Association Between Symptoms of Common Mental Disorders and Adolescent Smoking: A Cross-Sectional Study From Indonesia's Basic Health Research Survey.

Association Between Symptoms of Common Mental Disorders and Adolescent Smoking: A Cross‐Sectional Study From Indonesia's Basic Health Research Survey

ABSTRACT

Background and Aims

Evidence on the association between common mental disorder symptoms and adolescent smoking in a large nationally representative sample remains limited. We investigated this relationship among Indonesian adolescents using national survey data.

Methods

We analysed cross‐sectional data from the 2018 Indonesian Basic Health Research survey, a two‐stage stratified sample of 176,555 adolescents aged 10–18 years. Symptoms of common mental disorder were assessed in 70,969 adolescents aged 15–18 years using the Mini International Neuropsychiatric Interview (MINI) questionnaire and the Self‐Reporting Questionnaire‐20. We estimated the prevalence of adolescent smoking. We developed a directed acyclic graph (DAG) to identify the potential confounders and applied multivariable logistic regression models with survey weights to evaluate the association between common mental disorder symptoms and adolescent current smoking, controlling for measured confounders.

Results

The prevalence of adolescent current smoking aged 10–18 years in Indonesia was 9.10% (95% CI: 9.03, 9.16). The prevalence of adolescent current smoking among those completing common mental disorder symptom assessment was 18.14% (95% CI: 18.00, 18.18). Both cohorts showed higher prevalence in male, older, and rural groups. Adolescents with symptoms of common mental disorder had increased odds of current smoking: estimated odds ratio of 1.59 (95% CI: 1.40, 1.81). Important measured confounders of the estimation obtained from DAG analysis and adjusted for in the final model were age, sex, current education, place of residence, working status, and health behaviours (alcohol use, physical activity, and dietary patterns).

Conclusions

The prevalence of adolescent smokers in Indonesia is higher than reported previously. This study established the association between common mental disorder symptoms and adolescent current smoking. Incorporating mental disorder screening and support into prevention strategies could strengthen interventions and inform future policy to reduce the increased prevalence of youth smoking.

1Introduction

Tobacco consumption among adolescents is a critical public health concern, as most adult smokers begin smoking during adolescence, leading to lifelong nicotine dependence and addiction [1]. Early initiation is particularly harmful: starting at a younger age increases the likelihood of premature death from smoking‐related adverse health conditions such as cardiovascular diseases, chronic obstructive pulmonary disease (COPD), and various cancers, alongside short‐term impacts like impaired lung development and reduced physical fitness [2].

Globally, an estimated 11.3% adolescents are current smokers [3]. While many countries reported a decline in prevalence over the years, adolescent smoking prevalence in Indonesia has remained persistently high, with national survey data showing an increase among youth aged 10–18 years from 7.2% in 2013 [4] to 9.1% in 2018 [5], while recent Global Youth Tobacco Survey (GYTS) data indicates that tobacco use reached 19.2% among school‐going adolescents [6]. This has prompted the Indonesian government to prioritise reducing smoking prevalence among adolescents aged 10–18 years, given its potential long‐term burden on adult health and the associated risk to the nation's demographic dividend projected between 2030 and 2045 [7].

In Indonesia, 28.9% of individuals aged ≥ 10 years are active smokers [5]. Most initiated smoking at a very young age, 23.1% started smoking at 10%–14% and 52.1% at 15–19 [5]. To control different factors, the government has implemented programmes targeting adolescents, including smoke‐free areas, excise taxes and cigarette price increases (sin tax), and bans on advertising, promotion, and sponsorship [7]. These interventions primarily target environmental factors. However, individual‐level determinants, particularly mental disorder, have received little attention, despite adolescence being a period of major neurological and psychological changes that may influence risk behaviours [8].

In psychiatric epidemiology, symptoms of mental disorders in non‐clinical populations are conceptualised as common mental disorders, representing a continuum of psychological distress, such as depression, anxiety, and somatic symptoms that impairs functioning without requiring a formal clinical psychiatric diagnosis [9, 10, 11]. There are several compelling evidence that mental health problems like depression, stress, and anxiety are positively associated with enhanced smoking among adolescents, including earlier initiation and increased consumption [12, 13, 14, 15, 16]. Such adolescents are also less likely to quit smoking and more prone to relapse in adulthood [17]. Evidence on nicotine's interaction with neurotransmitter pathways supports the hypothesis that mental disorders may increase susceptibility to smoking by reinforcing a cycle of use and dependence [18, 19]. This supports the hypothesis that the presence of symptoms of common mental disorders is associated with increased likelihood of current smoking among Indonesian adolescents. However, many prior studies have relied on small cohorts [13, 15, 16]. Relatively few studies have examined the association between symptoms of common mental disorder and current smoking using large‐scale population‐based datasets [12, 14], particularly considering the complexity of the relationship in the presence of multifaceted risk factors. In addition, most studies have focused on adult populations, limiting their generalisability to adolescents.

