Tobacco smoking is one of the world’s largest health problems. Millions of people live in poor health because of smoking and researchers estimate that every year around 8 million people die an early death due to smoking.
It has been a major health problem for many decades. For the entire 20th century it is estimated that around 100 million people died prematurely because of smoking, most of them in rich countries.1
The share of smokers among the world population is falling and because smoking is such large health problem today this is one of the most positive developments in global health. It makes it possible that millions of people can live a longer and healthier life.
- Smoking is one the leading risks for early death – global studies estimate that about 8 million people die prematurely from smoking every year.
- This means that about 15% of global deaths are attributed to smoking. In some countries it’s more than 1-in-5 deaths.
- Smoking deaths typically affect older populations: more than half of deaths occurred in people over 70 years old; 93% were over 50 years.
- Death rates from smoking have fallen in most countries in recent decades.
- Nearly one-in-four adults in the world smoke tobacco. The share of smokers has fallen.
- Men are much more likely to smoke than women. As a consequence most victims are men too.
- The share of adults who smoke is falling in most countries around the world, regardless of income level.
- Taxing cigarettes, bans on advertising and support to help quit smoking are all critical to accelerate the decline of smoking.
All our charts on Smoking
The two regularly updated studies on the global death toll from tobacco use – published respectively by the World Health Organization and the Institute for Health Metrics and Evaluation – agree that around 8 million people die prematurely every year from tobacco use.
The aggregate statistics refer to ‘tobacco use’ (rather than smoking alone) because some other forms of consuming tobacco – for example chewing tobacco – also lead to premature deaths. The huge majority of deaths from tobacco use however are due to smoking. The death toll from smoking is very close to the total sum of deaths from tobacco use (higher than 99.9% in the case of the IHME estimates).
The WHO estimates that more than 8 million people die prematurely due to tobacco use each year. This is the latest available WHO estimate as of June 2021.
More than 7 million of those deaths are the result of direct tobacco use. About 1.2 million are non-smokers who are dying because they are exposed to second-hand smoke.
The IHME – in their annual Global Burden of Disease study – estimates that 8.7 million people die prematurely from tobacco use every year. As of June 2021 these are the latest estimates and refer to deaths in the year 2019.
7.7 million of those deaths are the result of smoking, 1.3 million are non-smokers who are dying because they are exposed to second-hand smoke. (An additional 56,000 people die annually from chewing tobacco.) The majority (71%) of those who die prematurely from smoking are men.
These estimates of the annual number of deaths attributed to a wide range of risk factors are shown in the interactive chart. This visualization is by default showing the global total, but can be explored for any country or region using the “change country” toggle. Smoking is a risk factor for several of the world’s leading causes of death, including lung and other forms of cancer, heart disease, and respiratory diseases. In the chart we see that it is the second leading risk factors for death globally.
There are two publications that are closely related to the annually updated Global Burden of Disease study:
While the overall GBD is published annually, the researchers from this same study also publish major publications that focus on particular risk factors less frequently. The last such GBD-related study that focused on the global health impact from smoking was published in The Lancet in early 2021.
In this study the researchers estimate that smoking tobacco accounted for 7.69 million deaths in 2019.
Reference: Reitsma et al. (2021) – Spatial, temporal, and demographic patterns in prevalence of smoking tobacco use and attributable disease burden in 204 countries and territories, 1990–2019: A systematic analysis from the Global Burden of Disease Study 2019. In The Lancet, 397(10292), 2337–2360.
The GBD estimates are also presented in yet another prominent publication. The Tobacco Atlas is published by the American Cancer Society and Vital Strategies and presents estimates for the global death toll from smoking taken from the Global Burden of Disease Study published by the Institute for Health Metrics and Evaluation (IHME).
Tobacco smoking has already been one of the world’s largest health problems for many decades. Over the course of the 20th century, it killed around 100 million people, most of them in today’s rich countries.2 The health burdens of smoking are now moving from high-income to low-to-middle income countries; some estimates have suggest that one billion people could die from tobacco over the 21st century.3,4
According to the estimates of the IHME about 13% of deaths worldwide was the result of direct smoking in 2017; a further 2% was the result of secondhand smoke. This means 15% – about 1-in-7 deaths – was the result of tobacco.
In the map here we see the share of deaths attributed to direct smoking across the world. In some countries this share was even greater than 13%.
There were several countries in 2017 where more than 1-in-5 deaths resulted from smoking, including China, Denmark, the Netherlands, Greece, Bosnia and Herzegovina, and Greenland.
Related chart– the share of deaths from secondhand smoke. This map shows the share of premature deaths attributed from secondhand smoke across the world.
