IntroductionMore than 1 billion people were estimated to be current tobacco smokers in 2019, hence, in the coming decades, the massive health and economic costs of the tobacco epidemic are likely to increase.1GBD 2019 Tobacco Collaborators
Spatial, temporal, and demographic patterns in tobacco smoking prevalence and attributable disease: a systematic analysis of 204 countries and territories from the Global Burden of Disease Study 2019., 2WHO
WHO report on the global tobacco epidemic 2019: offer help to quit tobacco use. Large cohort studies have found that, at minimum, 50% of long-term tobacco smokers will die from causes directly linked to smoking, and that tobacco smokers have an average life expectancy that is 10 years shorter than that of people who have never smoked.3Doll R Peto R Boreham J Sutherland I Mortality in relation to smoking: 50 years' observations on male British doctors., 4Jha P Ramasundarahettige C Landsman V et al.21st-century hazards of smoking and benefits of cessation in the United States. The broad effects of tobacco use and exposure, from reduced productivity to increasing inequality, make tobacco control a priority for countries working to meet the global progress targets of the UN Sustainable Development Goals (SDGs) and the WHO monitoring framework for non-communicable diseases.2WHO
WHO report on the global tobacco epidemic 2019: offer help to quit tobacco use., 5WHO Regional Office for South-East Asia
Tobacco control for sustainable development., 6WHO
Tobacco control for sustainable development., 7Goodchild M Nargis N Tursan d'Espaignet E Global economic cost of smoking-attributable diseases., 8Bilano V Gilmour S Moffiet T et al.Global trends and projections for tobacco use, 1990-2025: an analysis of smoking indicators from the WHO Comprehensive Information Systems for Tobacco Control.Among established tobacco smokers, cessation leads to a substantial reduction in risk, with the largest reductions among individuals who stop smoking before age 40 years.9National Center for Chronic Disease Prevention and Health Promotion (US) Office on Smoking and Health
The health consequences of smoking—50 years of progress: a report of the surgeon general. Although progress has been made in improving cessation treatments and services, achieving high rates of cessation remains elusive worldwide. Despite most smokers intending to quit, fewer than 10% do so successfully and only 32% of the global smoking population has access to evidence-based cessation support services.2WHO
WHO report on the global tobacco epidemic 2019: offer help to quit tobacco use., 10Babb S Malarcher A Schauer G Asman K Jamal A Quitting smoking among adults - United States, 2000–2015., 11Hajek P Phillips-Waller A Przulj D et al.A randomized trial of e-cigarettes versus nicotine-replacement therapy. The health loss associated with tobacco use will remain substantial for years to come, particularly in low-income and middle-income countries, unless countries are able to substantially reduce the number of new smokers initiating each year. Strong efforts to prevent smoking initiation through effective policy and targeted interventions can interrupt this pipeline and yield massive long-term health and economic benefits.1GBD 2019 Tobacco Collaborators
Spatial, temporal, and demographic patterns in tobacco smoking prevalence and attributable disease: a systematic analysis of 204 countries and territories from the Global Burden of Disease Study 2019., 7Goodchild M Nargis N Tursan d'Espaignet E Global economic cost of smoking-attributable diseases., 12Global effects of smoking, of quitting, and of taxing tobacco.Previous studies have found that initiation of smoking tobacco use occurs predominantly among young people.9National Center for Chronic Disease Prevention and Health Promotion (US) Office on Smoking and Health
The health consequences of smoking—50 years of progress: a report of the surgeon general., 13Freedman KS Nelson NM Feldman LL Smoking initiation among young adults in the United States and Canada, 1998–2010: a systematic review., 14Marcon A Pesce G Calciano L et al.Trends in smoking initiation in Europe over 40 years: a retrospective cohort study., 15National Center for Chronic Disease Prevention and Health Promotion (US) Office on Smoking and Health
The epidemiology of tobacco use among young people in the United States and worldwide. Behavioural and biological studies suggest that young people are particularly susceptible to addiction, and that most adult smokers regret starting smoking.16Kessler DA Natanblut SL Wilkenfeld JP et al.Nicotine addiction: a pediatric disease., 17Fong GT Hammond D Laux FL et al.The near-universal experience of regret among smokers in four countries: findings from the International Tobacco Control Policy Evaluation Survey., 18Sansone N Fong GT Lee WB et al.Comparing the experience of regret and its predictors among smokers in four Asian countries: findings from the ITC surveys in Thailand, South Korea, Malaysia, and China. Despite this abundant motivating evidence, a study of tobacco use among adolescents aged 13–15 years across 143 countries found variable progress in reducing prevalence in this age group.19Prevalence and trends in tobacco use among adolescents aged 13–15 years in 143 countries, 1999–2018: findings from the Global Youth Tobacco Surveys.Research in context

