Validity concerns of common heart-rate variability indices

Time- and frequency-domain analyses of HRV have provided researchers with important measures of cardiac vagal activity. Stationarity is of theoretical importance for such analyses in the frequency domain but may not be of practical significance in any particular data set. It has been argued that if a stationarity test is available, it should be used. On the other hand, it is also possible that the RSA is quite robust to nonstationarity; the spectral data support that contention. Furthermore, procedures that correct for nonstationary data segments may compromise the representativeness of the data set. With regard to the time domain, MSD is advantageous in that it is conceptually and computationally simple, does not require respiration data, and under many conditions may be a reasonable alternative to HF spectral power or RSA. All told, the selection of cardiac vagal control indices and use of correction procedures should be based upon experimental situation and availability to the researcher, rather than orthodox adherence to idealized standards.