The smoking benchmark

Before we rank anything against smoking, we need a clean number for what smoking costs. The cleanest estimate comes from Doll, Peto, Boreham, and Sutherland's 50-year follow-up of British doctors, published in BMJ in 2004 [1]. Across 34,439 male physicians followed from 1951 to 2001, lifelong smokers lost on average about 10 years of life expectancy compared to lifelong never-smokers. Half of all lifelong smokers were killed by their habit; a third died before age 70. Quitting at 50 recovered roughly six years; quitting at 30 recovered almost the full ten.

That 10-year figure is the benchmark every other risk factor in this article gets measured against. The ranking that follows uses two evidence inputs: pooled hazard ratios for all-cause mortality from large prospective cohorts (which tell you the strength of the association), and life-table modelling estimates (which translate that hazard into years lost). Both have honest uncertainty bands, and we have flagged them where the data does.

One framing note before we start. "Worse than smoking" is a hazard-ratio comparison, not a moral one. Smoking is uniquely addictive, uniquely marketed, and uniquely targeted at children. The point of this article is not to let smoking off the hook - it is to point out that several behaviours we treat as normal-life-stuff carry hazard burdens in the same ballpark, and they are quietly doing more population-level damage than the headlines suggest.

1. Low cardiorespiratory fitness - up to 12 years

The single most striking finding in modern preventive cardiology comes from Mandsager et al., JAMA Network Open, 2018 [2], a 122,007-patient observational study from the Cleveland Clinic. They sorted patients into fitness quintiles based on a maximal treadmill test, then watched what happened over a decade of follow-up.

The result that made the journals: the adjusted hazard ratio for all-cause mortality in the lowest fitness group versus the elite fitness group was 5.04. For reference, the smoking hazard ratio in the same model was 1.41. End-stage renal disease was 3.1. The authors quote-of-record: "Cardiorespiratory fitness is inversely associated with long-term mortality with no observed upper limit of benefit. Extremely high aerobic fitness was associated with the greatest survival." In life-expectancy terms, the gap between the lowest and highest fitness quintile in this cohort was modelled at 11–12 years by subsequent analyses - directly comparable to, and possibly exceeding, the smoking gap.

This effect is dose-dependent and modifiable. Moving from "low" to even "below-average" cuts mortality by roughly 50%. Most untrained 40-somethings can make that move with 12–16 weeks of structured aerobic training. We wrote the longer version on why VO₂ max predicts mortality so strongly, and on how to actually move it in your forties.

2. Untreated stage-2 hypertension - 5 to 7 years

Hypertension is the single largest contributor to global mortality, accounting for an estimated 10.8 million deaths per year per the Global Burden of Disease analysis [3]. The Prospective Studies Collaboration's pooled analysis of one million adults across 61 cohorts (Lewington et al., Lancet, 2002 [4]) found that mortality from ischemic heart disease and stroke doubles for every 20-mmHg increase in systolic blood pressure above 115 mmHg, across the full range from 115 to 180 mmHg, in every age band studied.

In years-of-life-expectancy terms: untreated stage-2 hypertension (systolic ≥ 160 mmHg) carries an estimated 5–7 year cost compared to optimally controlled blood pressure, modelled across the major Western cohorts. The reason it does not get more attention is that it is silent until it kills you - there is no cough, no inhaler, no visible smoke. The good news is that the cost is almost entirely reversible. Trials show that lowering systolic blood pressure by 10 mmHg cuts major cardiovascular events by roughly 20% [5].

The practical move: if you don't know your blood pressure to within 5 mmHg, you don't know whether this hazard applies to you. Measure it. Twice. If it's high, treat it.

3. Chronic loneliness - equivalent to ~15 cigarettes a day

This one surprises people. Holt-Lunstad et al., Perspectives on Psychological Science, 2015 [6] meta-analysed 70 prospective studies covering 3.4 million participants and found that loneliness, social isolation, and living alone each carried significant mortality hazards. The effect sizes (odds ratios of 1.26, 1.29, and 1.32 respectively) are comparable to the well-established mortality risk of smoking up to 15 cigarettes a day, and larger than the mortality risk of obesity.

In life-expectancy terms, the modelled cost of chronic, severe social isolation is in the 3–8 year range depending on the cohort, with women slightly less affected than men. The mechanism is not mysterious - chronic loneliness elevates systemic inflammation, raises cortisol, disrupts sleep, suppresses immune function, and quietly compounds every other risk factor in this list.

