VO2max: The Best Single Predictor of Lifespan

VO2max is the maximum amount of oxygen your body can consume per minute at peak effort, and across large cohorts it tracks mortality more closely than any conventional risk factor. In 122,007 adults tested at the Cleveland Clinic, the lowest-fitness group carried roughly five times the adjusted risk of death of the elite group, and no upper limit of benefit appeared — even the jump from "high" to "elite" bought additional survival. The honest qualifier matters as much as the headline: these are observational data, and association is not causation.

What the number actually represents#

Think of oxygen as fuel and the body as an engine. VO2max answers one question: how much fuel can the whole system move per minute when it is working flat out? The unit is millilitres of oxygen per kilogram of body mass per minute, so a typical figure looks like 42 mL/kg/min.

Behind that figure sits a supply chain, and the chain is only as good as its narrowest link. The lungs load oxygen into blood, and in a healthy person this step is rarely limiting. The heart moves the blood, and this is where the real constraint sits: cardiac output equals stroke volume times heart rate, and while maximum heart rate barely responds to training, stroke volume responds a great deal. Blood carries oxygen on haemoglobin, so anaemia can flatten VO2max regardless of fitness. Finally, muscle extracts and burns the oxygen; capillary density and mitochondrial content decide how much of what arrives gets used.

The consequence is worth stating plainly. VO2max is not a heart test or a lung test. It is an integrated function test, which is precisely why it predicts outcomes so well — one measurement compresses the state of several systems at once.

One more unit appears constantly in the literature: the MET, or metabolic equivalent, defined as resting metabolism, about 3.5 mL/kg/min. So 42 mL/kg/min equals 12 METs. Clinical treadmill tests usually report METs because they infer oxygen use from speed and grade rather than measuring it with a gas analyser.

What the evidence shows#

The study everyone cites is Mandsager and colleagues, published in JAMA Network Open in 2018: a retrospective cohort of 122,007 adults referred for symptom-limited treadmill testing at the Cleveland Clinic between 1991 and 2014, followed for a median of 8.4 years across 1.1 million person-years, with 13,637 deaths. Patients were sorted into five fitness bands matched for age and sex: low (below the 25th percentile), below average, above average, high, and elite (97.7th percentile and up).

After adjustment the pattern was stark. Elite versus low gave a hazard ratio of 0.20 — roughly a fivefold difference in mortality. Below average versus above average gave 1.41, meaning even one step within the ordinary range was expensive. For calibration, the same model returned 1.41 for smoking, 1.40 for diabetes and 1.29 for coronary artery disease. The gap between poor and merely average fitness was about the size of having diabetes, and the gap between poor and elite exceeded any single classical risk factor.

The second finding was the one the authors had set out to test. Sports cardiology was then debating whether very high training volumes damage the heart, citing myocardial fibrosis, coronary calcium and arrhythmias in endurance athletes. No such penalty appeared. Elite performers outlived even the high-fitness group (hazard ratio 0.77), and that advantage persisted in patients over 70 and in those with hypertension. People whose fitness sat two or more standard deviations above the age- and sex-matched mean had the lowest adjusted mortality of anyone.

Kodama and colleagues, writing in JAMA in 2009, supply the same picture quantitatively and in healthy people rather than referred patients. Pooling 33 cohorts and roughly 103,000 participants, each additional MET of aerobic capacity was associated with about 13 percent lower all-cause mortality and 15 percent fewer cardiovascular events. They also identified a practical threshold at 7.9 METs, about 28 mL/kg/min: below it, mortality ran 1.7 times higher than in people at 10.9 METs or more.

In 2016 the American Heart Association published a scientific statement titled "Importance of Assessing Cardiorespiratory Fitness in Clinical Practice: A Case for Fitness as a Clinical Vital Sign." The title is meant literally. The authors argue that fitness should be measured or at least estimated as routinely as blood pressure, because its prognostic value exceeds that of most laboratory panels.

The caveats that get left out#

An industry of promises has grown around this number, so the limits deserve equal space.

These are observational data. Nobody randomised anyone into a high or low VO2max for twenty years. Fit and unfit people differ in dozens of ways: smoking, sleep, education, income, access to care, body composition. Statistical adjustment removes part of that, never all of it.

Reverse causation explains part of the effect. This applies with particular force to the Cleveland Clinic cohort, because patients were not volunteers — they were referred for symptom-limited testing, typically for chest pain, breathlessness or suspected ischaemia. Someone with undiagnosed heart failure, early cancer, occult anaemia or severe arthritis will score poorly not because they trained too little but because they are ill, and that same illness raises their chance of dying soon. The disease depresses both fitness and survival, manufacturing an association without a causal arrow between them. Researchers push back by excluding early deaths and adjusting for known comorbidities, but an undiagnosed condition cannot, by definition, enter the model.

The headline number overstates what an individual should expect. "Five times" compares the extremes of a clinical referral population. Someone moving from the bottom third to the middle is looking at a much smaller absolute gain, spread across decades.

