Smoking, Alcohol, and Drugs: Exactly How Many Years Each One Costs You

Let us start with the uncomfortable truth: every substance you put into your body is either adding minutes to your life or subtracting them. Some substances are so thoroughly studied that we can calculate the cost per unit, per day, per year, right down to the minute.

Others remain frustratingly ambiguous because governments spent decades criminalising research instead of funding it.

This guide covers everything. We are going to walk through the peer-reviewed data on smoking, vaping, alcohol, cannabis, opioids, psychedelics, and caffeine. For each substance, we will give you the exact numbers: how many years it costs (or adds), what the dose-response curve looks like, and what happens when you quit.

No moralising. No scare tactics. Just research, numbers, and the occasional dark joke, because if you cannot laugh about your own mortality, you are probably already dead inside.

1. Smoking: The Clearest Data in All of Medicine

If there is one thing the entire medical establishment agrees on without reservation, without nuance, without a single dissenting voice, it is this: smoking kills you. Slowly, expensively, and with remarkable consistency across every population ever studied.

The data on smoking is so robust, so replicated, so overwhelmingly one-directional that it serves as the gold standard against which all other health risks are measured. When researchers want to communicate how dangerous loneliness is, they say it is "equivalent to smoking 15 cigarettes a day." Smoking is the unit of measurement for dying.

The Doll and Peto Study: 50 Years of Watching Doctors Die

In 1951, Sir Richard Doll and Sir Austin Bradford Hill began tracking the health outcomes of 34,439 male British doctors. They followed them for fifty years. Published in the BMJ in 2004, the results are devastating in their clarity.

Smokers who continued smoking throughout their lives died, on average, 10 years earlier than non-smokers. Not 10 months. Not a vague "increased risk." Ten years. Gone. Evaporated into carbon monoxide and tar.

But the study also delivered hope. Doctors who quit at age 60 gained back about 3 years. Those who quit at 50 gained back 6 years. Those who quit at 40 gained back nearly all 10 years. Those who quit at 30 had life expectancy curves virtually indistinguishable from people who never smoked at all.

The message: your body is remarkably forgiving, but it has a deadline for that forgiveness.

The 11-Minute Rule

A widely cited calculation from the BMJ (Shaw et al., 2000) found that each cigarette costs approximately 11 minutes of life. This is derived by dividing the total life expectancy reduction (roughly 10 years for a pack-a-day smoker over 40+ years) by the total number of cigarettes smoked.

At a pack per day (20 cigarettes), that is 220 minutes per day. Three hours and forty minutes of your remaining life, burned. Every day. For the price of looking cool at a bus stop in 1997.

The maths gets worse at higher consumption. Two packs per day (40 cigarettes) does not simply double the risk in a linear fashion; the dose-response curve steepens at higher consumption levels due to the body's increasingly overwhelmed repair mechanisms.

How Smoking Kills: The Mechanisms

Smoking does not kill you in one way. It kills you in approximately 27 documented ways, which is both impressive and horrifying:

Cardiovascular damage: Carbon monoxide displaces oxygen in your blood. Nicotine constricts blood vessels and raises blood pressure. Oxidative stress damages arterial walls. Result: doubled risk of heart attack, tripled risk of stroke.

Cancer: Tobacco smoke contains 70+ known carcinogens. Lung cancer is the headline (15x risk increase for heavy smokers), but smoking also significantly increases risk of cancers of the mouth, throat, oesophagus, stomach, pancreas, kidney, bladder, and cervix.

Respiratory destruction: Chronic obstructive pulmonary disease (COPD) is almost exclusively a smoker's disease. Progressive, irreversible destruction of lung tissue. You slowly suffocate over decades.

Immune suppression: Smokers have weaker immune responses, heal more slowly, and are more susceptible to infections including pneumonia and tuberculosis.

Secondhand Smoke: Killing People Who Made Better Choices

A meta-analysis of 37 epidemiological studies found that non-smokers living with smokers have a 20-30% increased risk of lung cancer and a 25-30% increased risk of coronary heart disease. The US Surgeon General estimates that secondhand smoke causes approximately 41,000 deaths per year among non-smoking adults in the United States alone.

Living with a pack-a-day smoker is roughly equivalent to actively smoking 1-2 cigarettes per day yourself. You are, in effect, involuntarily enrolled in a slow-motion poisoning programme.

Dose-Response: How Much Matters

The relationship between smoking quantity and mortality is not perfectly linear, but it is consistently dose-dependent:

1-4 cigarettes per day: Still triples cardiovascular mortality risk (Bjartveit & Tverdal, Tobacco Control 2005, Norwegian cohort n=42,722). There is no safe level of smoking.

5-14 cigarettes per day: 5-year life expectancy reduction.

15-24 cigarettes per day (approximately one pack): 8-10 year reduction.

25+ cigarettes per day: 10-12 year reduction, with some studies suggesting up to 14 years for the heaviest smokers.

The Global Death Toll

The WHO estimates that tobacco kills more than 8 million people per year worldwide. Of these, approximately 1.3 million are non-smokers exposed to secondhand smoke. Tobacco is the single largest preventable cause of death in human history, killing more people annually than HIV, malaria, and tuberculosis combined.

