Introduction

Tobacco use remains one of the leading causes of preventable morbidity and mortality worldwide, accounting for more than 8 million deaths annually, including approximately 1.3 million deaths attributable to second-hand exposure [1]. Although the global prevalence of conventional cigarette (CC) smoking has declined over recent decades, the tobacco epidemic continues to impose a substantial health and economic burden with global costs estimated at USD 1.4 trillion annually [2,3]. In response to tightening tobacco control policies, the tobacco industry has diversified its product portfolio, particularly through the introduction and expansion of electronic nicotine delivery system (ENDS), including electronic cigarettes (EC).

EC are battery-operated devices that aerosolise a liquid typically containing nicotine, propylene glycol, flavourings, and other chemical compounds. Marketed as a “safer” or “smoke-free” alternative to CC, EC have gained widespread acceptance, especially among adolescents and young adults [4,5]. However, emerging evidence indicates that EC aerosols contain multiple toxic substances, including formaldehyde, chromium, nickel, acetaldehyde, and tobacco-specific nitrosamines [6], which exert cytotoxic and oxidative effects and are associated with adverse respiratory outcomes [7,8]. These findings raise concerns that EC use may perpetuate nicotine dependence rather than facilitate cessation, in addition to causing health problems.

In Malaysia, a shift in nicotine consumption patterns is increasingly evident. While CC smoking prevalence declined from 21.3% in 2019 to 19.0% in 2023, EC use has risen from 0.8% in 2011 to 5.0% in 2023, with higher uptake observed among adolescents and young adults [9–11]. Among university students, EC use prevalence has been reported to range between 8% and 40%, indicating a substantial behavioural health concern within this population [12–15]. This demographic represents a critical transitional stage characterised by increased autonomy, academic demands, and psychosocial stressors, all of which are associated with engagement in health-risk behaviours, including nicotine use and physical inactivity [16–18].

Physical inactivity is a major contributor to global disease burden and is associated with non-communicable diseases, poor mental health, and reduced functional capacity [19,20]. In Malaysia, the prevalence of physical inactivity increased from 25.1% in 2019 to 29.9% in 2023 [10,11]. Among university students, participation in physical activity is often constrained by barriers such as time limitations, competing academic priorities, and environmental factors [21–25]. Evidence from CC smoking populations consistently demonstrates an inverse relationship between nicotine dependence and physical activity, whereby individuals with higher dependence are less likely to engage in regular physical activity [26,27]. However, findings specific to EC users remain limited and inconsistent [28,29].

EC use behaviours are multifaceted, encompassing variations in use frequency, nicotine content, and levels of nicotine dependence. These characteristics may influence engagement in health-promoting behaviours such as physical activity through physiological, behavioural, and cognitive mechanisms [30–32]. For instance, higher nicotine dependence may reduce physical activity participation due to withdrawal symptoms, fatigue, or prioritisation of nicotine-seeking behaviour. In addition, contextual factors – such as perceived barriers to physical activity, knowledge regarding EC use, and perceptions of harm and benefit – may further shape behavioural outcomes [33–35].

Previous studies, both internationally and within Malaysia, have examined components of EC use and health behaviours, but often in isolation [13,22,24,36–41]. Malaysian studies have focused primarily on prevalence, patterns of EC use, and associated sociodemographic factors among university students [12–15], while separate lines of research have investigated physical activity levels and their barriers within similar populations [21–25]. For example, studies conducted across Malaysian universities have reported substantial variability in EC use prevalence and highlighted associations with sociodemographic and behavioural factors, although these studies did not examine physical activity concurrently [12–15]. Internationally, a consistent inverse relationship has been established between CC smoking and physical activity [26,27], however, evidence specific to EC use remains limited and inconsistent, with some studies suggesting differing behavioural patterns compared to traditional smokers [28,29]. Importantly, few studies have simultaneously examined EC use characteristics – such as frequency, nicotine content, and dependence – alongside contextual factors to understand their association with physical activity. This lack of integrated analysis limits understanding of how nicotine-related behaviours interact with lifestyle behaviours within young adult populations.

