Population-Level Coexistence of Smoking and Vaping Across the Life Course in Australia: Implications for Midlife and Ageing-Related Health Burden

Applications of Study Findings

• Tobacco control strategies should not assume that vaping has displaced smoking among adults approaching older age.

Background and Rationale

Globally, tobacco use remains a leading cause of preventable mortality, with an estimated 1.3 billion users worldwide, while electronic cigarette use has increased rapidly, particularly among younger populations (World Health Organization, 2023). Although smoking prevalence has declined in many high-income countries, it remains substantial among adults in midlife and older age groups, while vaping is disproportionately concentrated among adolescents and young adults. This divergence raises important questions about how nicotine use is distributed across the life course and whether emerging products are reshaping long-term health risks.

Population ageing has intensified concern about how long-term behavioural risk factors are distributed across adulthood and carried into later life. Cigarette smoking remains a leading contributor to morbidity, disability, and premature mortality among older adults, reflecting cumulative exposure accrued over decades (Doll et al., 2004). Although smoking prevalence has declined in many high-income countries, substantial levels persist among adults in midlife and older age groups, ensuring that smoking-related health burdens will continue to shape gerontological outcomes for years to come (Australian Institute of Health and Welfare., 2025).

Over the past decade, the rapid expansion of electronic cigarettes (e-cigarettes) and other vaping devices has altered the broader nicotine landscape, complicating the interpretation of population health paths. While smoking has continued to decline on average, vaping prevalence has increased sharply, particularly among adolescents and young adults (Hammond et al., 2019; McNeill, 2018). These divergent trends have prompted debate about whether vaping is accelerating reductions in smoking-related harm among ageing cohorts through population-level substitution or whether it represents a parallel pattern of nicotine use that coexists with persistent smoking throughout adulthood.

Two competing interpretations dominate this debate. One frames vaping as a harm-reduction pathway that may displace smoking among some adults, thereby potentially reducing cumulative exposure and future disease burden as cohorts age. The alternative perspective emphasises age-structured coexistence, in which vaping is concentrated among younger cohorts while smoking remains prevalent into midlife and older adulthood (Glantz & Bareham, 2018; McKeganey et al., 2025). Under this configuration, population-average declines in smoking may obscure the continued concentration of smoking-related risk among older adults, while simultaneously introducing new forms of nicotine exposure earlier in the life course.

A central challenge in evaluating these interpretations is the reliance on aggregate prevalence trends. Population averages can mask substantial heterogeneity by age, sex and cohort, limiting their usefulness for understanding later life health risks. Declines in smoking prevalence may occur alongside rising vaping among younger populations without implying that smoking has been displaced among adults who already carry the greatest cumulative exposure and are entering ages of heightened vulnerability to chronic disease (Creamer et al., 2019; McNeill, 2018). For gerontological research, understanding how nicotine use is structured across adulthood is therefore more informative than population averages alone.

Population-Level Patterns of Nicotine Use Across Adulthood

From a population health perspective, prevalence reflects the distribution of behaviours across age groups rather than individual-level transitions between products. In most high-income countries, smoking and vaping display opposing age gradients: vaping prevalence peaks in adolescence and young adulthood and declines sharply with age, whereas smoking prevalence increases into midlife and remains comparatively high at older ages (East et al., 2018; Sun et al., 2026). These patterns suggest that vaping uptake has been cohort-specific rather than diffusing evenly across the adult population.

Australian surveillance data mirror this configuration. Although smoking prevalence has declined across successive cohorts, it remains concentrated among adults in midlife and older age groups, while vaping is disproportionately concentrated among adolescents and young adults (Australian Bureau of Statistics, 2022; Australian Institute of Health and Welfare., 2025). As a result, reductions in smoking prevalence at the population level may coexist with persistent smoking-related risk among ageing cohorts, and do not necessarily indicate population-level replacement of smoking by vaping.

Sex differences further reinforce the need for age-structured analysis. Smoking prevalence has historically been higher among males across adulthood, while sex differences in vaping are less consistent and often age-specific (Bahji, 2020). Joint consideration of age and sex is therefore essential for assessing whether nicotine use patterns reinforce or mitigate existing inequalities in later-life health.