To address the research gap, the present study investigated the association between symptoms of common mental disorder and current smoking among adolescents. Using nationally representative data from the 2018 Indonesian Basic Health Research Survey, we estimated the prevalence of adolescent current smoking in the 10–18‐year age group. We then constructed a directed acyclic graph (DAG) to identify potential confounders and incorporated these into multivariable models to estimate the association of symptoms of common mental disorder with adolescent current smoking.

2Methods

2.1Data Source

This cross‐sectional study used secondary data from the 2018 Indonesian Basic Health Research (IBHR) survey, a nationally representative survey of the Indonesian population. The IBHR survey collected information on a broad range of health domains, including health services, communicable and non‐communicable diseases, nutrition, reproductive and maternal health, dental and oral health, disabilities and injuries, mental health, and behavioural risk factors. Data were collected through interviews, measurements, and examinations. The survey was conducted in April and May 2018 based on sampling in March 2018 [5].

2.2Sampling Technique

The IBHR survey employed a two‐stage stratified sampling design [5]. In the first stage, from 720,000 census blocks across Indonesia, 180,000 (25%) were selected as the sampling frame using implicit stratification based on welfare strata. A census block is a small administrative unit representing a small part of a village, usually containing around 80–120 households. From this sampling frame, 30,000 census blocks were systematically selected using a probability proportional to size within each urban/rural stratum across districts and municipalities. In the second stage, ten households were selected from each census block using systematic sampling, stratified by the head of the household's education level. This yielded 300,000 households with all household members invited to participate in the survey.

In the smoking survey, the questionnaire was asked to respondents aged ≥ 10 years, and 818,507 individuals completed the survey. To estimate the prevalence of adolescent smoking, the present study considered 176,555 adolescent respondents aged 10–18 years [5]. Per the official 2018 IBHR survey protocol, common mental disorder symptom questionnaires (SRQ–20 and MINI) were administered exclusively to respondents aged 15 years and above, reflecting the validated age boundaries of these instruments in Indonesia. A total of 70,969 individuals aged 15–18 years completed the survey. This subset data represented the primary analytic cohort for evaluating the association between the presence of common mental disorder symptoms and current smoking. To evaluate potential sub‐sample selection bias, sociodemographic distributions of this analytic sub‐sample (15–18 years) were evaluated against the full adolescent sample (10–18 years).

The 2018 Indonesian Basic Health Research (IBHR) survey protocol was reviewed and ethically approved by the Health Research Ethics Committee of the Health Research and Development Agency (NIHRD/Litbangkes), Ministry of Health of the Republic of Indonesia (Ethical Approval Number: LB.02.01/2/KE.267/2017). Written informed consent was obtained from all adult participants and from parents/legal guardians for minor participants prior to data collection, along with informed assent obtained from the minor participants. This present study is a secondary analysis of the fully anonymized and de‐identified IBHR dataset. Official permission to access and use the dataset was granted by the Health Policy Agency (BKPK), Ministry of Health of the Republic of Indonesia (Data Permission Letter Number: IR.03.01/H.I/3663/2023). Because this study utilized non‐identifiable secondary data, additional institutional review board (IRB) approval was exempt.

2.3Study Variables

The outcome, current smoking, was defined as having smoked any tobacco or nicotine products, including manufactured filter cigarettes, clove cigarettes (kretek), hand‐rolled cigarettes (linting), electronic cigarettes, and shisha/hookah within the 30 days prior to the survey, in accordance with World Health Organization (WHO) Global Youth Tobacco Survey (GYTS) measurement standards [6]. Respondents who had not smoked within the past 30 days were classified as non‐smokers (1 = current smoker, 0 = non‐smoker).

The exposure was the presence of common mental disorder symptoms (binary), measured as a composite variable based on the presence of depression symptoms (assessed via the Mini International Neuropsychiatric Interview [MINI]) or emotional mental disorder symptoms (assessed via the Self Reporting Questionnaire‐20 [SRQ‐20] with a national validated cutoff ≥6) [5, 9, 20, 21]. These measures reflected the respondent's mental state in the 2 weeks prior to the survey, capturing population‐level psychological distress rather than a formal clinical psychiatric diagnosis [10, 11]. Detailed operational definitions, coding procedures, and categorizations for all study variables are summarized in Supplementary Table 2.

