Substance and claimed effects

Cigarette smoking — the daily, repeated inhalation of combusted tobacco smoke — is the most studied modifiable health behaviour in the medical literature. The substance combines pharmacologically active nicotine (the dependence-producing agent) with thousands of combustion products including tar, carbon monoxide, polycyclic aromatic hydrocarbons, tobacco-specific nitrosamines, hydrogen cyanide, formaldehyde and reactive oxidant species. The 2014 US Surgeon General's report USSG 2014 formally attributes causation to disease across nearly every organ system in the body. This entry covers the substance and its meaningful consequences across the dimensions named in the brief: cardiovascular disease, lung function (including COPD), cancer at multiple sites, skin and visible appearance, male and female fertility, all-cause lifespan, and the time-graded reversal of these risks after cessation. Smoking is an avoid entry — the action the reader takes is to not start, or to stop.

Evidence by addressing question

mechanism

Two parallel injury cascades drive almost everything smoking does to the body. First, the combustion stream delivers a high oxidant load directly into the lungs and, via blood, to systemic endothelium. Reactive oxygen species in smoke quench nitric oxide and damage the endothelial monolayer, reducing flow-mediated dilation and triggering the chronic vascular inflammation that produces atherosclerosis Gallucci 2020. The Surgeon General lists six interlocking cardiovascular mechanisms: endothelial damage, prothrombotic state (platelet activation, increased fibrinogen), systemic inflammation, dyslipidaemia (raised LDL, lowered HDL), increased myocardial oxygen demand from sympathetic activation, and decreased oxygen supply from carbon monoxide displacement of haemoglobin USSG 2014. Carbon monoxide binds haemoglobin with ~240× the affinity of oxygen; carboxyhaemoglobin levels in heavy smokers run 5–10% versus <1% in non-smokers, directly reducing oxygen delivery.

Second, the carcinogen panel — benzo[a]pyrene, N-nitrosamines (NNK, NNN), aromatic amines — forms covalent adducts with DNA. When repair fails, signature mutations accumulate in epithelia directly exposed (bronchial, oral, laryngeal, oesophageal) and in epithelia downstream of metabolised carcinogen excretion (bladder, kidney, pancreas). The mutational signature of tobacco smoke (Alexandrov SBS4) is detectable in lung tumours and rises monotonically with pack-years USSG 2014.

In the airway, smoke triggers chronic neutrophilic inflammation, protease–antiprotease imbalance (elastase versus α1-antitrypsin), goblet-cell hyperplasia, ciliary paralysis, and the small-airway remodelling and alveolar destruction that constitute COPD. Skin effects share roots with vascular pathology: chronic dermal microvascular constriction plus matrix metalloproteinase upregulation (MMP-1 in particular) accelerates collagen and elastin breakdown.

evidence

The evidence base on smoking is among the largest in human medicine — over a million deaths analysed across multiple large prospective cohorts with multi-decade follow-up.

All-cause mortality. The Jha et al. 2013 NEJM analysis of 202,248 US adults (National Health Interview Survey linked to mortality) found current smokers' all-cause death rate was approximately three times that of never-smokers, with life expectancy shortened by more than 10 years Jha 2013. The 50-year follow-up of the British Doctors' Study (34,439 men) found the same 10-year life-expectancy gap, with stopping at 60/50/40/30 returning 3/6/9/10 years of life respectively Doll 2004. The Million Women Study (1.3M UK women) reported a mortality rate ratio of 2.76 (95% CI 2.71–2.81); two-thirds of female smokers aged over 50 will die from smoking-related disease, but quitting before 40 averts about 90% of the excess risk Pirie 2013.

Cause-specific mortality. In the Jha cohort, hazard ratios for cause-specific death in current smokers were: lung cancer HR 17.8 (women) / 14.6 (men); other respiratory diseases (COPD-dominated) HR 8.5 / 9.0; ischaemic heart disease HR 3.5 / 3.2 Jha 2013. The Carter et al. 2015 NEJM pooled analysis of ~1 million participants identified additional causally-linked deaths beyond the previously-recognised set (renal failure, infections, ischemic bowel) that bring smoking-attributable mortality estimates up by ~17% Carter 2015.

