Possibilities and limitations of the polar RS800 in measuring heart rate variability at rest

A growing trend among clinical studies is the use of heart rate monitors (HRMs) for assessment of heart rate variability (HRV). These instruments offer a convenient alternative to traditional electrocardiographs (ECGs) for recording and processing of R-R data. Reports on the validity of such systems are, however, conflicting. This study aimed to assess the validity of a commercial HRM on a large study sample, with emphasis on gender and age. Simultaneous recordings of R-R intervals were conducted with the Polar RS800 HRM and a 3-lead ECG on 341 individuals. Data editing was performed with individually designated software for each instrument. Agreement on SDNN, RMSSD, and HF- and LF power was assessed with intraclass correlations (ICCs), standard errors of measurement (SEMs) and Bland and Altman plots. The HRM was not able to identify 18 observations with non-sinus beats. For men, agreement between instruments ranged from good to excellent (ICC >= 0.8) on all HRV measures, and SEMs were generally small. For women the results were weaker, with unacceptable agreement between instruments on SDNN. Women over 60 years did not reach a critical ICC value of 0.75 on any of the HRV measures. Bland and Altman plots demonstrated that the RS800 generally overestimated HRV, and that uncertainty increased with higher values. Since the Polar system did not identify errors satisfactorily, or return valid values of HRV for certain groups, it is concluded that, whenever possible, traditional ECGs should be used for both gathering and editing of HRV data.