Potential confounders were identified based on previous literature and mapped within the DAG conceptual framework. Demographic factors included sex (male or female), age (15, 16, 17, and 18 years), and place of residence (urban or rural), accounting for age‐graded smoking exposure, strongly gendered cultural norms in Indonesia, and geographic differences in tobacco accessibility [22, 23, 24]. Variables representing socioeconomic status were current education (five levels: not currently studying, elementary, junior high, senior high, and university) and working status (three levels: working, studying, not working or studying), capturing differential purchasing power and peer group exposures in academic versus workplace environments [25]. Behavioural and lifestyle covariates included physical activity (less, moderate, vigorous) based on the Global Physical Activity Questionnaire (GPAQ) calculation [26], dietary behaviour representing the amount and frequency of taking unhealthy foods and beverages (five levels: less or never, several times a month, several times a week, daily, and several times a day) according to the previous study categorisation [27], fruit and vegetable consumption (below or meets recommendation) based on WHO guidelines [28], and alcohol consumption (yes, no) based on the consumption of alcoholic drinks in the last 30 days. These lifestyle factors were included to account for broader health‐related behaviours and lifestyle patterns (lifestyle clustering) where health‐compromising habits co‐occur with psychological distress and substance use initiation [29, 30].

2.4Data Analyses

We summarised all categorical variables as frequencies and percentages. The prevalence of adolescent current smoking and the corresponding 95% confidence intervals (CI) were estimated for the full dataset (10–18 years) and the subset (15–18 years) for the overall population, and stratified by sex, age group, and place of residence. Prevalence estimates were obtained using a logistic regression model accounting for the survey weights.

To identify and visualize potential confounding structures, we developed a qualitative, domain‐knowledge‐driven Directed Acyclic Graph (DAG) using the open‐source software DAGitty (http://dagitty.net/) [31]. DAG was employed strictly as a conceptual structural framework to guide variable selection based on directional separation (d‐separation) rules, rather than a quantitative model‐fitting procedure; thus, goodness‐of‐fit metrics were not computed. Applying d‐separation criteria identified the minimal sufficient adjustment set required to block open backdoor paths between common mental disorder symptoms and current smoking. This adjustment set comprised sociodemographic variables (sex, age, residence, education, working status) and lifestyle behaviours (alcohol consumption, physical activity, fruit/vegetable intake, and unhealthy food and beverage consumption). Unobserved theoretical factors, specifically genetic predisposition, were explicitly included as latent (unobserved) nodes in the DAG (represented by hatched circles and dashed biasing paths in Figure 1) to transparently acknowledge residual confounding inherent in secondary survey data.

We fitted survey‐weighted multivariable logistic regression models of symptoms of common mental disorder on current smoking, adjusting for the identified confounders. Analyses were restricted to the subset of respondents aged 15–18 years who completed the common mental disorder questionnaire (n = 70,969). Missing data across the exposure, outcome, and covariates was negligible (< 0.003%, n < 20); thus, a complete case analysis approach was applied, excluding observations with missing values. Odds ratios (ORs) and 95% CIs were reported. All statistical tests were two‐sided with a type I error rate of 5%.

All data management and statistical analysis were conducted in the R Statistical Software environment (version 4.4.3). The survey package was used for fitting weighted logistic regression models, and the emmeans package was used to estimate the marginal means from the fitted model. The present study is reported based on the guidelines provided by the Consensus‐based Checklist for Reporting of Survey Studies (CROSS) [32], and the checklist is summarised in Supplementary Table 1.

3Results

3.1Summary Statistics

A total of 176,555 adolescents aged 10–18 years were included to estimate the overall national prevalence of current smoking. Supplementary Table 3 summarises the sociodemographic characteristics of this full adolescent sample. From the full dataset, an analytic sub‐sample of 70,969 respondents aged 15–18 years who completed the common mental disorder questionnaires was selected as the analytic sample for further modelling; their baseline characteristics are detailed in Table 1. Within this subset, a proportion of adolescents reported symptoms indicative of common mental disorders.

