Establishing Wrist-Based Cutpoints for the Actical Accelerometer in Elementary School-Aged Children

Abstract

The wrist has become a standard location for accelerometry (ACC) data collection, primarily to optimize compliance, yet interpretation of wrist ACC data is limited due to a lack of calibration studies. This study aimed to establish cutpoints for a wrist-mounted Actical accelerometer in 6- to 11-year-old children using 2 methods. Metabolic and ACC data (15-sec epoch) were collected during 8 activities in 22 children ages 6-11. Linear regression (LR) and Receiver Operator Characteristics (ROC) were used to examine the relationship between METs and ACC counts. Cutpoints were established at < 1.5, 1.5-2.99, 3-5.99, and ≥ 6 METs for sedentary, light, moderate, and vigorous activity, respectively. Cutpoints were applied to a large, multiday sample of children (n = 269) to examine differences in cutpoints on minutes of moderate to vigorous PA (MVPA). LR and ROC yielded moderate cutpoints of 574 and 388, respectively. When applied to the large sample, LR and ROC cutpoints resulted in an estimated 83 and 140 minutes of daily MVPA, respectively. This study established wrist-mounted Actical cutpoints for children using 2 methods. The differences in cutpoints and their effect on estimates of MVPA in an independent sample highlight challenges associated with establishing cutpoints, suggesting that standardized calibration procedures be developed.