To contextualise: 8 million deaths per year is equivalent to the entire population of Switzerland dying every single year, every year, indefinitely. It is the equivalent of forty fully loaded Boeing 747s crashing every day with no survivors. We have normalised an ongoing catastrophe of staggering proportions because it happens slowly, one person at a time, in hospital beds rather than in spectacular fashion.

The economic burden is equally staggering. The CDC estimates that smoking costs the United States over $300 billion annually in direct medical care and lost productivity. Globally, the WHO estimates the economic cost of tobacco at $1.4 trillion per year, approximately 1.8% of global GDP. Tobacco-related illness accounts for 12% of all healthcare spending worldwide.

Genetics and Smoking Risk

Not everyone who smokes gets lung cancer, and not everyone who gets lung cancer smoked. Genetic variation plays a significant role in individual susceptibility. Variants in the CHRNA5-CHRNA3-CHRNB4 gene cluster on chromosome 15 influence both nicotine dependence and lung cancer susceptibility. CYP1A1 and GSTM1 gene variants affect how the body metabolises tobacco carcinogens.

This genetic variation explains why some lifelong smokers live to 90 while some die of lung cancer at 55. But genetics is not destiny: even the most genetically resilient smoker is still taking 10 years off their average life expectancy. They are rolling better dice, but the house still wins.

The critical insight: even "light" or "social" smoking carries substantial mortality risk. The cardiovascular system is exquisitely sensitive to tobacco toxins; even a few cigarettes per week measurably increases heart attack risk.

2. Vaping and E-Cigarettes: What We Know and Don't Know

E-cigarettes arrived on the mass market around 2010-2012. This means we have approximately 14 years of widespread human use data. For context, the cigarette-cancer link took 30+ years of data to establish definitively. We are, in essence, running a population-level experiment in real time.

Here is what we know with reasonable confidence:

The 95% Figure

In 2015, Public Health England (PHE) published a landmark evidence review concluding that e-cigarettes are approximately 95% less harmful than combustible cigarettes. This figure refers to short-term toxicant exposure, not long-term mortality outcomes, because those outcomes literally do not exist yet.

The 95% figure has been both widely cited and widely criticised. The Lancet published an editorial arguing the evidence base was too thin to support such a precise estimate. PHE stood by its analysis, noting it was based on relative harm from known toxicants, not an absolute safety claim.

What E-Cigarettes Eliminate

Combustion is the primary killer in cigarette smoking. Burning tobacco produces carbon monoxide, tar, and thousands of toxic compounds. E-cigarettes eliminate combustion entirely. No tar. No carbon monoxide. No polycyclic aromatic hydrocarbons.

What remains: nicotine (addictive but not carcinogenic at typical doses), propylene glycol, vegetable glycerine, flavourings (some with unknown inhalation profiles), and trace metals from heating coils.

What We Don't Know

The honest answer is: we do not know the 20-year or 30-year mortality outcomes of vaping. We cannot know yet. The product has not existed long enough.

Specific unknowns include: long-term pulmonary effects of propylene glycol inhalation, cardiovascular impact of chronic nicotine delivery without other tobacco toxins, cancer risk from flavouring compounds (some contain diacetyl, linked to bronchiolitis obliterans in factory workers), and the effects of novel metallic nanoparticles from heating elements.

The EVALI Crisis

In 2019-2020, a sudden outbreak of severe lung injuries (EVALI: E-cigarette or Vaping product use Associated Lung Injury) killed 68 people and hospitalised over 2,800 in the United States. Investigation by the CDC identified vitamin E acetate, used as a cutting agent in illicit THC vape cartridges, as the primary cause.

EVALI demonstrated two things: first, that some vaping products can cause acute, severe harm; second, that the harm was traced to a specific adulterant in unregulated products, not to regulated nicotine e-cigarettes. The distinction matters for policy but not for the 68 dead.

The Verdict (For Now)

If you currently smoke: switching to vaping is almost certainly a significant harm reduction. The known toxicant reduction is substantial.

If you do not smoke: starting to vape exposes you to risk with zero benefit. Nicotine addiction is its own burden.

For Death Clock calculations: we estimate vaping at approximately -1 to -2 years of life expectancy impact (compared to -10 for smoking), heavily caveated as speculative pending long-term data.

3. Alcohol: The Great Debate (Now Settled)

For decades, the public health message on alcohol was muddled by an apparent paradox: moderate drinkers seemed to live longer than non-drinkers. This produced the famous "J-curve" where a glass or two of wine per day appeared protective, particularly for cardiovascular health.

The J-curve is now largely considered an artefact of flawed methodology. Here is what happened and what we actually know.

The J-Curve: How We Got Fooled

The J-curve appeared in dozens of observational studies throughout the 1990s and 2000s. Moderate drinkers (1-2 drinks per day) showed lower all-cause mortality than abstainers. Researchers proposed biological mechanisms: alcohol raises HDL cholesterol, reduces clotting factors, and the polyphenols in red wine have antioxidant properties.

The problem: the "abstainer" category in most studies included former heavy drinkers who quit due to health problems, people too sick to drink, and people whose religious or socioeconomic circumstances correlated with other health disadvantages. When you compare moderate drinkers against these groups, of course the drinkers look healthier. They started healthier.

