VO₂ Max

VO₂ max — the maximum rate at which the body can take in and use oxygen, measured in millilitres of oxygen per kilogram of body weight per minute (mL/kg/min) — is among the strongest predictors of all-cause mortality in modern epidemiology, strong enough that the American Heart Association has argued it should be measured as a clinical vital sign.^[1] In the largest cohort to put them side by side, being unfit carried a larger relative mortality risk than smoking, type 2 diabetes, or hypertension.^[2] The dose-response shows no clear upper limit of benefit — the highest-fit people have the lowest mortality — though the extra margin from "high" into "elite" shows up in older adults and in people with hypertension rather than across the board. This article walks the physiology, the mortality data, and the recent Mendelian randomization finding that complicates the simple "VO₂ max extends life" story.

For the training side — the interval protocols, the minimal effective dose, and how to progress at any age — see Training VO₂ max.

What VO₂ max actually measures

Evidence: Strong — established exercise physiology, and reference percentiles drawn from large testing registries, though the registries disagree with each other.

Per the Fick equation: VO₂ = cardiac output × the arteriovenous oxygen difference. In plainer terms, VO₂ max distils three systems into a single number — how much blood the heart can pump per beat (stroke volume × heart rate), how well-perfused the working muscle is (capillary density), and how well the muscle's mitochondria extract and burn oxygen from that blood.

Typical adult values in the fourth decade of life (mL/kg/min):

Read your number against the percentile for your age and sex, not against an absolute threshold — the same figure means something quite different at 35 and at 65.^[3] Be aware that the two most-used reference sets disagree substantially: the FRIEND registry, drawn from clinical exercise-testing laboratories, puts the female median around 30 mL/kg/min, while the Cooper Institute and American College of Sports Medicine percentiles reproduced above put it near 37. A woman reading 33 is below average against one standard and above average against the other. Treat any single band label with suspicion; treat your own change over time as the reliable signal.

VO₂ max declines roughly 10% per decade in men and women largely irrespective of activity level; sustained high-intensity training can cut that loss by up to about half in young and middle-aged men, but does not abolish it.^[4] The decline is also not linear: in the Baltimore Longitudinal Study of Aging it accelerated from ~3–6% per decade in the 20s and 30s to more than 20% per decade from the 70s onward, in every quartile of self-reported activity.^[5] Nor is it uniform across the oxygen-delivery chain — see the central-versus-peripheral split below.

The mortality data

Evidence: Strong (observational) for the association — large, well-replicated prospective cohorts with hard endpoints. Absent for the causal step: no randomised trial has shown that raising VO₂ max lowers all-cause mortality, and these cohorts cannot rule out reverse causation, since early undiagnosed disease lowers fitness before it kills.

A foundational 2009 meta-analysis in JAMA established that every 1-MET increase in fitness — one metabolic equivalent equals 3.5 mL/kg/min of oxygen consumption — associates with about 13% lower all-cause mortality, and about 15% lower risk of a coronary or cardiovascular event.^[6] The largest synthesis to date — a 2024 umbrella review of 199 cohorts (20.9 million observations) — refreshes that headline: the fittest had 53% lower all-cause mortality than the least fit (hazard ratio 0.47, 95% confidence interval 0.39–0.56 — the bracketed range is where the true effect most plausibly lies, and because it sits entirely below 1.0 the benefit is very unlikely to be a fluke; a range crossing 1.0 would be compatible with no effect), with each 1-MET higher fitness worth an 11–17% reduction, and the protective association extended across 12 incident conditions including cardiovascular disease, several cancers, depression, and dementia.^[7] The underlying evidence is observational throughout, and the authors graded its certainty as very low to moderate — with most of it toward the low end.