3.2Prevalence of Adolescent Smoking

Table 2 summarises the prevalence of adolescent smokers in Indonesia. The estimated prevalence of current smoking among Indonesian adolescents aged 10–18 years was 9.10% (95% CI: 9.03–9.16). Smoking prevalence was markedly higher among males (17.48%, 95% CI: 17.09–17.87) than females (0.25%, 95% CI: 0.20–0.31). Prevalence increased with age, from 0.58% (95% CI: 0.51–0.68) at 10–12 years to 22.83% (95% CI: 22.16–23.52) at 17–18 years. Adolescents in rural areas reported a slightly higher prevalence (9.48%, 95% CI: 9.21–9.75) than those in urban areas (8.76%, 95% CI: 8.44–9.09). In the subset dataset (15–18 years), smoking prevalence was higher at 18.14% (95% CI: 18.00–18.28), consistent with age‐related increases observed in the full dataset.

3.3Confounder Identification

To identify potential confounders in the pathway between symptoms of common mental disorder and current smoking, we employed a directed acyclic graph (DAG). In DAG terminology, a confounder is defined as a common cause of both the exposure and the outcome [31]. Using DAGitty software two alternative minimal adjustment sets were identified to estimate the total effect of common mental disorder symptoms on current smoking. The first set included age, sex, current education, place of residence, working status, race/ethnicity, social influence, tobacco‐related policy, genetics, health behaviours, and attitudes/beliefs. The second, more parsimonious set included age, sex, current education, place of residence, working status, genetics, and health behaviours (included as an observed representation of health literacy, attitudes and beliefs) (Figure 1).

We selected the second parsimonious adjustment set because it involved fewer unobserved (latent) variables. To address the potential confounding effect of latent constructs such as health literacy and attitudes/beliefs, we used proxy variables (e.g., health behaviours) that are strongly correlated with these constructs. It is expected that a surrogate variable strongly associated with an unobserved confounding variable would remove a large part of the unobserved confounding [33]. These adjustment sets effectively blocks all identified backdoor biasing paths while preserving exposure‐outcome path [31]. Consequently, genetics remained the sole unmeasured (latent) structural confounder in the final theoretical model.

3.4Association Between Symptoms of Common Mental Disorder and Adolescent Current Smoking

Figure 2 and Supplementary Table 4 show the association between symptoms of common mental disorder and current smoking in the subset of adolescents aged 15–18 years. Adolescents with symptoms of common mental disorder had significantly higher odds of current smoking compared with those without symptoms (aOR: 1.59, 95% CI: 1.40–1.81). The model adjusted for several confounding variables like sex, age, place of residence, current education, working status, and health behaviours (alcohol and food consumption, fruit and vegetable intake, and physical activity). Some confounders also showed strong associations with current smoking. Male adolescents were far more likely to smoke than females (aOR: 108.16, 95% CI: 86.04–135.96), reflecting the very low prevalence among females. Current smoking likelihood increased with age and was higher in rural than in urban areas. Adolescents who were working were more likely to smoke, while those still studying were less likely to smoke compared with those not working or studying. Unhealthy behaviours, including frequent consumption of unhealthy foods and beverages, low fruit and vegetable intake, and alcohol use, were associated with greater odds of current smoking.

4Discussion

Using the largest nationally representative survey dataset on the adolescent population in Indonesia, we estimated the prevalence of current smoking among those aged 10–18 at 9.10%. Male, older age, and rural residence were associated with higher current smoking prevalence. Adolescents reporting symptoms of common mental disorder had significantly higher odds of current smoking (OR: 1.59; 95% CI: 1.40–1.81), even after adjusting for confounders identified through DAG analysis. Smokers were also more likely to engage in other unhealthy behaviours, including alcohol use, consumption of unhealthy foods and beverages, and low fruit and vegetable intake.

Our study demonstrated an increase in adolescent current smoking in Indonesia, from 7.2% in 2013 [4] to 9.1% in 2018 (this study). Consistent with prior reports, Indonesia ranks among the countries with the highest prevalence of adolescent smoking worldwide and is one of the few countries where prevalence has not declined since 1999, along with Botswana, Zimbabwe, Bosnia, and Herzegovina [3]. Within the Southeast Asia region, Indonesia has the second‐highest prevalence of adolescent smoking after Timor Leste [34].

Prior studies reported positive associations between mental health problems such as depression, stress, or anxiety and adolescent smoking, including earlier initiation and heavier consumption [16, 35, 36]. National surveys in other contexts have similarly shown that adolescents with symptoms of mental disorder are more likely to initiate smoking, progress to daily use, and smoke more heavily [15]. Mental distress in adolescence reduces the likelihood of quitting smoking and increases relapse risk in adulthood [17]. Adolescents with asthma and comorbid depression and anxiety, following adjustment of potential demographic and clinical covariates, were more likely to be smokers and had increased susceptibility to initiation of smoking [37]. A study of common mental disorder symptomatology before and during the COVID‐19 pandemic also supported the finding that after adjusting for sociodemographic factors, increased anxiety and depression symptoms increased the likelihood of cigarette and e‐cigarette consumption [38]. A longitudinal study using 16 years of nationally representative data applied robust econometric methods and concluded that improved mental health reduced the likelihood of smoking [39].