Mendelian Randomisation: The Correction

Mendelian randomisation (MR) studies use genetic variants as natural experiments. Some people carry gene variants that make them drink less (e.g., ALDH2 variants common in East Asian populations). By comparing health outcomes between genetic groups, researchers can estimate causal effects of alcohol without the confounding that plagued observational studies.

Multiple MR studies (Millwood et al., Lancet 2019, n=512,715 Chinese adults; Holmes et al., BMJ 2014; Biddinger et al., JAMA Network Open 2022) have consistently found that the supposed cardiovascular benefit of moderate drinking disappears when confounders are properly controlled. The J-curve flattens into a straight line: less alcohol is always better.

Global Burden of Disease 2018: The Final Word

The Global Burden of Disease study (GBD 2018, published in The Lancet) analysed data from 28 million people across 195 countries. Its conclusion was unambiguous: the level of alcohol consumption that minimises total health risk is zero.

Not one drink per day. Not a glass of wine with dinner. Zero.

This does not mean a single glass of wine will measurably harm you. The risk at very low consumption is tiny. But the optimal dose for health outcomes is nil, and any framing that suggests moderate drinking is "healthy" is, per the best available evidence, wrong.

Dose-Response: How Much Costs How Much

From Lancet meta-analyses and cohort studies:

Light drinking (1-7 drinks per week): Minimal mortality impact. Approximately 0-1 year life expectancy reduction. Slight increase in breast cancer risk for women.

Moderate drinking (7-14 drinks per week): 1-2 years life expectancy reduction. Measurable increases in liver disease, several cancers, and accident risk.

Heavy drinking (14-28 drinks per week): 4-5 years life expectancy reduction. Significant liver, cardiovascular, and neurological damage accumulation.

Very heavy drinking (28+ drinks per week): 5-7+ years reduction. High probability of alcohol-related disease or accident.

Alcohol use disorder (AUD): Scandinavian population registry studies (Westman et al., Drug and Alcohol Dependence 2015) found that people hospitalised with AUD died 24-28 years earlier than the general population. This reflects both direct toxicity and the cascade of associated harms: accidents, violence, self-harm, nutritional deficiency, and medical neglect.

What Alcohol Does to Your Body

Liver: Alcohol is metabolised primarily by the liver. Chronic heavy use progresses through fatty liver (reversible), alcoholic hepatitis (partially reversible), fibrosis (slowly reversible), and cirrhosis (irreversible). Approximately 50% of cirrhosis deaths are alcohol-related.

Brain: Alcohol shrinks brain volume. A 2022 study in Nature Medicine (n=36,678) found that even moderate drinking (1-2 drinks per day) was associated with measurable brain volume reduction equivalent to approximately 2 years of ageing. Heavy drinking accelerates this dramatically.

Cancer: The WHO classifies alcohol as a Group 1 carcinogen (same category as asbestos and tobacco). It directly causes cancers of the mouth, throat, oesophagus, liver, breast, and colon. The mechanism is primarily through acetaldehyde, alcohol's first metabolic breakdown product, which damages DNA.

Cardiovascular: While low-dose alcohol may have mild anticoagulant properties, higher doses directly damage heart muscle (alcoholic cardiomyopathy), raise blood pressure, and increase arrhythmia risk. Binge drinking is particularly dangerous for acute cardiac events.

The "French Paradox" Debunked

The observation that France has relatively low heart disease despite high saturated fat and wine consumption was attributed to protective effects of red wine. Subsequent research has shown this was primarily explained by: Mediterranean dietary patterns independent of wine, delayed reporting of cardiac deaths in French statistics, and the confounding we discussed above.

Resveratrol, the supposedly magical compound in red wine, would require drinking approximately 100-1,000 bottles per day to achieve the doses shown to be active in laboratory studies. You would die of alcohol poisoning roughly 99 bottles before reaching a therapeutic dose.

Alcohol and Women: A Different Equation

Women metabolise alcohol differently from men due to lower levels of alcohol dehydrogenase (the primary enzyme that breaks down alcohol), higher body fat percentage (alcohol is water-soluble and distributes less in fat tissue), and hormonal interactions. The result: women reach higher blood alcohol concentrations from the same quantity of alcohol and sustain damage at lower consumption levels.

Breast cancer risk is particularly notable. A meta-analysis in the British Journal of Cancer (Chen et al., 2011) found that even one drink per day increases breast cancer risk by approximately 7-10%. At 3+ drinks per day, the risk increase exceeds 50%. Alcohol is estimated to cause approximately 6% of all breast cancers globally.

This is rarely communicated clearly to women. The same glass of wine that might carry minimal risk for a 90kg man carries measurably more risk for a 60kg woman, and the cancer risk specifically adds a dimension not present in the male data.

Alcohol and Sleep

One of the most insidious effects of alcohol is its destruction of sleep quality. While alcohol is a sedative that helps people fall asleep faster, it severely disrupts sleep architecture: it suppresses REM sleep (the stage critical for memory consolidation and emotional processing), fragments the second half of the night, increases sleep apnoea severity, and triggers early morning awakenings.

Matthew Walker's research at UC Berkeley has documented that even moderate alcohol consumption (2 drinks) reduces REM sleep by 20-30%. Since poor sleep independently increases all-cause mortality by 12-15%, alcohol's sleep-disrupting effect creates a compounding mortality pathway on top of its direct organ damage.

