Inter- and Intra-Day Comparisons of Smartphone-Derived Heart Rate Variability across Resistance Training Overload and Taper Microcycles

The purposes of this study were: (1) to determine if smartphone-derived heart rate variability (HRV) could detect changes in training load during an overload microcycle and taper, and (2) to determine the reliability of HRV measured in the morning and measured immediately prior to the testing session. Twelve powerlifters (male = 10, female = 2) completed a 3-week resistance training program consisting of an introduction microcycle, overload microcycle, and taper. Using a validated smartphone application, daily measures of resting, ultra-short natural logarithm of root mean square of successive differences were recorded in the morning (LnRMSSD(M)) and immediately before the test session (LnRMSSD(T)) following baseline, post-overload, and post-taper testing. LnRMSSD(M) decreased from baseline (82.9 ± 13.0) to post-overload (75.0 ± 9.9, p = 0.019), while post-taper LnRMSSD(M) (81.9 ± 7.1) was not different from post-overload (p = 0.056) or baseline (p = 0.998). No differences in LnRMSSD(T) (p < 0.05) were observed between baseline (78.3 ± 9.0), post-overload (74.4 ± 10.2), and post-taper (78.3 ± 8.0). LnRMSSD(M) and LnRMSSD(T) were strongly correlated at baseline (ICC = 0.71, p < 0.001) and post-overload (ICC = 0.65, p = 0.010), whereas there was no relationship at post-taper (ICC = 0.44, p = 0.054). Bland–Altman analyses suggest extremely wide limits of agreement (Bias ± 1.96 SD) between LnRMSSD(M) and LnRMSSD(T) at baseline (4.7 ± 15.2), post-overload (0.5 ± 16.9), and post-taper (3.7 ± 15.3). Smartphone-derived HRV, recorded upon waking, was sensitive to resistance training loads across an overload and taper microcycles in competitive strength athletes, whereas the HRV was taken immediately prior to the testing session was not.