Our study confirms that common mental disorder symptoms are independently associated with adolescent current smoking. These findings strengthen the view that symptoms of mental disorder may contribute to smoking status, while acknowledging potential endogeneity from reverse causality, unmeasured confounders, and measurement error [39]. However, the possibility of a bidirectional relationship between smoking and mental health symptoms has been discussed in previous studies [40, 41]. Reviews suggesting bidirectionality were largely based on studies conducted in high‐income settings, predominantly the United States, with relatively small sample sizes and substantial heterogeneity in study design and outcome definitions. In addition, several studies did not adequately adjust for important confounders, such as alcohol consumption and other health‐related behaviours. Hence, evidence supporting a clear causal direction from smoking to common mental disorder remains limited.

Biological plausibility further supports this direction of association. Self‐medication theory proposes that individuals with psychological distress smoke to alleviate symptoms or to facilitate coping with their mental illness [42]. Nicotine stimulates nicotinic acetylcholine receptors, releasing dopamine and other neurotransmitters involved in reward and stress regulation [43, 44]. However, these effects contribute to short‐term coping behaviour, which may provide short‐term relief from stress and depression but reinforce dependence on smoking, tending to self‐medicate [45]. Tobacco comprises various compounds, notably nicotine, which could elicit the release of various substances such as norepinephrine, serotonin, dopamine, acetylcholine, γ‐aminobutyric acid, and glutamate in the brain that can relieve stress, anxiety, and depression, but prolonged use increases the risk of addiction and may worsen vulnerability to depression and anxiety [36].

Our study also confirmed sociodemographic patterns: current smoking was more common among males, older adolescents, and those in rural areas [22]. In Indonesia, smoking is strongly tied to cultural constructions of masculinity [23], while it is considered taboo or morally flawed in females [46]. Rural populations tend to have higher acceptance of smoking, which may explain urban–rural differences. Older adolescents may be more exposed to peers and social pressures, financially independent and have easier access to cigarettes [22, 24]. Employment also increased smoking risk, likely due to exposure to smoking peers and financial independence, while students were less likely to smoke due to restrictions in educational environments [25].

Our approach of a robust inferential framework, combined with a DAG to guide confounder selection, strengthens the validity of our findings. From a policy perspective, these results have practical relevance. Adolescent smoking is a complex, multifactorial behaviour that requires coordinated action at individual, family, school, and community levels. Governments could therefore develop comprehensive policies and programmes that enable parents, schools, and community services to work collaboratively, building integrated interventions that foster supportive environments and promote both mental health and healthy behaviours. Embedding such approaches within healthcare and allied settings may help prevent escalation of adverse smoking addiction and the exacerbation of health issues in adolescents. Such integrated models have been implemented in Australia and the United States and have been reported as successful [47, 48]. Collaboration and integration would help identify and resolve this issue at the earliest opportunity, reducing longer‐term impacts as adolescents transition into adulthood.

This study has established the association of common mental disorder symptoms and current smoking in Indonesian adolescents, harnessing nationally representative cross‐sectional data. Nevertheless, the study has some limitations. First, although the dataset is large and nationally representative, it is cross‐sectional; exposure and outcome were measured simultaneously, meaning temporal sequence, directionality, and potential reverse causality cannot be established. Second, current smoking status was self‐reported, raising the possibility of underreporting, particularly among female adolescents. Third, common mental disorder was assessed using screening instruments (MINI and SRQ‐20), which capture symptoms but do not establish clinical diagnoses. Fourth, some relevant confounders such as family history (e.g., parental smoking), genetic predisposition, social influence, individual temperament, and health literacy were unobserved in the dataset, while broader cultural nuances were only approximated using socioeconomic and demographic variables.

To summarise, the prevalence of adolescent current smoking in Indonesia is higher than that reported earlier. Common mental disorder symptoms were associated with a 59% increase in the odds of current smoking among adolescents (OR: 1.59, 95% CI: 1.40–1.81). By providing nationally representative empirical evidence from a high‐burden LMIC setting, this study extends existing literature beyond traditional environmental models. Future tobacco‐control policies should integrate mental disorder screening among the adolescent population and support into prevention strategies, alongside existing environmental measures, to more effectively reduce adolescent smoking.

