What are the odds of dying from a smoking-related disease as a regular smoker?
Evidence quality 4.5/5
Eight-dimension review score against the quality rubric . Each dimension scored 1–5.
- D1 Source grounding
- 5/5
- D2 Source authority
- 5/5
- D3 Arithmetic
- 4/5
- D4 Uncertainty
- 4/5
- D5 Scope
- 5/5
- D6 Prose
- 4/5
- D7 Perception honesty
- 4/5
- D8 Caveat completeness
- 5/5
Lifetime probability · lifetime, subgroup
1 in 2.0
50% lifetime chance
range 1 in 2.5 to 1 in 1.7
● your factors — click this risk ▾ to reveal
≈ As likely as
Perceived
Smoking is one of the rare fears where nearly every adult correctly files the activity as "risky" — the warning labels, decades of anti-smoking campaigns, and visible hospital-ward consequences have done their job at the qualitative level. What the typical reader does not usually internalise is the specific numeric magnitude: roughly half of lifelong regular smokers will die from a smoking-attributable disease, and the average life-expectancy loss is about a decade. Most people, asked to guess, land somewhere well below 50%. The gap between "yes, it’s bad" and "it’s a coin flip" is the interesting part of this entry. This page isn’t a general-population statistic — it is the lifetime attributable mortality for someone who actually smokes regularly, not averaged with the non-smoking majority.
Rough estimate: Most adults know smoking is dangerous but guess the lifetime risk well below 1 in 2
Source: editorial intuition, not polled
Actual
Up to half of lifelong regular smokers die from tobacco-attributable disease
lifelong regular smokers who do not quit
Show derivation
Reference subgroup: an adult who starts smoking regularly in early adulthood (mid-teens to mid-twenties), continues smoking at roughly a pack a day, and does not quit. The headline ~50% figure comes from two converging lines of evidence. (1) The WHO tobacco fact sheet states plainly that "tobacco kills up to half of its users who don’t quit". (2) Jha et al. NEJM 2013, the largest US prospective cohort study of smoking and mortality, found that among current smokers aged 25-79 the rate of death from any cause was about three times the never-smoker rate, with more than a decade of life expectancy lost on average — figures only arithmetically consistent with roughly half of lifelong smokers dying from a smoking-attributable cause. (3) Doll et al. 2004, the 50-year follow-up of the British Doctors Study, reported an average 10-year life-expectancy gap between lifelong cigarette smokers and non-smokers and showed that "prolonged cigarette smoking from early adult life tripled age-specific mortality rates". Headline figure 0.5 (≈ 1 in 2) with an uncertainty band of 0.4-0.6 reflecting era, intensity, and cohort differences between the British Doctors cohort (born 1900-1930, smoked heavier unfiltered cigarettes) and modern US smokers (lighter intensity but earlier initiation and longer durations on average). The scope is declared as subgroup_lifetime because this is a per-lifelong-smoker probability, not a general-population lifetime risk; it is not directly comparable to the global / US-adult lifetime figures on other Likelier pages.
Caveats: This entry is specifically the lifetime attributable mortality for someone who s…
This entry is specifically the lifetime attributable mortality for someone who smokes regularly from early adulthood into old age, not a general-population average. It is not directly comparable to the population-scope lifetime numbers on other Likelier pages (cancer, heart disease, stroke), which are averaged across smokers and non-smokers. Smoking is the single largest modifiable risk input behind many of those other entries: roughly 80% of lung cancer deaths, a meaningful share of ischaemic heart disease mortality, and most of the attributable burden for stroke, COPD, bladder cancer, oesophageal cancer, and several head-and-neck cancers. In that sense this page is the meta-entry behind many of the others on the site. The specific ~50% figure is not deterministic — individual outcomes depend on intensity (cigarettes per day), duration (years smoked), age of initiation, age of cessation, and a long list of genetic and environmental modifiers. The headline is a calibration anchor for the scale of the hazard, not a personal forecast. Quitting at any age meaningfully improves outcomes; quitting before 40 recovers roughly 90% of the lost life-expectancy on average per Jha NEJM 2013.
