Smoking

The US National Health Interview Survey study followed 113,752 women and 88,496 men aged 25 or older, interviewed between 1997 and 2004, until the end of 2006. Among those aged 25 to 79, current smokers died at about three times the rate of never-smokers (hazard ratio 3.0 in women and 2.8 in men, adjusted for age, education, body fat and alcohol). The chance of surviving from age 25 to 79 was 70% in women who had never smoked compared with 38% in women who smoked, and 61% compared with 26% in men. Life expectancy was more than ten years shorter in current smokers [2].

The British Doctors Study recorded the smoking habits of 34,439 male doctors from 1951 and followed their causes of death until 2001. Men born between 1900 and 1930 who smoked only cigarettes and kept smoking died on average about ten years younger than lifelong non-smokers. The size of the hazard grew over the century as smokers started younger and smoked more heavily. Among men born in the 1920s, 43% of cigarette smokers died in middle age (35 to 69) compared with 15% of non-smokers, and the authors estimated that about two thirds of persistent cigarette smokers in that generation would eventually be killed by their habit, up from about half among men born in 1900 to 1909 [3].

Five pooled US studies from 2000 to 2010 showed that the risks for women who smoke had risen to match those of men: current smokers of both sexes aged 55 or older had about 2.8 times the all-cause mortality of never-smokers [7]. These estimates come from observational studies, in which people chose whether to smoke. They are adjusted for other factors, but the size, consistency and dose-dependence of the effect across generations and countries, together with the specific diseases involved, are why smoking is regarded as a cause of these deaths rather than a marker of other risks.

Smoking has its largest relative effects on the lungs. In the contemporary US data, current smokers aged 55 or older had about 25 times the risk of dying from lung cancer (25.0 in men and 25.7 in women) and 22 to 26 times the risk of dying from chronic obstructive pulmonary disease (COPD, a long-term lung disease that obstructs airflow), compared with never-smokers. Their risk of death from ischaemic heart disease was 2.5 to 2.9 times higher and from stroke about twice as high [7].

Smoking remains a strong risk factor for cardiovascular disease in older age. In individual data on 503,905 people aged 60 or older from 25 studies in the CHANCES consortium, current smokers had 2.07 times the cardiovascular mortality of never-smokers and former smokers 1.37 times. The authors expressed this as a risk advancement period: current smoking brought cardiovascular death forward by about 5.5 years. Excess risk rose with the number of cigarettes smoked and fell steadily with time since stopping [8].

Smoking is also linked to metabolic and brain disease. In 88 studies of 5,898,795 people, current smokers had 37% higher risk of developing type 2 diabetes than non-smokers, with risk rising with the amount smoked; the authors estimated that 11.7% of cases in men and 2.4% in women worldwide were attributable to smoking [9]. In a meta-analysis of 37 studies, current smokers had 30% higher risk of dementia, 40% higher risk of Alzheimer's disease and 38% higher risk of vascular dementia, while former smokers showed no increase [10]. Other risk factors for dementia are described under dementia risk factors, and the effects of smoking on the skin and mouth under skin aging and oral health.

A growing share of smokers smoke fewer than ten cigarettes a day. The NIH-AARP Diet and Health Study asked 290,215 adults aged 59 to 82 about their smoking at nine stages of life and followed their deaths through 2011. Compared with never-smokers, people who had consistently smoked fewer than one cigarette a day had 64% higher all-cause mortality (hazard ratio 1.64, 95% CI 1.07 to 2.51), and those who had consistently smoked one to ten a day had 87% higher mortality (1.87, 1.64 to 2.13) [4].

The association was strongest for lung cancer, with about 9 times the risk in people who smoked fewer than one cigarette a day and about 12 times the risk in those who smoked one to ten. Former light smokers had progressively lower risks the younger they had stopped. The authors concluded that the findings added to evidence that there is no risk-free level of exposure to tobacco smoke [4].

Secondhand smoke is the smoke that non-smokers breathe from other people's tobacco. A meta-analysis of 18 studies found that non-smokers exposed to it had 25% higher risk of coronary heart disease than unexposed non-smokers (relative risk 1.25, 95% CI 1.17 to 1.32). Risk rose with the amount of exposure: 23% higher with exposure to the smoke of 1 to 19 cigarettes a day and 31% higher with 20 or more [5].

