Karl Peltzer  ( Department for Management of Science and Technology Development, Ton Duc Thang University, Faculty of Pharmacy, Ton Duc Thang University, Ho Chi Minh City, Vietnam, South Africa. )
Supa Pengpid  ( ASEAN Institute for Health Development, Mahidol University, Salaya, Phutthamonthon, Nakhon Pathom, Thailand, Department of Research and Innovation, University of Limpopo, Polokwane, South Africa. )

Objective: To investigate the prevalence and correlates of tobacco use among school-going adolescents.
Methods: This cross-sectional study consisted of a secondary analysis of existing cross-sectional survey data from the Global School-based Health Survey from Cambodia and Vietnam in 2013, and comprised schoolchildren. Data was analysed using Stata 13.
Results: The study comprised 7,137 schoolchildren. The overall mean age was 15.5±1.4 years. Besides, 3,806(53.33%) were from Cambodia and 3,331(46.67%) from Vietnam. Overall, 279(5.1%) students reported currently using tobacco (smoking and/or smokeless), including 214(6.2%) boys and 65(1.6%) girls, and 127(3.6%) in Cambodia and 152(5.2%) in Vietnam. Current cigarette smoking was 239(4.5%) and smokeless tobacco use was 143(2.4%).
Conclusion: Tobacco use was found to be significantly high among schoolchildren in Cambodia and Vietnam.
Keywords: Tobacco use, Adolescents, Cambodia, Vietnam. (JPMA 67: 1024; 2017)


Introduction

Tobacco use has increased in the Association of Southeast Asian Nations (ASEAN) region, with a prevalence of adult smoking in men of 39.1% in Cambodia and 47.4% in Vietnam, respectively, and a prevalence of adult smokeless tobacco use in women of 12.5% and 2.3% in Cambodia and in Vietnam, respectively,¹ making ASEAN a big market and prime target for transnational tobacco companies to expand their market.² In adolescents (13-15 years) the prevalence of current tobacco use in Cambodia was 5.1% (7.2% in boys, 3.0% in girls) in 2003 and 6.3% (7.9% in boys and 5.0% in girls) in 2010, while 3.8% (6.5% in boys and 1.5% in girls) in 2007 in Vietnam.³ The average age of smoking initiation was 21.1 years in Cambodia and 19.8 years in Vietnam in 2011, being the oldest age to initiate smoking among ASEAN countries.¹ Also among adolescents, intentions to start smoking within the next year were the lowest within ASEAN, with  6.1%  in 2011 in Cambodia, and 6.6% in 2007 in Vietnam.^1,3 In Cambodia in 2015⁴ and in Vietnam in 2012⁵ tobacco control laws were ratified. There is a lack of more recent and contextual data on tobacco use among adolescents in Cambodia and Vietnam.
The prevalence of current tobacco use among adolescents in some of the other ASEAN countries ranged from 11.5% current smoking in 2012 in Malaysia,⁶ 12.7% (18.7% in boys and 6.0% in girls) in 2011 in Lao People\\\'s Democratic Republic (PDR), 13.7% (18.8% in boys and 9.3% in girls) in 2011 in the Philippines, 17.9% (26.9% in boys and 9.2% in girls) in 2009 in Thailand, and current tobacco use of 22.5% (41.1% in boys and 6.2% in girls) in 2009 in Indonesia.³ Sociodemographic factors and various risk factors have been identified to be associated with tobacco use among adolescents. Sociodemographic factors include being male,^6,7 older age,⁸ and low social status.⁹ Risk factors may include early initiation of smoking, exposure to passive smoking,⁶ current alcohol drinking,^6,10 illicit drug use,^6,10 consumption of energie and soft drinks,¹¹ eating fatty food and eating from fast food,¹² being overweight,¹³ having had sexual intercourse,¹⁰ being bullied,^6,14 and psychological distress.^7,10,14 The current study was planned to determine estimates of the prevalence and correlates of tobacco use among school adolescents in Cambodia and Vietnam using the most recent data (2013) available.