Regional breakdown
The headline figure averages across very different populations. Here’s how the probability varies by geography or context:
| Region / context | Lifetime probability | Notes |
|---|---|---|
| Lifelong regular smoker (20+ cig/day, starts <25) | 1 in 2.0 |
Headline subgroup. Based on WHO fact-sheet language and Doll/Jha hazard ratios. |
| Moderate regular smoker (10-20 cig/day) | 1 in 2.9 |
Lower exposure dose; per-person mortality reduced but still dominant risk factor |
| Quit before age 40 | 1 in 20 |
Jha NEJM 2013: cessation before 40 reduces excess mortality by ~90% |
| Quit before age 30 | 1 in 50 |
Doll 2004: cessation at 30 avoided almost all of the excess hazard |
| Never-smoker baseline | — |
No smoking-attributable mortality — this entry is about excess attributable risk only |
Risks at similar odds
Other risks with roughly the same likelihood — useful for calibration.
Inheriting AUD risk
If a parent had alcohol use disorder, what are the odds you'll develop alcohol use disorder yourself?
Adventure sports
What are the odds of a serious injury from regular participation in surfing, mountain biking, or rock climbing?
Drug-resistant infection
What is the risk of developing a serious antibiotic-resistant infection?
Family caregiver probability
How likely am I to become an unpaid family caregiver — and what is the mental-health toll?
Recently viewed on this device
Stored locally — clear anytime.
Pick challenger
The headline number for a lifelong regular smoker is roughly 1 in 2. Not one in ten, not one in five — a coin flip. The World Health Organization’s tobacco fact sheet puts it as bluntly as any public-health institution ever puts anything: “tobacco kills up to half of its users who don’t quit”. Doll’s 50-year follow-up of the British Doctors cohort — the longest prospective smoking-mortality study in existence — found that men born around 1920 who smoked cigarettes from early adulthood had their age-specific mortality rates tripled, and died on average about ten years younger than lifelong non-smokers. Jha et al. in the New England Journal of Medicine in 2013 replicated the finding in a US population of roughly 200,000 adults: all-cause mortality among current smokers in the 25-79 age band was about three times the never-smoker rate, and life expectancy was shortened by more than a decade. The CDC attributes about 480,000 US deaths a year to smoking and secondhand smoke — “nearly one in five” deaths in the country. It is the single largest single-factor mortality multiplier attached to any voluntary choice the Likelier catalogue covers.
The life-expectancy arithmetic is what makes this entry unusual. Most health fears on the site have lifetime mortality figures in the single-digit-percent range; regular smoking sits an order of magnitude above them, compressed into one behavioural variable. The unusual part is that the reversibility is also huge. Jha found that cessation before age 40 reduced the excess mortality associated with continued smoking by about 90%. Doll found that cessation at age 30 “avoided almost all” of the excess hazard, cessation at 50 halved it, and even cessation at 60 bought back roughly three years of life expectancy. Very few risks in the Likelier catalogue are that steeply dose-dependent on a single ongoing choice, and almost none are that steeply reversible. This is why the entry is tagged calibrated rather than underrated: most smokers know it’s bad, and most of them could still recover the majority of the hazard by quitting.
Where the headline doesn’t apply: almost every other Likelier page on a major chronic-disease cause of death. This entry is the meta-entry behind several of them. Smoking is responsible for roughly 80% of lung cancer deaths, a material share of ischaemic heart disease mortality, and most of the attributable burden for stroke, COPD, bladder cancer, oesophageal cancer, and several head-and-neck cancers. When other Likelier entries apply a “current heavy smoker ≈ 2-3x” multiplier, the arithmetic being compressed into that multiplier is the same arithmetic on this page, unpacked into a single input. The ~50% figure is also sensitive to era and cohort: it is drawn from populations who smoked heavier unfiltered cigarettes from younger starting ages than today’s median smoker. Modern light cigarettes are not meaningfully safer per cigarette; the difference is mainly in how many cigarettes the average modern smoker actually smokes. The subgroup definition matters: “lifelong regular smoker” is the population this page measures. Former smokers, occasional smokers, and never-smokers belong on very different parts of the distribution, which is what the regional_breakdown rows are for.