In the interactive map we show death rates from tobacco smoking across the world. Death rates measure the number of premature deaths from smoking per 100,000 people in a given country or region.
We see large differences in death rates across the world. Rates tend to be highest across Asia and Eastern Europe, where figures are often over 100, and in some cases over 150 deaths per 100,000 people.
Across some of the lowest-income countries in the world, such as Sudan or Nigeria where very few people smoke, death rates are around ten times lower.
Related chart – number of deaths from smoking. This map shows the annual number of deaths from tobacco smoking across the world.
When we look at the breakdown of deaths from smoking by age we see that it is mainly older populations that are affected.
In the visualizations we show the death rates from smoking by age category, and the share of annual deaths which occur in each age group.
Here we see that death rates from smoking are much higher in people older than 70 years old, followed by those aged 50 to 69. Death rates for younger adults and children are very low.
This is also reflected in the number of deaths by age: in 2017 just over half of the people who died prematurely from smoking were older than 70 years old, and around 93% were older than 50 years.
Globally, death rates from smoking have fallen from 146 per 100,000 people in 1990 to 90 per 100,000 in 2017.
But where in the world are death rates falling or rising?
In the scatterplot here we see the comparison of smoking death rates in 1990 (shown on the y-axis) versus the death rate in 2017 (on the x-axis). The grey line is the line of parity: countries which lie along this line had equal death rates in 1990 as in 2017. Countries which lie above the grey line had higher death rates in 1990; those which lie below the grey line had higher rates in 2017.
We see that almost all countries in the world lie above the grey line: this means that most countries have achieved a decline in death rates from smoking in recent decades.
The declining global trend is reflected almost everywhere across the world.
Death rates from smoking have been declining in most countries over the past few decades. But what is happening to the number of deaths from smoking?
Here, the story is largely divided along lines of income.
In the scatterplot here we see the comparison of the number of deaths from smoking in 1990 (shown on the y-axis) versus the number of deaths in 2017 (on the x-axis). The grey line is the line of parity: countries which lie along this line had equal number of deaths in 1990 as in 2017. Countries which lie above the grey line had more deaths in 1990; those which lie below the grey line had more deaths in 2017.
What we find is that the total number of deaths attributed to smoking are falling in many rich countries today. If you hover over the ‘high-income’ label on the interactive chart you see that many countries across Western Europe, the United States and Canada all lie above the grey line. This means more people were dying as a result of smoking a few decades ago versus today.
The opposite is true for most low and middle income countries today. Although death rates from smoking tend to be falling, population growth and ageing in these countries means the total number of deaths has continued to increase in recent decades.
If we are to reverse the trends in the total number of deaths in most countries, death rates will need to fall much more quickly than they have been.
Related chart – global deaths from smoking over time. This chart shows the global number of deaths from smoking and secondhand smoke each year. It can also be viewed by country.
Nearly one-in-four (23%) adults in the world smoke tobacco.
But where in the world is smoking most common?
In the map we see the share of adults, aged 15 years and older, who smoke tobacco.
There are a number of countries where at least 40% of population smoke, if not more. The places where many people smoke are clustered in two regions. South-East Asia and the Pacific islands and Europe – particularly the Balkan region – but also France, Germany, and Austria.
In some countries very few people smoke: in Ethiopia, Ghana, and Nigeria less than 5% do.
There are several factors which influence the prevalence of smoking. One is prosperity: if we look at the relationship between smoking prevalence and income we find that richer countries tend to smoke more. But as you see in this correlation there are very large differences at each level of income.
Smoking rates are high across many countries, but we know from experience that this can change quickly. Many of today’s high-income countries had much higher rates of smoking in the past, and have seen a dramatic reduction. In 2000, the UK had rates similar to Indonesia today – 38% of adults smoked. Now, less than 20% do. The rise, peak, then decline of smoking is one we see across many countries.
The prevalence of smoking also differs significantly between men and women. Here we look at sex differences in smoking across the world.
This visualization shows the share of adults, aged 15 years and older, who smoke every day.
Here we see that daily smoking is most common across Europe and Asia – more than 20% of adults in most countries in these regions smoke every day.
Related chart – number of daily smokers. This map shows the number of people who smoke every day across the world.
Nearly one-in-four adults in the world smokes tobacco. But there are large differences between men and women.
In almost all countries it is true that more men smoke. In the visualization we see the share of men who smoke (plotted on the y-axis) compared with the same metric for women(plotted on the x-axis).
The grey line in the plot represents equality in the prevalence: countries where smoking is more common in men will lie above this line; and countries where more women smoke lie below.