The single most-cited individual study is the 2018 Cleveland Clinic cohort of 122,007 adults referred for exercise treadmill testing.^[8] That referral is worth holding onto: these are patients sent for a clinical treadmill test, not a general population, and in any such cohort part of what low fitness marks is disease already present but not yet diagnosed. Three findings shaped how preventive cardiology now thinks about fitness:

• No upper limit of benefit. The highest-fitness category had the lowest mortality.
• "Elite" fitness was protective on top of "high" fitness. Adults two standard deviations above the age-and-sex mean had about five times lower all-cause mortality than the least-fit, and roughly 23% lower than the merely "high-fit" (hazard ratio 0.77, 95% CI 0.63–0.95). That extra margin held up in two groups — people aged 70 or older (hazard ratio 0.71) and people with hypertension (0.70) — and was absent in the younger age groups and in the diabetes, coronary-disease and hyperlipidaemia subgroups.
• Low fitness ranks alongside the biggest risk factors. The authors put it carefully: the mortality risk of being unfit was "comparable to, if not greater than" that of coronary artery disease, smoking, or diabetes. A larger veterans cohort reaches the same conclusion with chronic kidney disease in the comparison set.^[9]

A 2022 cohort of 750,302 US veterans confirms and extends this across the full age, sex, and race spectrum: each 1-MET higher fitness lowered mortality 14% (HR 0.86, 95% CI 0.85–0.87), the least-fit fifth carried ~4× the mortality of the extremely fit, and there was no excess mortality risk even at the highest fitness — reinforcing the no-upper-limit finding for that endpoint, though it is not a blanket all-clear: the same cohort drew published correspondence questioning how far the top of the curve can be read,^[10] and sustained very high lifetime endurance volumes carry their own atrial-fibrillation signal, covered in Training VO₂ max.^[11] Because the gain is age-relative, the same study gives concrete age-specific thresholds for roughly halving mortality risk: ~11 METs at 30–49, ~10 at 50–59, ~8 at 60–79, and ~7 METs for octogenarians — fitness is best read against an age-and-sex percentile, not one absolute number. (This cohort is ~94% male and uses treadmill-estimated METs rather than gas-exchange VO₂ max.) Keep those units apart when comparing figures across this page: a treadmill-estimated MET, a laboratory VO₂ max with a verified plateau, and a symptom-limited "peak VO₂" are three different measurements, and the last runs systematically lower than the first — which matters, because most of the mortality data comes from older, clinical, referred populations tested to symptoms rather than to a plateau.^[12]

The trajectory over time matters as much as the snapshot. A 15-year follow-up of middle-aged men found that maintaining just 1 mL/kg/min of VO₂ over an 11-year window associated with 9% lower all-cause mortality — and a 1-MET higher maintained fitness associated with a 29% lower mortality risk.^[13] The reverse is also true: complete training cessation can drop VO₂ max ~20% in 12 weeks in masters athletes.^[14] The 1966 Dallas Bed Rest Study put five healthy 20-year-old men through three weeks of strict bed rest and measured a steep fall in maximal oxygen uptake.^[15] The punchline came from the 30-year follow-up of the same five men: those three weeks of bed rest had a more profound impact on physical work capacity than the three decades of aging that followed.^[16]

A 2023 study tracking fitness change in 93,060 US veterans (two treadmill tests about six years apart, 94.6% men) sharpens the actionable message: among those with cardiovascular disease, losing more than 2 METs raised all-cause mortality by 37–74% depending on starting fitness, while gaining fitness lowered it stepwise.^[17] Starting fit does not exempt you: in this cohort the only group whose risk did not measurably rise after a large decline was the fittest quartile without cardiovascular disease. For the fittest quartile with cardiovascular disease, a decline of more than 2 METs still carried about 37% higher mortality. Improving fitness lowers risk at any baseline, and holding onto it matters most for the people who already have heart disease.