Lung cancer. A 2012 meta-analysis of 287 studies put the relative risk of lung cancer at 8.43 for current smokers versus never-smokers, with intensity-stratified ORs ranging up to 18.3 for small-cell carcinoma in the heaviest-smoking categories (≥30 cig/day) Lee 2012.

Cardiovascular. Smoking approximately doubles 10-year risk of fatal coronary events; the effect is dose-dependent but with a striking near-linear extension into the low-exposure range — even <5 cig/day produces ~40–50% of the cardiovascular risk of pack-a-day smoking Gallucci 2020.

Pregnancy. Maternal smoking increases stillbirth odds by 47% (OR 1.47, 95% CI 1.37–1.57), with a clear dose-response: 1–9 cig/day raises odds 9%, ≥10/day raises odds 52% Marufu 2015. Low birthweight OR is 1.89 (55-study meta-analysis) Quan 2021.

Fertility. A 2016 meta-analysis of 5,865 men found significantly reduced sperm count, motility and normal morphology in smokers, with effect size scaling with smoking intensity and exceeding the WHO lower reference limit for clinical infertility in heavy smokers Sharma 2016. Erectile dysfunction risk: current-smoker OR 1.51, former-smoker OR 1.29 versus never-smokers in a meta-analysis of 28,586 men Cao 2013.

Lung function recovery. The Lung Health Study (Anthonisen et al., 5,887 middle-aged smokers with mild airflow obstruction, 14-year follow-up) is the canonical RCT of cessation: sustained quitters had FEV₁ decline 28 mL/year, intermittent quitters 48 mL/year, continuing smokers 62 mL/year. The development of clinically significant airway obstruction was largely prevented in those who quit Anthonisen 1994.

Visible aging. Model's 1985 BMJ characterisation of "smoker's face" — accentuated nasolabial and perioral wrinkling, gaunt cheeks, greyish complexion — has been replicated by multiple controlled image-analysis studies; smokers in their 40s commonly show wrinkle counts equivalent to non-smokers in their 60s Model 1985.

protocol

The protocol is cessation. The 2020 Surgeon General's report on smoking cessation is the consensus reference: pharmacotherapy plus behavioural support roughly doubles or triples successful quit rates over willpower-alone USSG 2020. The Cochrane component network meta-analysis (Lindson et al. 2023) ranks first-line pharmacotherapy:

• Varenicline: RR for sustained abstinence 2.32 (95% CI 2.15–2.51) versus placebo (41 trials, 17,395 participants). Most effective single agent.
• Combination NRT (long-acting patch + short-acting gum/lozenge/inhaler/spray): RR comparable to varenicline.
• Single-form NRT: RR ~1.55–1.60 versus placebo; varenicline beats single-form NRT (RR 1.25, 95% CI 1.14–1.37).
• Bupropion: RR ~1.60 versus placebo; varenicline beats bupropion (RR 1.36, 95% CI 1.25–1.49).
• Cytisine: emerging evidence comparable to varenicline; widely used in Eastern Europe.
• Nicotine e-cigarettes: now rated by Cochrane as having high-certainty evidence of greater quit-rate efficacy than NRT (RR 1.59, 95% CI 1.29–1.93), with the caveat that long-term safety beyond ~2 years is unestablished Lindson 2023.

Behavioural support — text-message programmes, telephone quitlines, structured counselling, group programmes — adds an independent ~10–15% to quit rates and combines additively with pharmacotherapy.

Cessation timeline. The Surgeon General's reports document a graded recovery USSG 2014 USSG 2020:

• 20 minutes: heart rate and blood pressure drop toward baseline.
• 12 hours: blood carbon monoxide normalises; haemoglobin oxygen-carrying capacity restored.
• 2 weeks – 3 months: circulation improves; FEV₁ begins improving.
• 1–9 months: cilia regrow, cough and dyspnoea decrease.
• 1 year: excess coronary heart disease risk halved.
• 5 years: stroke risk approaches never-smoker baseline; oral, throat, oesophageal, bladder cancer risk halved.
• 10 years: lung cancer mortality risk roughly halved versus continuing smokers (never returns fully to never-smoker baseline).
• 15 years: coronary heart disease risk approaches never-smoker baseline.