Subjects and Methods

This cross-sectional study consisted of a secondary analysis of existing cross-sectional survey data from the Global School-based Health Survey (GSHS) from Cambodia and Vietnam. Details and data of the GSHS can be accessed online.¹⁵ The year of data collection was 2013 in both Cambodia and Vietnam. A two-stage cluster sample design was used to collect data to represent all students in grades 6-10 in each country.¹⁵ At the first stage of sampling, schools were selected with probability proportional to their reported enrolment size.¹⁵ In the second stage, classes in the selected schools were randomly selected and all students in selected classes were eligible to participate irrespective of their actual ages.¹⁵ Students self-completed the questionnaire under the supervision of trained research assistants.¹⁵ To calculate the sample, prevalence estimation was 4%, sensitivity 0.1 and specificity 0.7, confidence 95% and desired precision of 0.05. Using EpiTools the sample size calculated for an assumed large (infinite) population was 6,592.
The study instruments used were from the GSHS15 (Table-1).

Body weight and height were assessed by self-report. The international age- and gender-specific child body mass index (BMI), calculated as weight/height2 (kg/m²), cut-points were used to define overweight and obesity.¹⁶ Adolescents were categorised as overweight or obese if their BMI was > +1 standard deviation (SD) and > +2 SD, respectively, from the median for BMI for age and gender.¹⁶ Psychological distress was assessed with three items (no close friends, feeling often or always lonely, and suicidal ideation).
Data was analysed using STATA software version 13 (Stata Corporation, College Station, Texas, the United States). Associations between socio-demographics, risk behaviour and current tobacco use among schoolchildren were evaluated calculating prevalence ratios (PR). Poisson regression estimated the impact of independent variables on current tobacco use (binary dependent variable). Independent variables found significant (p<0.05) in relation to the outcome variable in bivariate analysis were included in the final multivariate model. In the analysis, weighted percentages were reported. The reported sample size reflects the sample that was asked the target question. P<0.05 indicated statistical significance. Both the p-value and the reported 95% confidence intervals were adjusted for the multistage stratified cluster sample design of the study.


Results

Of the 7,137 schoolchildren, 3,318(46.5%) were boys and 3,804(53.3%) were girls. The overall mean age was 15.5±1.4 years. Moreover, 3,806(53.33%) participants were from Cambodia and 3,331(46.67%) were from Vietnam. The overall response rate was 3,325(85%) in Cambodia and 3,198(96%) in Vietnam. Overall, 279(5.1%) students reported they were currently using tobacco, including 214(6.2%) boys and 65(1.6%) girls. The prevalence of current cigarette smoking was 239(4.5%) and current smokeless tobacco use was 143(2.4%). The prevalence of current tobacco use was 127(3.6%) in Cambodia and 152(5.2%) in Vietnam (Table-2).


In univariate analysis, socio-demographic factors (older age: crude prevalence ratio [CPR] = 2.96, 95% confidence interval (CI) =1.63-5.39 and being male: CPR=5.21, 95% CI=2.94-9.23) and risk behaviour (smoked before the age of 14 years: CPR=8.48, 95% CI=6.11-11.78, exposure to passive smoking in the past 7 days: CPR=4.02, 95% CI=2.38-6.80, current alcohol use: CPR=7.83, 95% CI=5.83-10.52, ever illicit drug use: CPR: 10.54, 95% CI=6.59-16.85, having had once or more daily soft drinks: CPR: 1.80, 95% CI=1.30-2.50, having had twice or more times a week fast food: CPR: 2.05, 95% CI=1.35-3.11, ever had sex: CPR=6.72, 95% CI=4.88-9.26, and psychological distress: CPR: 3.70, 95% CI=1.70-5.19) were found to be associated with current tobacco use.
Compared to respondents of age 13 to 14 years, those who were 17 to 18 years and those who were 15 to 16 years of age were more likely to be tobacco users, adjusted prevalence ratio (APR)=2.01, 95% CI=1.22-3.31) and APR=1.66, 95% CI=1.13-2.44, respectively. Compared to female participants, males were 3.47 (95% CI=1.98-6.10) times more likely current tobacco users. Tobacco-related risk factors that were identified in this analysis were having started smoking before the age of 14 years (APR=3.45, 95% CI=2.65-4.49) and exposure to passive smoking in the past 7 days (APR=3.35, 95% CI=1.83-6.13). Compared with participants who did not currently consume alcohol, those who consumed alcohol were 3.47 (95% CI=2.45-4.92) times more likely to currently use tobacco. Participants who had ever taken illicit drugs were 2.55 (95% CI=1.49-4.36) times more likely to use tobacco compared to those who had never taken illicit drugs. Respondents who ever had sex were 2.69 (95% CI=1.91-3.78) times more likely to be current tobacco users compared to those who never had sexual intercourse (Table-3).