It is the hardest of the seven to dose. You cannot prescribe a friend. But the data is clear that the number of close, trusted relationships you actively maintain in mid-life is one of the single strongest predictors of how long you stay healthy. The Harvard Study of Adult Development, now in its ninth decade, finds that the quality of relationships at age 50 predicts physical health at age 80 better than cholesterol does.

4. Sedentary behaviour (sitting > 8 hours/day) - 3 to 5 years

"Sitting is the new smoking" is too tidy a slogan, but the underlying signal is real and independent of exercise. The most important paper here is Ekelund et al., Lancet, 2016 [7], a harmonised meta-analysis of more than one million adults. The key finding: sitting more than 8 hours a day carried a meaningful all-cause mortality hazard even after adjusting for total physical activity, and that hazard was only fully offset by 60–75 minutes per day of moderate-intensity activity.

In years-of-life-expectancy terms, the modelled gap between high-sitting (> 10 h/day) and low-sitting (< 4 h/day) sedentary behaviour is about 3–5 years across the major Western cohorts. The mechanism is metabolic: prolonged inactivity dramatically suppresses lipoprotein lipase activity in skeletal muscle, leading to worse lipid profiles, worse glycaemic control, and accelerated cardiovascular aging.

The practical interpretation: a 30-minute morning run does not buy you the right to sit for the other 15 hours. Frequent low-grade movement - standing, walking phone calls, walking after meals - has independent value beyond the exercise block.

5. Short sleep (chronic < 6 hours/night) - 3 to 5 years

Sleep duration has a U-shaped mortality curve, with the floor sitting around seven hours per night. The largest pooled analysis is Cappuccio et al., Sleep, 2010 [8], which combined 16 prospective studies covering 1.38 million participants. Short sleep (≤ 6 hours) carried a 12% increased mortality risk versus 7–8 hours; long sleep (≥ 9 hours) carried a 30% increase, though that signal is partly reverse causation (sick people sleep more).

More recent work using objective wearable data (Yin et al., Journal of the American Heart Association, 2017 [9]) found cleaner dose-response curves and broadly the same finding: chronically sleeping less than six hours is independently associated with cardiovascular, metabolic, cognitive, and immune harm. The modelled life-expectancy cost of habitual short sleep is in the 3–5 year range depending on the cohort.

The signal is not noise - short sleep raises blood pressure, worsens insulin sensitivity, increases inflammatory cytokines, accelerates beta-amyloid accumulation (the protein implicated in Alzheimer's), and depresses immune function. Sleep is doing real biological work. The cost of skipping it compounds.

6. Type 2 diabetes - 6 to 10 years

The Emerging Risk Factors Collaboration, NEJM, 2011 [10] pooled 102 prospective studies covering 698,782 adults. Type 2 diabetes at age 50 was associated with a ~6-year reduction in life expectancy; the cost rose to 8–10 years for diabetes diagnosed earlier in life. The mechanism is cumulative microvascular and macrovascular damage: diabetes accelerates atherosclerosis, doubles cardiovascular mortality, and raises mortality from a long list of cancers, infections, and kidney disease.

The point worth emphasising for a longevity audience: by the time you have a diabetes diagnosis, you have been pre-diabetic for years. The transition is gradual, and the upstream signal - fasting glucose creeping up, HbA1c trending up, central adiposity increasing - is visible long before any clinical threshold is crossed. The Diabetes Prevention Program randomised trial [11] showed that structured lifestyle intervention (modest weight loss, increased activity) cut the conversion rate from pre-diabetes to diabetes by 58% - better than the metformin arm.

If you are tracking biological age, fasting glucose and HbA1c are two of the most actionable single-number signals you can pay attention to. They sit upstream of most of the other risks on this list.

7. Heavy alcohol use (> 35 units/week) - 4 to 7 years

The Global Burden of Disease alcohol analysis (GBD 2016 Alcohol Collaborators, Lancet, 2018 [12]) put the field on a firmer footing: alcohol is a leading risk factor for premature death globally, and the safest level of consumption for overall health is zero. The dose-response is non-linear - light drinking (≤ 10 g/day) carries small but real harm; heavy drinking (> 50 g/day, roughly 5 standard drinks) carries a steeply rising hazard.