Author Contributions

Bagas Abdillah: conceptualization, methodology, software, data curation, investigation, validation, formal analysis, visualization, writing – original draft, writing – review and editing. Mintu Nath: conceptualization, methodology, software, investigation, validation, formal analysis, supervision, visualization, writing – review and editing.

Ethics Statement

Informed consent was obtained during data collection, which considered aspects of the data collection procedure, the voluntary nature, and confidentiality. The anonymous data were shared with the authors by the Indonesian Government (permission number: IR.03.01/H.I/3663/3023).

Conflicts of Interest

The authors declare no conflicts of interest.

Transparency Statement

The lead authors Bagas Abdillah and Mintu Nath affirm that this manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained.

Supporting information

References

1D. K. Hatsukami , L. F. Stead , and P. C. Gupta , “Tobacco Addiction ,” Lancet
371 (2008 ): 2027 –2038 , 10.1016/S0140-6736(08)60871-5 .PMC4732578 18555914

2B. Thomson , N. A. Rojas , B. Lacey , et al., “Association of Childhood Smoking and Adult Mortality: Prospective Study of 120 000 Cuban Adults ,” Lancet Global Health
8 (2020 ): e850 –e857 , 10.1016/S2214-109X(20)30221-7 .PMC7248573 32446350

3C. Ma , B. Xi , Z. Li , et al., “Prevalence and Trends in Tobacco use among Adolescents Aged 13–15 Years in 143 Countries, 1999–2018: Findings From the Global Youth Tobacco Surveys ,” Lancet Child & Adolescent Health
5 (2021 ): 245 –255 , 10.1016/S2352-4642(20)30390-4 .33545071

4National Institute of Health Research and Development (NIHRD)
. Indonesian Basic Health Research 2013 (Riskesdas 2013) National Report. Jakarta: 2013.

5National Institute of Health Research and Development (NIHRD)
. Indonesia Basic Health Research 2018 (Riskesdas 2018) National Report. Jakarta: 2019.

6World Health Organization
., Ministry of Health of the Republic of Indonesia. GYTS│GLOBAL YOUTH TOBACCO SURVEY FACT SHEET INDONESIA 2019. 2020.

7Directorate General of Disease Prevention and Control TM of H of the R of I
. Action Plan Program 2020‐2024 Directorate General of Disease Prevention and Control. Jakarta: 2020.

8S. J. Blakemore and K. L. Mills , “Is Adolescence a Sensitive Period for Sociocultural Processing? ,” Annual Review of Psychology
65 (2014 ): 187 –207 , 10.1146/annurev-psych-010213-115202 .24016274

9M. Beusenberg and J. A. Orley , “User's.” Guide to The Self Reporting Questionnaire
SRQ: (1994 ).

10Z. Steel , C. Marnane , C. Iranpour , et al., “The Global Prevalence of Common Mental Disorders: A Systematic Review and Meta‐Analysis 1980–2013 ,” International Journal of Epidemiology
43 (2014 ): 476 –493 , 10.1093/ije/dyu038 .PMC3997379 24648481

11V. Patel , S. Saxena , C. Lund , et al., “The Lancet Commission on Global Mental Health and Sustainable Development ,” Lancet
392 (2018 ): 1553 –1598 , 10.1016/S0140-6736(18)31612-X .30314863

12X. Du , R. Wu , L. Kang , L. Zhao , and C. Li , “Tobacco Smoking and Depressive Symptoms in Chinese Middle‐Aged and Older Adults: Handling Missing Values in Panel Data with Multiple Imputation ,” Frontiers in Public Health
10 (2022 ): 1 –12 , 10.3389/fpubh.2022.913636 .PMC9458966 36091567

13A. Sumbe , A. V. Wilkinson , S. L. Clendennen , et al., “Association of Tobacco and Marijuana use with Symptoms of Depression and Anxiety among Adolescents and Young Adults in Texas ,” Tobacco Prevention & Cessation
8 (2022 ): 03 , 10.18332/TPC/144500 .PMC8792993 35128214

14M. A. You , J. Choi , and Y. J. Son , “Associations of Dual use of Tobacco Cigarettes and E‐Cigarettes, Sleep Duration, Physical Activity and Depressive Symptoms among Middle‐Aged and Older Korean Adults ,” Nursing Open
10 (2023 ): 4071 –4082 , 10.1002/nop2.1667 .PMC10170944 36929137