To address this gap, the present study adopts a structured conceptual framework in which EC use characteristics (frequency of use, nicotine content, and nicotine dependence) are treated as primary exposures, physical activity level is specified as the primary outcome, and sociodemographic variables are included as background covariates. Contextual factors, including barriers to physical activity, EC-related knowledge, and perceived harms and benefits, are incorporated as modifying influences that may affect engagement in physical activity (Fig 1).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Research study conceptual framework.

https://doi.org/10.1371/journal.pone.0354336.g001

Accordingly, the primary objective of this study is to examine the association between EC use characteristics and physical activity level among Malaysian university students. Secondary analyses are conducted to characterise contextual and behavioural factors associated with physical activity; however, these are exploratory and intended to support interpretation of the primary association. This approach calls for more integrated behavioural analyses of EC use within Southeast Asian populations, where patterns of use and lifestyle behaviours may differ from Western contexts.

By explicitly prioritising physical activity as the primary outcome and situating EC use characteristics within a defined analytical framework, this study aims to provide a clearer understanding of the behavioural relationship between nicotine use and physical activity. The findings are expected to inform the development of targeted, non-pharmacological strategies to address the dual burden of EC use and physical inactivity among university students in Malaysia.

Methods

Results

The study included 564 Malaysian university students who reported current EC use. Overall, 64.9% (n = 366) were classified as physically active, while 35.1% (n = 198) were inactive based on IPAQ classification. Descriptive characteristics of the sample, including sociodemographic profile, EC use patterns, and physical activity levels and barriers, are presented in Table 1. Participants reason for EC use is summarised in Table 2, while perceived benefits, harms, and risk of EC use are detailed in Table 3. Associations between EC use characteristics, contextual factors and physical activity are examined using crude and adjusted ratios in Table 4, with item-level analyses of EC knowledge and physical activity barriers provided in S1 Table and S2 Table, respectively. Lastly, perceived side effects associated with EC use are illustrated in Fig 2.

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Sociodemographic, E-Cigarette Use Behaviour and Physical Activity (N = 564).

https://doi.org/10.1371/journal.pone.0354336.t001

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Reasons for Using E-Cigarette (N = 564).

https://doi.org/10.1371/journal.pone.0354336.t002

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. E-cigarette User Perceived Benefits, Harms and Risks (N = 564).

https://doi.org/10.1371/journal.pone.0354336.t003

Download:

• PNG
  larger image
• TIFF
  original image

Table 4. Crude and Adjusted Odds Ratio for Factors Associated with Physical Activity among E-cigarette users (N = 564). (Outcome: Physical Activity Level; Inactive = event category, Active = reference category).

https://doi.org/10.1371/journal.pone.0354336.t004

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. E-cigarette use potential side-effects.

https://doi.org/10.1371/journal.pone.0354336.g002

Discussion

This study examined association between EC use characteristics, behavioural factors, and physical activity among Malaysian university students who use EC. Interpreted through the conceptual framework (Fig 1), the findings suggest that physical activity behaviour is shaped primarily through behavioural and contextual pathways – particularly nicotine dependence and perceived barriers – rather than by sociodemographic or knowledge factors alone. These associations remain correlational, given the cross-sectional design and the potential for reverse causation and residual confounding. Consequently, it cannot be determined whether higher nicotine dependence and greater perceived barriers contribute to lower physical activity participation, or whether lower levels of physical activity are associated with increased nicotine dependence and more pronounced perceived barriers. Therefore, the observed relationships should be interpreted as associations rather than evidence of causal pathways.

A key contribution of this study lies in positioning of EC use within a broader behavioural profile. The association between nicotine use and physical activity is not linear or uniform across populations. Prior evidence suggests that higher nicotine dependence is associated with lower engagement in physical activity, particularly when behavioural intensity is considered [31,32]. In contrast, studies that do not account for dependence often report mixed findings, including higher activity levels among EC users compared with non-users [48]. This inconsistency is reflected in the present findings, where nicotine dependence, rather than frequency of use, nicotine content or dual-use status, emerged as the primary EC-related correlate of physical inactivity. Other EC variables were excluded from the adjusted model due to multicollinearity. This supports the interpretation that these variables cluster around a common behavioural construct – namely dependence, rather than acting independently.

Although nicotine dependence emerged as an important behavioural correlate, the strongest associations observed in the multivariable model were related to perceived barriers to physical activity, particularly within the personal domain. Participants reporting high personal barriers were more than twenty times more likely to be physically inactive compared to those reporting low barriers. These barriers encompassed factors such as lack of motivation, low confidence, perceived lack of benefit, embarrassment, and limited energy for physical activity. These findings suggest that contextual and self-regulatory constraints may exert a greater influence on physical activity participation than EC use characteristics alone thus highlighting the importance of considering physical activity behaviour within a broader behavioural and environmental context rather than solely through nicotine-related pathways.

