih.gov/pubmed/17156180

29. DiFranza JR and Wellman R. Early dating and smoking initiation: some thoughts about a common cause [editorial]. Addiction 2006;101:1682-3.

30. Tang H, Greenwood G, Cowling D, Lloyd J, Roeseler A and Bal D. Cigarette smoking among lesbians, gays, and bisexuals: how serious a problem? (United States). Cancer Causes & Control 2004;15:797-803. Available from: http://www.ncbi.nlm.nih.gov/pubmed/15456993

31. Easton A, Jackson K, Mowery P, Corneau D and Sell R. Adolescent same-sex and both-sex romantic attractions and relationships: implications for smoking. American Journal of Public Health 2008;98:462-7. Available from: http://www.ncbi.nlm.nih.gov/pubmed/18235075

32. US Department of Health and Human Services. Women and smoking: a report of the US Surgeon General. Atlanta, Georgia: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health, 2001. Available from: http://www.cdc.gov/tobacco/data_statistics/sgr/sgr_2001/sgr_women_chapters.htm

33. Australian Institute of Health and Welfare. 2004 National Drug Strategy Household Survey: detailed findings. Drug statistics series no.16, AIHW cat. no. PHE 66. Canberra: AIHW, 2005. Available from: http://www.aihw.gov.au/publications/index.cfm/title/10190

34. Sargent J and DiFranza J. Tobacco control for clinicians who treat adolescents. CA A Cancer Journal for Clinicians 2003;53(2):102-23. Available from: http://caonline.amcancersoc.org/cgi/reprint/53/2/102

35. de Meer G, Crone M and Reijneveld S. Gender differences in the association between pre-adolescent smoking initiation and emotional or behavioural problems. BMC Public Health 2010;10(1):615. Available from: http://www.biomedcentral.com/1471-2458/10/615

36. Koval J, Pederson L and Chan S. Psychosocial variables in a cohort of students in grades 8 and 11: a comparison of current and never smokers. Preventive Medicine 2004;39:1017-25. Available from: http://www.ncbi.nlm.nih.gov/pubmed/15475037

37. Mercken L, Snijders T, Steglich C, Vertiainen E and de Vries H. Smoking-based selection and influence in gender-segregated friendship networks: a social network analysis of adolescent smoking. Addiction 2010;105(7):1280-9. Available from: http://www.ncbi.nlm.nih.gov/pubmed/20456296

38. Snow P and Bruce D. Cigarette smoking in teenage girls: exploring the role of peer reputations, self-concept and coping. Health Education Research 2003;18:439-52. Available from: http://www.ncbi.nlm.nih.gov/pubmed/12939126

39. Thornton W, Douglas G and Houghton S. Transition through stages of smoking: the effect of gender and self-concept on adolescent smoking behavior. Journal of Adolescent Health 1999;25:284-9. Available from: http://www.ncbi.nlm.nih.gov/pubmed/10505846

40. Ernster V. Mixed messages for women. A social history of cigarette smoking and advertising. NY State Journal of Medicine 1985;85:335-40. Available from: http://www.ncbi.nlm.nih.gov/pubmed/3900827

41. Amos A and Haglund M. From social taboo to 'torch of freedom': the marketing of cigarettes to women. Tobacco Control 2000;9:3-8. Available from: http://tobaccocontrol.bmj.com/content/9/1/3.full?sid=14378807-5d29-45f6-927c-40b199822b0b

42. Potter B, Pederson L, Chan S, Aubut J-A and Koval J. Does a relationship exist between body weight, concerns about weight, and smoking among adolescents? An integration of the literature with an emphasis on gender. Nicotine & Tobacco Research 2004;6:397-425. Available from: http://www.ncbi.nlm.nih.gov/pubmed/15203775

43. Groth S and Morrison-Beedy D. Smoking, substance use, and mental health correlates in urban adolescent girls. Journal of Community Health 2010;36(4):552-8. Available from: http://www.ncbi.nlm.nih.gov/pubmed/21107998

44. Kendzor D, Adams C, Stewart D, Baillie L and Copeland A. Cigarette smoking is associated with body shape concerns and bulimia symptoms among young adult females. Eating Behaviors 2009;10(1):56–8. Available from: http://www.ncbi.nlm.nih.gov/pubmed/19171320

45. Seo DC and Jiang N. Associations between smoking and extreme dieting among adolescents. Journal of Youth and Adolescence 2009;38(10):1364–73. Available from: http://www.springerlink.com/content/081u0480g38280n2/

46. Hong T, Rice J and Johnson C. Social environmental and individual factors associated with smoking among a panel of adolescent girls. Women & Health 2011;51(3):187–203. Available from: http://www.ncbi.nlm.nih.gov/pubmed/21547857

47. Duncan A, Lessov-Schlaggar C, Nelson E, Pergadia M, Madden P and Heath A. Body mass index and regular smoking in young adult women. Addictive Behaviors 2010;35(11):983–8. Available from: http://www.ncbi.nlm.nih.gov/pubmed/20634004

48. Rees D and Sabia J. Body weight and smoking initiation: evidence from Add Health. Journal of Health Economics 2010;29(5):774–7. Available from: http://www.ncbi.nlm.nih.gov/pubmed/20708283

49. Koval J, Pederson L, Zhang X, Mowery P and McKenna M. Can young adult smoking status be predicted from concern about body weight and self-reported BMI among adolescents? Results from a ten-year cohort study. Nicotine & Tobacco Research 2008 10(9):1449–55. Available from: http://www.informaworld.com/smpp/content~db=all?content=10.1080/14622200802239140

50. Larsen J, Otten R and Engels R. Adolescent depressive symptoms and smoking behavior: the gender-specific role of weight concern and dieting. Journal of Psychosomatic Research 2009;66(4):305–8. Available from: http://www.jpsychores.com/article/PIIS0022399908004868/fulltext