Evidence before this study

Previous studies found that initiation of smoking tobacco use occurs predominantly among young people. An analysis of adolescents aged 13–15 years across 143 countries using Global Youth Tobacco Surveys found mixed progress in reducing prevalence of smoking tobacco use in this age group. Additional evidence suggests that adolescents and young people are more susceptible to addiction than older age groups and multi-country surveys consistently find that most adult smokers regret their decision to begin smoking.

Added value of this study

Using data from the Global Burden of Diseases, Injuries, and Risk Factors Study 2019, we estimated the prevalence of smoking tobacco use among young people (aged 15–24 years), by sex, for 204 countries and territories from 1990 to 2019. Our results are based on an analysis of more than 3000 surveys and provide timely estimates of the current situation and historical trends using a modelling framework that was consistently applied across countries. Our estimates of prevalence among people aged 15–24 years complement estimates of prevalence among the aged 13–15 years group, which might underestimate the magnitude of prevalence because initiation continues to occur after age 15 years. Additionally, we estimated the mean age of initiation of smoking tobacco use and a complete distribution of initiation ages among current smokers for all countries. Using this distribution, we identified the crucial age window during which young people transition to becoming regular smokers in each country, providing locally relevant estimates of a novel indicator that can be used to target interventions and policies to prevent initiation.

Implications of all the available evidence

Across all countries and both sexes, we found clear and consistent evidence that young people transition to become regular smokers in a narrow age window spanning adolescence and young adulthood. With 89·1% of eventual smokers initiating smoking by age 25 years, ensuring that young people remain smoke-free through their mid-twenties should result in substantial reductions in the prevalence of smoking in the next generation. Considered alongside behavioural and biological evidence on nicotine addiction and the documented feeling of regret among most adult smokers, our findings refute tobacco industry claims that smoking is an informed choice. To protect susceptible young people from lifelong nicotine addiction, governments and societies must intensify their efforts at prevention and imminently adopt, implement, and enforce strong tobacco control policies and interventions that target prevention of initiation of smoking tobacco use among young people.

We aimed to examine patterns of smoking tobacco use among young people, by sex, for 204 countries and territories. We analysed prevalence of current smoking among people aged 15–24 years. We also estimated the mean age of initiation and characterised the full distribution of age at initiation among current smokers aged 20–54 years in 2019 to identify the age window in which people are becoming addicted to nicotine across the globe, and link results to opportunities for intervention that can change the course of the tobacco epidemic for the next generation.

ResultsIn 2019, an estimated 155 million (95% UI 150–160) individuals aged 15–24 years globally were tobacco smokers. Prevalence of smoking tobacco use in this age group was 20·1% (19·4–20·8) among males and 4·95% (4·64–5·29) among females (appendix 2 pp 9–17). Prevalence of smoking tobacco use exceeded 20% among males aged 15–24 years in 120 countries and exceeded 20% among females aged 15–24 years in 43 countries (figure 1; appendix 2 pp 9–17). In 12 locations in 2019, more than 33% of people aged 15–24 years were current smokers. Five of these locations were islands in the Pacific (Kiribati, Federated States of Micronesia, Papua New Guinea, Cook Islands, and Nauru), four were in Europe (Bulgaria, Croatia, Latvia, and France), and the last three locations were Chile, Turkey, and Greenland. Among countries with a population of more than 1 million people, prevalence of smoking tobacco use among males was highest in Timor-Leste (52·8% [48·8–57·0]), Papua New Guinea (46·7% [41·3–52·0]), and Turkey (45·6% [42·2–48·9]), and among females prevalence was highest in Bulgaria (39·7% [32·5–47·1]), Chile (39·2% [29·8–49·2]), and Croatia (34·9% [28·8–41·0]; appendix 2 pp 9–17). Among all 204 countries and territories included in this analysis, Federated States of Micronesia, Kiribati, Timor-Leste, Greenland, and Papua New Guinea had the highest prevalence among males aged 15–24 years, and Greenland, Nauru, Bulgaria, Chile, and Federated States of Micronesia had the highest prevalence among females aged 15–24 years (appendix 2 pp 9–17). The countries with the largest numbers of current tobacco smokers aged 15–24 years in 2019 were China (26·5 million [23·8–29·4]), India (19·8 million [16·7–23·2]), and Indonesia (9·91 million [9·18–10·7]; appendix 2 pp 18–26). The ten countries with the largest number of tobacco smokers aged 15–24 years in 2019 accounted for 55·9% of total tobacco smokers in this age group (table).

