The primary aim of this study was to assess the validity of VO2max estimations from Apple Watch compared to indirect calorimetry, with a secondary aim to evaluate the accuracy of heart rate and energy expenditure estimations.
Thirty recreationally active adults wore an Apple Watch for a period of 5-10 days to generate an estimated VO2max value. Subsequently, each participant underwent a maximal exercise treadmill test, in accordance with the modified Astrand protocol. Reference standard VO2max and energy expenditure measurements were obtained using indirect calorimetry, while the Polar H10 served as the criterion measure for heart rate. Apple Watch heart rate and energy expenditure values were recorded during exercise testing. Bland-Altman analyses were conducted to evaluate the agreement between Apple Watch and reference standard measurements. Mean differences, mean absolute errors (MAE), and mean absolute average percentage errors (MAPE) were calculated with corresponding 95% confidence intervals.
Apple Watch underestimated VO2max in comparison to indirect calorimetry, with a mean difference of 5.51 ml/kg/min (SD 1.19; [95% CI 3.18 to 7.85). Bland-Altman limits of agreement (LoA) indicated variability between measurement methods (lower LoA: -7.27; upper LoA: 18.30). The MAPE was 13.14% (95% CI 10.6 to 16.11) and the MAE was 6.82 ml/min/kg (95% CI 5.00 to 8.64). Apple Watch also underestimated energy expenditure (mean difference: 7.5 kcal [SD 5.62; 95% CI -3.52 to 18.52]). Wide limits of agreement were observed (lower LoA: -52.84; upper LoA: 67.84). The MAPE was 13.97% (95% CI -3.51 to 18.52), and the MAE was 24.5 kcal (95% CI 17.17 to 31.83). Heart rate measurements demonstrated the closest agreement among the variables, with a mean difference of 1.1 bpm (95% CI 0.76 to 1.43).
Apple Watch tended to underestimate both VO2max and energy expenditure compared to indirect calorimetry. Significant variability was evident for energy expenditure particularly. However, heart rate measurements demonstrated close agreement with criterion measures. Future research should endeavour to validate modern devices among clinical populations to investigate its capability for cardiorespiratory fitness assessment.
While Apple Watch provides accurate heart rate measurements, its VO2max and energy expenditure estimates require further refinement before they can be used in clinical practice. This emphasises the importance of continuously validating wearable devices as their measurement capabilities develop and healthcare professionals consider their use.
Physiological measurement
Cardiorespiratory fitness
