Use of Objective Measures to Estimate Sedentary Time in Youth

Purpose The purpose of this study was to examine the use of the ActivPAL (AP) and ActiGraph GT3X (AG) for estimating SB during structured and free-living physical activity (PA) in youth.

Methods Twenty-nine boys and 23 girls (mean±SD; age, 12±0.9 yrs; BMI 23.3±5.8 kg.m-2) performed up to 8 structured activities and approximately 2-hrs of free-living PA. Structured activities were split into 4 categories: lying (5 min), sitting (3 min), standing activities (3 min), and walk/run (3 min). Activity data was collected using an AG, positioned on right hip, AP, and energy expenditure (METRMR; activity VO2 divided by resting VO2) was measured using a Cosmed K4b2. For the AG, two methods to estimate SB were examined: 1) inclinometer function, and 2) < 75 vector magnitude (VM) counts/10 sec, < 25 vertical axis (VA) counts/10 sec, and < 50, 100, 150, and 200 VA counts/min. For the structured activities, the AG and AP outcomes were compared to actual time spent in each activity category and for the free-living activity the AG and AP estimates of SB were compared to the Cosmed (<1.5 METRMR).

Results Structured PA Categories: The AG inclinometer significantly underestimated actual lying, sitting, standing, and walk/run time by 52.0%, 45.4%, 26.3% and 0.1%, respectively (P<0.05). The AP was within 1.4% of measured standing and walk/run time (P>0.05), but significantly underestimated measured lying/ sitting time by 16.8% (P<0.05). The AG SB cut-points per 10 sec and <50 counts/ min significantly underestimated measured lying time by 3.6-11.9% (P<0.05) and all AG SB cut-points except <150 and 200 counts/min significantly underestimated measured sitting time by 4.2-13.8% (P<0.05). Free-Living PA: On average, there were 28.2 measured minutes of SB time. All AG and AP estimates of SB time were within ±44.5% of measured SB time (P>0.05). The AG inclinometer and AP were within0.7% and 4.8% of measured SB time, respectively. The closest AG SB cut-points,compared to measured SB time were <25 VA counts/10 sec (2.2%) and <50 counts/ min (4.3%).

Conclusion The AP, AG inclinometer, and AG SB cut-points of <25 VA counts/10 sec and < 50 counts/min provided the closest estimates of measured SB time during free-living PA in youth. However, the AP was better at discriminating between lying/sitting time and standing time. Study supported by NIH grant 5R21HL093407