VO2max as a Digital Biomarker

VO2max is the maximum rate at which the body can take up and use oxygen during exercise, expressed in millilitres per kilogram per minute. It is the standard index of cardiorespiratory fitness, and cardiorespiratory fitness is among the strongest predictors of mortality available, which is why a professional society has argued for treating it as a clinical vital sign.

True measurement requires a graded exercise test to volitional exhaustion with expired gas analysis, in a laboratory, with staff present. That is accurate and impractical at scale, which is why estimation has a long history: submaximal protocols and nomograms for predicting fitness from exercise test performance date back decades.

Consumer wearables extend that estimation chain further. They infer VO2max from the relationship between heart rate and running or walking speed during ordinary activity, without any test at all.

The reference method is cardiopulmonary exercise testing. Reference standards derived from large registries of directly measured tests allow an individual value to be placed against age and sex matched norms, which is what makes the number interpretable.

Wearable estimation works from submaximal data. During activity the device pairs heart rate with external workload, usually pace derived from satellite positioning, models the heart rate to workload relationship, and extrapolates to the intensity that would correspond to maximal effort. Accuracy therefore depends on the accuracy of heart rate, of pace, and of an assumed maximum heart rate that is itself usually predicted from age.

Systematic review with meta-analysis of consumer wearable estimates against laboratory testing has quantified how much error that chain accumulates.

In research, fitness is used as an outcome in exercise and cardiac rehabilitation trials, as a prognostic stratifier in cardiovascular and oncology cohorts, and as a covariate in studies where fitness confounds the relationship of interest.

The practical attraction of the wearable estimate is coverage. A laboratory test gives one precise value at one timepoint; a wearable gives a rough value every week for years, and for questions about trajectory that trade can be worth making.

The measure sits alongside supervised exercise tests in this library, including the Bruce treadmill protocol and the Harvard step test, which are the graded protocols estimation was originally built on, and alongside the Duke Activity Status Index, which asks the person what activities they can manage and produces its own functional capacity estimate.

Wearable estimates carry substantial error against laboratory measurement, and the error is systematic rather than random in some groups: estimation assumes a running or walking relationship between heart rate and workload, so it degrades in people who cannot exercise at the required intensity, which includes most clinical populations of interest.

Values are not comparable between devices, since each manufacturer uses a different proprietary model, and they shift with firmware updates.

The dependence on predicted maximum heart rate is a further weakness, because age based prediction has wide individual variation and is distorted by rate limiting medication such as beta blockers. In a cardiology population, that is not an edge case. Wearable VO2max is best read as a within person fitness index, not as a physiological measurement.

• Ross R, et al. Importance of assessing cardiorespiratory fitness in clinical practice: a case for fitness as a clinical vital sign. Circulation. 2016. pubmed.ncbi.nlm.nih.gov
• Kaminsky LA, et al. Updated reference standards for cardiorespiratory fitness measured with cardiopulmonary exercise testing. Mayo Clin Proc. 2022. pubmed.ncbi.nlm.nih.gov
• Molina-Garcia P, et al. Validity of estimating the maximal oxygen consumption by consumer wearables: a systematic review with meta-analysis. Sports Med. 2022. pubmed.ncbi.nlm.nih.gov
• Bruce RA, et al. Maximal oxygen intake and nomographic assessment of functional aerobic impairment in cardiovascular disease. Am Heart J. 1973. pubmed.ncbi.nlm.nih.gov