Discussion

The study investigated in large, nationally representative samples of school-going adolescents in Cambodia and Vietnam the prevalence and correlates of tobacco use. Compared to previous Global Youth Tobacco Surveys (GYTS)³ in Cambodia, the prevalence of current tobacco use among school-going adolescents reduced from 5.1% in 2003, 6.3% in 2010 to 3.8% in 2013, while it increased from 3.8% in 2007 to 5.2% in 2013 in Vietnam.³
It is not clear why tobacco use among adolescents decreased in Cambodia and increased in Vietnam, especially since tobacco control legislation was implemented from 2012 in Vietnam and only from 2015 in Cambodia.^4,5 Yet, current tobacco use prevalence rates among adolescents in Cambodia and Vietnam seem to be still lower compared to other ASEAN countries (e.g., Indonesia, Lao PDR, Philippines, Thailand).^3,6 A possible reason for the relatively low tobacco use prevalence among adolescents in Cambodia and Vietnam is the "late" initiation of smoking (on average 20 years and more) in the latter countries compared to other ASEAN countries.¹ Apart from tobacco control effects during adolescents, Cambodia and Vietnam should strengthen tobacco control efforts during early adulthood.
In agreement with previous studies,^6-8 this study found an association between being male and older age and current tobacco use. Having experienced hunger (as a proxy for low socio-economic status) was not associated with tobacco use, as found previously.⁹ Further, as also observed previously,^6,10 this study found strong associations between substance use variables (early initiation of smoking, exposure to passive smoking, current alcohol drinking and illicit drug use) and tobacco use. The high exposure to passive smoking in this study may explain that tobacco use behaviour may be modelled by parents or guardians or others who smoke in the presence of the adolescent.⁶ In addition, having had sexual intercourse was another risk behaviour, which was found associated with tobacco use, as found in a previous study.¹⁰ These two behaviours (having sexual intercourse and tobacco use) can be considered as pointers of a risk-taking lifestyle or sensation-seeking activities.^18-20
Unlike in some previous studies,^11,12 this study did not find an association between the consumption of soft drinks, fast food and tobacco use in the final model. Psychological distress was in bivariate analysis associated with tobacco use, as found in the previous studies.^7,10,14 Unlike in some previous studies,^6,14 bullying victimisation was not found to be associated with tobacco use.
Study limitations include the cross-sectional nature of the study, which means one point in time we would have not been able to assess the tobacco use of those students who were absent that day. Moreover, data was collected by self-report, which may have biased the results.


Conclusion

A large number of adolescents in Cambodia and Vietnam used a significant proportion of tobacco. Several socio-demographic and a number of risk factors were identified, which can help guide interventions to prevent tobacco use in this population. In particular, tobacco control programmes should incorporate seemingly clustering other risk behaviours, such as alcohol and illicit drug use and sexual behaviour.


Acknowledgements

We are grateful to the World Health Organisation (WHO) and the Centres for Disease Control and Prevention for making the data publicly available for this analysis. We also thank the country coordinators from Cambodia (Chher Tepirou) and Vietnam (LE Thi Hoan) for collecting the GSHS data. The ministries of education and health and the study participants in the GSHS in the two ASEAN countries are acknowledged. The governments of the respective study countries and the World Health Organisation did not influence the analysis nor did they have an influence on the decision to publish these findings.

Disclaimer: None.
Conflict of Interest: None.
Source of Funding: None.


