Date of Award

2018

Document Type

Open Access Thesis

Department

Exercise Science

Sub-Department

The Norman J. Arnold School of Public Health

First Advisor

Russell Pate

Abstract

Obesity prevalence is continuing to rise in children in the United States. Decreased physical activity has long been thought to be a cause of this trend. As such, much of the accelerometer study has been focused on moderate to vigorous physical activity (MVPA). More recently, time spent in sedentary behavior has received much attention as well. There is an abundance of time spent in light intensity physical activity, and yet comparatively little focus has been placed on it in the field of accelerometer study. The purpose of this study was to generate an accelerometer count cut-point for the ActiGraph GT3X+ monitor that distinguishes sedentary behaviors from very light intensity physical activity, and very light intensity physical activity from light intensity activity in 10-12-year old children.

Eighteen children wore accelerometers on the hip and non-dominant wrist, along with a COSMED portable metabolic system. Participants engaged in nine structured activities in a laboratory setting. Respiratory gases and oxygen consumption were measured on a breath by breath basis and accelerometer data was collected at 15 second intervals. ROC curve analyses were used to generate count cut-points for very light intensity physical activity at the hip and non-dominant wrist sites.

Correlation between VO2 (ml/kg/min) and accelerometer counts was strong for both the hip (r=0.95) and non-dominant wrist (r=0.82) across all activities. Cut-points for the hip site were identified as 10 counts/15 seconds for sedentary and 131 counts/15 seconds for very light intensity activity. Cut-points for the non-dominant wrist site were identified as 180 counts/15 seconds and 305 counts/15 seconds for sedentary and very light intensity activity, respectively.

This study was the first to identify accelerometer count cut-points for very light intensity activity for this age range. Findings suggest that the non-dominant wrist site is more effective at differentiating between the sedentary and very light intensity activities performed in this study than the hip site. The very light activities performed in this study required little trunk movement, which limited the effectiveness of the hip site in differentiating between sedentary and very light intensity. These findings provide a starting point for the refinement of accelerometer-based study of light intensity activities.

Rights

© 2018, Joseph S. Gorab

Download

Share

COinS