Figure 1Prevalence of smoking tobacco use among females (A) and males (B) aged 15–24 years, in 2019

TableCurrent smoking tobacco use in 2019 and initiation patterns in the ten countries with the largest populations of smokers aged 15–24 years in 2019, by sex

Count data, prevalence data, and mean age at initiation are given to three significant figures. Data in parentheses are 95% uncertainty intervals. Prevalence and number of smokers are reported for individuals aged 15–24 years in 2019. Mean age of initiation window is reported for current smokers aged 20–54 years in 2019. The age window of initiation is defined as the range between the 10th and 90th percentiles of age at initiation.

Prevalence of smoking tobacco use decreased significantly between 1990 and 2019 among both males (32·9% [95% UI 29·9 to 35·9] decrease) and females (37·6% [32·2 to 42·7] decrease). Progress in reducing the prevalence of smoking varied substantially across countries (figure 2; appendix 2 pp 27–35). Of 204 countries and territories included in our analysis, only 81 (40%) had significant decreases in prevalence among individuals aged 15–24 years since 1990. The largest absolute decreases in the prevalence of smoking in this age group between 1990 and 2019 were observed in Norway (from 38·9% [36·0 to 41·7] in 1990 to 15·3% [12·8 to 18·3] in 2019; absolute change of –23·5 percentage points [–19·5 to –27·3]), Australia (from 35·8% [34·6 to 36·9] in 1990 to 15·1% [12·3 to 18·8] in 2019; absolute change of –20·6 percentage points [–16·9 to –27·3]), and Brazil (from 27·5% [25·2 to 30·0] in 1990 to 7·01% [5·85 to 8·31] in 2019; absolute change of –20·5 percentage points [–17·9 to –23·1]). 12 countries (Albania, Bosnia and Herzegovina, Serbia, Afghanistan, Lebanon, Bahrain, Lesotho, Saudi Arabia, El Salvador, Zambia, Antigua and Barbuda, and Uzbekistan) had significant increases in the prevalence of smoking in this age group between 1990 and 2019. The remaining 111 countries had no significant changes in prevalence over the past 30 years (appendix 2 pp 27–35).

Figure 2Prevalence of smoking tobacco use in 2019 vs percentage change in prevalence between 1990 and 2019, among individuals aged 15–24 years, by sex

Show full caption

Percentage change exceeded 100% for five countries (Albania, El Salvador, Afghanistan, Zambia, and Zimbabwe) for females and one country (Afghanistan) for males. These values were capped at 100% for visualisation purposes. To avoid visual clutter, only locations with significant increases, significant and large decreases, and high prevalence are labelled.

More countries had significant reductions in the prevalence of smoking tobacco use between 1990 and 2019 in males (n=83) than had significant reductions in the prevalence of smoking in females (n=24; figure 2; appendix 2 pp 27–35). Significant reductions in the prevalence of smoking were observed in countries across all initial levels of prevalence in 1990. The effect of population growth was relatively greater than reductions in prevalence in smoking tobacco use in 72 countries for males and 41 countries for females, resulting in significant increases in the number of smokers aged 15–24 years between 1990 to 2019 (appendix 2 pp 36–44). India, Egypt, and Indonesia had the largest absolute increases in the number of male smokers aged 15–24 years (4·67 million [95% UI 1·31–8·12] in India, 1·24 million [0·864–1·59] in Egypt, and 1·22 million [0·206–2·22] in Indonesia). The largest absolute increases in the number of females smokers aged 15–24 years were seen in Turkey (468 000 [127–840]), Jordan (116 000 [66·2–174]), and Zambia (110 000 [51·1–183]; (appendix 2 pp 45–53).Globally, 89·1% (95% UI 88·5–89·7) of current smokers aged 20–54 years in 2019 began smoking tobacco regularly by age 25 years, and 82·6% (82·1–83·1) began smoking in the 12-year window between ages 14 and 25 years (figure 3). Across all countries, the mean age at initiation of regular smoking of tobacco was 19·2 years (19·1–19·4). An estimated 18·5% (17·7–19·3) of current smokers aged 20–54 years in 2019 began smoking regularly by age 15 years, while 65·5% (64·3–66·5) began smoking by age 20 years. Across countries, the mean proportion of current smokers aged 20–54 years in 2019 who initiated smoking by age 21 years was 76·6% (59·0–97·5). The youngest mean ages at initiation were observed in Europe and the Americas. The oldest mean ages at initiation were seen in east and south Asia and sub-Saharan Africa (figure 3). Across countries and for both sexes combined, the mean age at initiation ranged from 16·4 years (16·2–16·7) in Denmark to 22·5 years (22·0–23·1) in Togo (figure 4; appendix 2 pp 54–62). Among the ten countries with the largest numbers of smokers aged 15–24 years in 2019, the lowest mean ages at initiation were observed in Russia (16·6 years [16·3–16·9]) for males and in the USA (17·5 years [17·1–17·9]) for females. The highest mean ages at initiation were observed in India (20·6 years [20·1–21·1]) for males and Bangladesh (26·3 years [24·8–28·0]) for females (table).

