Prevalence of adolescent dual tobacco use and its associated factors: findings from a cross-sectional study

Abstract

Objective

To determine the prevalence and its associated factors of dual tobacco use among Malaysian adolescents in developing effective public health strategies.

Design

Cross-sectional study.

Setting

This study analysed data from the National Health and Morbidity Survey (NHMS) 2022: Adolescent Health Survey (AHS), a nationwide cross-sectional survey conducted among secondary school students across Malaysia.

Participants

This cross-sectional study used data from the NHMS 2022: AHS, which included a representative national sample of Malaysian adolescents between the ages of 13 and 17. Using a two-stage stratified sampling approach, this study included participation from 33 523 adolescents from 240 schools nationwide.

Primary and secondary outcome measures

The primary outcome was the prevalence of dual tobacco use among Malaysian school-going adolescents aged 13–17, defined as self-reported concurrent use of conventional cigarettes and e-cigarettes in the past 30 days. Secondary outcomes included the factors associated with dual tobacco use, such as sociodemographic characteristics, dual tobacco use behaviour, marital status of parents, tobacco use of parent or guardian, supervision, connectedness, bonding and respect for privacy of parent or guardian, along with peer support and truancy. A complex sample analysis was performed using SPSS V.27.0, maintaining a 95% CI and multiple logistic regression was applied.

Results

Out of 33 380 Malaysian adolescents surveyed, 1728 (5.5%; 95% CI: 4.88% to 6.28%) reported dual tobacco use. The prevalence was significantly higher among males (9.7%; 95% CI: 8.54% to 10.98%) compared with females (1.4%; 95% CI: 1.15% to 1.70%) and highest among 17 years (8.6%; 95% CI: 7.05% to 10.41%), with a decreasing trend in younger age groups. Predominant vapers constituted the largest proportion (9.9%, 95% CI: 8.66% to 10.13%) among adolescents who reported dual tobacco use, followed by predominant smokers (6.2%, 95% CI: 5.53% to 7.06%), dual daily users (4.5%, 95% CI: 3.18% to 5.64%) and non-daily dual users (3.9%, 95% CI: 2.79% to 4.39%). Dual tobacco use was more prevalent among adolescents whose parents or guardians used tobacco (8.1%; 95% CI: 7.10% to 9.20%). Multiple logistic regression analysis suggested that significant predictors of dual tobacco use included older age (adjusted OR (AOR) for 17 years: 2.92; 95% CI: 2.19 to 3.89), male sex (AOR: 8.53; 95% CI: 6.78 to 10.74), ‘others’ ethnicity (AOR: 1.82; 95% CI: 1.40 to 2.35), predominant vapers (AOR 2.65, 95% CI: 1.96 to 2.72), separated or widowed parental status (AOR: 1.69; 95% CI: 1.42 to 2.02), parental or guardian tobacco use (AOR: 2.47; 95% CI: 2.08 to 2.94), lack of bonding (AOR: 1.79; 95% CI: 1.49 to 2.14), lack of privacy respect (AOR: 1.53; 95% CI: 1.29 to 1.81), lack of peer support (AOR: 1.65; 95% CI: 1.39 to 1.96) and truancy (AOR: 1.81; 95% CI: 1.55 to 2.10).

Conclusions

This study highlights a concerning prevalence of dual tobacco use (5.5%) among Malaysian adolescents, which is notably higher than the current prevalence of exclusive cigarette smoking (6.2%) and closely approaches that of e-cigarette use (14.9%). Compared with previous national surveys and international benchmarks, the prevalence of dual use among Malaysian youth appears to be rising and represents a significant public health concern. These findings underscore the urgent need for targeted prevention and control strategies addressing both conventional and emerging tobacco products in this population.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• This study used a large-scale, nationally representative survey of school-going adolescents in Malaysia, allowing the findings to be generalised to most Malaysian adolescents.

• This study followed a robust methodology aligned with national survey standards, and no existing literature was found to indicate significant methodological bias affecting the associations observed.

• The analysis relied on self-reported data, which may be subject to recall or response bias.

• As a cross-sectional study, it does not allow for the establishment of causal relationships.

• Tobacco use was not biochemically verified, which may have led to under-reporting of dual tobacco use.

