Covariation and reliability of ECG parameters during 24-hour monitoring

Although diverse ECG parameters are used in psychophysiology, relatively little is known of the covariation and reliability of these parameters. In the present study with 31 male cardiac patients, 24-h ambulatory ECG monitoring was used to describe the covariation and reliability of 18 ECG parameters. Analysis of the ECG parameters was done beat-by-beat and averaged on a 5-min basis. To describe different aspects of covariation, partitioning of covariance was used, yielding within-subjects, within-time periods, and residual correlations. Moreover, reliability was assessed for different time lags between 5 min and 22 h. Results indicate that the ECG parameters are only partly redundant, since several correlation clusters emerged with low correlations between these clusters. In order to utilize more information from the ECG, at least heart rate, heart rate variability, ST-amplitude, QRS-duration, and one of the main ECG waves (P-, R-, T-wave) would be necessary. Reliability was high for all parameters except heart rate variability measures.