The reversal signal is large. In 9,777 men given two treadmill tests about five years apart, those who moved from unfit to fit between the two exams had a 44% lower all-cause death rate (95% CI 25–59%) than men who were unfit at both — and every extra minute of maximal treadmill time between exams tracked with ~8% lower mortality.^[18]

Beyond all-cause mortality: cancer, brain, and sex

The protection is not limited to cardiovascular death. In over a million Swedish conscripts, higher youth fitness tracked linearly with lower risk of nine cancers — for lung cancer, about 42% lower in the fittest versus least fit (HR 0.58).^[19] A separate Swedish cohort linked higher fitness to lower colon cancer incidence and lower lung and prostate cancer mortality; one wrinkle — moderate fitness associated with slightly higher prostate cancer incidence — most likely reflects screening-detection bias (fitter men get screened more), not true harm.^[20] A 2025 re-analysis of the conscript registry settles that reading from inside the data: comparing brothers with each other, which strips out shared family and childhood confounding, the fittest quartile's slightly higher rate of cancer diagnosis disappeared entirely, while the lower cancer mortality survived the comparison.^[21] Detection, not disease. For the brain, a cohort of 649,605 veterans free of dementia found fitness inversely and stepwise associated with incident Alzheimer's disease and related dementias — about a third lower risk in the fittest group (hazard ratio 0.67), with risk falling stepwise across the five fitness bands.^[22]

Sex matters for interpretation: in a cohort of 57,284 adults, women reached the same survival benefit at a lower absolute fitness level than men, while the mortality reduction per unit of fitness was similar in both sexes.^[23] The gradient does appear to flatten at the top end in women: a spline analysis of 17,901 women found no further mortality reduction beyond about 11 METs.^[24] That is another argument for reading fitness as an age-and-sex percentile rather than as one absolute number.

The causality complication: closer to a receipt than a drug

Evidence: Weak-to-moderate — one Mendelian randomization study on a small number of outcomes, plus family-study evidence on how much of the baseline number is inherited. No trial has tested the causal question directly.

The simple reading of all this data is "VO₂ max extends life." A more careful one came from a 2025 Mendelian randomization study published in the Journal of Clinical Endocrinology and Metabolism.^[25] Mendelian randomization is a technique that uses random inherited genetic variation as a stand-in for a randomized trial — if a gene that raises VO₂ max also extends life, that's strong evidence for a causal relationship.

The study found:

• Genetically predicted higher physical activity, more lean mass, and lower body fat are causally linked to higher VO₂ max.
• Genetically predicted VO₂ max itself showed no causal association with longevity or with type 2 diabetes.

That null comes from nine genome-wide-significant markers in about 70,000 people, and a null result on an instrument that weak is an absence of evidence rather than evidence of absence.

The translation: VO₂ max looks more like an indicator than a causal agent — at least on the evidence available, which is one Mendelian randomization study testing a small number of outcomes. Its life-extending power appears to come from the behaviours required to build it: sustained aerobic work, vascular remodelling, mitochondrial proliferation, body-composition optimisation, lean-mass retention. The caveat is that the number is not purely earned. In 429 people measured in the untrained state, at least half the variation between them was familial.^[26] So a high reading does not by itself prove the work was done, and a modest one does not prove it wasn't. What a rising reading proves is that something changed, and the only things that reliably change it are the behaviours above.

How much of the number is movable is its own question, and the answer is "a useful amount, but not unlimited" — Training VO₂ max covers the trainability ceiling, the apparent non-responders, and the finding that most apparent non-response dissolves once the training dose goes up.

This isn't a reason to demote VO₂ max — it's actually clarifying. A high VO₂ max is the most reliable biological receipt of life-extending behaviour we currently have.

There is a second reason to hold the claim carefully. Everything above is observational: it shows that people who have a high VO₂ max live longer, not that raising one extends life. The trials that tested the second question have not shown it. Generation 100 randomised 1,567 adults aged 70 to 77 to five years of supervised interval training, supervised moderate training, or following the national activity guidelines, and reported that all-cause mortality did not differ between the training and control groups.^[27] HF-ACTION, in 2,331 heart-failure patients, found essentially no difference in death from any cause — a 4% reduction well inside the range expected from chance (hazard ratio 0.96, 95% confidence interval 0.79–1.17).^[28] Look AHEAD produced larger fitness gains in its intervention arm and was stopped early for futility on cardiovascular events.^[29] The pooled Cochrane synthesis of exercise-based cardiac rehabilitation finds a likely small reduction in all-cause mortality at six to twelve months — about 13% (risk ratio 0.87, 95% confidence interval 0.73–1.04) — which is no longer detectable at medium- or long-term follow-up.^[30]

That is not a case against training. The same Cochrane review finds cardiovascular mortality does fall over the long term, by around 40% (RR 0.58, 95% CI 0.43–0.78), and the trials above were mostly short, in old or already-ill populations, and powered for other endpoints. But it does mean the honest version of this page's claim is narrower than the slogan: fitness is the best marker we have, the behaviours that build it are well evidenced, and the specific proposition that raising a healthy adult's VO₂ max extends their life has never been tested to a mortality endpoint.