The most-cited life-expectancy estimate comes from Wood et al., Lancet, 2018 [13], a pooled analysis of 599,912 current drinkers across 83 prospective studies. The headline finding: alcohol consumption above 100 g/week (roughly 10 standard UK units) was associated with measurable life-expectancy reduction, rising to 4–5 years lost at > 350 g/week compared to abstainers. The mechanisms are familiar: cardiovascular disease, multiple cancers (particularly oesophageal, liver, and breast), liver cirrhosis, and a disproportionate share of accidental deaths.

This is the risk factor on the list where the cultural narrative is furthest behind the data. The "red wine is good for you" finding from the 1990s did not survive the larger, better-controlled cohorts of the 2010s - it was almost entirely confounded by the fact that moderate drinkers tend to be wealthier and healthier than abstainers. The honest 2020s read of the literature is that less is better, and that anything above ~10 units a week is starting to carry meaningful hazard.

The combined picture

The seven factors in this article are not independent. They cluster. Sedentary behaviour drives low fitness, which drives obesity, which drives type 2 diabetes and hypertension. Short sleep drives insulin resistance, which drives weight gain and pre-diabetes. Loneliness drives all of the above by depressing motivation, raising cortisol, and reducing the social accountability that holds healthy habits in place.

This is why the biggest mortality wins in published cohort data come from combining the moves, not from optimising any single one. The Nurses' Health Study and Health Professionals Follow-up Study analysis (Li et al., Circulation, 2018 [14]) found that adopting five low-risk lifestyle factors at age 50 - never smoking, healthy BMI, 30+ minutes of daily activity, moderate alcohol, healthy diet - was associated with 14 years of additional life expectancy for women and 12 for men. That is a larger swing than any single intervention in the medical pharmacopoeia.

The implication for anyone tracking biological age: the lever that moves the number most reliably is not a single supplement or a single workout protocol. It is the boring combination of fitness work, blood-pressure control, sleep hygiene, sane alcohol intake, glucose awareness, and the social connections that hold the whole stack together. If you want the deeper read on how those inputs are weighted into a single composite score, our companion piece on how bio age is actually computed walks through the math and the honest uncertainty bands.

What to do with this list

A few practical takeaways once you have stopped reading mortality tables:

  • Rank your own risks, not the population's. If you are a non-smoking, non-drinking, well-slept, well-connected, hypertensive sedentary 50-year-old, factors 1, 2, and 4 are your real bill. Don't waste cycles on the ones that don't apply.
  • Measure before you treat. Blood pressure, HbA1c, resting heart rate, VO₂ max, and sleep duration are all measurable with equipment you almost certainly already own. The hazards on this list are mostly invisible; the inputs that drive them are not.
  • Combine moves wherever possible. A 45-minute walk with a friend after dinner moves four of these levers at once - activity, sedentary time, social connection, and post-meal glucose. The compounding is real.
  • The recovery curves are kind. Quitting smoking at 50 recovered six of the ten years. Treating hypertension at any age cuts cardiovascular events by ~20% per 10 mmHg. Most of the hazards on this list are reversible if you start the work this year.

The takeaway

Smoking deserves its reputation as the cleanest example of a self-inflicted life-expectancy cost. The point of this article is not that smoking is fine - it is that several extremely common, extremely tolerated behaviours quietly carry the same or larger hazard burden, and the population-level damage they do is enormous because the prevalence is so high. Low cardiorespiratory fitness, untreated hypertension, chronic loneliness, sedentary behaviour, short sleep, type 2 diabetes, and heavy alcohol are the seven that the cohort data picks out most cleanly.

Most of them are measurable. Most of them are modifiable. And almost all of them respond to the same boring stack of inputs that move biological age in the right direction - aerobic base, strength, sleep, blood-pressure control, alcohol moderation, social connection. If you want those inputs tracked, scored, and explained from the data you already have on your wrist, that is what Thier does.

Frequently asked questions

Is it really true that some things age you more than smoking?

Yes - and the published cohort evidence is unambiguous. The Cleveland Clinic's 122,000-patient treadmill study (Mandsager et al., JAMA Network Open 2018) found that low cardiorespiratory fitness carried a higher all-cause mortality hazard than smoking, hypertension, or type 2 diabetes. Severe loneliness (Holt-Lunstad meta-analysis, 2015) carries roughly the same mortality risk as smoking 15 cigarettes a day. The framing "X is worse than smoking" is not clickbait - it is how the hazard ratios actually stack up.

How many years of life does smoking actually cost?