A later meta-analysis of 24 studies estimated relative risks, for exposed compared with unexposed non-smokers, of 1.27 for ischaemic heart disease, 1.35 for stroke and 1.66 for COPD, with higher risks in women than in men for all three [11]. In seven studies of 156,439 never-smokers, exposure to secondhand smoke was associated with 22% higher risk of type 2 diabetes [9].

The benefit of stopping depends on the age at which it happens. In the US survey data, people who stopped at 25 to 34, 35 to 44 and 45 to 54 gained about 10, 9 and 6 years of life compared with those who continued [2]. In the British Doctors Study, stopping at 60, 50, 40 and 30 gained about 3, 6, 9 and 10 years of life expectancy; stopping at 50 halved the excess hazard and stopping at 30 avoided almost all of it [3].

Lung cancer risk follows the same pattern. In UK data from 1990, the cumulative risk of death from lung cancer by age 75 was 16% in men who continued to smoke cigarettes, and 10%, 6%, 3% and 2% in men who stopped at 60, 50, 40 and 30 [12]. The excess cardiovascular risk in older adults fell continuously with time since stopping [8].

Weight gain after stopping does not cancel the benefit. In three large US studies, recent quitters had a temporary rise in type 2 diabetes risk that was proportional to the weight they gained, peaking 5 to 7 years after stopping. They had no rise in mortality: compared with continuing smokers, cardiovascular mortality was 31% lower in recent quitters who gained no weight and 67% lower in those who gained more than 10 kg, and similar associations were seen for death from any cause [13]. In a meta-analysis of 26 studies, stopping smoking was also associated with reduced depression, anxiety and stress and with improved mood and quality of life compared with continuing to smoke [14].

Several treatments increase the chance of stopping. A 2023 Cochrane network meta-analysis of 319 randomised trials with 157,179 participants compared stop-smoking medicines and e-cigarettes with placebo or no medication, measuring abstinence after at least six months. Nicotine e-cigarettes (odds ratio 2.37), varenicline (2.33) and cytisine (2.21) had the largest effects, each with high-certainty evidence, equivalent to about seven or eight additional quitters per 100 people treated. Combination nicotine replacement therapy (a patch plus a fast-acting form such as gum or lozenge) had an estimated odds ratio of 1.93. A nicotine patch alone (1.37), fast-acting nicotine replacement alone (1.41) and bupropion (1.43) were also effective. Serious adverse events were uncommon, averaging 3% across treatments [15].

E-cigarettes heat a liquid, usually containing nicotine, to produce an aerosol that is inhaled. The 2024 Cochrane review of 88 studies with 27,235 participants found high-certainty evidence that nicotine e-cigarettes help more people stop smoking than nicotine replacement therapy (risk ratio 1.59, 95% CI 1.29 to 1.93), about four additional quitters per 100. The most common adverse effects were throat or mouth irritation, headache, cough and nausea. No serious harm was detected, but the longest follow-up was two years [16].

The long-term effects of vaping are unknown. A 2026 meta-analysis of 15 observational studies found that current e-cigarette users had about twice the odds of having COPD (odds ratio 2.03), including users who had never smoked cigarettes (2.09); because the studies measured both at one point in time, they cannot show which came first [17].

Lung cancer is the leading cause of cancer death in the US, with an estimated 135,720 deaths in 2020 [18]. The National Lung Screening Trial enrolled 53,454 people at high risk of lung cancer at 33 US centres and randomly assigned them to three yearly scans with low-dose computed tomography (CT) or with chest X-ray. Low-dose CT reduced deaths from lung cancer by 20.0% (95% CI 6.8 to 26.7) and deaths from any cause by 6.7%. Of the positive CT results, 96.4% were false positives [6].

The US Preventive Services Task Force recommends yearly low-dose CT screening for adults aged 50 to 80 who have a 20 pack-year smoking history and who still smoke or stopped within the past 15 years. A pack-year is the equivalent of smoking one pack a day for one year. Screening stops once a person has not smoked for 15 years [18]. Alcohol, the other main recreational drug covered in this section, is described under alcohol.

Full deep dive: Smoking and life expectancy.