At the same time, the broader literature suggests that EC users do not represent a uniform risk group. For example, youth who use EC have been shown to be more physically active than traditional smokers, while still engaging in nicotine use [48]. Longitudinal evidence further complicates this relationship, showing that while higher physical activity may be associated with lower EC use cross-sectionally, yet may also predict future uptake, potentially due to perceptions of reduced harm [49]. Similarly, dual users and smokers are less likely to meet physical activity recommendations compared to EC-only users [28], while recent evidence suggests that EC use may reduce the likelihood of achieving high physical activity levels [34]. These findings indicate that the relationship between EC use and physical activity is heterogenous and context dependent, and is better understood when behavioural intensity and user profiles are considered.

Within this behavioural context, the present study provides further insight into the role of knowledge and perception. Although overall EC knowledge was high, it was not independently associated with physical activity. This aligns with studies showing that EC users often report higher subjective knowledge while simultaneously holding more favourable perceptions of EC use, including beliefs that it is safer, cost-effective, and useful for smoking cessation [36,37,50,51]. This suggests that knowledge may coexist with cognitive rationalisation, rather than directly promoting protective behaviour. In the present study, perception-related variables were attenuated after adjustment, supporting the interpretation that their influence is indirect and mediated through behavioural factors such as dependence and habitual use. Within the conceptual framework (Fig 1), knowledge and perceptions functions as intermediate constructs, influencing behaviour through more proximal mechanisms rather than acting independently.

An unexpected finding was that participants who reported obtaining a “definite nicotine hit” from EC use had lower odds of physical inactivity after adjustment. This result should be interpreted cautiously. The crude analysis demonstrated the opposite pattern, with perceived nicotine satisfaction associated with higher odds of inactivity, suggesting that the adjusted association may reflect the complex interrelationship between nicotine dependence, user perceptions, and behavioural characteristics rather than a direct protective effect of nicotine exposure itself. One possible explanation is that participants who perceive a clear nicotine effect may represent a subgroup of more experienced or intentional EC users who are nevertheless physically active and use ECs within recreational or social contexts rather than as highly dependent users. Previous studies have shown that EC users are a heterogeneous population, with some users maintaining active lifestyles despite ongoing nicotine use, particularly among younger adults and university students [28,48].

Alternatively, results of the multiple regression models may reflect a statistical suppression effect arising from simultaneous adjustment for nicotine dependence and multiple correlated perception variables. Perceived nicotine satisfaction is conceptually related to dependence, craving, satisfaction, and frequency of use, several of which demonstrated significant crude associations with physical inactivity. When highly correlated constructs are entered into the same multivariable model, coefficient reversal can occur, resulting in associations that differ substantially from unadjusted estimates. The reversal observed between crude (cOR = 2.06) and adjusted (aOR = 0.51) estimates is consistent with this possibility and suggests that the variable may be acting as a marker of a broader user profile rather than exerting an independent behavioural effect. Given the cross-sectional design and potential residual confounding, this finding should not be interpreted as evidence that nicotine reinforcement promotes physical activity. Rather, it highlights the complexity of the relationship between EC-related perceptions, dependence, and health behaviours, warranting further investigation using longitudinal designs and a more detailed measures of EC use patterns.

The Malaysian context further supports this interpretation. Wan Puteh et al (2018) [13] reported that EC use among Malaysian university students was strongly influenced by social norms, peer influence, and perceived benefits, including smoking cessation, self-image, and social engagement. Importantly, despite reporting symptoms such as dizziness, cough and headaches, many users did not perceive significant health risks. EC-related side effects were commonly reported across multiple physiological domains; however these variables were not analysed in relation to physical activity and therefore should not be interpreted as behavioural determinants in this study. Nonetheless, existing evidence indicates that EC use is associated with a broad range of symptoms, including respiratory, neurological, and gastrointestinal effects, likely mediated through inflammatory, oxidative, and neurophysiological pathways [38,52,53]. The current literature remains inconclusive, with predominantly cross-sectional evidence and substantial heterogeneity limiting causal inference. Within this context, EC use may be embedded within socially reinforced environment where perceived benefits, such as smoking substitution, nicotine satisfaction, and social acceptability outweigh perceived harms in shaping behaviour [13].