Introduction

Dual tobacco use, which refers to the simultaneous use of two tobacco products, exposes a variety of hazards that affect health and well-being. Current definitions of dual tobacco use often treat the behaviour as homogeneous, but emerging literature emphasises significant heterogeneity in usage patterns. For example, Borland et al¹ identified four distinct dual tobacco user profiles: predominant smokers, dual daily users, predominant vapers and concurrent non-daily users. Incorporating these distinctions into adolescent tobacco surveillance is crucial, as different combinations of use may imply different intervention needs. Baig and Giovenco² highlighted that dual tobacco use behaviour is heterogeneous, shaped by variations in both the frequency of smoking and vaping, as well as by sociodemographic factors. Similarly, Coleman et al,³ in their comprehensive review of existing literature, emphasised the critical need to examine specific patterns of dual tobacco use in order to better understand exposure to toxicants and the associated long-term health outcomes.

These patterns are associated with differing health risks, dependence trajectories and cessation outcomes. Adolescents who are dual tobacco users are exposed to more harmful chemicals than those who use a single product.⁴ The current use of both conventional cigarette and e-cigarette products in the past 30 days intensifies exposure to nicotine, carcinogens and other toxic substances, increasing the risk of developing respiratory issues, cardiovascular diseases and cancer.⁵ Importantly, nicotine exposure during adolescence, a critical period for brain development, can lead to addiction and long-term changes in brain function and potentially affect attention, learning and susceptibility to addiction in adulthood.⁶ Dual tobacco use can also worsen the risks of mental health issues, including depression and anxiety, potentially due to the combined effects of nicotine and other harmful substances on the developing brain.⁷

The increasing trend of dual tobacco use has been reported in many countries by recent Global Youth Tobacco Surveys (GYTS) (2014–2019), which showed the prevalence of dual tobacco use with significant differences across different geographic locations, from as low as 0.1% in Cambodia to 13.6% in Poland. In 15 out of 67 countries or territories, the prevalence of dual tobacco use exceeded 5%.⁸ In Malaysia, according to the National Health and Morbidity Survey (NHMS) 2022: Adolescent Health Survey (AHS), the prevalence of current e-cigarette use among school-going adolescents aged 13–17 years was 14.9%, while 6.2% reported current use of conventional cigarettes only. These figures suggest that e-cigarette use is now more common than conventional cigarette use among Malaysian adolescents, a shift that mirrors trends observed in several other countries.⁹ However, a significant subgroup engages in concurrent use of both products, warranting closer examination due to potential compounded health risks and higher levels of nicotine dependence associated with dual tobacco use.

It shows the importance of conducting specific research to explore regional differences and trends in dual tobacco use. The observable differences in dual tobacco use among adolescents across various geographic locations highlight distinct public health challenges faced by regions. The rise towards dual tobacco use among adolescents poses critical public health implications as it represents a potential gateway to increased exposure to harmful substances, thereby increasing health risks beyond single tobacco product use.¹⁰ In addition, dual use of tobacco products among adolescents is also associated with heightened nicotine dependence. Evidence suggests that individuals who use both cigarettes and e-cigarettes may have higher total nicotine exposure compared with single-product users, increasing the risk of addiction and making cessation more difficult.¹¹ In a longitudinal study, dual tobacco users exhibited higher levels of nicotine dependence symptoms and were less likely to quit smoking or vaping over time compared with exclusive users.¹² This is particularly concerning during adolescence, a critical developmental stage, as nicotine exposure can result in long-term alterations in neural circuits related to attention, learning and susceptibility to other substance use.^{6 13} Therefore, dual tobacco use not only sustains nicotine addiction but may also entrench long-term tobacco use trajectories and dependency into adulthood.

Internationally, several studies have identified a consistent set of factors associated with dual tobacco use among adolescents. Common predictors include older age, male gender, peer influence, low parental monitoring, truancy and exposure to tobacco use within the household.14,16 In the USA, Owusu et al¹² found that dual tobacco users were more likely to report higher levels of nicotine dependence, lower perceived harm of e-cigarettes and greater susceptibility to continued smoking. In Jordan and other parts of the Middle East, waterpipe smoking combined with cigarette use has been reported among male adolescents, often associated with low parental involvement and high peer pressure.¹⁷ These findings underscore the importance of examining not only individual behaviours but also familial and social contexts, which may vary regionally. Given the rising popularity of e-cigarettes among Malaysian youth, examining these predictors in a local context is vital for developing culturally relevant public health interventions.