What survives all of this. Randomised trials of exercise consistently improve blood pressure, insulin sensitivity, lipids, vascular function, body composition and cognition, and training reliably raises VO2max — that part is interventional evidence, not correlation. The advice to move more rests on solid ground even where the fivefold figure demands caution.

Читайте также: Health Checkups by Age: What Screening Is Actually Worth Doing

Reference values by age and sex#

An absolute figure means little on its own: 35 is unremarkable for a man of twenty-five and impressive for a woman of sixty-five. What matters is your percentile within your own age and sex group. The tables below come from the FRIEND registry, built from laboratory tests with directly measured oxygen uptake — not estimates — in people free of cardiovascular disease.

Age	25th percentile	50th (median)	75th percentile	90th percentile
20–29	40.1	48.0	55.2	61.8
30–39	35.9	42.4	49.2	56.5
40–49	31.9	37.8	45.0	52.1
50–59	27.1	32.6	39.7	45.6
60–69	23.7	28.2	34.5	40.3
70–79	20.4	24.4	30.4	36.6

Age	25th percentile	50th (median)	75th percentile	90th percentile
20–29	30.5	37.6	44.7	51.3
30–39	25.3	30.2	36.1	41.4
40–49	22.1	26.7	32.4	38.4
50–59	19.9	23.4	27.6	32.0
60–69	17.2	20.0	23.8	27.0
70–79	15.6	18.3	20.8	23.1

Three things are worth reading out of these tables. Capacity falls by roughly 10 percent per decade — the male median halves from 48.0 in the twenties to 24.4 in the seventies. The decline is unavoidable; its steepness is not. The sex difference is mostly body composition and haemoglobin: VO2max is normalised to total body weight rather than lean mass, so a higher fat percentage plus lower haemoglobin accounts for most of the gap, which shrinks considerably per kilogram of fat-free mass. And protocols are not interchangeable — cycle ergometry typically returns 5 to 10 percent less than a treadmill because less muscle mass is recruited.

One more landmark from the Kodama analysis: the 7.9 MET threshold, about 28 mL/kg/min. For a 60-year-old man that is roughly the median; for a 40-year-old woman it sits above the 75th percentile. Crude, but it works as a warning line.

How to measure it#

Cardiopulmonary exercise testing#

The gold standard puts you on a treadmill or cycle ergometer with a graded protocol to volitional exhaustion, wearing a mask that analyses inspired and expired gas. Oxygen uptake is measured directly rather than inferred.

A test counts as valid when a true maximum is reached: oxygen uptake plateaus despite rising workload, the respiratory exchange ratio passes about 1.10, and heart rate approaches the age-predicted maximum. The loaded portion should last eight to twelve minutes — longer understates the result through fatigue, shorter does not give the system time to reach its ceiling. The bonus is everything besides the headline number: ventilatory thresholds that define your training zones, plus an ECG under load.

Field tests#

Laboratories are not available to everyone, and for practical purposes an estimate is enough.

The Cooper test. Cover as much ground as you can in 12 minutes on a track or flat course. VO2max is approximately (distance in metres − 504.9) ÷ 44.73, so 2,400 metres works out to about 42 mL/kg/min. Simple, but unforgiving of pacing errors: going out too fast and starting too cautiously both depress the result.

The 20-metre shuttle run, or beep test. Run back and forth between markers in time with signals that accelerate each level, until you can no longer keep up; level and shuttle count convert to VO2max via standard tables. It needs no track and works well for groups, but turning technique and sheer willingness to suffer bleed into the score.

The Rockport one-mile walk. Walk a mile as fast as you comfortably can, record the time and your finishing heart rate, then feed those into a formula alongside age, sex and weight. It suits people who cannot run and is far safer than any maximal effort.

Submaximal cycle protocols such as the Åstrand test infer VO2max from heart rate at a fixed workload: wider error, moderate exertion.

Expect around 10 to 15 percent error from any of these. Their value is not the absolute number but repeatability: the same test in three months under the same conditions will tell you honestly whether anything changed.

What your watch is doing#

Sports watches do not measure VO2max, they infer it. The algorithm looks at the relationship between GPS pace and heart rate: the faster you run at a given heart rate, the higher the modelled capacity.

Four practical consequences follow. The error is comparable to a field test, roughly 10 to 15 percent, and grows for people with unusual body mass or an atypical maximum heart rate. The output depends on the profile you entered — a device using 220 minus age inherits every flaw of that formula. Wrist optical sensors misread during intervals and in the cold, and a wrong heart rate corrupts the calculation. And because every manufacturer uses a different model, numbers from different brands are not comparable at all.

Treat the figure as an index of your own progress rather than a measurement. A six-month trend on one device is informative. Your number against your friend's number on a different watch is noise.

Читайте также: Heart Rate Zones: How to Calculate and Why Zone 2 Matters

How to raise it#

Both physiology and endurance-sport practice point the same way: you need the two ends of the intensity range. A lot of easy work, a little very hard work. The comfortable middle delivers the least per hour invested.