Many people use alcohol as a sleep aid without realising it is making their sleep measurably worse. It is the equivalent of treating a headache by hitting yourself with a smaller hammer.

The Complete Substance Impact Table

Note: Estimates marked "est." have wide confidence intervals due to limited long-term data. All figures represent averages; individual outcomes vary based on genetics, dose, duration, and concurrent factors. Use the Death Clock calculator to see your personal impact.

4. Cannabis: The Evidence Gap

Cannabis research has been systematically undermined by decades of Schedule I classification in the United States and equivalent restrictions globally. This classification made large-scale, long-term epidemiological studies nearly impossible to conduct legally. As a result, our understanding of cannabis and mortality is far less precise than for tobacco or alcohol.

Here is what we can say with confidence, and what remains genuinely unknown:

What the Limited Data Shows

A 2016 systematic review in the International Journal of Drug Policy examined cannabis and all-cause mortality across available cohort studies. The findings were mixed: some studies found small increases in mortality among heavy cannabis users, while others found no significant association after controlling for concurrent tobacco use.

The fundamental methodological challenge: most long-term cannabis users in existing studies also smoke tobacco, drink alcohol, or both. Isolating the independent effect of cannabis on mortality is extremely difficult with observational data.

Cardiovascular Concerns

A 2022 study in the Canadian Medical Association Journal (n=37,000+) found that cannabis use was associated with increased risk of heart attack and stroke, particularly in younger users. THC acutely increases heart rate and can trigger arrhythmias in vulnerable individuals.

Case reports document myocardial infarctions in otherwise healthy young cannabis users, though the absolute risk remains low. The mechanism appears to involve both direct cardiovascular effects of THC and carbon monoxide exposure from smoking.

Respiratory Effects

Smoking cannabis produces many of the same combustion byproducts as tobacco: tar, carbon monoxide, and particulate matter. Cannabis smokers tend to inhale more deeply and hold smoke longer, increasing lung deposition.

However, cannabis smoke does not appear to cause the same pattern of COPD or lung cancer seen with tobacco, possibly because the total lifetime exposure is typically much lower (a heavy cannabis smoker might use 3-5 joints per day versus 20-40 cigarettes for a heavy tobacco smoker). A large cohort study (Pletcher et al., JAMA 2012, n=5,115) found no significant association between cannabis use and impaired lung function at typical use levels.

Mental Health and Indirect Mortality

The strongest evidence for cannabis harm relates to mental health: cannabis use during adolescence is associated with increased risk of psychotic disorders, particularly in individuals with genetic predisposition (Di Forti et al., Lancet Psychiatry 2019). Psychotic disorders carry significant excess mortality.

Heavy cannabis use is also associated with motivational deficits, reduced economic attainment, and impaired driving, all of which can indirectly affect life expectancy through socioeconomic and accident pathways.

Our Best Estimate

For Death Clock calculations, we estimate cannabis impact as: light/occasional use (-0 to -0.5 years), regular daily use (-1 to -2 years, primarily from respiratory and cardiovascular effects), heavy daily use (-2 to -3 years, including mental health and accident risk).

These estimates carry wide confidence intervals. Better data will emerge as legalisation enables proper longitudinal research.

5. Opioids: The Modern Plague

The opioid crisis is the deadliest drug epidemic in recorded history. It has killed over 600,000 Americans since 1999, reduced US life expectancy at the national level for three consecutive years (2015-2017), and continues to kill at accelerating rates due to the infiltration of illicit fentanyl.

The Numbers

In 2022, approximately 81,806 Americans died from opioid overdoses (CDC WONDER database). That is more than car accidents and gun deaths combined. It is equivalent to a fully loaded Boeing 737 crashing every single day with no survivors.

Globally, the WHO estimates 115,000 opioid overdose deaths per year, though this likely underestimates true mortality due to poor reporting in many countries.

Life Expectancy Impact

For individuals with active opioid use disorder (OUD), life expectancy reduction is catastrophic. Studies of heroin users in the pre-fentanyl era found standardised mortality ratios of 6-20 (meaning death rates 6-20 times higher than age-matched peers). Translated to life expectancy: untreated OUD reduces lifespan by approximately 15-25 years.

With fentanyl contamination of the current drug supply, mortality risk per use episode has increased dramatically. Fentanyl is 50-100 times more potent than morphine, and dosing inconsistency in illicit preparations means overdose can occur even among experienced users.

The Three Waves

Wave 1 (1990s-2010): Prescription opioid overprescribing. OxyContin marketing. Approximately 30% of chronic pain patients prescribed opioids developed OUD.

Wave 2 (2010-2013): Heroin resurgence as prescription access tightened. Cheaper, stronger, and increasingly available as users transitioned from pills.

Wave 3 (2013-present): Synthetic opioids, primarily illicit fentanyl and its analogues. Fentanyl now contaminates cocaine, methamphetamine, and counterfeit pills, killing people who never intended to use opioids.

Treatment and Recovery

Medication-assisted treatment (MAT) with buprenorphine or methadone reduces all-cause mortality in OUD by approximately 50% (Sordo et al., BMJ 2017, systematic review). This makes MAT one of the most effective life-saving medical interventions available, comparable to antiretrovirals for HIV.

