Smoking and lifespan: how many years does each cigarette cost?
Smoking remains the largest single modifiable cause of premature death globally. We examine the dose-response data and the evidence for lifespan recovery after quitting.
Search for a command to run...
Longevity research is not only about what extends life. Understanding the highest-impact risk factors — and their reversibility — is equally important. These are not moral judgments; they are epidemiological data.
Smoking remains the largest single modifiable cause of premature death globally. We examine the dose-response data and the evidence for lifespan recovery after quitting.
For decades, moderate drinking appeared protective. The more recent Mendelian randomisation data tell a different story.
Chronic psychological stress shortens telomeres, accelerates epigenetic aging, and drives inflammation. We examine the mechanisms and the evidence-grounded interventions.
Social isolation is as harmful as smoking 15 cigarettes per day, according to a meta-analysis of 148 studies. We examine the biology of loneliness and the evidence for social connection.
Estimated years of life lost (YLL) from population-level epidemiological data. Figures represent median estimates across major cohort studies; individual variation is substantial.
| Risk factor | Years lost (median) | Primary mechanism | Reversible? | Evidence tier |
|---|---|---|---|---|
| Heavy smoking (>20/day) | 10–12 | Oxidative stress, DNA damage | Yes — 10yr cessation restores ~50% | I |
| Social isolation | 3–8 | Chronic inflammation, HPA dysreg. | Yes | I |
| Heavy alcohol (>4 drinks/day) | 5–7 | Liver fibrosis, epigenetic aging | Partial | I |
| Chronic stress (untreated) | 2–3 | Cortisol, telomere attrition | Yes | II |
| High sugar diet (>25g added/day) | 3–5 | Glycation, insulin resistance | Yes | II |
| Physical inactivity | 3–5 | Mitochondrial dysfunction, sarcopenia | Yes | I |
| Chronic sleep deprivation (<6h) | 2–4 | Immune suppression, metabolic dysreg. | Yes | I |
Sources: GBD 2019 Risk Factors Collaborators; Holt-Lunstad et al. (2015); Rehm et al. (2017); Kivimäki et al. (2012). Tier I = multiple RCTs or large prospective cohorts; Tier II = observational with plausible mechanism.