VO2 Max

VO2 max is the highest rate at which your body can take in, deliver and use oxygen during all-out exercise, usually reported in milliliters of oxygen per kilogram of body weight per minute (mL/kg/min). It is the standard measure of cardiorespiratory fitness, and in large cohort studies it predicts long-term survival as strongly as smoking or diabetes.

If you wear an Apple Watch, you already have a VO2 max number. The Health app calls it Cardio Fitness. That number is an estimate from walks and runs, and it behaves differently from a lab result. This entry covers what VO2 max measures, how labs and watches arrive at it, how far the Apple Watch estimate sits from a lab test, what is normal for your age and sex, what the mortality research shows, how to raise it, and how Titan uses it. The wearable data quality guide sets the watch estimate beside other wearable metrics by accuracy.

01Units and what sets the ceiling

VO2 max comes in two forms. Absolute VO2 max, in liters per minute, is the total oxygen your body consumes at peak effort. Relative VO2 max divides that by body weight, so mL/kg/min = L/min × 1000 ÷ kg. Relative values are what watches, norm tables and most studies report, because moving your own body weight is what running and walking demand.

Body weight changes the relative number without any change in your heart or muscles. A 70 kg runner with an absolute VO2 max of 3.5 L/min scores 50 mL/kg/min. At 65 kg with the same heart, the same runner scores 53.8. The reverse holds too, so weight gain can hide a real fitness gain.

Clinical papers often express fitness in METs, metabolic equivalents. One MET is about 3.5 mL/kg/min (Apple, 2021). A treadmill result of 10 METs corresponds to roughly 35 mL/kg/min.

The Fick equation defines what VO2 max is made of. VO2 max = maximal cardiac output × maximal arteriovenous oxygen difference. Cardiac output is heart rate times stroke volume, the blood pumped per beat. The arteriovenous difference is how much oxygen your muscles pull from each liter of blood. Bassett and Howley (2000) reviewed the evidence and concluded that in healthy people exercising with large muscle groups, oxygen delivery is the main limit. Blood doping raises VO2 max, hypoxia and beta-blockers lower it, and training raises it mainly by increasing maximal cardiac output. Muscle adaptations matter more for how long you can hold a pace below the ceiling, which is the job of your lactate threshold.

02Lab tests and VO2 peak

A lab measures VO2 max with a cardiopulmonary exercise test. You run on a treadmill or ride a bike ergometer while the workload rises every minute or two, wearing a mask that samples every breath. The analyzer calculates oxygen consumed from the difference between inhaled and exhaled air. The test ends when you cannot continue.

A true VO2 max shows as a plateau, where the workload keeps rising and oxygen uptake stops following. Many people quit before that plateau appears. The highest value reached is then a VO2 peak. Poole and Jones (2017) argued that the usual backup criteria, such as a respiratory exchange ratio above a cutoff or reaching a predicted maximum heart rate, can be chosen arbitrarily and produce "grossly inaccurate" results. Their fix is a second, short bout at about 110 percent of the final ramp workload. If oxygen uptake does not rise further, the first result was a true maximum.

The mode matters. In the FRIEND registry of U.S. lab tests with measured oxygen uptake, treadmill values ran higher than cycle values in every age group for both sexes (Kaminsky et al., 2022). Cycling uses less muscle mass, so a cyclist's lab result will look low against a treadmill table.

03Field estimates

Field tests predict VO2 max from a performance and a regression equation. They trade accuracy for access.

The Rockport walk test asks you to walk one mile as fast as you can and record your finishing heart rate and time. Kline and colleagues (1987) built the equation from 343 adults aged 30 to 69 with treadmill-measured VO2 max. In a separate cross-validation group of 169, predicted and measured values correlated at r = 0.92, with a standard error of 0.355 L/min.

The heart rate ratio method uses no exercise at all. Uth and colleagues (2004) showed that VO2 max ≈ 15 × maximum heart rate ÷ resting heart rate. In well-trained men, it estimated VO2 max within a standard error of 2.7 mL/kg/min when maximum heart rate was measured, and 4.7 mL/kg/min when it was predicted from age. The authors tested it only in well-trained men, so its accuracy in other groups is unknown. It also shows why resting heart rate and maximum heart rate carry fitness information.

A wearable estimate is an automated submaximal field test that runs every time you walk or run.