Conceptual Framework: Substitution Versus Age-Structured Coexistence

Guided by life-course Elder (1998) and population gerontology perspectives, this study adopts a conceptual framework that distinguishes behavioural substitution from age-structured coexistence at the population level (Online Appendix Figure 1). Population-level substitution implies that vaping displaces smoking within the same adult age groups over time, such that vaping prevalence exceeds smoking prevalence across adulthood, consistent with reduced cumulative exposure among ageing cohorts (East et al., 2018; Hammond et al., 2019). Age-structured coexistence, by contrast, describes a configuration in which different nicotine products dominate at different stages of the life course, reflecting cohort-specific uptake rather than widespread substitution among established smokers and allowing smoking-related harm to persist into later life (Creamer et al., 2019; Hammond et al., 2019).

This framework is grounded in life-course theory (Elder, 1998), which emphasises how health behaviours and exposures accumulate across different stages of life and shape later-life health outcomes. Smoking represents a cumulative exposure with effects that intensify over time, while vaping reflects a more recent, cohort-specific behaviour. While individual-level frameworks such as the Health Belief Model Janz and Becker (1984); Rosenstock (1974) explain behavioural uptake emphasising perceived susceptibility, perceived severity, perceived benefits, and barriers in shaping health-related decision-making, the present study adopts a population-level life-course perspective more suited to analysing age-structured patterns of nicotine exposure.

For conceptual clarity, smoking refers to the use of combustible tobacco products, vaping refers to the use of electronic cigarettes or related devices, dual use refers to the concurrent use of both products, and ageing-related health burden refers to the cumulative impact of long-term exposure on morbidity, disability, and health service demand in later life, consistent with established definitions in the tobacco control and public health literature (e.g., World Health Organisation, 2023).

Within this framework, smoking and vaping are conceptualised as key exposure variables, while age group or life stage functions as a moderating factor shaping their distribution and cumulative impact. The dependent outcome is ageing-related health burden, linking patterns of nicotine exposure to their long-term implications for population health. Smoking and vaping behaviours are expected to interact across the life course in ways that shape cumulative exposure. Where vaping substitutes for smoking, overall harm may decline; however, where both behaviours coexist or are distributed across different age cohorts, cumulative exposure remains concentrated among individuals entering later life (Dai et al., 2022).

From a gerontological perspective, population prevalence represents the aggregation of age and cohort-specific behaviours rather than a direct indicator of individual switching pathways (Elder, 1998). High vaping prevalence among younger cohorts alongside persistent smoking among older adults therefore signals coexistence rather than replacement and implies continued cumulative risk among ageing populations. Making this distinction explicit is critical for evaluating claims about harm reduction, prevention, and future health-care burden in ageing societies.

To summarise the analytical structure, smoking and vaping are treated as the primary independent variables, age group (life stage) functions as a moderating variable, and ageing-related health burden represents the dependent outcome. This structure provides a clear basis for distinguishing between population-level substitution and age-structured coexistence.

Study Aims

Building on this framework, the present study provides a population-level assessment of smoking and vaping prevalence across the adult life course in Australia, with particular attention to midlife and older age groups where cumulative exposure shapes later-life health outcomes. Using nationally representative cross-sectional survey data, the analysis examines age- and sex-specific prevalence patterns to evaluate whether vaping predominates in the population segments in which smoking has historically been most prevalent. In doing so, the study addresses a key gap in the literature, the lack of integrated life-course analysis of smoking and vaping within a unified framework, particularly in the Australian context. Existing studies have rarely examined these behaviours jointly across the life course, with limited attention to midlife and ageing populations where cumulative exposure is most consequential. Importantly, the analysis is descriptive and does not identify individual-level behavioural transitions such as substitution or cessation. This approach aligns with gerontological surveillance priorities and provides an empirical basis for assessing whether observed nicotine-use patterns are consistent with population-level substitution or alternatively, with the persistence of age-structured risk into later life.

Data and Methodology

Analytical Approach

The analysis focuses on describing how smoking and vaping prevalence are distributed across age and sex groups at the population level. The study adopts a descriptive analytical approach based on population-weighted prevalence estimation, absolute difference calculations, and ratio analysis as the primary tools of analysis. Two complementary descriptive measures are calculated for each age–sex group. First, absolute differences (percentage points) between vaping and smoking prevalence are used to quantify the magnitude and direction of divergence between the two behaviours. Second, vaping-to-smoking prevalence ratios are used to assess the relative dominance of vaping versus smoking within each age group.