15D. Lawrence , S. R. Zubrick , F. Mitrou , and M. G. Sawyer , “Smoking Status, Mental Disorders and Emotional and Behavioural Problems in Young People: Child and Adolescent Component of the National Survey of Mental Health and Wellbeing ,” Australian and New Zealand Journal of Psychiatry
44 (2010 ): 805 –814 .10.3109/00048674.2010.482921 20815667

16S. S. Chung and K. H. Joung , “Risk Factors for Smoking Behaviors Among Adolescents ,” Journal of School Nursing
30 (2014 ): 262 –271 , 10.1177/1059840513505222 .24051583

17A. H. Weinberger , R. S. Kashan , D. M. Shpigel , et al., “Depression and Cigarette Smoking Behavior: A Critical Review of Population‐Based Studies ,” American Journal of Drug and Alcohol Abuse
43 (2017 ): 416 –431 , 10.3109/00952990.2016.1171327 .27286288

18R. K. Tiwari , V. Sharma , R. K. Pandey , and S. S. Shukla , “Nicotine Addiction: Neurobiology and Mechanism ,” Journal of Pharmacopuncture
23 (2020 ): 1 –7 , 10.3831/KPI.2020.23.001 .PMC7163392 32322429

19N. L. Benowitz , “Neurobiology of Nicotine Addiction: Implications for Smoking Cessation Treatment ,” American Journal of Medicine
121 (2008 ): S3 –S10 , 10.1016/j.amjmed.2008.01.015 .18342164

20D. Sheehan , J. Janavs , R. Baker , et al., “M.I.N.I DSM‐IV (Mini International Neuropsychiatric Interview English Version 5.0.0),” (2006 ).

21D. H. Tjandrarini , H. Ashar , J. P. Nari , et al., “Disabilities and Depression in Young Adolescents Living in Underdeveloped Areas of Indonesia: Results from the 2018 Indonesia Basic Health Survey ,” BMJ Open
15 (2025 ): e089737, 10.1136/bmjopen-2024-089737 .PMC12056641 40328647

22N. Kusumawardani , I. Tarigan , I. Suparmi , and A. Schlotheuber , “Socio‐Economic, Demographic and Geographic Correlates of Cigarette Smoking Among Indonesian Adolescents: Results from the 2013 Indonesian Basic Health Research (RISKESDAS) Survey ,” Global Health Action
11 (2018 ): 54 –62 , 10.1080/16549716.2018.1467605 .PMC5990951 29855228

23N. Ng , L. Weinehall , and A. Ohman , “‘If I Don't Smoke, I'm Not a Real Man’—Indonesian Teenage Boys' Views about Smoking ,” Health Education Research
22 (2006 ): 794 –804 , 10.1093/her/cyl104 .16987943

24F. Fithria , M. Adlim , S. R. Jannah , and T. Tahlil , “Indonesian Adolescents' Perspectives on Smoking Habits: A Qualitative Study ,” BMC Public Health
21 (2021 ): 82, 10.1186/s12889-020-10090-z .PMC7791848 33413232

25E. L. Amelia and N. Adnan , “Implementation of Smoke‐Free Zones in Schools in Indonesia ,” Journal of Social Research
4 (2025 ): 662 –670 .

26World Health Organization
., Global Physical Activity Questionnaire (GPAQ) Analysis Guide (Geneva: World Health Organization, 2021 ).

27C. Hartmann , M. Siegrist , and K. Van Der Horst , “Snack Frequency: Associations With Healthy and Unhealthy Food Choices ,” Public Health Nutrition
16 (2013 ): 1487 –1496 , 10.1017/S1368980012003771 .PMC10271874 22894987

28World Health Organization
., “WHO Steps Surveillance Manual.” WHO Stepwise Approach to Chronic Disease Risk Factor Surveillance (WHO, 2005 ).

29N. Meader , K. King , T. Moe‐Byrne , et al., “A Systematic Review on the Clustering and Co‐Occurrence of Multiple Risk Behaviours ,” BMC Public Health
16 (2016 ): 657, 10.1186/s12889-016-3373-6 .PMC4966774 27473458

30V. Busch , A. Loyen , M. Lodder , A. J. P. Schrijvers , T. A. van Yperen , and J. R. J. de Leeuw , “The Effects of Adolescent Health‐Related Behavior on Academic Performance ,” Review of Educational Research
84 (2014 ): 245 –274 , 10.3102/0034654313518441 .

31J. Textor , “Drawing and Analyzing Causal DAGs with DAGitty.” 2015.