Yet only approximately 20% of people with OUD in the US receive any form of evidence-based treatment, due to stigma, access barriers, and policy failures.

The Age Factor

Age of first opioid use dramatically affects outcomes. Adolescents who begin using opioids before age 18 have significantly worse trajectories than those who begin in adulthood: faster progression to dependence, higher rates of concurrent psychiatric disorders, and lower treatment completion rates. The developing brain is exquisitely vulnerable to opioid-induced neuroplastic changes that entrench addictive patterns.

Conversely, older adults face unique opioid risks. Slower metabolism means standard doses produce higher blood levels. Polypharmacy (multiple medications) creates dangerous interactions, particularly with benzodiazepines. Falls while impaired carry catastrophic consequences for elderly patients with osteoporosis. The combination of an ageing population and legacy prescribing practices has created a largely invisible opioid crisis among the over-65 population.

Methamphetamine: The Other Crisis

While media attention focuses on opioids, methamphetamine-related deaths have quadrupled since 2015 in the United States. Methamphetamine causes severe cardiovascular toxicity: cardiomyopathy, pulmonary hypertension, aortic dissection, and stroke. Unlike opioids, there is no reversal agent (no "Narcan for meth") and no FDA-approved medication-assisted treatment.

Chronic methamphetamine use is associated with approximately 15-20 years of life expectancy reduction, driven by cardiovascular catastrophe, severe dental disease (reducing nutrition and increasing systemic infection risk), neurotoxicity, and the profoundly destabilised lifestyle that accompanies severe stimulant addiction.

The intersection of the two crises is particularly deadly: fentanyl-contaminated methamphetamine now kills users who have no opioid tolerance whatsoever, adding overdose risk to an already lethal substance.

Cocaine: The Cardiovascular Gamble

Cocaine's primary life-shortening mechanism is cardiovascular: it causes coronary artery spasm, accelerated atherosclerosis, aortic dissection, and cardiomyopathy. Unlike most substances where risk accumulates gradually over decades, cocaine can kill on any given use through acute myocardial infarction or arrhythmia, even in young, otherwise healthy users.

A study in Circulation (Qureshi et al., 2001) found that cocaine use increases stroke risk by a factor of 6 in the 24 hours after use. Regular cocaine users in their 30s and 40s show coronary artery calcification scores typical of non-users in their 60s and 70s. The drug effectively ages your cardiovascular system by 20-30 years.

Life expectancy impact estimates for regular cocaine use range from -5 to -10 years, with high individual variability depending on frequency, route of administration (smoking crack cocaine is more harmful than insufflation), and concurrent substance use. The addition of fentanyl to the cocaine supply has dramatically increased acute mortality risk in recent years.

6. Psychedelics: Rewriting the Harm Rankings

In 2010, David Nutt published a landmark paper in The Lancet that ranked 20 drugs by overall harm (to self and others). The results challenged decades of drug scheduling:

Alcohol ranked as the most harmful drug overall (when considering harm to both the user and others). Heroin and crack cocaine ranked highest for harm to self. Mushrooms and LSD ranked among the least harmful substances studied, below cannabis, tobacco, and alcohol by significant margins.

Safety Profile

Classic psychedelics (psilocybin, LSD, DMT, mescaline) have remarkably low physiological toxicity. There are no confirmed cases of fatal overdose from psilocybin or LSD alone in the medical literature. The LD50 (dose required to kill 50% of test animals) for psilocybin is estimated at approximately 280mg/kg, a dose roughly 1,000 times a typical recreational dose.

Deaths associated with psychedelic use are almost exclusively from behavioural impairment (falls, drowning, traffic) or pre-existing cardiac conditions exacerbated by acute sympathetic activation. The drugs themselves do not kill in the way that opioids, alcohol, or even aspirin can.

Therapeutic Potential

FDA-designated breakthrough therapy status has been granted for psilocybin (treatment-resistant depression, COMPASS Pathways and Usona Institute) and MDMA (PTSD, MAPS/Lykos Therapeutics). Phase III trials show significant efficacy.

If these therapies successfully treat depression and PTSD, the indirect mortality benefit could be substantial, given that both conditions significantly increase all-cause mortality (depression: 40-60% excess mortality; PTSD: 50-100% excess mortality for combat veterans).

MDMA: The Nuanced Case

MDMA (3,4-methylenedioxymethamphetamine) occupies a unique position. Unlike classic psychedelics, it has documented neurotoxicity at high or repeated doses: serotonergic neuron damage in animal models and measurable serotonin transporter reductions in heavy human users (de Win et al., Archives of General Psychiatry 2008). Hyperthermia (overheating) is the primary acute risk, particularly in hot environments like nightclubs where dehydration and physical exertion compound the drug's thermogenic effects.

Deaths from MDMA, while rare per dose (estimated at 1 per 50,000-100,000 uses), do occur. They are primarily caused by hyperthermia, hyponatraemia (water intoxication from excessive fluid intake as a misguided precaution), and serotonin syndrome when combined with other serotonergic drugs.