References

1. Lian IY, Dorotheo U. The ASEAN tobacco control atlas, 2014. Bangkok: Southeast Asia Tobacco Control Alliance, 2014; pp. 1-88.
2. Slama K, Callard C, Saloojee Y, Eithuphakdee B. Effective health promotion against  tobacco use. In: McQueen DV, Jones CM, eds. Global perspectives on health promotion effectiveness. New York: Springer, 2007; pp. 150-63.
3. Global Youth Tobacco Survey (GYTS) Fact sheets. [Online] [Cited 2016 April 10]. Available from URL: http://nccd.cdc.gov/ GTSSData/Ancillary/DataReports.aspx-CAID=1
4. WHO Framework Convention On Tobacco Control. Cambodia: new law on tobacco control passed, 2015. [Online] [Cited 2016 April 10]. Available from URL: http://apps.who.int/fctc/ implementation/database/groups/cambodia-new-law-tobacco-control-passed
5. Vietnam Steering Committee on Smoking and Health (VINACOSH) and Southeast Asia Tobacco Control Alliance (SEATCA). The ASEAN tobacco control report. [Online] [Cited 2016 April 10]. Available from URL: http://seatca.org/dmdocuments/ATCRC,May%202015.pdf.
6. Tee GH, Kaur G. Correlates of current smoking among Malaysian secondary school children. Asia Pac J Public Health. 2014; 26:70S-80S.
7. Zhang F, Zhang T, Xiong J, Zhang L, Chen J.  Analysis on prevalence and influence factors of smoking and drinking among middle school students of Tujia and Miao in Xiangxi Area. Wei Sheng Yan Jiu. 2015; 44:750-5.
8. Siziya S, Muula AS, Rudatsikira E. Prevalence and correlates of current cigarette smoking among adolescents in East Timor-Leste. Indian Pediatr. 2008; 45:963-8.
9. Lampert T, Thamm M. [Consumption of tobacco, alcohol and drugs among adolescents in Germany. Results of the German Health Interview and Examination Survey for Children and Adolescents (KiGGS)]. Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz. 2007; 50: 600-8.
10. Takakura M, Ueji M, Sakihara S. Covariation of cigarette smoking and other health-risk behaviors among Japanese high school students: a preliminary study. J Epidemiol. 2001; 11: 224-8.
11. Terry-McElrath YM, O’Malley PM, Johnston LD. Energy drinks, soft drinks, and substance use among United States secondary school students. J Addict Med. 2014; 8:6-13.
12. Nouira A, Maatoug J, Harrabi I, Hmad S, Belkacem M, Slama S, et al. Clustering of risk factors in the smoking habits of school children in Sousse, Tunisia. Int J Adolesc Med Health. 2014; 26: 267-73.
13. Potter BK, Pederson LL, Chan SS, Aubut JA, Koval JJ. 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 Tob Res. 2004; 6:397-425.
14. Guo L, Hong L, Gao X, Zhou J, Lu C, Zhang WH. Associations between depression risk, bullying and current smoking among Chinese adolescents: Modulated by gender. Psych Res. 2016; 237:282-9.
15. Centers for Disease Control (CDC). The Global School and Health Survey background, 2015. [Online] [cited 2016 April 10]. Available from: URL: http://www.cdc.gov/gshs/background/index.
16. Cole TJ1, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide. BMJ 2000; 320:1240-3.
17. The World Bank. Countries and economies [Online] 2015[cited 2016 April 10]. Available from URL: http://data.worldbank.org/country.
18. Graves KL, Leigh BC. The relationship of substance use to sexual activity among young adults in the United States. Fam Plann Perspect. 1995; 27:18-22.
19. Peltzer K, Pengpid S. Risk and protective factors affecting sexual risk behavior among school-aged adolescents in Fiji, Kiribati, Samoa, and Vanuatu. Asia Pac J Public Health. 2016; 28:404-15.
20. Zuckerman M. Sensation seeking: Beyond the optimal level of arousal. Hillsdale, NJ: Lawrence Erlbaum Associates, 1979.