Figure 3Individual-level distribution of age at initiation among current tobacco smokers aged 20–54 years in 2019, by super-region and sex

Show full caption

The sizes of the bars indicate the total number of current tobacco smokers initiating at a given age. Data for males are displayed to the left of the vertical dashed line and data for females are displayed to the right.

Figure 4Mean age at initiation of regular smoking of tobacco among current smokers aged 20–54 years in 2019, both sexes combined

Males began smoking earlier than females in 194 (95%) of 204 countries, although the magnitude of between-sex difference varies by region. The difference in mean age of initiation between males and females is as small as 0·3 years (95% UI 0·2–0·4) in countries in the high-income super-region and as large as 3·8 years (3·1–4·6) in Southeast Asia, east Asia, and Oceania super-region (appendix 2 pp 52–62). Countries with the largest difference between males and females in the mean age at initiation also had the largest difference in prevalence of smoking tobacco use between males and females (appendix 2 pp 27–35, 63). Notably, in countries where prevalence of smoking tobacco use among people aged 15–24 years has decreased significantly, the mean age of initiation has remained constant over time. Of 81 locations that had significant decreases in prevalence among this age group, mean age of initiation did not change significantly in 76 locations and mean age of initiation increased by less than 1 year in the remaining five locations.Discussion

Despite decades of accumulated evidence of the harmful lifelong consequences of smoking tobacco use, the prevalence of smoking among young people remains high in many parts of the world. In 2019, global prevalence of smoking tobacco use among males aged 15–24 years was 20·1% (95% UI 19·4–20·8) and among females was 4·95% (4·64–5·29). 65·5% (64·3–66·5) of all current smokers aged 20–54 years in 2019 began smoking by age 20 years, with 82·6% (82·1–83·1) of all current smokers initiating during a 12-year window between the ages of 14 and 25 years. This age range is the crucial window during which individuals develop nicotine addiction and transition to becoming established smokers. Tobacco will remain a problem for generations to come if initiation of smoking tobacco use in this age window is not substantially reduced.