However, in clinical settings with controlled doses, medical monitoring, and therapeutic protocols, MDMA has shown remarkable efficacy for PTSD treatment. The Phase III MAPS trials showed 67% of participants no longer met diagnostic criteria for PTSD after three MDMA-assisted therapy sessions, compared to 32% with therapy alone. If approved, MDMA therapy could prevent the substantial excess mortality associated with untreated PTSD.

The harm profile of MDMA is thus highly context-dependent: recreational use in uncontrolled settings with unknown doses and adulterants carries meaningful risk, while clinical use under medical supervision appears to have a favourable risk-benefit ratio for severe PTSD.

Our Estimate

For Death Clock purposes: occasional psychedelic use has no measurable negative impact on life expectancy based on available data. The primary risk is acute behavioural impairment and extremely rare psychiatric emergencies in vulnerable individuals. We assign 0 to -0.5 years for occasional use in otherwise healthy adults.

7. Caffeine: The Drug That Extends Life

In a world of substances that shorten your life, caffeine stands as the glorious exception. Coffee, the delivery mechanism of choice for the world's most popular psychoactive drug, is associated with longer life in virtually every large-scale study ever conducted.

The NEJM Mega-Study

In 2012, Freedman et al. published results from the NIH-AARP Diet and Health Study in the New England Journal of Medicine. The cohort: 402,260 participants followed for 12-13 years. The findings: compared to non-drinkers, coffee consumers had 10-15% lower all-cause mortality, with the greatest benefit at 4-5 cups per day.

The inverse association held across causes of death: heart disease, respiratory disease, stroke, injuries, accidents, diabetes, and infections all showed reduced risk with coffee consumption. The only exception was a possible slight increase in lung cancer, likely confounded by the historical association between coffee drinking and smoking.

Mechanisms

Coffee's benefits likely come not from caffeine itself but from the 1,000+ bioactive compounds in the beverage:

Chlorogenic acids: Powerful antioxidants that reduce oxidative stress and inflammation.

Diterpenes (cafestol and kahweol): Anti-cancer and anti-inflammatory properties (though they also raise cholesterol slightly).

Trigonelline: Neuroprotective compound that may reduce Alzheimer's risk.

Melanoidins: Formed during roasting, these have prebiotic properties supporting gut health.

Evidence that caffeine itself is not the sole driver: decaffeinated coffee also shows mortality benefit in most studies, though the effect is typically smaller.

The Dose-Response Curve

From pooled analyses of multiple cohorts:

1-2 cups/day: 8-12% lower all-cause mortality.

3-5 cups/day: 12-15% lower all-cause mortality (the sweet spot in most studies).

6+ cups/day: Benefits plateau or slightly attenuate. No increase in mortality, but no additional benefit.

Very heavy consumption (8+ cups): Limited data, but generally not harmful for most people. Individual tolerance varies based on CYP1A2 genetic variants (fast vs slow caffeine metabolisers).

Coffee and Specific Diseases

Type 2 diabetes: 25-30% lower risk with 3-4 cups/day (Ding et al., Diabetes Care 2014, meta-analysis of 28 prospective studies, n=1,109,272).

Parkinson's disease: 25% lower risk in men, dose-dependent (Ross et al., JAMA 2000; Hernon et al., Annals of Neurology 2002).

Liver disease: 40-50% lower risk of cirrhosis with 2+ cups/day (Kennedy et al., Alimentary Pharmacology & Therapeutics 2016).

Depression: 20% lower risk with 4+ cups/day (Lucas et al., Archives of Internal Medicine 2011, n=50,739 women).

Alzheimer's and dementia: 27% lower risk with moderate consumption (Santos et al., Journal of Alzheimer's Disease 2010).

Our Estimate

For Death Clock calculations: regular moderate coffee consumption (3-5 cups/day) is associated with approximately +0.5 to +1.5 years of life expectancy benefit. We include this as a positive factor in our algorithm because the evidence is overwhelming, consistent, and biologically plausible.

If you are reading this with a coffee in hand: congratulations. You are currently extending your life. Every sip is a tiny rebellion against death. Keep going.

8. Quitting Timelines: How Fast You Recover

One of the most powerful motivators for behaviour change is knowing exactly how quickly your body repairs itself. The human body is a remarkably resilient machine. Stop poisoning it and it begins fixing itself almost immediately.

Smoking Cessation Timeline

20 minutes: Heart rate and blood pressure drop to normal levels.

12 hours: Carbon monoxide level in blood drops to normal. Oxygen levels normalise.

2-12 weeks: Circulation improves. Lung function increases by up to 30%.

1-9 months: Coughing and shortness of breath decrease. Cilia in lungs begin regrowing and resume clearing mucus.

1 year: Risk of coronary heart disease drops by 50% compared to continuing smoking.

5 years: Risk of stroke equals that of a non-smoker. Risk of cancers of the mouth, throat, and oesophagus drops by 50%.

10 years: Lung cancer death risk drops to approximately half that of a continuing smoker. Risk of pancreatic and laryngeal cancer decreases.

15 years: Risk of coronary heart disease equals that of a non-smoker. All-cause mortality risk approaches (but may not fully reach) non-smoker levels.

Source: US Surgeon General's Reports; American Cancer Society; NHS Smokefree data.

