Prospective association between descriptive accelerometer-derived physical behaviour metrics and cardiometabolic risk...

To better understand which aspects of physical behavior are most critical for health, this study investigated prospective associations between various accelerometer-derived physical activity (PA) and sedentary behaviour (SB) metrics with cardiometabolic risk indicators. We analyzed waist-worn accelerometer data from 114 Dutch children (50 girls), age 8.6 ± 0.4 years, in the Amsterdam Born Children and Development (ABCD) cohort. Physical behavior metrics included volume-based metric (counts per minute [CPM]), time-use metrics (e.g., time spent in different physical behaviour intensities), pattern-based metrics (e.g., fragmentation indices), and intensity gradient. Health indicators, assessed at age 11-12 years, included BMI-standard deviation score (SDS) and a cardiometabolic risk score (MetScore) calculated from BMI, systolic and diastolic blood pressure, high-density lipoprotein, total cholesterol, and triglycerides. Regression analyses were conducted to examine associations between physical behaviour metrics and both BMI-SDS and MetScore. Linear regression analyses, adjusted for gender and wear time, revealed that more time spent in moderate-to-vigorous physical activity (MVPA), a conventional time-use metric, was significantly associated with lower BMI-SDS (β = -0.02; 95 % CI: 0.03, 0), but not with MetScore. Other physical behavior metrics, including volume-based, pattern-based, and intensity gradient measures, did not show significant associations with either BMI-SDS or MetScore in this cohort. These findings indicate that conventional time-use metrics, particularly MVPA, may be more effective in detecting health associations in small homogenous populations. Future research with larger populations is necessary to determine if alternative metrics provide additional insight into the association between physical behavior and cardiometabolic health.