AbstractsComputer Science

VALIDITY AND RELIABILITY OF A NOVEL COMPUTER-ANIMATED SELF-REPORT PHYSICAL ACTIVITY INSTRUMENT IN OLDER ADULTS

by James Janssen




Institution: Wake Forest University
Department:
Year: 2014
Keywords: Accelerometry
Record ID: 2025803
Full text PDF: http://hdl.handle.net/10339/39265


Abstract

The purpose of this study was to investigate the reliability, construct validity and responsiveness to a physical activity intervention of the MAT-PA (Mobility Assessment Tool-Physical Activity) physical activity questionnaire (PAQ) for walking in older adults. The MAT-PA is a novel computer-animation assisted self-report questionnaire created for the assessment of physical activity in older adults. We recruited 185 older adults (67.0 ± 4.8yrs, 72% female) at risk for cardiovascular disease as part of the Cooperative Lifestyle Intervention Program-II (CLIP-II) study. Participants completed the MAT-PA with supervision on three occasions, twice at baseline (~14 days apart) and once at six-months. Gait speed at usual and fast pace was measured using a GAITRite instrumented carpet, and 7-day accelerometry were collected at baseline and 6-months. The 400m walk test and a modified version of the CHAMPS PAQ were both completed at baseline. The MAT-PA walking metrics displayed acceptable test-retest reliability (rs = 0.63, p<0.01), good correlation with accelerometry moderate intensity physical activity minutes (rs=0.60, p<0.01) and mCHAMPS5 (rs=0.54, p<0.01), and was responsive to change from a physical activity intervention at 6-months when comparing the CLIP-II walking group with controls (F(1,68)= 31.253, p<0.0001, Eta²=0.32). Older adults were able to discriminate between walking speed animations. These findings suggest that the MAT-PA has promising psychometric properties for the assessment of walking as a mode of physical activity training.