04How Apple Watch estimates Cardio Fitness

Apple Watch Series 3 and later estimate VO2 max from your heart rate response to movement. The core idea is that a fitter heart needs fewer beats for the same pace. Walking briskly at 110 bpm implies a higher VO2 max than walking at the same speed at 140 bpm. The watch extrapolates that relationship toward a predicted maximum heart rate, which is why it does not need an all-out effort.

Apple's 2021 white paper lists the conditions for an estimate. You need an Outdoor Walk, Outdoor Run or Hiking workout on ground with less than a 5 percent grade, adequate GPS, a clean heart rate signal, and exertion that raises your heart rate by roughly 30 percent of the range between resting and maximum. The first qualifying workout does not produce an estimate. The watch also takes passive readings when you walk outdoors without starting a workout. Indoor workouts never count. The model uses your age, sex, weight, height and any heart rate-limiting medications entered in Health. Apple's support page gives a supported range of 14 to 65 mL/kg/min and says the first estimate can take at least 24 hours of wear plus several outdoor workouts.

For people 20 and older, the Health app classifies the estimate by age and sex. You can opt in to a Low Cardio Fitness notification. For ages 20 to 59, it fires when your estimate sits in the lowest fifth of FRIEND values for your sex and age decade. From 60 on, Apple uses fixed thresholds of 18 mL/kg/min for men and 15 for women, the levels Paterson and colleagues (1999) linked with independent living at age 85.

How close it gets to a lab test

Apple validated the algorithm on 221 people with a mean age of 55 and an average reference VO2 max of 29.7 mL/kg/min. The watch's final estimate read 1.4 ± 4.7 mL/kg/min above the reference on average, and repeat estimates at least 28 days apart agreed with an intraclass correlation of 0.86. The reference in that study was itself a projection from submaximal lab stages. Apple did not measure VO2 max at exhaustion.

Independent studies that did measure it found the watch reading low, especially in fit people.

The error grew with fitness. In the Series 7 study, the watch read 3.8 mL/kg/min high for the three least fit participants and 12 mL/kg/min low for the five fittest. In the Series 10 study, it read 7.8 low for people at or above the 80th FRIEND percentile and 2.4 low for those at or below the 70th. It read 8.6 low for women and 3.8 low for men. Only 14 percent of participants landed in the correct FRIEND percentile band, and another 34 percent landed one band away.

A 2022 meta-analysis of 14 studies and 403 participants across brands found almost no average bias for exercise-based wearable estimates, −0.09 mL/kg/min, with 95 percent limits of agreement of ±9.8 mL/kg/min (Molina-Garcia et al., 2022). Estimates from resting data overestimated by 2.2 on average and spread even wider. The authors concluded that exercise-based estimates work at the population level, while the error for any one person is large.

Taken together, the watch comes closest for older and less fit users and underestimates young, fit users, by 4 to 12 points in these studies. The 65 mL/kg/min ceiling also caps the fittest athletes. Use the watch number to follow your own direction over months, and a lab test when you need the absolute value.

Getting a cleaner estimate

Apple's white paper lists what distorts the estimate. Wrong age, sex or weight in Health produces consistently wrong results. Carrying a heavy pack or a child, walking on sand, pushing a stroller, heat, dehydration, caffeine and a recent move to altitude all raise heart rate for a given pace and push the estimate down. Heart rate-limiting drugs push it up if you leave them out. In Apple's validation group, people taking beta-blockers or calcium channel blockers saw their error fall from 11.8 ± 4.0 to 1.6 ± 3.1 mL/kg/min once the medication was entered in Health.

To help the watch, keep your Health details current, record frequent outdoor walks or runs on flat ground, push the pace hard enough to raise your heart rate well above easy, and wear the watch through the day.

05Norms by age and sex

The FRIEND registry reference standards come from 7,783 maximal treadmill tests with gas analysis, run in eight U.S. labs on adults without known heart disease (Kaminsky et al., 2015). Values are relative VO2 max in mL/kg/min.

VO2 max fell about 10 percent per decade across these age groups, and men averaged 27 percent higher than women, which is why each sex has its own table. The 2022 FRIEND update, with 16,278 treadmill tests, came out 1.5 to 4.6 mL/kg/min lower than these 2015 treadmill values and added cycle standards. It put the average decline at 4.0 mL/kg/min per decade on the treadmill (Kaminsky et al., 2022).