The life-expectancy gain from quitting (Jha 2013, Doll 2004 corroborate): quitting by age 30 returns ~10 years; by 40 returns ~9 years; by 50 returns ~6 years; by 60 returns ~3 years. Even quitting late offers meaningful gain; the message is that earlier is dramatically better but it is never too late Jha 2013 Doll 2004.

contraindications

Standard pharmacotherapy contraindications are well-mapped USSG 2020:

• Bupropion: contraindicated in seizure disorders, eating-disorder history, current MAOI use. Caution in bipolar disorder.
• Varenicline: dose-reduce in severe renal impairment; the historical FDA black-box warning for neuropsychiatric events was removed in 2016 after EAGLES (8,144-participant trial) found no excess versus placebo, NRT, or bupropion.
• NRT in pregnancy: lower nicotine doses than smoking deliver; behavioural support is first-line in pregnancy but NRT is preferred over continued smoking when needed.
• E-cigarettes in pregnancy and youth: not recommended; data inadequate.

Smoking itself has no clinical contraindication — the behaviour is harmful in every population studied. But certain comorbidities (recent MI, pregnancy, COPD, post-surgical recovery) raise the urgency of cessation from years-of-life to immediate-clinical-outcome scale.

misconceptions

• "Light smoking is safe." Pirie 2013 found that women smoking <10 cig/day still had double the 12-year mortality of never-smokers Pirie 2013. The dose-response is steep at the low end for cardiovascular endpoints — there is no safe threshold.
• "Quitting later is pointless / damage is done." Doll 2004 specifically refutes this: quitting at 60 still returns ~3 years of life Doll 2004. Cardiovascular risk halves within a year of cessation regardless of age.
• "I'd gain too much weight if I quit." Average post-cessation weight gain is 4–5 kg, plateaus around year 1, and is dwarfed by the cardiovascular and oncologic benefits of cessation in every cohort analysis.
• "Hand-rolled / 'natural' / menthol / low-tar cigarettes are safer." None of these is safer in any large cohort; some (menthol) facilitate deeper inhalation and may be worse.
• "It's the nicotine that kills you." Nicotine is the addictive agent but is not the principal driver of cancer or cardiovascular mortality — the combustion products are. This is the conceptual basis for harm reduction with vaping/NRT.
• "Vaping is just as bad." Acute toxicity profile is substantially lower than combusted tobacco; long-term outcomes remain under investigation, but Cochrane now rates nicotine e-cigarettes as effective cessation aids with high-certainty evidence Lindson 2023.

failure-modes

The most common failure mode is single-modality willpower-only attempts. Unaided quit attempts succeed long-term at roughly 3–7% per attempt; pharmacotherapy plus behavioural support roughly triples that USSG 2020. The median successful quitter makes ~6–11 attempts before achieving sustained abstinence — relapse is the norm, not a personal failure, and reattempt within weeks of relapse is the protocol. Other failure modes: stopping pharmacotherapy too early (the 8–12 week course matters), under-dosing NRT (most users use too little patch + no short-acting), trying to quit during an acute stressor (work crisis, bereavement) without scheduled support, and bidirectional triggers (alcohol co-use is the highest-risk relapse context).

stakes

The reader-relevant stakes are: a typical 35-year-old smoker who continues is choosing, in expectation, to lose about a decade of life — most of it the final retirement decade — and to spend years before that with diminished function (breathlessness on stairs, reduced exercise capacity, impotence risk, worsening skin) Doll 2004 Jha 2013. The Pirie cohort shows the effect is symmetric in women Pirie 2013. For partners, fertility delays of months to years are typical; for the next generation, pregnancy smoking elevates stillbirth and low-birthweight risk substantially Marufu 2015 Quan 2021.