Absolute differences are emphasised alongside relative measures because absolute prevalence gaps provide a more direct indication of population health burden and service implications, particularly in ageing populations where small relative differences may correspond to substantial absolute risk (Fuller et al., 2013; Harper & Lynch, 2007). Ratios are interpreted descriptively, with substantive emphasis placed on prevalence levels rather than ratio magnitudes alone, especially where prevalence is low at older ages.

Age and sex gradients in smoking and vaping prevalence are summarised using fractional response models, which are appropriate for proportion data bounded between zero and one (Papke & Wooldridge, 1996, 2008). In this study, these models are used strictly as a descriptive smoothing and summarisation tool to characterise age and sex patterns in bounded prevalence data, rather than for causal estimation or hypothesis testing. These models are applied to aggregated prevalence estimates and are used solely to characterise population-level patterns rather than to support causal inference or individual-level interpretation. This approach is appropriate for proportion outcomes and allows flexible modelling of non-linear age gradients while remaining consistent with the descriptive aims of the study (Mood, 2010). Models include age-group indicators, sex indicators, and their interactions to capture differences in prevalence across the life course and between females and males.

Results

Discussion and Conclusion

Implications for Midlife and Later-Life Health

A central implication of the observed age-structured coexistence pattern is the persistence of smoking-related risk into midlife and older adulthood. Although vaping prevalence is highest at younger ages, it does not exceed smoking prevalence beyond adolescence and declines rapidly across adulthood. In contrast, smoking prevalence rises into midlife and remains present at older ages. From a gerontological standpoint, this indicates that newer nicotine products have not displaced smoking among cohorts approaching later life, where cumulative exposure, rather than recent initiation, is the dominant determinant of health outcomes.

From a life-course perspective, this indicates that newer nicotine products have not displaced smoking among cohorts approaching later life, where cumulative exposure is the dominant determinant of health outcomes.

Midlife represents a critical stage in the life course: it is the period when the clinical consequences of long-term smoking begin to emerge and when cessation can still substantially reduce later-life morbidity. The finding that smoking prevalence far exceeds vaping prevalence among adults aged 45–64 years suggests that population-level patterns do not reflect widespread harm-reduction through substitution in these age groups. Instead, smoking appears entrenched among cohorts entering older age, reinforcing cumulative disadvantage and increasing the likelihood of multimorbidity and functional decline later in life.

These patterns have direct implications for ageing-related health burden and service demand. Smoking remains a major contributor to conditions that disproportionately affect older adults, including cardiovascular disease, chronic respiratory illness, cancer, and functional decline. The persistence of smoking into later adulthood therefore implies continued pressure on health and aged-care systems, even if overall smoking prevalence continues to decline. From a gerontological planning perspective, improvements in population averages may therefore mask substantial future demand for chronic disease management, disability support, and aged-care services among cohorts now entering older age.

The limited uptake of vaping among older adults also raises questions about the reach of harm-reduction strategies across the life course. While debates about vaping often centre on youth uptake, the low prevalence of vaping in midlife and older age groups suggests that vaping has not functioned as a widely adopted cessation or substitution pathway for established smokers approaching older age. In addition, the absence of information on dual use (concurrent smoking and vaping) limits the ability to assess whether vaping complements rather than replaces smoking among some individuals. This finding challenges assumptions that the availability of alternative nicotine products will automatically translate into reduced smoking-related harm among ageing populations. Instead, it underscores the importance of cessation strategies explicitly designed for adults in midlife and later life.

These findings challenge assumptions that the availability of alternative nicotine products will automatically translate into reduced smoking-related harm among ageing populations.

Implications for Tobacco Control and Public Health Policy

Interpreted through the substitution versus age-structured coexistence framework, these findings have important implications for tobacco control and public health policy in ageing societies. At the population level, the results provide limited support for the claim that vaping has replaced smoking among adults. Instead, vaping appears layered onto existing age-specific patterns of nicotine use, with minimal penetration among older cohorts who continue to bear the greatest burden of smoking-related morbidity and mortality. This coexistence creates a policy tension rather than a transition: youth-focused vaping prevention and adult smoking cessation must be addressed simultaneously rather than sequentially.