Cross-sectional tables smooth over how aging works in one person. Fleg and colleagues (2005) retested 810 adults in the Baltimore Longitudinal Study of Aging over a median of 7.9 years. Peak VO2 fell 3 to 6 percent per decade in people in their 20s and 30s, and the decline accelerated to more than 20 percent per decade in their 70s and beyond, regardless of reported activity.

Two cautions apply when you hold a watch number against this table. FRIEND measured oxygen directly, so if the watch underestimates you by 6 points, your percentile looks lower than it is. And a percentile band is a coarse label. A value of 41 and one of 42 differ by test noise even when they fall on either side of a cut point.

06VO2 max and mortality

A 2016 American Heart Association scientific statement reviewed the evidence linking low cardiorespiratory fitness with cardiovascular disease, all-cause mortality and deaths from several cancers. It called for fitness to be treated as a clinical vital sign (Ross et al., 2016).

Kodama and colleagues (2009) pooled 33 studies with 102,980 participants. Each 1-MET higher fitness level, about 3.5 mL/kg/min, was associated with 13 percent lower all-cause mortality and 15 percent fewer heart disease or cardiovascular events. People below 7.9 METs, about 28 mL/kg/min, had substantially higher rates of both than people above it.

Mandsager and colleagues (2018) followed 122,007 patients for a median of 8.4 years after treadmill testing. Compared with the least fit quarter for their age and sex, the elite group, at or above the 97.7th percentile, had 80 percent lower adjusted mortality (hazard ratio 0.20). Elite fitness still beat high fitness (hazard ratio 0.77), so the study found no point beyond which more fitness stopped helping. Below average fitness, compared with above average, carried a hazard ratio of 1.41, the same size as smoking (1.41) and larger than diabetes (1.40) or coronary artery disease (1.29).

Kokkinos and colleagues (2022) tested 750,302 U.S. veterans aged 30 to 95 and followed them for a median of 10.2 years. The least fit fifth had about four times the mortality of the fittest group (hazard ratio 4.09). Risk was lowest near 14 METs, about 49 mL/kg/min, for both men and women, with no rise in risk at extremely high fitness. The pattern held for septuagenarians, octogenarians, women and every racial group studied.

Change counts as well as level. Laukkanen and colleagues (2016) measured VO2 max by gas analysis in 579 Finnish men, then again 11 years later. Each 1 mL/kg/min smaller decline over that period was associated with 9 percent lower all-cause mortality.

These studies are observational. Fit people differ from unfit people in many ways that adjustment only partly captures. Both of the largest cohorts estimated METs from treadmill time without measuring oxygen. What they establish is that low fitness marks high risk across ages, sexes and races, and that the gradient does not flatten at the top.

07How to raise it

Training raises VO2 max reliably in healthy adults. Milanović and colleagues (2015) pooled 28 controlled trials in healthy adults aged 18 to 45, starting at an average of 40.8 mL/kg/min. Compared with no exercise, continuous endurance training raised VO2 max by 4.9 mL/kg/min and high-intensity interval training by 5.5. Both produced larger gains in people who started less fit and in longer programs.

Intensity changes the result when total work is held equal. Helgerud and colleagues (2007) had moderately trained men train three times a week for 8 weeks at matched total oxygen cost. Long slow running and threshold running did not raise VO2 max significantly. The 15 seconds on, 15 seconds off intervals raised it 5.5 percent, and four 4-minute intervals at 90 to 95 percent of maximum heart rate raised it 7.2 percent. The VO2 max gains tracked stroke volume, which rose about 10 percent after interval training.

Dose matters more than the label "non-responder" suggests. Montero and Lundby (2017) assigned 78 healthy adults to one to five 60-minute sessions per week for 6 weeks. The share who did not improve fell from 69 percent at one session to 40, 29, 0 and 0 percent at two through five. Everyone who had not improved did so after 6 more weeks with two extra sessions per week. Genes still shape the size of the gain. In the HERITAGE Family Study, the same 20-week program raised VO2 max by an average of about 400 mL/min, from almost nothing in some people to more than 1 L/min in others, with heritability of the response estimated at 47 percent (Bouchard et al., 1999).