Our findings are important for two reasons: first, smoking tobacco use is probably the only risk factor among the top five global risk factors for mortality for which such an opportunity for intervention during a short window of time in a person's life exists. Unlike other risk factors, such as obesity, diet, and hypertension, if an individual does not become a regular smoker by age 25 years, then they are unlikely to become a smoker. This finding alone highlights the unique opportunity for interventions that target prevention of smoking initiation to individuals aged 25 years and younger. Second, a comprehensive package of evidence-based tobacco control interventions exists that have been shown to be effective among young people.22Comprehensive tobacco marketing restrictions: promotion, packaging, price and place., 23The effect of cigarette prices on youth smoking., 24Lantz PM Jacobson PD Warner KE et al.Investing in youth tobacco control: a review of smoking prevention and control strategies., 25Strategies for reducing youth access to tobacco: a framework for understanding empirical findings on youth access policies., 26Francis DB Mason N Ross JC Noar SM Impact of tobacco-pack pictorial warnings on youth and young adults: a systematic review of experimental studies., 27Robertson L Cameron C McGee R Marsh L Hoek J Point-of-sale tobacco promotion and youth smoking: a meta-analysis., 28Song AV Dutra LM Neilands TB Glantz SA Association of smoke-free laws with lower percentages of new and current smokers among adolescents and young adults: an 11-year longitudinal study. Hence, again, the risk factor of smoking tobacco use is unique compared with other risk factors, such as obesity, for which effective interventions have not yet been identified or implemented at a population level. Implementation and enforcement of strong tobacco control policies have led to substantial progress in protecting young people and reducing the number of young smokers in some countries. For example, Brazil had the largest decrease in prevalence of smoking tobacco use among individuals aged 15–24 years, with a decrease in prevalence of 74·5% (95% UI 69·0–78·9) from 27·5% (25·2–30·0) in 1990 to 7·01% (5·85–8·31) in 2019.29Portes LH Machado CV Turci SRB Figueiredo VC Cavalcante TM Silva VLDCE Tobacco control policies in Brazil: a 30-year assessment. The fact that mean age of initiation has remained consistent across time in countries that have experienced significant decreases in prevalence is encouraging evidence that interventions prevent smoking altogether, as opposed to only delaying the age at which people start smoking. Yet, persistently high prevalence of smoking tobacco use in young people in most countries, along with the increased use of e-cigarettes and vaping products that risk reversing progress,30de Andrade M Hastings G Angus K Promotion of electronic cigarettes: tobacco marketing reinvented?. highlight an urgent need to invest more strongly in tobacco control to protect young people from nicotine addiction.Effective interventions can be broadly grouped into two categories, demand reduction policies and supply-side restrictions.24Lantz PM Jacobson PD Warner KE et al.Investing in youth tobacco control: a review of smoking prevention and control strategies. Strong implementation and enforcement of interventions are crucial to success in all settings.24Lantz PM Jacobson PD Warner KE et al.Investing in youth tobacco control: a review of smoking prevention and control strategies., 25Strategies for reducing youth access to tobacco: a framework for understanding empirical findings on youth access policies. The MPOWER framework was introduced by WHO in 2008 to guide implementation of effective demand reduction interventions, following the WHO Framework Convention on Tobacco Control entering into force and becoming international binding law in 2005.2WHO
WHO report on the global tobacco epidemic 2019: offer help to quit tobacco use. The framework recommends that countries enact comprehensive smoke-free policies; use health warning labels and anti-tobacco mass media campaigns; ban tobacco advertising, promotion, and sponsorship; and decrease the affordability of tobacco products to reduce demand and prevalence of smoking tobacco use. The recommended interventions have been shown to reduce initiation of smoking among young people, and these strategies are even more effective when implemented jointly.22Comprehensive tobacco marketing restrictions: promotion, packaging, price and place., 23The effect of cigarette prices on youth smoking., 24Lantz PM Jacobson PD Warner KE et al.Investing in youth tobacco control: a review of smoking prevention and control strategies., 25Strategies for reducing youth access to tobacco: a framework for understanding empirical findings on youth access policies., 26Francis DB Mason N Ross JC Noar SM Impact of tobacco-pack pictorial warnings on youth and young adults: a systematic review of experimental studies., 27Robertson L Cameron C McGee R Marsh L Hoek J Point-of-sale tobacco promotion and youth smoking: a meta-analysis., 28Song AV Dutra LM Neilands TB Glantz SA Association of smoke-free laws with lower percentages of new and current smokers among adolescents and young adults: an 11-year longitudinal study. Despite notable progress in some countries, interference from the tobacco industry and waning political commitment have resulted in a large and persistent gap between knowledge and action on global tobacco control.2WHO
WHO report on the global tobacco epidemic 2019: offer help to quit tobacco use.Among tobacco control strategies, decreasing the affordability of tobacco products is one of the most effective mechanisms for reducing overall prevalence of smoking tobacco use, especially among young people who are particularly responsive to price changes.23The effect of cigarette prices on youth smoking., 31Youth are more sensitive to price changes in cigarettes than adults. Over the past decade, globally, 50% of countries decreased the affordability of tobacco products.2WHO
WHO report on the global tobacco epidemic 2019: offer help to quit tobacco use. Over the next decade, as countries look to accelerate progress towards to the SDGs, expanding and strengthening fiscal policy to reduce affordability of tobacco products is an essential component to preventing young people from starting to smoke.As the tobacco industry innovates new ways to market their products, including leveraging social media to reach young people through marketing campaigns and so-called influencers, tobacco control strategies must also evolve.30de Andrade M Hastings G Angus K Promotion of electronic cigarettes: tobacco marketing reinvented?., 32New media and tobacco control., 33O'Brien EK Hoffman L Navarro MA Ganz O Social media use by leading US e-cigarette, cigarette, smokeless tobacco, cigar and hookah brands. In the digital age, bans on advertising, promotion, and sponsorship recommended by MPOWER must extend to internet-based media, but as of 2018 only 25% of countries have comprehensively banned all forms of direct and indirect advertising.