Previous studies emphasise the need for comprehensive research into the factors influencing adolescent dual tobacco use to shape effective interventions and policies. Understanding the factors, this research provides critical insights into the current state of dual tobacco use among adolescents in Malaysia, a country combating the global challenge of tobacco use among adolescents. This knowledge enables the tailoring of interventions to the needs of these vulnerable groups, enhancing the potential effectiveness of prevention and cessation programmes by pinpointing the significant factors of adolescent tobacco use. Gaining insight into these factors enables the development of more focused educational content aimed at adolescents. Therefore, this research aims to determine the prevalence and its associated factors of dual tobacco use within a nationally representative sample of Malaysian adolescents. This research finding offers up-to-date statistics on dual tobacco use among adolescents for policy-makers, healthcare professionals and educators with the actual evidence needed to assess this issue.

Methods

Data sources

This study used data from the NHMS 2022: AHS, a cross-sectional data collected between June and July 2022, focusing on national representation through a two-stage stratified sampling design. The survey included both public and private secondary schools from 13 states and three federal territories in Malaysia that were registered with the Ministry of Education (MOE) and the Ministry of Rural and Regional Development in 2021. This study employs a two-stage stratified cluster sampling method to ensure that the resulting sample is representative of the entire country. Data collection took place during school hours in classrooms under the supervision of trained research assistants and teachers. The survey was conducted anonymously to ensure confidentiality and to encourage honest responses. Prior to participation, both parental or guardian consent and student assent were obtained. Information sheets and consent forms were distributed to parents or guardians in advance, and only students who returned signed parental consent forms and provided verbal assent on the day of the survey were included. Further details of the NHMS 2022: AHS study have been published.^{9 18}

A total of 240 schools were selected, and 36 000 students were selected from the selected schools. 33 523 students participated in this survey, with an overall response rate of 89.0%. This response rate for the survey is considered high for school-based cross-sectional studies. Although a formal non-response bias analysis was not performed, the NHMS 2022: AHS employed a rigorous poststratification weighting procedure. This adjustment was based on known population parameters (eg, age, sex and school location) to minimise potential bias due to differential response rates.^{9 18} This approach is consistent with established best practices in large-scale AHS, where high response rates combined with statistical weighting are generally effective in reducing the impact of non-response bias on prevalence estimates.

The questionnaires were designed to capture sociodemographic information, parents’ marital status, parents’ current tobacco use and protective factors. The questionnaires comprised multiple-choice questions and participants answered them using an optical mark recognition (OMR) answer sheet. These sheets were subsequently scanned and processed using OMR software, allowing for efficient and accurate data capture. For the present study, 143 respondents who answered ‘I don’t know’ for the parental marital status variable were excluded from the analysis, resulting in a final analytic sample of 33 380 adolescents. This decision was made to ensure consistency in sociodemographic categorisation, as the ‘I don’t know’ responses could not be reliably classified within the marital status subgroups. While the number of exclusions was relatively small (0.4% of the total sample), we acknowledge the potential for bias if the excluded respondents systematically differed from those included. However, due to the low proportion of missing data and the use of weighted analyses, we believe the overall impact on the study findings is minimal.19,21

Patient and public involvement

Patients or the public were not involved in the design, or conduct, or reporting, or dissemination plans of our research.

Measure

Dependent variables

This study uses current dual tobacco use as a dependent variable. Current dual tobacco use was defined as the concurrent use of any conventional cigarettes (including traditional hand-rolled cigarettes, roll-your-own cigarettes with papers, manufactured cigarettes, cigars or cigarillos) and e-cigarettes (or vapes) in the past 30 days. Two questions were used to assess dual tobacco use: (1) ‘During the past 30 days, how many days did you smoke cigarettes?’ and (2) ‘During the past 30 days, how many days did you use electronic cigarettes or vape?’. Current dual tobacco use was defined as the use of both conventional cigarettes and e-cigarettes on at least 1 day in the past 30 days.