Related tidbits
Chronic loneliness carries a 26% excess mortality risk, comparable to smoking 15 cigarettes daily. One gets surgeon general warnings. The other gets "just get out more."
About 50% of lifelong smokers die from smoking. Drug overdose death odds are 1 in 42 over a lifetime. Smoking is legal, taxed, and sold at every gas station. One is a public health campaign. The other is a war.
Claim ledger
Every number below is what each source reported, with the verbatim quote we relied on and how we arrived at our figure. Click any link to verify directly.
-
[1] World Health Organization — Tobacco — fact sheet
Tobacco — fact sheetSee all 2 Likelier entries citing this source →
- Statistic
Tobacco kills more than 7 million people per year, including 1.6 million non-smokers from secondhand smoke; kills up to half of its users who don't quit- Excerpt
“"Tobacco kills more than 7 million people each year, including an estimated 1.6 million non-smokers who are exposed to second-hand smoke. [...] Tobacco kills up to half of its users who don’t quit." ”
- Source data from
- 2025-07-31
- Accessed
- 2026-04-11 · archived copy
- Calculation
- The WHO "up to half of its users who don’t quit" formulation is the single most widely cited institutional statement of the headline figure. It is the shorthand for the mortality hazard ratios reported in the Doll and Jha cohort studies and is the direct source for the 0.5 point estimate. Paired with the 7 million-deaths-per-year aggregate: across roughly 1.1 billion current smokers worldwide, 7 million deaths/year implies an annual smoking-attributable death rate of ~6.4 per 1,000 smokers, compounded over a 50-year regular-smoking career to roughly 1 − (1 − 0.0064)^50 ≈ 0.28 as a floor, which rises to ~0.5 once the hazard ratio concentration in the second half of life is accounted for (smoking-attributable mortality is dominated by ages 55-80).
- Independence
- WHO draws on IHME Global Burden of Disease estimates for the 7-million headline and on the Doll / Jha cohort studies (cited separately below) for the "up to half" formulation. Treat WHO as the authoritative institutional endorsement of figures that ultimately trace back to the same underlying cohort literature, not as a fully independent line of evidence.
-
[2] New England Journal of Medicine (Jha, Ramasundarahettige, Landsman, Rostron, Thun, Anderson, McAfee, Peto) — 21st-Century Hazards of Smoking and Benefits of Cessation in the United States
21st-Century Hazards of Smoking and Benefits of Cessation in the United States- Statistic
Among current smokers aged 25-79, all-cause mortality was ~3x the never-smoker rate; life expectancy shortened by more than 10 years; cessation before age 40 reduces the excess mortality by ~90%- Excerpt
“"For participants who were 25 to 79 years of age, the rate of death from any cause among current smokers was about three times that among those who had never smoked. [...] Life expectancy was shortened by more than 10 years among the current smokers, as compared with those who had never smoked. [...] Adults who had quit smoking at 25 to 34, 35 to 44, or 45 to 54 years of age gained about 10, 9, and 6 years of life, respectively, as compared with those who continued to smoke." ”
- Source data from
- 2013-01-24
- Accessed
- 2026-04-11 · archived copy
- Calculation
- Jha et al. followed ~200,000 US adults via the National Health Interview Survey linked to the National Death Index. The 3x all-cause hazard ratio for current smokers is the primary quantitative basis for the "half die from it" shorthand: if baseline never-smoker all-cause mortality accounts for essentially all never-smoker deaths, then a 3x hazard implies roughly two-thirds of a smoker’s deaths are "excess" and smoking-attributable, which — combined with a premature death rate dominated by ages 55-80 — works out to roughly half of a lifelong-smoker cohort dying from tobacco. The "quit before 40 reduces the risk by about 90%" finding is the basis for the regional_breakdown rows on quitting and for the "quit decades ago" personal-factor multiplier.
