Alaa Khushhal · Sports Medicine International Open 2017 · Cross-sectional validation and reliability study · n=21

Validity and Reliability of the Apple Watch for Measuring Heart Rate During Exercise

Cited 95 times in the scientific literature.

Level 3 - non-randomized controlled study

Validation study comparing a consumer device to a criterion reference standard across controlled exercise conditions.

OpenAlex W2766894515 · doi:10.1055/s-0043-120195 · record verified 2026-08-29

What was done

Twenty-one males performed treadmill exercise while wearing an Apple Watch on each wrist and a Polar S810i heart rate monitor as the criterion standard. Participants completed 5-minute exercise bouts consisting of walking (4 km/h), jogging (7 km/h), and running (10 km/h), with 11-minute rest intervals between bouts to evaluate validity and inter-device reliability during exercise and recovery.

What was found

Mean bias was trivial across all exercise intensities. Correlation with the criterion was very good during walking (left: r = 0.97; right: r = 0.97), good during jogging (left: r = 0.93; right: r = 0.92), and poor/good during running (left: r = 0.81; right: r = 0.86). Standardized typical error of the estimate was small during walking, moderate during jogging, and moderate to large during running. Recovery correlations were good after walking but poor after jogging and running. The percentage of heart rate data points recorded decreased as exercise intensity increased. Inter-device standardized typical errors were small to moderate with very good to nearly perfect intraclass correlations.

Why it matters

This study shows that while the Apple Watch provides valid heart rate monitoring at lower movement intensities, its accuracy and data acquisition reliability diminish at higher running speeds and during recovery.

Limits

The study evaluated a small sample of 21 participants and included only males. Exercise testing was limited to three fixed treadmill speeds up to 10 km/h, leaving higher speeds, variable outdoor running, and non-running activities unassessed. The abstract does not specify participant demographics such as skin tone or the specific generation of Apple Watch tested.

Cited by