[Skip to Content]
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
[Skip to Content Landing]
Figure.
Bland-Altman Plots and 95% Limits of Agreement With Electrocardiographic Measured Heart Rate (HR)
Bland-Altman Plots and 95% Limits of Agreement With Electrocardiographic Measured Heart Rate (HR)
Table.  
Concordance Correlation Coefficients for Each Heart Rate Monitor
Concordance Correlation Coefficients for Each Heart Rate Monitor
1.
Case  MA, Burwick  HA, Volpp  KG, Patel  MS.  Accuracy of smartphone applications and wearable devices for tracking physical activity data.  JAMA. 2015;313(6):625-626.PubMedGoogle ScholarCrossref
2.
Murakami  H, Kawakami  R, Nakae  S,  et al.  Accuracy of wearable devices for estimating total energy expenditure: comparison with metabolic chamber and doubly labeled water method.  JAMA Intern Med. 2016;176(5):702-703.PubMedGoogle ScholarCrossref
3.
Laukkanen  RMT, Virtanen  PK.  Heart rate monitors: state of the art.  J Sports Sci. 1998;16(suppl):S3-S7.PubMedGoogle ScholarCrossref
4.
Terbizan  DJ, Dolezal  BA, Albano  C.  Validity of seven commercially available heart rate monitors.  Meas Phys Educ Exerc Sci. 2002;6:243-247.Google ScholarCrossref
5.
El-Amrawy  F, Nounou  MI.  Are currently available wearable devices for activity tracking and heart rate monitoring accurate, precise, and medically beneficial?  Healthc Inform Res. 2015;21(4):315-320.PubMedGoogle ScholarCrossref
6.
Patel  MS, Asch  DA, Volpp  KG.  Wearable devices as facilitators, not drivers, of health behavior change.  JAMA. 2015;313(5):459-460.PubMedGoogle ScholarCrossref
Research Letter
January 2017

Accuracy of Wrist-Worn Heart Rate Monitors

Author Affiliations
  • 1The Heart and Vascular Institute, Cleveland Clinic, Cleveland, Ohio
JAMA Cardiol. 2017;2(1):104-106. doi:10.1001/jamacardio.2016.3340

Wrist-worn fitness and heart rate (HR) monitors are popular.1,2 While the accuracy of chest strap, electrode-based HR monitors has been confirmed,3,4 the accuracy of wrist-worn, optically based HR monitors is uncertain.5,6 Assessment of the monitors’ accuracy is important for individuals who use them to guide their physical activity and for physicians to whom these individuals report HR readings. The objective of this study was to assess the accuracy of 4 popular wrist-worn HR monitors under conditions of varying physical exertion.

Methods

This study recruited 50 healthy adults. The mean (SD) age was 37 (11.3) years; the mean (SD) body mass index (calculated as weight in kilograms divided by height in meters squared) was 23.5 (3.1); 28 participants were women (58%), and 7 participants were African American (14%). Exclusion criteria included cardiovascular disease, pacemakers, and treatment with heart rhythm medications. The protocol was approved by the Cleveland Clinic institutional review board. Participants provided written informed consent.

Participants wore standard electrocardiographic limb leads and a Polar H7 chest strap monitor secured tightly to ensure skin contact. Each participant was randomly assigned to wear 2 different wrist-worn HR monitors placed tightly above the ulnar styloid. Four wrist-worn monitors were assessed: Fitbit Charge HR (Fitbit), Apple Watch (Apple), Mio Alpha (Mio Global), and Basis Peak (Basis).

Heart rate was assessed with the participant on a treadmill at rest and at 2 mph, 3 mph, 4 mph, 5 mph, and 6 mph. Participants exercised at each setting for 3 minutes to achieve a steady state; HR was recorded instantaneously at the 3-minute point. After completion of the treadmill protocol, HR was recorded at 30, 60, and 90 seconds’ recovery.

Results

Across all devices, 1773 HR values were recorded. Twenty-seven data points were not obtained because of inability to complete the treadmill protocol or monitors losing skin contact. Heart rate ranged from 49 bpm to 200 bpm.

