Application of methods in physical activity measurement.

摘要

It is broadly accepted that physical activity provides substantial health benefits. Despite strong evidence, approximately 60% to 95% of US adults are insufficiently active to obtain these health benefits. This dissertation explored five projects that examined the measurement properties and methodology for a variety of physical activity assessment methods. Project one identified validity evidence for the new MyWellness Key accelerometer in sixteen adults. The MyWellness Key demonstrated acceptable validity evidence when compared to a criterion accelerometer during graded treadmill walking and in free-living settings. This supports the use of the MyWellness Key accelerometer to measure physical activity. Project two evaluated validity (study 1) and test-retest reliability evidence (study 2) of the Global Physical Activity Questionnaire (GPAQ) in a two part study. The GPAQ was compared to direct and indirect criterion measures including object and subjective physical activity instruments. These data provided preliminary validity and reliability evidence for the GPAQ that support its use to assess physical activity. Project three investigated the optimal h·d-1 of accelerometer wear time needed to assess daily physical activity. Using a semi-simulation approach, data from 124 participants were used to compare 10--13 h·d-1 to the criterion 14 h·d-1. This study suggested that a minimum accelerometer wear time of 13 h·d-1 is needed to provide a valid measure of daily physical activity. Project four evaluated validity and reliability evidence of a novel method (Movement and Activity in Physical Space [MAPS] score) that combines accelerometer and GPS data to assess person-environment interactions. Seventy-five healthy adults wore an accelerometer and GPS receiver for three days to provide MAPS scores. This study provided evidence for use of a MAPS score for future research and clinical use. Project five used accelerometer data from 1000 participants from the 2005--2006 National Health and Nutrition Examination Study. A semi-simulation approach was used to assess the effect of accelerometer wear time (10--14 h·d-1) on physical activity data. These data showed wearing for 12 h·d-1 or less may underestimate time spent in various intensities of physical activity.