How VO₂ max protects you, mechanistically

Evidence: Moderate — the vascular biology is well characterised in humans and animals, but the chain from a single mechanism to a mortality endpoint is inference rather than demonstration.

Whether the protection runs partly through glycaemic control is genuinely unsettled. The Mendelian randomization study above found no causal link between genetically predicted VO₂ max and type 2 diabetes, using nine genetic markers in about 70,000 people.^[31] A larger analysis using 160 fitness-associated genetic markers in 450,000 UK Biobank participants reached the opposite conclusion — that higher genetically predicted fitness does causally lower type 2 diabetes risk, independent of body fat — and closes by arguing for fitness training as diabetes prevention.^[32] The two disagree on instruments as much as on findings, and neither settles it. What is not in dispute is the vascular half of the story, which is where most of the mechanistic evidence sits.

1. Mechanotransduction and shear stress. Repeated high cardiac output during exercise pushes blood through the arteries at higher velocity and pulsatility. Endothelial cells lining the artery walls sense that physical force and convert it to a biochemical signal — phosphorylating endothelial nitric oxide synthase (eNOS) and increasing nitric oxide production.^[33]
2. Arterial compliance. Sustained nitric oxide relaxes vascular smooth muscle, keeps arteries flexible, and reduces the afterload the heart pumps against. Sustained over decades, this is thought to preserve left-ventricular function — though no trial has compared lifelong exercise against a drug on that endpoint, and the timescale makes one unlikely.
3. Anti-thrombotic effects. Nitric oxide inhibits platelet aggregation through cyclic GMP, lowering the risk that a ruptured plaque produces an occluding clot — the step between a plaque failing and a myocardial infarction or ischaemic stroke.
4. Endothelial preservation. Fit individuals develop arteries that physically resist atherosclerotic plaque deposition. This is why a high-fit smoker can have lower mortality than a low-fit non-smoker, and why elite-fit hypertensives outlive merely-fit ones.^[34]

These pathways run alongside the metabolic-syndrome story; how far the two are independent of each other is exactly the question the genetic evidence above leaves open. Both halves of the puzzle matter.

Why VO₂ max falls with age: the peripheral cascade

Evidence: Moderate — repeated-measures longitudinal data in small samples, converging with a much larger cross-sectional literature.

The intuitive story is that the aging heart is the bottleneck. The measurements say otherwise. In 99 community-dwelling adults from the Baltimore Longitudinal Study of Aging, followed for an average of 12.6 years with repeated gated blood-pool scans, the longitudinal declines in peak VO₂ and in the arteriovenous oxygen difference — the muscle's ability to strip oxygen out of the blood delivered to it — both steepened with advancing age. Peak cardiac output declined too, but at a steady rate — it did not steepen. Age-related impairment of peripheral oxygen utilisation imposed the stronger limit on peak VO₂, and the association was strongest in those over 50 — though that age-group difference did not reach statistical significance in this small sample.^[35]

The Dallas 30-year follow-up says the same thing in miniature. Across three decades in the same five men, maximal cardiac output was essentially unchanged (20.0 → 21.4 L/min) — a lower maximal heart rate was fully offset by a larger stroke volume — while the maximal arteriovenous oxygen difference fell 15% and accounted for the entire decline in aerobic capacity.^[36] The broader synthesis is that reduced oxygen delivery still dominates up to late middle age, while falling skeletal-muscle oxidative capacity takes over as the limiting factor in advanced old age.^[37] Either way, the muscle catches up to the heart as a source of failure.