The strongest and most consistent findings in this study relate to perceived barriers to physical activity, which showed the largest and most robust associations. Within the conceptual framework, these barriers represent proximal determinants that directly constrain behaviour through limitations in motivation, time, energy, and competing priorities. This aligns with established behavioural models in which capability, opportunity, and motivation interact to determine behaviour [24,25,54].Systematic evidence indicates that environmental context, social influence, and goal prioritisation are among the most important domains influencing physical activity in university populations [54], while studies in Malaysia have similarly identified time constraints, academic demands, and motivational barriers as key determinants [21,55–57].

The present findings extend this evidence by demonstrating that personal barriers – particularly those related to motivation and perceived effort, are more strongly associated with inactivity than environmental barriers. This suggests that internal constraints may outweigh structural limitations in this population, even in contexts where access to facilities is available. EC use may indirectly reinforce these barriers. Wan Puteh et al (2018) [13] highlight that EC use is often linked with behavioural routines and coping mechanism, such as stress relief and social interaction, which may compete with engagement in physical activity. This supports the interpretation that the relationship between EC use and physical activity is not solely physiological but also behaviourally embedded in daily routines.

The interaction between EC use and physical activity may also be influenced by perceived health effects. Participants most frequently reported gastrointestinal, respiratory, and neurological symptoms, which may affect physical activity through discomfort or reduced exercise tolerance. These findings are consistent with previous evidence linking EC use to respiratory and systemic health effects [58,59] and with qualitative reports of perceived reductions in physical performance during physical activity among EC users [33]. Although self-reported, such experiences may discourage participation in physical activity and contribute to the gradual adoption of more sedentary behaviours over time.

Sociodemographic variables showed limited independent associations in this study, with only selected ethnic differences persisting after adjustment. This pattern is consistent with broader epidemiological findings in which demographic effects diminish once behavioural and contextual variables are considered [60,61]. The predominantly male sample reflects national patterns of EC use in Malaysia [11,62], but limits generalisability. Prior research suggests that female students may experience greater perceived barriers to physical activity, particularly in social and environmental domains, indicating that these associations may differ in more sex-balanced populations [63–65].

These findings support the conceptual framework (Fig 1) as a multi-layered behavioural system, in which EC use characteristics operate through behavioural mechanisms (nicotine dependence and routines), contextual constraints act as the strongest proximal determinants of physical activity, and knowledge and perceptions exert indirect influence, often mediated through perceived benefits and behavioural reinforcement. This integrated perspective suggests that physical activity among EC users is shaped by interaction between behavioural intensity, contextual constraints, and cognitive framing rather than isolated factors.

This study has several strengths, including a relatively large multi-university sample, validated instruments, and a theory-informed framework. However, several limitations should be acknowledged. The cross-sectional design limits causal inference and raises the possibility of reverse causation. Self-reported measures may introduce recall and social desirability bias. The lack of biomechanical verification of nicotine exposure limits interpretation of dose-response relationships, and unmeasured confounders such as mental health, academic stress, and sleep patterns may influence both EC use and physical activity. In addition, variability in EC products, usage patterns, and institutional environments may contribute to heterogeneity in observed associations.

Overall, the findings suggest that physical activity among EC users is shaped less by sociodemographic characteristics or knowledge and more by behavioural patterns (nicotine dependence and routines) and perceived environmental barriers. Future research should adopt longitudinal and mixed-methods approaches to better elucidate these pathways and to determine whether modifying dependence, perceptions, or contextual barriers leads to meaningful improvements in physical activity.

Conclusion

In this cross-sectional study of Malaysian university students who use EC, physical activity patterns were more consistently associated with behavioural and contextual factors – particularly nicotine dependence and perceived personal barriers, than with sociodemographic characteristics or knowledge-related variables. Although participants generally demonstrated awareness of EC-related harms, this did not consistently correspond to more favourable physical activity patterns, suggesting that knowledge alone may be insufficient to influence behaviour in the presence of reinforcing perceptions and established routines.

From a practical standpoint, these findings indicate that interventions targeting university students who use EC may benefit from prioritising behavioural and contextual determinants, including strategies to address nicotine dependence and reduce key personal barriers such as motivation, time constraints, and competing academic or lifestyle demands. Approaches that incorporate behavioural support and context specific opportunities, such as accessible campus-based physical activity initiatives may be more effective than education-only strategies.

However, given the cross-sectional design and the potential for residual confounding and reverse causation, these implications should be interpreted as hypothesis-generating rather than prescriptive. Future longitudinal and intervention-based studies are needed to determine whether modifying nicotine dependence and contextual barriers can lead to sustained improvements in physical activity among EC users.

Supporting information