Alcohol Cessation Timeline

1-7 days: Withdrawal symptoms peak (can be medically dangerous for heavy drinkers; delirium tremens requires medical supervision). Sleep begins to improve toward the end of the first week.

2-4 weeks: Liver fat begins clearing. Blood pressure starts normalising. Skin improves. Mental clarity increases.

1-3 months: Liver enzyme levels normalise if no permanent damage. Brain volume begins recovering (Gazdzinski et al., Alcoholism: Clinical and Experimental Research 2010).

6-12 months: Liver fibrosis (not cirrhosis) can substantially reverse. Cardiovascular risk markers continue improving.

1-2 years: Cardiovascular risk drops significantly. Brain white matter integrity improves substantially.

5+ years: Cancer risk from alcohol begins declining but may take decades to fully normalise, as epigenetic changes from alcohol persist.

Cannabis Cessation

1-3 days: THC withdrawal peaks. Irritability, insomnia, decreased appetite, and sometimes vivid dreams. Withdrawal is real but not medically dangerous (unlike alcohol or benzodiazepine withdrawal).

1-2 weeks: Sleep quality begins improving. Appetite normalises. Mood stabilises.

1-3 months: Cognitive function, particularly short-term memory and attention, measurably improves (Schuster et al., Journal of Clinical Psychiatry 2018). Motivation levels increase.

3-6 months: Respiratory symptoms (chronic cough, phlegm) clear if smoked cannabis was the source. Lung function improves.

1 year+: Cardiovascular risk markers normalise. Any depressive or anxious symptoms that were cannabis-maintained begin resolving.

Opioid Recovery Timeline

With MAT (buprenorphine/methadone): Overdose risk drops immediately. Mortality drops 50% compared to untreated OUD. Neurocognitive function begins recovering within months. Social functioning, employment, and relationship stability improve over 1-2 years.

Without MAT: Acute withdrawal lasts 5-10 days (heroin) or 2-4 weeks (methadone). Post-acute withdrawal syndrome (PAWS) can persist for months: insomnia, anxiety, dysphoria, anhedonia. Relapse rates without MAT are approximately 80-90% within the first year, and each relapse carries overdose risk due to lost tolerance.

Critical danger period: The highest overdose risk occurs in the first 2 weeks after release from incarceration or completion of detox, because tolerance drops rapidly during abstinence but craving does not. A dose that was routine before abstinence can now be lethal.

The Key Insight

Recovery is not linear and it is not complete for all substances. Tobacco: near-complete recovery if you quit young enough. Alcohol: substantial but not complete recovery, depending on the extent of organ damage already sustained. Opioids: immediate cessation of overdose risk with MAT, but long-term recovery of neurocognitive function takes years.

The best time to quit was yesterday. The second best time is today. The worst time is tomorrow, because tomorrow's version of you will say the same thing.

9. Harm Reduction: What Actually Works

If you are not ready to quit entirely, harm reduction strategies can significantly reduce your mortality risk while you work toward cessation. The perfect should not be the enemy of the good.

Smoking Harm Reduction

Nicotine Replacement Therapy (NRT): Patches, gums, lozenges, inhalers. Meta-analyses show NRT increases quit rates by 50-60% compared to placebo (Hartmann-Boyce et al., Cochrane Review 2018). Combining two forms (e.g., patch + gum) is more effective than either alone.

Varenicline (Champix/Chantix): The most effective single pharmacotherapy for smoking cessation. Approximately doubles quit rates compared to placebo and is superior to NRT alone (Cahill et al., Cochrane Review 2013). Partial nicotinic agonist that reduces cravings and withdrawal while blocking the rewarding effects of nicotine.

Bupropion (Zyban/Wellbutrin): Antidepressant with demonstrated smoking cessation efficacy. Particularly useful for smokers with concurrent depression.

Switch to vaping: As discussed, approximately 95% harm reduction. Not zero risk, but dramatically less harmful than continuing to smoke combustible cigarettes.

Reduce quantity: Every cigarette not smoked is 11 minutes saved. Cutting from 20 per day to 10 per day halves your daily life-loss rate. Not as good as quitting, but better than nothing.

Alcohol Harm Reduction

Naltrexone: Opioid antagonist that reduces the pleasurable effects of alcohol. Reduces heavy drinking days by approximately 25% (Jonas et al., JAMA 2014). Can be used as daily oral or monthly injection (Vivitrol).

Acamprosate (Campral): Modulates glutamate signalling to reduce cravings. Most effective for maintaining abstinence after initial detoxification.

The Sinclair Method: Taking naltrexone one hour before drinking (not daily). Clinical trials show approximately 78% reduction in drinking over 3-6 months. Based on pharmacological extinction of the learned alcohol-reward association.

Drink-free days: The UK Chief Medical Officers recommend at least 2-3 alcohol-free days per week. This gives the liver recovery time and breaks habitual patterns.

Lower-strength alternatives: Switching from spirits to beer, or from regular beer to low-alcohol alternatives (0.5%), progressively reduces exposure while maintaining social rituals.

Opioid Harm Reduction

Naloxone (Narcan): Opioid overdose reversal drug. Available over-the-counter in many jurisdictions. If someone in your life uses opioids, having naloxone available is the single most impactful harm reduction measure. It reverses overdose within minutes.