The peripheral cascade has four converging components:

1. Sarcopenia — age-related muscle loss that preferentially strips the fast-twitch Type II fibres, reducing the total tissue available to consume oxygen.
2. Mitochondrial decay — fewer mitochondria per fibre and lower activity of oxidative enzymes (citrate synthase, succinate dehydrogenase).
3. Capillary rarefaction — thinning of the capillary networks that bring blood to individual muscle fibres, increasing the diffusion distance oxygen has to cross.
4. Interstitial fibrosis — connective-tissue stiffening of the extracellular matrix between fibres, slowing oxygen movement into the cell.

The good news: these are exactly the systems that training rebuilds, and the peripheral machinery remains trainable into late life — the largest trial in this age group put adults aged 70 to 77 through five years of supervised training and did raise their fitness, even though it did not move all-cause mortality.^[38] Resistance training covers the muscle-mass half; Zone 2 covers the capillary-and-mitochondria half; and Training VO₂ max puts the whole program together.

What about Apple Watch / Garmin VO₂ estimates?

Evidence: Moderate for the absolute inaccuracy — repeatedly measured, though in small validation samples. Weak for the trend premise, which has been tested once by a manufacturer and never independently.

Wrist-based estimates are best treated as a trend tool rather than a measurement. The absolute figure is unreliable, overestimating the unfit and underestimating the fit;^[39] the commonly quoted 10–20% gap rests on small studies and varies with the watch and the reference test. The trend half of the claim is the weaker one: the premise is that a consistent bias cancels out when you compare a reading against your own earlier readings, and no independent group has verified that it does. The lab gold standard is direct gas-exchange testing (cardiopulmonary exercise testing, CPET), worth doing once if you're training enough to care about the difference — it is a maximal, symptom-limited effort, so it belongs in a supervised setting and has its own contraindications,^[40] though the measured risk is small: across 5,060 tests in patients with high-risk cardiac diagnoses, eight adverse events occurred and none were fatal.^[41] Worth a distinction, though: an imprecise value and a useless predictor are not the same thing — a meta-analysis of estimated versus measured fitness found that even simple non-exercise equations predict mortality about as well as objectively measured fitness, at roughly 15–19% lower mortality per MET (relative risk 0.81–0.85).^[42] The number on your wrist is fuzzy; the prognostic signal underneath it is not. See Wearable VO₂ max estimates for device-by-device accuracy, the confounders that distort the number, and how to get a cleaner reading.

