Individual calibration of accelerometers in children and their health-related implications

This study compared children's physical activity (PA) levels, the prevalence of children meeting current guidelines of ≥60 minutes of daily moderate to vigorous PA (MVPA), and PA-health associations using individually calibrated (IC) and empirical accelerometer cutpoints. Data from 75 (n = 32 boys) 10-12 year old children were included in this study. Clustered cardiometabolic (CM) risk, directly measured cardiorespiratory fitness (CRF), anthropometric and 7 day accelerometer data were included within analysis. PA data were classified using Froude anchored IC, Evenson et al. (Evenson, K. R., Catellier, D. J., Gill, K., Ondrak, K. S., & McMurray, R. G. (2008). Calibration of two objective measures of physical activity for children. Journal of Sports Sciences, 26(14), 1557-1565. doi:10.1080/02640410802334196) (Ev) and Mackintosh et al. (Mackintosh, K. A., Fairclough, S. J., Stratton, G., & Ridgers, N. D. (2012). A calibration protocol for population-specific accelerometer cutpoints in children. PLoS One, 7(5), e36919. doi:10.1371/journal.pone.0036919) (Mack) cutpoints. The proportion of the cohort meeting ≥60mins MVPA/day ranged from 37%-56% depending on the cutpoints used. Reported PA differed significantly across the cutpoint sets. IC LPA and MPA were predictors of CRF (LPA: standardised β = 0.32, p = 0.002, MPA: standardised β = 0.27 p = 0.013). IC MPA also predicted BMI Z-score (standardised β = -0.35, p = 0.004). Ev VPA was a predictor of BMI Z-score (standardised β = -0.33, p = 0.012). Cutpoint choice has a substantial impact on reported PA levels though no significant associations with CM risk were observed. Froude IC cutpoints represent a promising approach towards classifying children's PA data.