Fentanyl test strips: Allow users to test substances for fentanyl contamination before use. Reduces unintentional fentanyl exposure.

Supervised consumption sites: Where legal, these facilities provide sterile equipment, trained observers, and immediate overdose intervention. Studies in Vancouver (Insite) show approximately 35% reduction in overdose deaths in surrounding areas.

MAT: Buprenorphine and methadone are the gold standard. They are not "replacing one addiction with another." They are evidence-based medical treatment that cuts mortality risk by 50%.

The Psychology of Quitting: Why Willpower Fails

Understanding why people fail to quit substances is as important as understanding the harm those substances cause. The dominant cultural narrative, that quitting is simply a matter of willpower and personal choice, is contradicted by decades of neuroscience research.

Addiction involves fundamental changes in brain reward circuitry, particularly the mesolimbic dopamine pathway. Repeated substance use downregulates dopamine receptors (reducing the capacity to experience pleasure from normal activities) while simultaneously strengthening the associative memories linking the substance to relief. The result: the addicted brain experiences abstinence not as neutral but as actively painful, while remembering the substance as the only reliable source of relief.

This is why shame-based approaches to cessation fail. Telling someone with alcohol use disorder to "just stop drinking" is neurologically equivalent to telling someone with clinical depression to "just cheer up." The intent may be sincere but the instruction is biologically illiterate.

Effective cessation strategies work with neuroscience, not against it. Pharmacotherapies like varenicline (smoking), naltrexone (alcohol), and buprenorphine (opioids) directly address the neurochemical imbalances that make quitting so difficult. Cognitive behavioural therapy helps restructure the associative memories that drive cravings. Contingency management (providing tangible rewards for abstinence) leverages the same dopamine system that addiction hijacked.

The most effective approaches combine pharmacotherapy with behavioural support. For smoking cessation, combining varenicline with brief counselling roughly triples quit rates compared to willpower alone. For alcohol, combining naltrexone with cognitive behavioural therapy approximately doubles the rate of sustained reduction.

If you have tried to quit a substance and failed, that failure is not a character defect. It is a predictable outcome of insufficient pharmacological support. Try again with medication. The evidence says your odds improve dramatically.

Social Environment: The Underestimated Factor

Your substance use is profoundly shaped by your social environment. Research consistently shows that people who drink, smoke, or use drugs tend to cluster socially, and that these social networks both initiate and maintain substance use patterns.

The Framingham Heart Study's social network analysis (Christakis & Fowler, NEJM 2008) found that a person's chance of smoking increased by 36% if a friend smoked, 25% if a sibling smoked, and 8% if a neighbour smoked. Similar clustering effects have been demonstrated for alcohol consumption and obesity.

The implication is stark: if you want to change your substance use, you may need to change your social environment. This is brutally difficult advice, particularly for people whose social lives revolve around drinking, but the data is clear. People who successfully quit substances almost always restructure their social connections in the process, whether through mutual aid groups (AA, NA, SMART Recovery), new hobby communities, or intentional friendship shifts.

This is not about judging your friends. It is about acknowledging that humans are social animals who unconsciously calibrate their behaviour to match their peer group. If your peer group's baseline is 4 pints on a Tuesday, your willpower is fighting against the most powerful force in human psychology: the desire to belong.

10. Frequently Asked Questions

The Bottom Line: Your Substance Audit

Let us be blunt about what we have learned across 35+ peer-reviewed studies and millions of research participants.

If you smoke cigarettes, quitting is the single most impactful health decision you can make. Nothing else comes close. Not diet, not exercise, not meditation, not supplements. Quitting smoking adds more years to your life than any other single intervention studied in the history of medicine. And if you quit before 40, you recover nearly all of the lost time. Your body wants to forgive you. Let it.

If you drink alcohol, less is always better. The comfortable myth that a glass of wine with dinner is "good for your heart" has been systematically dismantled by Mendelian randomisation studies. This does not mean you must become a teetotaller. It means you should be honest with yourself about what you are choosing. You are not drinking for your health. You are drinking despite your health, and that is a different thing entirely.

If you use opioids recreationally or have developed dependence, please know that medication-assisted treatment cuts your mortality risk in half. It is not weakness to seek treatment. It is not "replacing one addiction with another." It is using evidence-based medicine to keep yourself alive long enough to build the life you want. If you know someone with opioid use disorder, having naloxone available is the single most impactful thing you can do. It costs less than a dinner out and it saves lives.

If you drink coffee, keep drinking coffee. You are, against all intuition, consuming one of the most well-documented longevity-promoting substances on Earth. Every cup is a small act of defiance against entropy. Enjoy it.

The Death Clock algorithm incorporates all of these substance impacts into your personalised life expectancy calculation. It uses the same peer-reviewed data cited throughout this article: the Doll and Peto figures for smoking, the GBD 2018 data for alcohol, the Freedman et al. NEJM study for coffee, and our best estimates (clearly marked as such) for substances with less robust long-term data.

Your death date is not fixed. It is a moving target. Every cigarette not smoked, every drink not poured, every day of recovery from harder substances, shifts that date further into the future. The maths is relentless and it works in both directions.

The clock is ticking. But how fast it ticks is, to a remarkable degree, up to you.