First, the concentration of vaping at younger ages reinforces the need for sustained efforts to prevent youth uptake, particularly given evidence that many young people who vape have no prior smoking history. Youth-focused regulatory strategies remain essential for protecting long-term population health. However, a policy focus dominated by youth vaping risks diverting attention from the substantially larger burden of smoking-related disease among adults approaching older age. Second, the persistence of smoking across adulthood highlights the continued importance of effective smoking cessation strategies targeted at established smokers, particularly in midlife and older age groups. These include primary care-based cessation programs, subsidised pharmacotherapy such as nicotine replacement therapy, and tailored behavioural interventions for adults aged 45–64 years. Policy approaches that focus primarily on vaping risk diverting attention from the principal driver of tobacco-related disease burden.

More broadly, the findings caution against treating vaping as a uniform population phenomenon. Instead, smoking and vaping occupy different positions at different stages of the life course, suggesting that policy responses should reflect this heterogeneity. A dual-track policy approach combining strong protections against youth uptake with intensified, age-appropriate cessation support for adults in midlife and older age is more consistent with observed population patterns and with the goals of healthy ageing. More broadly, the findings demonstrate that smoking and vaping occupy different positions across the life course, requiring differentiated policy responses rather than uniform approaches.

Beyond tobacco control policy, the findings have important implications for healthcare delivery and clinical practice. Given the persistence of smoking into midlife and older adulthood, routine smoking assessment and cessation support should remain central components of primary care, chronic disease management, and preventive health services. Adults aged 45–64 years represent a particularly important target group, as smoking cessation during midlife can substantially reduce later-life morbidity and mortality risks. Clinicians should therefore prioritise evidence-based cessation approaches, including behavioural counselling, nicotine replacement therapy, and referral to specialised cessation services where appropriate. More broadly, health systems planning should recognise that declining population smoking averages may not translate into immediate reductions in ageing-related disease burden if smoking remains concentrated among cohorts entering later life.

Limitations and Future Directions

Several limitations warrant consideration. First, the analysis relies on cross-sectional prevalence data and cannot identify individual-level transitions between smoking and vaping, such as initiation, cessation, or switching. Second, measures of smoking and vaping capture current use but do not reflect intensity, duration, or dual-use patterns, which may have distinct health implications. Third, estimates are based on published survey proportions and may be subject to reporting bias or under-coverage of specific subpopulations.

Future research using longitudinal and cohort data is needed to examine individual-level transitions between smoking and vaping and to assess how these transitions vary by age, sex, and cohort. Such studies would complement the population-level evidence presented here by clarifying the mechanisms underlying age-structured prevalence patterns. Further work linking cohort-specific nicotine-use patterns to later-life morbidity, functional outcomes, and health and aged-care service utilisation would be particularly valuable for gerontological research and policy. Future research should also examine how smoking and vaping patterns influence healthcare utilisation, multimorbidity, and ageing-related outcomes across cohorts, thereby improving forecasting of future health system demand in ageing populations.

Conclusion

This study demonstrates that smoking and vaping in Australia follow sharply divergent age distributions. Vaping prevalence is concentrated at younger ages and declines rapidly across adulthood, whereas smoking prevalence increases into midlife and remains substantial at older ages. The relative dominance of vaping over smoking is confined to adolescence and early adulthood, while smoking remains the dominant form of nicotine use across most of the adult life course.

The key contribution of this study is to show that nicotine use is structured by age in a way that reflects age-structured coexistence rather than population-level substitution. From a population gerontology perspective, these findings indicate that vaping has not replaced smoking at scale among adults and therefore has limited potential, at the population level, to reduce smoking-related harm among cohorts entering later life. These conclusions should be interpreted in light of the study’s descriptive design and the absence of data on individual transitions and dual use. Nonetheless, recognising the persistence of smoking among midlife and older cohorts is essential for interpreting population trends accurately and for designing tobacco control strategies that address both youth uptake and the ongoing burden of smoking-related risk in ageing populations.

Supplemental Material

Supplemental Material