Independent variables

The sociodemographic variables were used as categorical variables in analyses. Age was captured in five categories (13 years old, 14 years old, 15 years old, 16 years old and 17 years old); sex in two categories (male and female) and ethnicity in four categories (Malay, Chinese, Indian and others). There are four types of concurrent use: predominant smokers (daily smokers with non-daily vaping); dual users (daily use of both products); predominant vapers (daily vapers and non-daily smokers) and non-daily concurrent users (non-daily use of both). We further categorised dual tobacco users based on their frequency of use of both conventional cigarettes and e-cigarettes in the past 30 days to provide a more nuanced understanding of dual tobacco use behaviour. These four subgroups were defined: predominant smokers as respondents who reported frequent use of cigarettes (≥20 days) and infrequent e-cigarette use (1–9 days), dual daily users as respondents who reported frequent use of both cigarettes and e-cigarettes (≥20 days each), predominant vapers as respondents who reported frequent e-cigarette use (≥20 days) and infrequent cigarette use (1–9 days) and concurrent non-daily users as respondents who reported infrequent use of both products (1–9 days each). These subgroups were derived using self-reported responses to items assessing the number of days each product was used in the past 30 days. Marital status of parents, parent or guardian tobacco use, parent or guardian supervision, connectedness, bonding and respect for privacy were determined along with peer support and truancy. The marital status of parents was categorised into two categories as married and divorced or widowed. Parent or guardian supervision is defined as parents or guardians always or most of the time, checking to see if their homework was done in the past 30 days and parent or guardian connectedness is defined as parents or guardians always or most of the time, understanding their problems and worries in the past 30 days. Meanwhile, parental or guardian bonding is defined as parents or guardians always or most of the time really knowing what they were doing with their free time in the past 30 days. Parental or guardian respect for privacy is defined as parents or guardians having never or rarely gone through their things without their approval in the past 30 days. Peer support is defined as adolescents in their school being kind and helpful most of the time or always during the past 30 days. The responses were combined and categorised as: yes (includes always and most of the time) and no (includes sometimes, rarely and never). Truancy is defined as missed class or school without permission for at least 1 day in the past 30 days. The responses were combined and categorised as: yes (includes 1 or 2 days, 3–5 days, 6–9 days and 10 or more days) and no (includes 0 days/never).

Data analysis

This study’s statistical analysis employed univariate and bivariate methods, using SPSS V.27.0 software for data analysis. The univariate analysis presented the frequency and percentage of all variables (both dependent and independent). The bivariate analysis, executed through multiple logistic regression, investigated the relationships between independent and dependent variables. To illustrate respondent characteristics, including age, sex, ethnicity, parents’ marital status, parent or guardian tobacco use, parental or guardian supervision, connectedness, bonding and respect for privacy, complex sample descriptive statistics were applied. This approach considered sample weights and adhered to a complex sampling framework, with weighting for estimation derived from the inverse probability of sampling, a non-response adjustment factor and poststratification adjustments. The factor associated with dual tobacco use was identified through multiple logistic regression analysis, adjusted for sampling design. The study specifically sought to pinpoint factors associated with dual tobacco use among adolescents, reporting findings as ORs and adjusted ORs (AOR) with 95% CIs, identifying statistically significant characteristics at a p<0.05.

Results

Out of the 33 523 students surveyed, 33 380 were included, with 1728 identified as dual tobacco users, accounting for a prevalence rate of 5% (95% CI: 4.88% to 6.28%). The analysis highlights a descending trend in dual tobacco use for age groups; it was highest among 17- years at 8.6% (95% CI: 7.05% to 10.41%) and gradually reduced. In terms of sex differences, male adolescents showed a significantly higher tendency towards dual tobacco use (9.7%, 95% CI: 8.54% to 10.98%). Among different ethnic groups, the prevalence of dual tobacco use was highest among those classified as ‘others’ (9.7%, 95% CI: 7.84% to 11.86%), followed by Malays (5.9%, 95% CI: 5.12% to 6.83%), Indians (3.5%, 95% CI: 2.32% to 5.13%) and Chinese (2.0%, 95% CI: 1.36% to 2.83%). Among dual tobacco users, the prevalence of predominant vapers was highest (9.9%, 95% CI: 8.66% to 10.13%), followed by predominant smokers (6.2%, 95% CI: 5.53% to 7.06%), dual daily users (4.5%, 95% CI: 3.18% to 5.64%) and non-daily dual users (3.9%, 95% CI: 2.79% to 4.39%). The prevalence is also higher among adolescents with separated or widowed parents (8.1%, 95% CI: 6.80% to 9.58%). Furthermore, adolescents who have a parent or guardian who uses tobacco showed a higher prevalence of dual tobacco use (8.1%, 95% CI: 7.10% to 9.20%). The lack of parental or guardian supervision showed a higher dual tobacco use rate (5.5%, 95% CI: 4.84% to 6.28%). Similarly, the lack of connectedness with parents or guardians showed a higher prevalence (5.7%, 95% CI: 5.05% to 6.54%) and the lack of bonding showed higher rates of dual tobacco use (6.3%, 95% CI: 5.58% to 7.16%). Moreover, adolescents experiencing the lack of privacy respect from parents or guardians had higher rates for dual tobacco use (7.1%, 95% CI: 6.05% to 8.42%). The lack of peer support resulted in higher prevalence (7.0%, 95% CI: 6.23% to 7.97%) and truancy shows higher rates of dual tobacco use (9.4%, 95% CI: 8.31% to 10.52%) (table 1).

