The purpose of this study was to see if there were any correlations between linear or nonlinear HRV parameters and physical fatigue scores under physical workload.
Fifteen construction workers who were at least 18 years old (mean age, 32.3 and SD, 3.2Y), were chosen to participate in the study. The HRV was monitored using a textile-based wearable sensor during the 60 minutes of physical workload. At baseline, 15, 30, 45, and 60 minutes of the task, subjects were administered the Borg-20 rating of perceived exertion to determine their subjective levels of fatigue. Parameters of HRV were derived from both the time- and frequency-domains, as well as from a nonlinear approach. Additionally, blood lactate levels were collected at baseline and at the end of physical workload task to objectively assess physical fatigue. The correlations between linear or non-linear metrics and fatigue scores, as well as between linear and non-linear metrics alone, were analyzed using Pearson's correlations coefficient. The important predictor of physical fatigue while doing physical workload tasks was determined using multiple linear regression analysis.
Several time-domain (r = 0.27 to 0.70), frequency-domain (r = 0.37 to 0.68), and nonlinear metrics (r = 0.38 to 0.69) were significantly correlated with physical fatigue during physical workload task (p ≤ 0.05). Furthermore, most of the nonlinear variables were strongly associated to variables in both the time domain and the frequency domain (p ≤ 0.05). Multiple linear regression analysis showed that several HRV variables, such as pNN50 (count), HFperc (%), LF/HF ratio, VLF (Hz), SD2 (ms), and SDNN (ms), were independently associated to physical fatigue scores (r2 = 64.5, P ≤ 0.05).
The findings demonstrate the feasibility of evaluating physical fatigue by analyzing changes in HRV parameters. The responses of HRV indices were related to fatigue levels during physical workload. The nonlinear parameters may be substituted for the time and frequency domains to quantify physical fatigue during physical workload. Additionally, the results indicated that the combination of linear and nonlinear HRV analysis metrics yielded the highest accuracy in predicting outcomes.
We observed that physical fatigue caused major changes in HRV parameters, indicating a change in the relative activity of the parasympathetic and sympathetic nervous systems in relation to physical fatigue. Variations in HRV values during intensive workloads can function as digital biomarkers to distinguish between individuals with and without cardiovascular illnesses.
Heart rate variability
Workplace safety
