Predicting Heart Rate Variability Parameters in Healthy Korean Adults: A Preliminary Study

Area of Science:

• Physiology
• Biostatistics
• Cardiovascular Health

Background:

• Heart rate variability (HRV) is a crucial indicator of autonomic nervous system function and cardiovascular health.
• Traditional HRV assessment often requires specialized equipment, limiting accessibility in certain settings.
• Developing predictive models using easily measurable variables can enhance HRV monitoring capabilities.

Purpose of the Study:

• To develop a multiple linear regression model for estimating heart rate variability (HRV) parameters.
• To utilize readily available independent variables (e.g., sex, age, anthropometrics, heart rate) for HRV prediction.
• To assess the model's efficacy in estimating both time-domain and frequency-domain HRV parameters.

Main Methods:

• A multiple linear regression model was constructed using the backward elimination technique.
• Seventy-five healthy adults provided data for HRV parameters and independent variables.
• Key HRV metrics including SDNN, RMSSD, NN50, pNN50, TP, VLF, LF, and HF were analyzed.

Main Results:

• The regression model demonstrated a high coefficient of determination (adjusted R²: 69.8%-92.1%) for time-domain HRV parameters (SDNN, RMSSD, NN50, pNN50).
• Moderate coefficients of determination (adjusted R²: 40.3%-72.6%) were observed for frequency-domain HRV parameters (TP, VLF, LF, HF).
• The model significantly predicted HRV parameters (P < .001) using the selected independent variables.

Conclusions:

• Simple physiological variables can effectively predict heart rate variability (HRV) parameters.
• The developed multiple linear regression model offers a viable method for estimating time-domain HRV.
• This approach complements traditional HRV measurement methods, potentially improving accessibility and application in diverse settings.