Table 1. Current dual tobacco users among Malaysian adolescents aged 13–17 years old by characteristics (N=33 380).

Characteristics	n	N	% (95% CI)
Overall	1728	114 562	5.5 (4.88 to 6.28)
Age
13 years old	209	15 044	3.4 (2.76 to 4.23)
14 years old	265	18 334	4.3 (3.51 to 5.16)
15 years old	318	23 395	5.6 (4.64 to 6.63)
16 years old	450	24 697	6.3 (5.13 to 7.77)
17 years old	486	33 093	8.6 (7.05 to 10.41)
Sex
Male	1506	100 080	9.7 (8.54 to 10.98)
Female	222	14 483	1.4 (1.15 to 1.70)
Ethnicity
Malay	1234	77 071	5.9 (5.12 to 6.83
Chinese	86	7386	2.0 (1.36 to 2.83)
Indian	55	4257	3.5 (2.32 to 5.13)
Others	353	25 849	9.7 (7.84 to 11.86)
Dual tobacco use behaviour
Predominant smoker	486	39 102	6.2 (5.53 to 7.06)
Dual daily user	268	14 562	4.5 (3.18 to 5.64)
Predominant vaper	797	57 109	9.9 (8.66 to 10.13)
Dual non-daily use	177	28 312	3.9 (2.79 to 4.39)
Marital status of parents
Married	1303	86 727	5.0 (4.40 to 5.72)
Separated or widowed	371	24 368	8.1 (6.80 to 9.58)
Parent or guardians are tobacco users
Yes	1097	74 357	8.1 (7.10 to 9.20)
No	552	35 622	3.2 (2.76 to 3.82)
Lack of parental or guardian supervision
Yes	1535	101 586	5.5 (4.84 to 6.28)
No	163	10 421	5.1 (3.99 to 6.57)
Lack of parental or guardian connectedness
Yes	1348	89 168	5.7 (5.05 to 6.54)
No	352	23 076	4.7 (3.87 to 5.61)
Lack of parental or guardian bonding
Yes	1309	86 119	6.3 (5.58 to 7.16)
No	391	26 036	3.8 (3.17 to 4.58)
Lack of parental or guardian respect for privacy
Yes	385	26 660	7.1 (6.05 to 8.42)
No	1313	85 370	5.1 (4.49 to 5.81)
Lack of peer support
Yes	1168	77 872	7.0 (6.23 to 7.97)
No	531	34 338	3.6 (3.07 to 4.34)
Truancy
Yes	725	49 033	9.4 (8.31 to 10.52)
No	976	63 297	4.2 (3.57 to 4.84)

n, count (unweighted); N, estimated population (weighted).

The likelihood of being a dual tobacco user was highest among male adolescents (AOR 8.53, 95% CI: 6.78 to 10.74) indicating a strong relationship between sex and dual tobacco use among Malaysian adolescents. The results obtained show that older adolescents also have an association with dual tobacco use. The ‘others’ ethnicity and Chinese groups were more likely to become dual tobacco users compared with other ethnic groups. Among dual tobacco users, the largest subgroup was predominant vapers, underscoring the heterogeneity in dual tobacco use behaviour and suggesting that most adolescents are 2.65 times more likely to favour e-cigarettes over cigarettes. Among parental characteristics, adolescents with parents or guardians who use tobacco are 2.47 times more likely to become dual tobacco users. Adolescents with separated or widowed parents or guardians are 1.69 times more likely to engage in dual tobacco. The findings indicate that the lack of bonding with parents or guardians, the lack of respect for privacy by parents or guardians, the lack of peer support and truancy are also associated with dual tobacco use among Malaysian adolescents. However, the lack of parental or guardian supervision and the lack of parental or guardian connectedness were not likely associated with the dual tobacco use among adolescents (table 2).

Table 2. Association factors of dual tobacco use among Malaysian adolescents aged 13–17 years old (N=33 380).