We see that almost all countries lie above the grey line, meaning a higher share of men smoke. But there are a few exceptions: in the Pacific island-state of Nauru, more women smoke than men; and smoking rates in Denmark and Sweden show almost no sex difference.
In many countries – particularly across Asia and Africa – the differences are very large. We see these countries clustered on the far left, where smoking rates for women are very low – typically less than 5%. In Indonesia, 70% of men smoke but only 5% of women; in China it’s 48% of men versus 2% of women; and in Egypt half of men smoke whilst almost no women (0.4%) do.
The fact that men are more likely than women to smoke is reflected health statistics: particularly lung cancer, for which smoking is a primary risk factor. We see that in every country in the world, men are more likely to die from lung cancer.
Related chart – number of daily smokers by sex. This chart details the number of men and women who smoke daily.
The extent of smoking is not only determined by the prevalence of populations who smoke, but also by the intensity of smoking. This is measured as the average number of cigarettes consumed by smokers.
In the visualization here we see differences in the average number of cigarettes consumed by smokers each day across the world.
Across much of Asia, Eastern Europe, North America and Oceania, the average is around 20 to 25 cigarettes per day. Rates across Latin America, Africa and Western Europe tend to be slightly lower.
The smoking of cigarettes first saw a significant rise across today’s rich countries in the early 20th century. Since then, trends in smoking have undergone a century-long cycle of peak and decline.
In the visualization we see the average number of cigarettes sold per adult per day across many of today’s rich countries, including the United States and Europe.
Here we see that all countries followed a very similar trajectory: a steep rise in cigarette consumption during the early-to-mid 1900s; peaking from the mid-to-latter half of the century; before entering into a steep decline in the decades which followed. The path for almost all rich countries today was very similar.
What is notable is that this rise-peak-fall pathway took around a century in all cases. This long trajectory has had major health impacts for the populations of rich countries today.
The rise, peak then decline of smoking in rich countries took around a century. A long trajectory with severe health impacts.
This timescale is important when we consider low-to-middle income countries today: if they are to avoid the scale of the health consequences seen by rich countries, they must move through this pathway much quicker. The positive news is that most have – smoking is already falling in most countries today.
In the visualization here we see the share of adults who smoking in the year 2007 (on the y-axis) and 2018 (on the x-axis). The grey line here shows parity: countries that lie along this line would have the same prevalence of smoking in 2007 as in 2018. Countries which lie above this line had higher smoking prevalence in 2007; those below had lower prevalence in 2018.
We see that most countries lie above the grey line: this means the share of adults who smoke has declined in most countries in the world over the past decade. This is a surprising fact to many, since it means smoking prevalence is not only falling in high-income countries, but also at low-to-middle incomes.5
Low-to-middle income countries have effectively ‘leapfrogged’ the century-long rise-peak-decline pathway of rich countries. Almost everywhere, smoking is on the decline.
This chart shows death rates from lung cancer in men in the US and Spain since 1950. It is possible to add many more countries to this chart.
In many countries we see a significant rise, peak and then decline in lung cancer death rates in the 20th century. In the United States, the death rate peaked in the 1980s in men. In Spain this peak was later, only in the 1990s.
These trends are driven by the trends in smoking. The other chart shows the sales of cigarettes per person. Smoking is the biggest risk factor for lung cancer and we see that the trends in lung cancer follow those in smoking with a lag of around 20 years.
In 2017 7 Million people globally died a premature death because of smoking. The fact that smoking causes lung cancer is the major reason for the high death toll of smoking.
It is possible to add the data to lung cancer in women in the US to this chart. In the US it was once much more common for men to smoke so that the peaks of lung cancer for men are much higher. Smoking became more common for women only later so that lung cancer death rates for women peaked later.
The world map shows the Global Burden of Disease estimates of the share of cancer deaths that can be attributed to smoking.
Globally more than one in five cancer deaths (22% in 2016) are attributed to smoking – switch to the chart tab to see the global estimate.
In most richer countries the share is higher – the average in high-income countries is 28% in 2016.
In poor countries, where fewer people were smoking in the past, tobacco is responsible for a much smaller fraction of cancer deaths.
- Data: Consumption of tobacco products, prevalence of smoking and mortality
- Geographical coverage: 30 countries
- Time span: Often spanning back 100 years
- Available at: Online here
- Data: Death rates and absolute number of premature deaths from smoking and secondhand-smoke
- Geographical coverage: Global, across all regions and countries
- Time span: 1990 onwards
- Available at: Online here
- Data: Smoking prevalence, prices, taxes and policy support
- Geographical coverage: Global, across all regions and countries
- Available at: Online here