The base: easy aerobic volume#

Long, relaxed sessions grow cardiac stroke volume, capillary density and mitochondrial content, expanding delivery and extraction at once. These adaptations respond to accumulated time under load, not to how unpleasant the session felt. In practice that means three to five hours a week of easy cardio: brisk walking, easy running, cycling, swimming, rowing. The reliable marker of correct intensity is that you can speak in full sentences; how that maps onto measured heart rate zones is covered in detail elsewhere on the site and is not repeated here.

The floor is set by the WHO guidelines: 150 to 300 minutes of moderate aerobic activity a week, or 75 to 150 minutes of vigorous, plus two strength sessions. That is a public health minimum, not an optimum for aerobic capacity.

The stimulus: high-intensity intervals#

Easy volume builds the infrastructure. Raising the ceiling itself requires the heart to spend time near maximal output, which means intervals.

The Norwegian 4 × 4. Warm up for 10 minutes. Work for 4 minutes at 90 to 95 percent of maximum heart rate — hard enough that talking is impossible, but not a sprint. Recover actively for 3 minutes. Repeat four times, then cool down for 5 minutes. Around 40 minutes in total, once or twice a week. This is the most studied protocol for raising VO2max.

Short intervals, 30/15. Thirty seconds fast, fifteen seconds easy, in blocks of 10 to 13 repetitions, two or three blocks with 3 minutes between them. Psychologically easier than long efforts and gentler on joints. Five by three minutes with equal recovery sits between the two.

The governing rule is dosage: one or two hard sessions a week, three at most for people carrying a large total volume. Intense work is expensive to recover from, and attempting it every other day ends in accumulated fatigue or injury. Roughly 80 percent easy and 20 percent hard is a workable split for a recreational athlete.

Everything else that moves the number#

Losing fat mass raises VO2max mechanically, because the score is normalised to body weight: shed 5 kg while holding performance steady and 35 mL/kg/min becomes about 37. The unglamorous inputs matter too — adequate iron and ferritin (deficiency depresses VO2max independent of training), sleep, stopping smoking, and treating anaemia or sleep apnoea.

How much, and how fast#

Expectations deserve calibrating, because this is where marketing gets loudest.

Untrained adults typically gain 10 to 20 percent over eight to twelve weeks. The Bacon meta-analysis pooled 37 interval-training studies lasting six to thirteen weeks across 334 participants and found a mean increase of about 0.51 L/min, rising to roughly 0.8 to 0.9 L/min in the subset using longer intervals. For a 75 kg person, 0.5 L/min is about 7 mL/kg/min — the difference between 32 and 39, or one to two percentile bands.

Trained people have far less headroom. After a first year of systematic training, gains shrink to low single digits per season, and performance depends increasingly on movement economy and lactate threshold rather than VO2max itself.

Individual variability is enormous. In classical training studies, participants on identical programmes ranged from 30 to 40 percent gains down to essentially nothing, with a large genetic component. One detail deserves emphasis: even in low responders by VO2max, blood pressure, insulin sensitivity and lipids improved. A flat number does not mean a wasted programme.

Detraining is faster than training. Stop entirely and a meaningful share of the gain disappears within two to four weeks, mostly through falling plasma and stroke volume. Holding a level takes far less work than building it — but not zero.

Strength is a separate account#

Resistance training has been rightly rehabilitated in recent years: muscle mass, grip strength and the ability to rise from a chair unaided all predict survival and independence in old age. That has led some people to a wrong conclusion — that the gym is enough.

Physiologically, lifting barely raises VO2max. The efforts are short, local and separated by long rests, so the heart never accumulates time near peak output, and the central adaptations that matter — left ventricular volume, plasma volume — need sustained load. A purely resistance-based programme moves aerobic capacity by a few percent in an untrained person and then stalls.

The reverse is equally true: cardio does not build muscle and does not protect against sarcopenia. These are two distinct reserves of ageing, and they fail in different ways. Aerobic capacity governs functional reserve — climbing five flights, tolerating surgery, recovering from pneumonia. Strength and muscle mass govern fall resistance, metabolic health (muscle is the main sink for glucose), bone density and everyday autonomy.

This is exactly why the WHO guidelines list both: aerobic activity plus muscle-strengthening work on all major groups at least twice a week. Not one or the other.

Читайте также: Strength Training for Beginners: Your First Three Months

Where to start#

Establish a baseline: a laboratory test if you can get one, otherwise a Cooper test if you have no contraindications, or the Rockport walk if running is off the table. Note the conditions — location, weather, footwear — so the retest is comparable. Then find your percentile in the tables above rather than fixating on the raw figure.

Build easy volume first: three to five hours a week at an intensity where you can hold a conversation, and add nothing intense for four to six weeks. Then add one interval session a week, 4 × 4 minutes at 90 to 95 percent of maximum; after a month, a second is reasonable. Keep lifting throughout — two sessions a week covering the major muscle groups, because that is the part cardio will never do for you. Retest at twelve weeks with the same method under the same conditions, expecting 10 to 20 percent from a low base and considerably less if you were already fit.

The single most useful thing to take from the whole evidence base is where the leverage sits. It is not the move from good to excellent. It is leaving the bottom quartile. If your result sits below the 25th percentile, the first few months of consistent movement will do more for you than any amount of protocol tuning afterwards.