Characteristics	OR (95% CI)	P value	OR (95% CI)	P value
Age
13 years old	1		1
14 years old	1.26 (0.97 to 1.64)	0.085	1.22 (0.92 to 1.61)	0.177
15 years old	1.66 (1.28 to 2.16)	<0.001*	1.67 (1.26 to 2.21)	<0.001*
16 years old	1.91 (1.40 to 2.59)	<0.001*	2.12 (1.57 to 2.87)	<0.001*
17 years old	2.65 (1.98 to 3.56)	<0.001*	2.92 (2.19 to 3.89)	<0.001*
Sex
Male	7.56 (6.08 to 9.41)	<0.001*	8.53 (6.78 to 10.74)	<0.001*
Female	1		1
Ethnicity
Malay	1		1
Chinese	0.32 (0.22 to 0.47)	<0.001*	0.38 (0.25 to 0.58)	<0.001*
Indian	0.57 (0.37 to 0.87)	0.009	0.83 (0.54 to 1.25)	0.365
Others	1.70 (1.31 to 2.21)	<0.001*	1.82 (1.40 to 2.35)	<0.001*
Dual tobacco use behaviour
Predominant smoker	1.59 (1.27 to 2.10)	<0.001*	1.56 (1.26 to 2.07)	<0.001*
Dual daily user	1.15 (0.97 to 1.67)	<0.001*	1.21 (0.99 to 1.83)	<0.001*
Predominant vaper	2.54 (1.78 to 2.84)	<0.001*	2.65 (1.96 to 2.72)	<0.001*
Dual non-daily use	1		1
Marital status of parents
Married	1		1
Separated or widowed	1.66 (1.41 to 196)	<0.001*	1.69 (1.42 to 2.02)	<0.001*
Parent or guardians are tobacco user
Yes	2.62 (2.24 to 3.08)	<0.001*	2.47 (2.08 to 2.94)	<0.001*
No	1		1
Lack of parental or guardian supervision
Yes	1.08 (0.84 to 1.39)	0.557	0.89 (0.67 to 1.17)	0.391
No	1		1
Lack of parental or guardian connectedness
Yes	1.25 (1.06 to 1.47)	0.009	1.02 (0.83 to 1.24)	0.876
No	1		1.00
Lack of parental or guardian bonding
Yes	1.70 (1.45 to 2.00)	<0.001*	1.79 (1.49 to 2.14)	<0.001*
No	1		1
Lack of parental or guardian respect for privacy
Yes	1.43 (1.23 to 1.66)	<0.001*	1.53 (1.29 to 1.81)	<0.001*
No	1		1
Lack of peer support
Yes	2.00 (1.71 to 2.34)	<0.001*	1.65 (1.39 to 1.96)	<0.001*
No	1		1
Truancy
Yes	2.38 (2.07 to 2.74)	<0.001*	1.81 (1.55 to 2.10)	<0.001*
No	1		1

^*Statistically significant different at α=0.05.

Discussion

In this study, the prevalence rate of dual tobacco use was 5.5%. The results are in line with global trends, which report a prevalence of 4.9%. Among the upper-middle-income countries, Malaysia reported a lower prevalence of dual tobacco use compared with Indonesia (8.5%), while Thailand recorded a much lower prevalence (1.8%), followed by other Southeast Asia countries such as Cambodia (0.1%) and the Philippines (0.0%).⁸ The findings of this study highlight that, despite some exceptions, dual tobacco use among Malaysian adolescents remains a significant public health concern in Southeast Asia. These results emphasise the need for targeted strategies to address this ongoing public health challenge.

We found that males were eight times more likely than females to become dual tobacco users. This strong association is similarly aligned with previous studies, which showed high dual tobacco use rates in male adolescents.^{16 17 22} Furthermore, this study identifies the sex differences in dual tobacco use, with male adolescents showing a greater rise towards such behaviours than female adolescents, a trend significantly driven by sex roles and the societal perception of conventional cigarettes or e-cigarettes.^{23 24} This tendency makes male adolescents at higher risk of nicotine addiction and tobacco-related health issues as they age.^{15 25}

This study found an association between older adolescents and the increasing prevalence of dual tobacco use as they age. Previous findings have shown that young adolescents’ early initiation significantly increases the risk of regular smoking later in life.²⁶ Dual tobacco users may begin smoking before age 13, and nicotine dependence tends to increase with cigarette smoking.²⁷ Delaying the initiation of smoking can reduce nicotine dependency and boost the chances of cessation.²⁸ We suggest that dual tobacco use is affected by both sex and age, with smoking rates among male adolescents increasing as they get older. It is important to note that by delaying the onset of smoking in younger male adolescents, the intensity of nicotine dependence can be reduced, thereby improving the likelihood of successfully quitting smoking.

