Changes in Japanese heart rate variability over time as seen in ALLSTAR big data

Abstract

Holter ECG is the most popular and successful wearable sensor, which records continuous ECG under daily activities. Many parameters, including heart rate variability (HRV), have been proposed to extract the various features contained in Holter ECG signals, but there would still be a lot of unused information. In 2010, we started the Holter ECG database construction project called the Allostatic State Mapping by Ambulatory ECG Repository (ALLSTAR) to mine and utilize the information of Holter ECG. So far, more than 600,000 Holter ECG data have been registered from all over Japan. Since the data collection period exceeded 10 years, we analyzed the changes in Japanese HRV over time in the past 10 years. After adjusting for the effects of age and season (month) and place (prefecture) of recoding, we observed slight but progressive shortening of mean N-N interval, increase in low-frequency power, and decrease in high-frequency power in both sexes.