payoff

Cessation produces the most rapidly-reversing risk profile in modern preventive medicine. The first day delivers a measurable cardiovascular signal (heart rate, blood pressure, CO clearance); the first year halves the excess coronary risk; the first decade brings lung cancer risk down by half; by 15 years the cardiovascular profile is largely indistinguishable from never-smokers. The Jha and Doll cohorts independently quantify the life-expectancy reclamation: quit by 30, gain ~10 years; quit by 40, gain ~9 years; quit by 50, gain ~6 years Jha 2013 Doll 2004. Subjective gains the reader can expect: taste and smell return within weeks, exertion tolerance improves over months, skin tone improves perceptibly over 6–12 months (improved microcirculation), and the chronic morning cough resolves within months.

practicalities

A pack-a-day habit at US/UK prices runs roughly $3,000–$5,500/year. Pharmacotherapy is comparatively cheap: varenicline generic ~$200–400/12-week course, NRT patches plus gum ~$150–300/course, bupropion generic ~$50–100/course. Most US insurance plans now cover cessation pharmacotherapy with no copay (Affordable Care Act preventive-services mandate); UK NHS provides free cessation pharmacotherapy and behavioural support through stop-smoking services. Quitline access (1-800-QUIT-NOW in the US, NHS Smokefree in the UK) is free.

The credibility range

Optimist case (for cessation). Smoking is the most preventable cause of death in modern medicine; the evidence base is the largest and most internally consistent of any modifiable risk factor. Multiple million-participant cohorts converge on the same numbers across cultures and decades. The pharmacotherapy toolkit is mature and inexpensive; insurance coverage is broad. Even partial success — switching from combusted tobacco to nicotine e-cigarettes — produces a substantial expected health gain, and Cochrane now endorses this on high-certainty evidence. A reader who quits by 40 has near-complete reclamation of life-expectancy.

Skeptic case. The cessation literature concentrates on the easy outcomes — sustained abstinence at 6–12 months — and the success rates remain modest (15–35% with optimal pharmacotherapy at 6 months, falling further at 5 years). E-cigarette long-term safety beyond ~2 years is not established; the vaping-induced lung injury (EVALI) episode of 2019 was traced to additive contamination rather than to nicotine vaping per se, but the lack of decades-long cohort data is a genuine gap. The dose-response near zero is real but uncertain; whether ultra-light social smoking carries the same proportional risk as steady low-volume daily smoking is not fully settled.

Author's call. This is one of the clearest, highest-evidence calls in the catalogue: avoid starting, and if smoking, quit. The marginal cost-benefit of quitting outweighs nearly every other preventive intervention available. Controversy score is essentially zero (no credible scientific position defends smoking continuation); evidence score is at the ceiling. The action is unambiguous; the difficulty is behavioural, not epistemic.

Stakeholder and incentive map

• Tobacco industry. Commercial incentive to retain smokers and recruit new ones; well-documented historical suppression of internal evidence (the Master Settlement Agreement documents). Now diversified into heat-not-burn and nicotine-pouch products.
• E-cigarette industry. Mixed — some firms are independent harm-reduction advocates, others are tobacco subsidiaries (Juul–Altria, Vuse–RJ Reynolds, Blu–Imperial). Incentive structure is complicated.
• Pharma. Modest commercial interest in varenicline (Pfizer, generic since 2021), NRT (J&J Nicorette franchise, GSK Nicoderm). Bupropion is generic and cheap.
• Public health bodies. CDC, NHS, WHO, USPSTF, FDA all aligned on cessation. Disagreement is real between US public-health bodies (cautious on e-cigarettes for cessation) and UK bodies (Public Health England, NICE — more permissive on e-cigarettes as harm reduction). This affects the practical messaging more than the underlying evidence.
• Patients. Strong cultural identity component in some smoking communities; cessation is sometimes resisted as identity loss, not just chemical withdrawal.