Additionally, we found the differences in the rates and likelihood of dual tobacco use among ‘others’ ethnic groups. Previous study showed that ‘others’ ethnic groups of adolescents were particularly likely to be dual tobacco users in 2017. It was discussed that cultural attitudes towards e-cigarettes, which are newer products compared with conventional cigarettes, and thus are less widely used among other ethnic groups.²⁹ Further study should be conducted to investigate factors related to ethnic variation on a combination of substance use among adolescents.

This study revealed heterogeneity in dual tobacco use patterns among Malaysian adolescents, with predominant vapers representing the largest subgroup. The predominance of e-cigarette use over cigarette smoking among dual tobacco users is consistent with emerging international trends, where vaping has become more socially acceptable and perceived as less harmful among youth.^{30 31} The relatively high proportion of predominant vapers may reflect the widespread appeal of flavoured vape products, aggressive marketing via social media and perceived lower risk associated with e-cigarettes.³² These patterns raise public health concerns, as research indicates that adolescents who begin with or heavily rely on e-cigarettes are at increased risk of progressing to regular cigarette smoking, developing nicotine addiction and sustaining long-term dual tobacco use.^{24 33} The distribution of use patterns observed in this study reinforces the importance of differentiated intervention strategies that target not just dual tobacco use as a whole, but also address dominant behaviours, such as vaping, that may serve as a gateway to sustained nicotine dependence.

Previous studies showed that parental behaviours significantly influenced adolescents’ dual tobacco use, especially among those who live with smoking parents.^{14 34} Adolescents with smoking parents are more likely to adopt similar behaviours, as tobacco use becomes normalised within the household.³⁵ The normalisation of tobacco in the household context fosters an environment where adolescents tend to adopt similar behaviours.¹⁴ This phenomenon is attributed to modelling behaviours, where parental behaviours set precedents for children’s actions. A study highlights the influential role of parents in shaping early children’s behaviours, including ‘pretend smoking’ and attitudes towards smoking.³⁶ Undoubtedly, smoking behaviours, practices and parenting styles profoundly influence adolescent dual tobacco use advancing to regular smoking in adulthood. Hence, implementing smoke-free policies in homes is necessary to minimise adolescent exposure and counteract the social normalisation of tobacco use.

This study also found that adolescents from separated or widowed family backgrounds had a higher prevalence of being dual tobacco users. This finding is probably due to factors such as family structures significantly influencing adolescent behaviour patterns, including tobacco use.³⁷ Parent marital status appears to indirectly impact adolescent tobacco use. However, direct evidence associating marital status with dual tobacco use specifically remains limited.¹⁴ The lack of bonding was also found to be associated with dual tobacco use in this study. Strong parent–child relationships generally reduce risky behaviours like tobacco use.³⁸ Nevertheless, factors like divorce can weaken parental bonds, increasing tobacco use risk.³⁹ Then, the lack of parental respect for privacy is also associated with dual tobacco use. This finding might be due to the increased openness towards smoking behaviours that older adolescents experience, which could elevate their risk of engaging in behaviours like smoking. In Asian countries like Malaysia, parents or guardians may offer more protection to younger adolescents compared with older ones.²⁹ Accordingly, future research is needed to fully understand the impact of why familial factors and the lack of protective factors increase the risk of dual tobacco use.

In addition to family-related factors, peer dynamics also play a significant role in shaping adolescent health behaviours. This study found that adolescents who reported a lack of peer support had significantly higher odds of being dual tobacco users. This finding aligns with existing literature indicating that lower levels of perceived peer support are associated with increased susceptibility to risk-taking behaviours, including tobacco use. Rather than being in contrast to familial influences, peer-related factors operate alongside them as complementary predictors of adolescent tobacco use. Interventions that leverage positive peer influence, such as programmes encouraging leadership and smoke-free lifestyles, can effectively reduce dual tobacco use.^{38 40}

Truant adolescents often skip school together with peers who engage in similar delinquent behaviours, potentially steering them towards dual tobacco use.^{41 42} We also found differences in the prevalence and odds of being a dual tobacco user among the adolescents who commit truancy. A previous study showed that truancy is a precursor both for early and late onset smoking. Adolescents who skip school are 3.32 times more likely to start smoking early than those who attend school regularly.^{43 44} Addressing the lack of peer support and truancy is essential for reducing dual tobacco use and promoting healthier behaviours among adolescents.

