JMIR Publications

ABSTRACT

Background:

Accurate, continuous heart rate (HR) measurements are important for health assessment, physical activity and sporting performance and the integration of HR to wearable devices has extended its accessibility. Whilst the use of photoplethysmography (PPG) technology is not new, the available data relating to the validity of measurement are limited and the range of activities being performed is often restricted to one exercise domain and/or limited intensities.

Objective:

The aim of the present study was to assess the validity of the Polar OH1 and Fitbit Charge 3 devices for measuring HR during rest, light, moderate, vigorous and sprint-type exercise.

Methods:

Twenty healthy adults (9 female; height: 1.73 ± 0.1 m; body mass: 71.6 ± 11.0 kg; Age: 40 ± 10 years) volunteered and provided written informed consent to participate in the study consisting of two trials. Trial 1 was split into 3 components: 1) 15 minutes sedentary/resting activities; 2) 10 min cycle on a bicycle ergometer; and 3) Incremental exercise test to exhaustion on a motorised treadmill (18-42 min). Trial 2 was split into 2 components: 1) 4 × 15s supramaximal sprints on a cycle ergometer; 2) 4 × 30-50 m sprints on a non-motorised resistance treadmill. Data from the three devices were time-aligned and validity of the Polar OH1 and Fitbit Charge 3 were assessed against the Polar H10 (criterion device). Validity was evaluated using Pearson moment correlation coefficient, Bland and Altman analysis and the mean absolute percentage error (MAPE).

Results:

Overall, there was only a small mean bias and very good correlation between the Polar OH1 and the Polar H10 device (bias = -1 beats·min-1, r = 0.95, P <.001). However, the limits of agreement (LoA) were wide (-20, 19). The Fitbit Charge 3 device underestimated HR by 7 beats·min-1 compared to the Polar H10, with large LoA and poor correlation (-46, 33, r = 0.8, P <.001). The MAPE for both devices were deemed acceptable (<5%). The Polar OH1 performed well across each phase of trial 1, but validity was poor for trial 2 (r < 0.8). The Fitbit Charge 3 only performed well during rest and non-sprint-based treadmill activities.

Conclusions:

Compared with our criterion device, the Polar OH1 was accurate at assessing HR, but the accuracy of the Fitbit Charge 3 was generally poor. The Polar OH1 performed worse during trial 2 compared to the activities in trial 1 and validity of the Fitbit Charge 3 device was particularly poor during cycling exercise.