32A. Sharma , N. T. Minh Duc , T. Luu Lam Thang , et al., “A Consensus‐Based Checklist for Reporting of Survey Studies (CROSS) ,” Journal of General Internal Medicine
36 (2021 ): 3179 –3187 , 10.1007/s11606-021-06737-1 .PMC8481359 33886027

33M. A. Hernan , “Causal Knowledge as a Prerequisite for Confounding Evaluation: An Application to Birth Defects Epidemiology ,” American Journal of Epidemiology
155 (2002 ): 176 –184 .10.1093/aje/155.2.176 11790682

34J. Kishun , A. Kumar , U. Singh , and S. Kumar , “Comparison of Common Predictors of Current Cigarette Smoking among Adolescents: Across South East Asian Countries ,” Clin Epidemiol Glob Health
17 (2022 ): 101112, 10.1016/j.cegh.2022.101112 .

35T. Jahnel , S. G. Ferguson , S. Shiffman , and B. Schüz , “Daily Stress as Link between Disadvantage and Smoking: An Ecological Momentary Assessment Study ,” BMC Public Health
19 (2019 ): 1284, 10.1186/s12889-019-7631-2 .PMC6790015 31606051

36Z. Wu , Q. Yue , Z. Zhao , et al. “A Cross‐Sectional Study of Smoking and Depression Among US Adults: NHANES (2005–2018),” (2023).10.3389/fpubh.2023.1081706 PMC9922891 36794066

37T. Bush , L. Richardson , W. Katon , et al., “Anxiety and Depressive Disorders Are Associated With Smoking in Adolescents With Asthma ,” Journal of Adolescent Health
40 (2007 ): 425 –432 , 10.1016/j.jadohealth.2006.11.145 .17448400

38S. L. Clendennen , B. Chen , A. Sumbe , and M. B. Harrell , “Patterns in Mental Health Symptomatology and Cigarette, E‐Cigarette, and Marijuana Use Among Texas Youth and Young Adults Amid the Coronavirus Disease 2019 Pandemic ,” Nicotine & Tobacco Research
25 (2023 ): 266 –273 , 10.1093/ntr/ntac205 .PMC9825349 36018816

39L. Yang and V. Zikos , “Mental Health and Smoking Behavior ,” Economic Modelling
126 (2023 ): 106407, 10.1016/j.econmod.2023.106407 .

40M. Fluharty , A. E. Taylor , M. Grabski , and M. R. Munafò , “The Association of Cigarette Smoking with Depression and Anxiety: A Systematic Review ,” Nicotine & Tobacco Research
19 (2017 ): 3 –13 , 10.1093/ntr/ntw140 .PMC5157710 27199385

41M. O. Chaiton , J. E. Cohen , J. O'Loughlin , and J. Rehm , “A Systematic Review of Longitudinal Studies on the Association between Depression and Smoking in Adolescents ,” BMC Public Health
9 (2009 ): 356, 10.1186/1471-2458-9-356 .PMC2758872 19772635

42F. Manzella , “Smoking in Schizophrenic Patients: A Critique of the Self‐Medication Hypothesis ,” World Journal of Psychiatry
5 (2015 ): 35, 10.5498/wjp.v5.i1.35 .PMC4369547 25815253

43A. L. Brody , A. G. Mukhin , M. S. Mamoun , et al., “Brain Nicotinic Acetylcholine Receptor Availability and Response to Smoking Cessation Treatment a Randomized Trial ,” JAMA Psychiatry
71 (2014 ): 797 –805 , 10.1001/jamapsychiatry.2014.138 .PMC4634637 24850280

44M. Besson and B. Forget , “Cognitive Dysfunction, Affective States, and Vulnerability to Nicotine Addiction: A Multifactorial Perspective ,” Frontiers in Psychiatry
7 (2016 ): 160, 10.3389/fpsyt.2016.00160 .PMC5030478 27708591

45G. Chang , L. Sherritt , and J. R. Knight , “Adolescent Cigarette Smoking and Mental Health Symptoms ,” Journal of Adolescent Health
36 (2005 ): 517 –522 , 10.1016/j.jadohealth.2004.05.008 .15901517

46S. Barraclough , “Women and Tobacco in Indonesia ,” Tobacco Control
8 (1999 ): 327 –332 , 10.1136/tc.8.3.327 .PMC1763945 10599579

47B. Vanfrank , E. Pasalic , B. Oliver , et al., “Tobacco‐Related Clinical Services and Tobacco‐Free Policies in Behavioral Health Treatment Facilities‐United States, 2023,” (2025).10.15585/mmwr.mm7414a3 PMC12021074 40273088

48Government of Western Australia North Metropolitan Health Service Mental Health PH and DS
. Smoke Free Mental Health Project Summary Report October 2020.