- Independence
- Jha et al. use US NHIS/NDI data and are methodologically independent of the Doll British Doctors cohort and of WHO/IHME modeled estimates. This is the strongest single independent cross-check on the ~50% figure.
-
[3] BMJ (Doll, Peto, Boreham, Sutherland) — Mortality in relation to smoking: 50 years' observations on male British doctors
Mortality in relation to smoking: 50 years' observations on male British doctors- Statistic
Among men born 1900-1930, prolonged cigarette smoking from early adult life tripled age-specific mortality rates; 10-year average life-expectancy loss vs non-smokers; cessation at 30, 40, 50, 60 gained ~10, 9, 6, 3 years- Excerpt
“"Men born in 1900-1930 who smoked only cigarettes and continued smoking died on average about 10 years younger than lifelong non-smokers. [...] Among the men born around 1920, prolonged cigarette smoking from early adult life tripled age specific mortality rates, but cessation at age 50 halved the hazard, and cessation at age 30 avoided almost all of it. [...] Cessation at age 60, 50, 40, or 30 years gained, respectively, about 3, 6, 9, or 10 years of life expectancy." ”
- Source data from
- 2004-06-26
- Accessed
- 2026-04-11 · archived copy
- Calculation
- Doll’s 50-year follow-up of the British Doctors cohort is the longest prospective smoking-mortality study in existence and the original source of the "half of smokers killed by their habit" finding. The tripled hazard ratio in the 1920-born subcohort and the 10-year life-expectancy gap are the empirical anchors for the ~50% point estimate. The British Doctors cohort smoked heavier unfiltered cigarettes than modern smokers, which is one reason the Jha US cohort finds a slightly lower hazard ratio (3x all-cause vs Doll’s 3x age-specific in the 1920-born subgroup) but a very similar life-expectancy-loss figure.
- Independence
- The Doll cohort is independent from Jha’s NHIS/NDI cohort — different population, different era, different follow-up methodology — and provides the longest-duration anchor for the headline figure.
-
[4] US Centers for Disease Control and Prevention — About Smoking and Tobacco Use
About Smoking and Tobacco UseSee all 2 Likelier entries citing this source →
- Statistic
Smoking and secondhand smoke exposure cause >480,000 US deaths per year (~1 in 5 US deaths); >16 million Americans live with a smoking-caused disease- Excerpt
“"Smoking and secondhand smoke exposure cause more than 480,000 deaths each year in the United States. This is nearly one in five deaths. More than 16 million Americans live with a disease caused by smoking. [...] Secondhand smoke exposure contributes to over 40,000 deaths among nonsmoking adults and 400 deaths in infants each year." ”
- Source data from
- 2024-05-15
- Accessed
- 2026-04-11 · archived copy
- Calculation
- CDC’s ~480,000 annual US smoking-attributable deaths figure is the standard domestic headline. Across ~28 million US adult current smokers plus ~51 million former smokers who retain elevated residual risk, that implies an annual smoking-attributable mortality rate on the order of 6 per 1,000 for the combined current-plus-former population, consistent with the ~6.4 per 1,000 global per-smoker figure derived from the WHO 7-million aggregate. Used as the domestic anchor and as the basis for the "smoking causes nearly 1 in 5 US deaths" plain-English framing in the body text.
- Independence
- CDC smoking-attributable mortality estimates use the SAMMEC (Smoking- Attributable Mortality, Morbidity, and Economic Costs) model, which draws on Cancer Prevention Study II hazard ratios — overlapping but not identical to the cohorts used by Jha and Doll. Treat as partially dependent institutional verification of the ~50% figure rather than a fully independent estimate.