Identification and validity of accelerometer cut-points for toddlers.

Area of Science:

• Pediatric physical activity monitoring
• Biomedical engineering
• Exercise science

Background:

• Accurate measurement of physical activity (PA) intensity in toddlers is crucial for understanding health behaviors.
• Existing ActiGraph cut-points, often derived from older children or adults, may not accurately reflect PA intensity in toddlers.
• The Children's Activity Rating Scale (CARS) provides a validated method for coding PA intensity from video.

Purpose of the Study:

• To derive and validate ActiGraph cut-points for sedentary (SED), light-intensity physical activity (LPA), and moderate-to-vigorous physical activity (MVPA) specifically for toddlers.
• To compare the predictive validity of these new toddler cut-points with established preschool cut-points.

Main Methods:

• Twenty-two toddlers wore ActiGraph accelerometers during a 20-minute videotaped play session.
• Videos were coded for PA intensity using the modified CARS.
• Receiver operating characteristic (ROC) curve analyses determined optimal cut-points, which were then validated in an independent sample of 18 toddlers.

Main Results:

• Derived cut-points for 15-second epochs were: SED (0-48 counts), LPA (49-418 counts), and MVPA (>418 counts).
• Classification accuracy was fair for SED (ROC-AUC = 0.74) and excellent for MVPA (ROC-AUC = 0.90).
• Both toddler-derived and preschool cut-points overestimated SED and underestimated LPA, but accurately estimated MVPA in the validation sample.

Conclusions:

• ActiGraph accelerometers can provide reliable group-level estimates of MVPA in toddlers.
• The established Pate cut-point of 420 counts/15 s is supported for assessing MVPA in this age group.
• Further research may be needed to refine cut-points for SED and LPA in toddlers.