Evaluation of youth pedometer-determined physical activity guidelines using receiver operator characteristic curves

Prev Med. 2008 May;46(5):419-24. doi: 10.1016/j.ypmed.2007.12.017. Epub 2008 Jan 4.

Abstract

Objective: Evidence-based pedometer guidelines have not been sufficiently validated in youth. The purpose of this paper was to evaluate the utility of current pedometer-determined physical activity guidelines for youth.

Methods: Participants (n=812, mean age 9.7 years) were from two Midwestern communities during Fall 2005. Participants completed 7 days of pedometer monitoring using a Digiwalker 200-SW. ROC analyses were utilized to assess the utility of the cutpoints in reference to weight status. The percentage of subjects meeting physical activity recommendations (steps/day) was determined using two published recommendations.

Results: The mean (SD) steps/day for boys, girls, and sexes combined were 12,709 (3,384), 10,834 (2562), and 11,665 (3028), respectively. Depending on the recommendation, approximately 55-75% of children did not meet the recommended number of steps/day. Sensitivity was high and specificity was low for both current guidelines evaluated. The ROC analysis indicated that lowering the steps/day cutpoint for boys and girls may increase accuracy.

Conclusions: This study demonstrates that a large percentage of children from two Midwestern communities do not meet current pedometer-based physical activity recommendations. Because the percentage of children categorized as active or inactive depends on the criteria that are employed, it is recommended that researchers select cutpoints based on the preferred context of application.

Publication types

  • Comparative Study
  • Evaluation Study
  • Multicenter Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Body Mass Index
  • Child
  • Exercise / physiology*
  • Female
  • Humans
  • Male
  • Midwestern United States
  • Monitoring, Ambulatory / instrumentation*
  • Obesity / prevention & control*
  • Practice Guidelines as Topic*
  • ROC Curve
  • Reference Values
  • Sensitivity and Specificity
  • Walking / physiology*