When compared with electrocardiogram, the HR monitors had variable accuracy as assessed by the concordance correlation coefficient (rc) (Table). While the Basis Peak overestimated HR during moderate exercise, with median differences of −8.9 and −7.3 bpm at 2 mph (P < .001) and 3 mph (P = .001), respectively, the Fitbit Charge HR underestimated HR during more vigorous exercise, with median differences of 7.2 and 6.4 bpm at 4 mph (P < .001) and 6 mph (P < .001), respectively (Wilcoxon signed-rank test comparing each device with electrocardiogram).

Bland-Altman analysis showed that variability occurred across the spectrum of midrange HRs during exercise, with less variability at the tail ends (Figure). The Apple Watch and Mio Fuse had 95% of differences fall within −27 bpm and +29 bpm of the electrocardiogram, while Fitbit Charge HR had 95% of values within −34 bpm and +39 bpm and the corresponding values for the Basis Peak were within −39 bpm and +33 bpm.

Body mass index, age, and sex did not influence monitor accuracy.

Discussion

We found variable accuracy among wrist-worn HR monitors; none achieved the accuracy of a chest strap–based monitor. In general, accuracy of wrist-worn monitors was best at rest and diminished with exercise.

This study is limited by examination of a convenience sample of 4 HR monitors in young, healthy adults exercising on a treadmill. Results should be confirmed with different types of exercise and with other devices. Continuous HR assessment, which is currently not feasible with all devices, would enable more detailed comparisons.

Electrode-containing chest monitors should be used when accurate HR measurement is imperative. While wrist-worn HR monitors are often used recreationally to track fitness, their accuracy varies; 2 of 4 monitors had suboptimal accuracy during moderate exercise. Because cardiac patients increasingly rely on these monitors to stay within physician-recommended, safe HR thresholds during rehabilitation and exercise, appropriate validation of these devices in this group is imperative.

Back to top
Article Information

Corresponding Author: Marc Gillinov, MD, Department of Thoracic and Cardiovascular Surgery, Cleveland Clinic, Desk J4-1, Cleveland, OH 44195 (gillinom@ccf.org).

Published Online: October 12, 2016. doi:10.1001/jamacardio.2016.3340

Author Contributions: Drs M. Gillinov and Wang had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Wang, Blackburn, Desai, Phelan, L. Gillinov, M. Gillinov.

Acquisition, analysis, or interpretation of data: Blackburn, Desai, Phelan, L. Gillinov, Houghtaling, M. Gillinov.

Drafting of the manuscript: Wang, Blackburn, Desai, L. Gillinov, Houghtaling, M. Gillinov.

Critical revision of the manuscript for important intellectual content: Blackburn, Desai, Phelan, M. Gillinov.

Statistical analysis: Desai, Houghtaling.

Administrative, technical, or material support: Wang.

Study supervision: Wang, Blackburn, Desai, Phelan, M. Gillinov.

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.

Funding/Support: The research was supported by The Mary Elizabeth Holdsworth Fund at the Cleveland Clinic.

Role of the Funder/Sponsor: The Mary Elizabeth Holdsworth Fund had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

References
1.
Case  MA, Burwick  HA, Volpp  KG, Patel  MS.  Accuracy of smartphone applications and wearable devices for tracking physical activity data.  JAMA. 2015;313(6):625-626.PubMedGoogle ScholarCrossref
2.
Murakami  H, Kawakami  R, Nakae  S,  et al.  Accuracy of wearable devices for estimating total energy expenditure: comparison with metabolic chamber and doubly labeled water method.  JAMA Intern Med. 2016;176(5):702-703.PubMedGoogle ScholarCrossref
3.
Laukkanen  RMT, Virtanen  PK.  Heart rate monitors: state of the art.  J Sports Sci. 1998;16(suppl):S3-S7.PubMedGoogle ScholarCrossref
4.
Terbizan  DJ, Dolezal  BA, Albano  C.  Validity of seven commercially available heart rate monitors.  Meas Phys Educ Exerc Sci. 2002;6:243-247.Google ScholarCrossref
5.
El-Amrawy  F, Nounou  MI.  Are currently available wearable devices for activity tracking and heart rate monitoring accurate, precise, and medically beneficial?  Healthc Inform Res. 2015;21(4):315-320.PubMedGoogle ScholarCrossref
6.
Patel  MS, Asch  DA, Volpp  KG.  Wearable devices as facilitators, not drivers, of health behavior change.  JAMA. 2015;313(5):459-460.PubMedGoogle ScholarCrossref
×