Fitness also fades. Coyle and colleagues (1984) found that endurance athletes lost 7 percent of their VO2 max in the first 21 days after stopping training and 16 percent by day 56, where it stabilized. See detraining for the time course.

Helgerud's four 4-minute intervals are a well-tested template. A typical week pairs that kind of session with a larger volume of easy aerobic work. Zone 2 training and a polarized distribution are two ways to organize the easy part. Losing excess body fat raises relative VO2 max arithmetically. It adds nothing to your heart's output, so track absolute values or performance alongside it. The VO2 max and aerobic capacity guide covers session design in detail, and the performance physiology guide explains how VO2 max, lactate threshold and economy combine to set race performance.

08How Titan uses VO2 max

Titan does not calculate its own VO2 max. It reads the most recent VO2 max sample in Apple Health, which for most people is the Cardio Fitness estimate from Apple Watch. Titan does not filter by source, so a value written by another app, or entered by hand after a lab test, counts until a newer sample replaces it.

VO2 max appears in three places. Under You, Your Fitness shows your latest value, its change over the last 30 days with a sparkline, and a rating bar labeled Poor, Average, Good or Excellent for your sex and age decade, with the norm group printed underneath, for example men aged 40 to 49. The bands are Titan's own cut points. They do not match FRIEND percentiles. For men in their 40s, Excellent begins above 42 mL/kg/min, which falls between FRIEND's 50th and 75th percentiles. In Trends, the Cardio fitness chart plots VO2 max over time and filters by activity type. When you pick Running, for example, Titan keeps only the days on which you logged a run.

The third place is Titan Age, where VO2 max is the largest input in the Fitness group and carries about 16 percent of the model. Titan uses the latest reading in the 30 days before the week ends. The neutral reference is 24 mL/kg/min, which adds zero years. Titan measures your distance from it in steps of 4 mL/kg/min, and values above it lower the estimate. Titan softens large distances so each extra point is worth less than the one before. With full data coverage, before weekly smoothing, the contributions look like this.

Climbing from 24 to 34 is worth about 1.2 years. Climbing from 34 to 44 is worth about 0.6. The neutral value is the same for every age and sex, because Titan Age starts from your actual age and adjusts from there. The age- and sex-specific rating bar in the expanded row is for context only and does not change the years. Titan smooths each week against the last, so a single new reading moves your displayed Titan Age only part of the way. If no reading exists in the 30-day window, Titan fills the row from the median of your recorded weeks in the past 26, once you have 6, or from the neutral value. The Fitness in Titan Age article explains how to find a missing reading.

VO2 max plays no part in Titan's daily Recovery score, which reads your overnight HRV and sleeping heart rate.

Because the watch tends to underestimate fit users, a lower-than-expected number does not always mean lower fitness. Watch the direction of your own readings over months, keep your Health details and medications current, and remember that Titan Age reads whatever value Apple Health holds.