Further reading

• Kodama S et al. Cardiorespiratory Fitness as a Quantitative Predictor of All-Cause Mortality and Cardiovascular Events in Healthy Men and Women (meta-analysis). JAMA 2009.^[43]
• Mandsager K et al. Association of Cardiorespiratory Fitness With Long-term Mortality Among Adults Undergoing Exercise Treadmill Testing. JAMA Network Open 2018.^[44]
• Lang JJ et al. Cardiorespiratory fitness is a strong and consistent predictor of morbidity and mortality — overview of meta-analyses (199 cohorts, 20.9 million observations). BJSM 2024.^[45]
• Kokkinos P et al. Cardiorespiratory Fitness and Mortality Risk Across the Spectra of Age, Race, and Sex (n=750,302). JACC 2022.^[46]
• Kokkinos P et al. Changes in Cardiorespiratory Fitness and Survival in Patients With or Without Cardiovascular Disease (n=93,060). JACC 2023.^[47]
• Cheng Y et al. Cardiorespiratory fitness and risk of Alzheimer's disease and related dementias among American veterans (n=649,605). Alzheimers Dement 2023.^[48]
• Ballin M et al. Adolescent cardiorespiratory fitness and risk of cancer in late adulthood: a nationwide sibling-controlled cohort study in Sweden. PLoS Med 2025.^[49]
• Al-Mallah MH et al. Sex differences in cardiorespiratory fitness and all-cause mortality: the Henry Ford ExercIse Testing (FIT) Project. Mayo Clin Proc 2016.^[50]
• Farrell SW et al. Examining the gradient of all-cause mortality risk in women across the cardiorespiratory fitness continuum. Med Sci Sports Exerc 2022.^[51]
• Blair SN et al. Changes in physical fitness and all-cause mortality — a prospective study of healthy and unhealthy men. JAMA 1995.^[52]
• Laukkanen JA et al. Long-term Change in Cardiorespiratory Fitness and All-Cause Mortality. Mayo Clin Proc 2016.^[53]
• Saltin B et al. Response to exercise after bed rest and after training (the Dallas Bed Rest Study). Circulation 1968.^[54]
• McGuire DK et al. A 30-year follow-up of the Dallas Bedrest and Training Study. Circulation 2001.^[55]
• Kjaergaard AD et al. Cardiorespiratory Fitness, Body Composition, Diabetes, and Longevity — a Mendelian randomization study. J Clin Endocrinol Metab 2025.^[56]
• Cai L et al. Causal associations between cardiorespiratory fitness and type 2 diabetes. Nat Commun 2023.^[57]
• Bouchard C et al. Familial resemblance for VO₂ max in the sedentary state: the HERITAGE Family Study. Med Sci Sports Exerc 1998.^[58]
• Stensvold D et al. Effect of exercise training for five years on all cause mortality in older adults — the Generation 100 study: randomised controlled trial. BMJ 2020.^[59]
• O'Connor CM et al. Efficacy and safety of exercise training in patients with chronic heart failure: HF-ACTION randomized controlled trial. JAMA 2009.^[60]
• Wing RR et al. Cardiovascular effects of intensive lifestyle intervention in type 2 diabetes (Look AHEAD). N Engl J Med 2013.^[61]
• Dibben G et al. Exercise-based cardiac rehabilitation for coronary heart disease. Cochrane Database Syst Rev 2021.^[62]
• Ross R et al. Importance of assessing cardiorespiratory fitness in clinical practice: a case for fitness as a clinical vital sign — a scientific statement from the American Heart Association. Circulation 2016.^[63]
• Pan A et al. The relationship between cardiorespiratory fitness and mortality (correspondence). J Am Coll Cardiol 2022.^[64]
• Hawkins S, Wiswell R. Rate and mechanism of maximal oxygen consumption decline with aging. Sports Medicine 2003.^[65]
• Fleg JL et al. Accelerated longitudinal decline of aerobic capacity in healthy older adults. Circulation 2005.^[66]
• AlGhatrif M et al. Longitudinal decline in peak VO₂ with aging is associated with reduced peripheral oxygen utilization but not cardiac output. Am J Physiol Heart Circ Physiol 2024.^[67]
• Betik AC, Hepple RT. Determinants of VO₂ max decline with aging — an integrated perspective. Appl Physiol Nutr Metab 2008.^[68]
• Burtscher J et al. The impact of training on the loss of cardiorespiratory fitness in aging masters endurance athletes. Int J Environ Res Public Health 2022.^[69]
• Golbidi S, Laher I. Exercise and the aging endothelium. J Diabetes Res 2013.^[70]
• Poole DC, Jones AM. Measurement of the maximum oxygen uptake VO₂ max: VO₂ peak is no longer acceptable. J Appl Physiol 2017.^[71]
• Rossi Neto JM et al. Cardiorespiratory fitness data from 18,189 participants who underwent treadmill cardiopulmonary exercise testing in a Brazilian population. PLoS One 2019.^[72]
• Singh B et al. Comparison of objectively measured and estimated cardiorespiratory fitness to predict all-cause and cardiovascular disease mortality in adults: a systematic review and meta-analysis. J Sport Health Sci 2025.^[73]
• Lambe R et al. Accuracy of VO₂ max estimates from Apple Watch Series 10. Mayo Clin Proc Digit Health 2026.^[74]
• Fletcher GF et al. Exercise standards for testing and training: a scientific statement from the American Heart Association. Circulation 2013.^[75]
• Skalski J et al. The safety of cardiopulmonary exercise testing in a population with high-risk cardiovascular diseases. Circulation 2012.^[76]