Population variability

• Sex. Women appear at slightly higher proportional risk for some smoking-related outcomes (lung cancer, especially adenocarcinoma) at equivalent exposure, possibly due to differences in carcinogen metabolism and lung architecture USSG 2014. Otherwise the hazard profile is largely symmetric.
• Age at start. Earlier initiation (teenage) produces greater lifetime hazard; nicotine dependence consolidates faster in adolescent brains. Three-quarters of US adult smokers started before age 18.
• Age at cessation. Sharply graded benefit by age at quit (Doll 2004, Jha 2013): quitting before 40 recovers ~90% of excess mortality risk; quitting at 60 still recovers ~30%.
• Genetic susceptibility. α1-antitrypsin deficiency dramatically accelerates COPD onset in smokers; some CYP1A1/GSTM1 variants modulate lung cancer susceptibility. Clinically these are minor compared to the smoking signal itself.
• Pregnancy. Special-population because effects are on both mother and fetus, and the timescale matters (any smoking during pregnancy carries elevated risk; quitting at any gestational age helps).
• Mental illness. Smoking prevalence runs 2–3× higher in schizophrenia, bipolar disorder, and severe depression. Cessation is achievable in these populations but requires more intensive support; varenicline and bupropion both tested as safe (EAGLES trial).
• Pre-existing cardiovascular disease. Effects of cessation are largest in this group — secondary-prevention cessation produces ~36% relative reduction in mortality post-MI, a larger effect than most cardiac pharmacotherapy.

Knowledge gaps

Long-term (decade+) outcomes of exclusive nicotine e-cigarette use without prior smoking history remain unestablished — the cohorts are too young. The dose-response curve at very low exposure (1–4 cig/day, intermittent social smoking) is still debated for cardiovascular endpoints; some data suggest near-linearity all the way to zero, others suggest a small floor. Effective cessation strategies for the highest-difficulty subpopulations (severe mental illness, chronic substance co-use) are less developed. The relative oncologic versus cardiovascular hazard of new combustible-tobacco alternatives (heat-not-burn) is still under investigation; current evidence suggests intermediate risk but with substantial uncertainty.

Category placement. Slotted under Breathing & Air because the route of exposure is respiratory and the most proximate organ damage is pulmonary. Medical & Healthcare was a runner-up but reads as too clinically-anchored for a behavioural entry; Lookmaxxing is a real-but-secondary consequence and would understate the longevity signal.

Scope coverage vs. brief. The brief named cardiovascular, lung function, multi-site cancer, skin, fertility, lifespan and post-cessation recovery. The article covers all of these end-to-end; cancer is summarised by site list rather than per-site breakdowns to avoid bloat, but each of the dimension scores reflects the holistic substance effect (per ./entry.md §1a).

Hard scoring calls.

• focus = 1. The literature is mixed: acute nicotine has measurable cognitive effects, chronic smoking has hypoxia-mediated decrement, and quit-attempt withdrawal disrupts focus. Net cessation benefit is real but small. Held at 1 rather than 0 to keep the dimension honest about a real but minor signal.
• cost_burden = 1 rather than 0. Quitting saves money in net terms, but the action itself (pharmacotherapy course) carries a few hundred dollars of cost when not covered. A reader without insurance coverage should see this as a real number.
• beauty_cumulative = 4, not 5. The "smoker's face" data is strong but the trajectory is comparable to chronic sun exposure rather than transforming the underlying aging clock. Reserved 5 for interventions with a more singular aesthetic profile.
• longevity = 5, evidence = 5. Both at ceiling and earned — this is the canonical maximal-evidence longevity hazard in modern medicine.

Separate-entry candidates.

• Nicotine vaping / e-cigarettes — distinct substance, distinct (and evolving) risk profile, increasingly important harm-reduction route. Currently signposted from out-of-scope; warrants its own entry.
• Secondhand smoke exposure — different population (non-smoker partners and children), different evidence base, deserves its own treatment.
• Smokeless tobacco / snus — different substance, different cancer risk profile (oral > lung), Scandinavian harm-reduction debate is its own conversation.
• Smoking cessation in pregnancy — overlapping but population-specific protocol (behavioural-first, NRT preferred over smoking, no e-cigs).

Future links. Once the above exist, this entry should cross-link to: nicotine vaping, alcohol (the highest-risk relapse co-substance), exercise (cardiovascular co-benefit), and any future entry on smoking cessation pharmacotherapy specifically.

Stakeholder caveat. The UK (NICE, former PHE) and US (CDC, FDA) public-health bodies disagree meaningfully on e-cigarette messaging — UK permissive, US cautious. Article takes the Cochrane evidence-based middle (effective cessation aid; long-term safety not fully established) rather than picking a side.