09References

• Apple Inc. (2021). Using Apple Watch to estimate cardio fitness with VO2 max. Apple Healthcare white paper. https://www.apple.com/healthcare/docs/site/UsingAppleWatchtoEstimateCardioFitnesswithVO2_max.pdf
• Apple Inc. About cardio fitness on Apple Watch. Apple Support, accessed September 2026. https://support.apple.com/en-us/108790
• Bassett DR, Howley ET (2000). Limiting factors for maximum oxygen uptake and determinants of endurance performance. Medicine and Science in Sports and Exercise 32(1):70-84. https://doi.org/10.1097/00005768-200001000-00012
• Bouchard C et al. (1999). Familial aggregation of VO2max response to exercise training, results from the HERITAGE Family Study. Journal of Applied Physiology 87(3):1003-1008. https://doi.org/10.1152/jappl.1999.87.3.1003
• Caserman P et al. (2024). Assessing the accuracy of smartwatch-based estimation of maximum oxygen uptake using the Apple Watch Series 7, validation study. JMIR Biomedical Engineering 9:e59459. https://doi.org/10.2196/59459
• Coyle EF et al. (1984). Time course of loss of adaptations after stopping prolonged intense endurance training. Journal of Applied Physiology 57(6):1857-1864. https://doi.org/10.1152/jappl.1984.57.6.1857
• Fleg JL et al. (2005). Accelerated longitudinal decline of aerobic capacity in healthy older adults. Circulation 112(5):674-682. https://doi.org/10.1161/CIRCULATIONAHA.105.545459
• Helgerud J et al. (2007). Aerobic high-intensity intervals improve VO2max more than moderate training. Medicine and Science in Sports and Exercise 39(4):665-671. https://doi.org/10.1249/mss.0b013e3180304570
• Kaminsky LA et al. (2015). Reference standards for cardiorespiratory fitness measured with cardiopulmonary exercise testing, data from the Fitness Registry and the Importance of Exercise National Database. Mayo Clinic Proceedings 90(11):1515-1523. https://doi.org/10.1016/j.mayocp.2015.07.026
• Kaminsky LA et al. (2022). Updated reference standards for cardiorespiratory fitness measured with cardiopulmonary exercise testing, data from the Fitness Registry and the Importance of Exercise National Database (FRIEND). Mayo Clinic Proceedings 97(2):285-293. https://doi.org/10.1016/j.mayocp.2021.08.020
• Kline GM et al. (1987). Estimation of VO2max from a one-mile track walk, gender, age, and body weight. Medicine and Science in Sports and Exercise 19(3):253-259. https://doi.org/10.1249/00005768-198706000-00012
• Kodama S et al. (2009). Cardiorespiratory fitness as a quantitative predictor of all-cause mortality and cardiovascular events in healthy men and women, a meta-analysis. JAMA 301(19):2024-2035. https://doi.org/10.1001/jama.2009.681
• Kokkinos P et al. (2022). Cardiorespiratory fitness and mortality risk across the spectra of age, race, and sex. Journal of the American College of Cardiology 80(6):598-609. https://doi.org/10.1016/j.jacc.2022.05.031
• Lambe R et al. (2025). Investigating the accuracy of Apple Watch VO2 max measurements, a validation study. PLOS One 20(5):e0323741. https://doi.org/10.1371/journal.pone.0323741
• Lambe R et al. (2026). Accuracy of VO2 max estimates from Apple Watch Series 10. Mayo Clinic Proceedings: Digital Health 4(2):100357. https://doi.org/10.1016/j.mcpdig.2026.100357
• Laukkanen JA et al. (2016). Long-term change in cardiorespiratory fitness and all-cause mortality, a population-based follow-up study. Mayo Clinic Proceedings 91(9):1183-1188. https://doi.org/10.1016/j.mayocp.2016.05.014
• Mandsager K et al. (2018). Association of cardiorespiratory fitness with long-term mortality among adults undergoing exercise treadmill testing. JAMA Network Open 1(6):e183605. https://doi.org/10.1001/jamanetworkopen.2018.3605
• Milanović Z, Sporiš G, Weston M (2015). Effectiveness of high-intensity interval training (HIT) and continuous endurance training for VO2max improvements, a systematic review and meta-analysis of controlled trials. Sports Medicine 45(10):1469-1481. https://doi.org/10.1007/s40279-015-0365-0
• Molina-Garcia P et al. (2022). Validity of estimating the maximal oxygen consumption by consumer wearables, a systematic review with meta-analysis and expert statement of the INTERLIVE network. Sports Medicine 52(7):1577-1597. https://doi.org/10.1007/s40279-021-01639-y
• Montero D, Lundby C (2017). Refuting the myth of non-response to exercise training, "non-responders" do respond to higher dose of training. Journal of Physiology 595(11):3377-3387. https://doi.org/10.1113/JP273480
• Paterson DH et al. (1999). Aerobic fitness in a population of independently living men and women aged 55-86 years. Medicine and Science in Sports and Exercise 31(12):1813-1820. https://doi.org/10.1097/00005768-199912000-00018
• Poole DC, Jones AM (2017). Measurement of the maximum oxygen uptake VO2max, VO2peak is no longer acceptable. Journal of Applied Physiology 122(4):997-1002. https://doi.org/10.1152/japplphysiol.01063.2016
• Ross R et al. (2016). 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 134(24):e653-e699. https://doi.org/10.1161/CIR.0000000000000461
• Uth N et al. (2004). Estimation of VO2max from the ratio between HRmax and HRrest, the Heart Rate Ratio Method. European Journal of Applied Physiology 91(1):111-115. https://doi.org/10.1007/s00421-003-0988-y