About 10 years of life expectancy for lifelong smokers compared to never-smokers, per the landmark 50-year British Doctors Study (Doll et al., BMJ 2004). That number is the benchmark every other risk factor in this article gets compared against.

Is sitting really the new smoking?

The slogan is too tidy, but the underlying signal is real. Pooled analyses (Ekelund et al., Lancet 2016) show that prolonged sedentary time independently raises mortality risk, and the effect is only partially offset by exercise. The honest version: sitting all day is a serious independent risk factor, not a replacement metaphor for smoking.

Which of these factors is easiest to change?

Cardiorespiratory fitness is the highest-leverage modifiable factor - most adults can move out of the lowest fitness quintile in 12–16 weeks of structured aerobic training, which cuts all-cause mortality risk by roughly 50%. Blood pressure responds to lifestyle and (if needed) medication within months. Sleep duration and social connection are harder but free.

Why aren't ultra-processed foods on the list?

They are absolutely a real signal - recent meta-analyses (Lane et al., BMJ 2024) link very high ultra-processed food intake to higher all-cause mortality. We left them off the top-seven because the years-of-life-expectancy estimate is still uncertain and the effect is largely mediated through the other factors on this list (hypertension, type 2 diabetes, body composition). It is an important upstream lever, but harder to express as a single hazard ratio.

References

  1. Doll R, Peto R, Boreham J, Sutherland I. Mortality in relation to smoking: 50 years' observations on male British doctors. BMJ. 2004;328(7455):1519. PubMed
  2. Mandsager K, Harb S, Cremer P, et al. Association of Cardiorespiratory Fitness With Long-term Mortality Among Adults Undergoing Exercise Treadmill Testing. JAMA Network Open. 2018;1(6):e183605. PubMed
  3. GBD 2019 Risk Factors Collaborators. Global burden of 87 risk factors in 204 countries and territories, 1990–2019. Lancet. 2020;396(10258):1223-1249. PubMed
  4. Lewington S, Clarke R, Qizilbash N, et al. Age-specific relevance of usual blood pressure to vascular mortality: a meta-analysis of individual data for one million adults in 61 prospective studies. Lancet. 2002;360(9349):1903-1913. PubMed
  5. Ettehad D, Emdin CA, Kiran A, et al. Blood pressure lowering for prevention of cardiovascular disease and death: a systematic review and meta-analysis. Lancet. 2016;387(10022):957-967. PubMed
  6. Holt-Lunstad J, Smith TB, Baker M, Harris T, Stephenson D. Loneliness and social isolation as risk factors for mortality: a meta-analytic review. Perspectives on Psychological Science. 2015;10(2):227-237. PubMed
  7. Ekelund U, Steene-Johannessen J, Brown WJ, et al. Does physical activity attenuate, or even eliminate, the detrimental association of sitting time with mortality? A harmonised meta-analysis of data from more than 1 million men and women. Lancet. 2016;388(10051):1302-1310. PubMed
  8. Cappuccio FP, D'Elia L, Strazzullo P, Miller MA. Sleep duration and all-cause mortality: a systematic review and meta-analysis of prospective studies. Sleep. 2010;33(5):585-592. PubMed
  9. Yin J, Jin X, Shan Z, et al. Relationship of Sleep Duration With All-Cause Mortality and Cardiovascular Events: A Systematic Review and Dose-Response Meta-Analysis of Prospective Cohort Studies. Journal of the American Heart Association. 2017;6(9):e005947. PubMed
  10. Emerging Risk Factors Collaboration. Diabetes mellitus, fasting glucose, and risk of cause-specific death. New England Journal of Medicine. 2011;364(9):829-841. PubMed
  11. Knowler WC, Barrett-Connor E, Fowler SE, et al. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. New England Journal of Medicine. 2002;346(6):393-403. PubMed
  12. GBD 2016 Alcohol Collaborators. Alcohol use and burden for 195 countries and territories, 1990-2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet. 2018;392(10152):1015-1035. PubMed
  13. Wood AM, Kaptoge S, Butterworth AS, et al. Risk thresholds for alcohol consumption: combined analysis of individual-participant data for 599 912 current drinkers in 83 prospective studies. Lancet. 2018;391(10129):1513-1523. PubMed
  14. Li Y, Pan A, Wang DD, et al. Impact of Healthy Lifestyle Factors on Life Expectancies in the US Population. Circulation. 2018;138(4):345-355. PubMed