The findings of this study have several important implications for strengthening tobacco control efforts in Malaysia. While the survey did not collect direct indicators of socioeconomic status, it is well-established in the literature that adolescents from lower socioeconomic status backgrounds are at greater risk of engaging in tobacco use, including dual use, due to factors such as limited access to health information, targeted marketing and reduced parental supervision.^{45 46}

Accessibility and affordability of tobacco products, particularly e-cigarettes, are also likely contributors to dual tobacco use. Despite regulations under Malaysia’s recent Control of Smoking Products for Public Health Act 2023, enforcement and coverage remain uneven, especially in informal or online markets. E-cigarettes remain widely available and often priced lower than conventional cigarettes, making them more accessible to adolescents.⁴⁷ Stronger regulation of pricing, taxation and point-of-sale restrictions, including online platforms, is essential.

The role of digital marketing and social media in influencing adolescent perceptions and use of tobacco products cannot be overstated. Exposure to online promotions, influencer content and unregulated advertising of e-cigarettes is linked to increased likelihood of initiation and continued use among youth.⁴⁸ Comprehensive bans on tobacco advertising, including for e-cigarettes across digital platforms, should be a policy priority. To respond effectively, multisectoral collaboration is critical. Healthcare professionals, educators, policy-makers and community leaders must work together to implement integrated tobacco control strategies. For example, the National Adolescent Health Policy 2001 was developed to address health challenges faced by young people, but it requires strengthening in several areas. Specifically, the policy should include more robust surveillance mechanisms for emerging tobacco products, incorporate digital literacy components related to tobacco marketing and increase funding for school-based interventions.

School-based programmes should be enhanced by shifting from one-time awareness campaigns to sustained, curriculum-integrated modules that cover both the physical and psychological harms of tobacco, digital misinformation and resistance skills training. These programmes should begin in lower secondary school and involve both teachers and peer leaders for greater impact.⁴⁹

Family and peer support mechanisms can be fostered through parent engagement workshops, peer-led antitobacco clubs and community health campaigns. Evidence shows that adolescents with strong parental bonds and positive peer influences are significantly less likely to initiate tobacco use.^{38 40} Community-level programmes that build these protective networks can reinforce national policy efforts.

The strength of our study depends on its large-scale survey with a large sample size which is enough to ensure adequate statistical power and reliability. One strength of this study is its strong methodological design, which helps reduce bias in the reported associations. As one of the first studies to explore dual tobacco use among Malaysian adolescents using nationally representative data, we found no evidence in the literature that common survey methods significantly affect results in similar studies. Combined with the high response rate (89%) and standardised data collection, this supports the reliability and validity of our findings.^{9 18} This is supported by established health surveillance tools such as GYTS and the Youth Risk Behaviour Surveillance System, which have demonstrated consistency and reliability across multiple settings using self-administered, school-based questionnaires.^{20 50} Although the study used a nationally representative sample of school-going adolescents, it excluded those not enrolled in school. As such, the findings may not fully capture the tobacco use behaviours of out-of-school youth, who may be at higher risk for health-compromising behaviours, including tobacco use. For Malaysia specifically, data from the MOE show that school enrolment rates among adolescents are high, often exceeding 90% for secondary school-aged children. Therefore, the impact on generalisability is likely minimal, but this should be stated transparently and cautiously.^{9 51}

However, we acknowledge several limitations in this study. First, the data on tobacco use were self-reported; this may cause recall or response bias, though previous research shows that anonymous, school-based surveys like this one generally yield reliable results.^{20 52} This cross-sectional study shows associations but cannot establish cause-and-effect relationships due to the lack of temporal data.^{53 54} Lastly, this study has no biochemical validation to verify any tobacco use. As a result, there may have been under-reporting of dual tobacco use. However, a previous study in Malaysia found consistency between self-reported data and exhaled carbon monoxide levels when respondents’ anonymity and confidentiality were assured.⁵⁵

Conclusions

Dual tobacco use among adolescents in Malaysia represents a significant public health challenge, underscoring the need for targeted intervention strategies that consider the sociodemographic characteristics of adolescents. This study showed a concerning prevalence of dual tobacco use, emphasising the urgency for a comprehensive approach to address this increasing trend. We can make significant progress by integrating prevention, education and rigorous study, along with the collaborative efforts of policy-makers, health professionals and educators, in protecting adolescents.

Data availability statement

No data are available.