(1) Background: This study compared manually-counted treadmill walking steps from the hip-worn DigiwalkerSW200 and OmronHJ720ITC, and hip and wrist-worn ActiGraph GT3X+ and GT9X; determined brand-specific acceleration amplitude (g) and/or frequency (Hz) step-detection thresholds; and quantified key features of the acceleration signal during walking. (2) Methods: Twenty participants (Age: 26.7 ± 4.9 years) performed treadmill walking between 0.89-to-1.79 m/s (2–4 mph) while wearing a hip-worn DigiwalkerSW200, OmronHJ720ITC, GT3X+ and GT9X, and a wrist-worn GT3X+ and GT9X. A DigiwalkerSW200 and OmronHJ720ITC underwent shaker testing to determine device-specific frequency and amplitude step-detection thresholds. Simulated signal testing was used to determine thresholds for the ActiGraph step algorithm. Steps during human testing were compared using bias and confidence intervals. (3) Results: The OmronHJ720ITC was most accurate during treadmill walking. Hip and wrist-worn ActiGraph outputs were significantly different from the criterion. The DigiwalkerSW200 records steps for movements with a total acceleration of ≥1.21 g. The OmronHJ720ITC detects a step when movement has an acceleration ≥0.10 g with a dominant frequency of ≥1 Hz. The step-threshold for the ActiLife algorithm is variable based on signal frequency. Acceleration signals at the hip and wrist have distinctive patterns during treadmill walking. (4) Conclusions: Three common research-grade physical activity monitors employ different step-detection strategies, which causes variability in step output.

This study tested the Wellness Enhancing Physical Activity in Young Children (WE PLAY) program, a 4-week online preschool teacher training, on children's moderate-to-vigorous physical activity (MVPA). In this cluster RCT, six Head Start preschools were randomized to an intervention and comparison group. Children's MVPA was measured using accelerometers at pre-and posttest. The magnitude of the difference in MVPA between groups at posttest was small, but in the expected direction: ⌬ min/hour ϭ 1.60, 95% CI [Ϫ0.97, 4.18], p ϭ .22, Cohen's d ϭ 0.32. We observed a pre/post within group increase in average minutes per hour of MVPA in school with a medium effect size for the intervention group: ⌬ mean min/hour ϭ 2.09, 95% CI [0.51, 3.67], p ϭ .0096, Cohen's d ϭ 0.42. An increase was not seen for the comparison group: ⌬ mean min/hour ϭ 0.44, 95% CI [Ϫ0.70, 1.59], p ϭ .45, Cohen's d ϭ 0.07. WE PLAY children in 6 hr/day programs gained 63 min of MVPA per week in school, providing preliminary evidence of the benefits of WE PLAY on children's physical activity levels. WE PLAY deserves further testing with larger groups of children and teachers. Impact and ImplicationsPhysical activity is critical for children's physical and emotional well-being and cognitive functioning, yet many children need help from adults to be more active. The Wellness Enhancing Physical Activity for Young Children (WE PLAY) online teacher professional development program was developed to support the knowledge, confidence, and skills of early childhood educators so they can lead active play every day with their students. This study found that children whose teachers completed WE PLAY training engaged in higher levels of moderateto-vigorous physical activity at posttest relative to pretest, and those changes were not observed in a comparison group.

Our study investigated the performance of proximity sensor-based wear-time detection using the GT9X under laboratory and free-living settings. Fifty-two volunteers (23.2 ± 3.8 y; 23.2 ± 3.7 kg/m) participated in either a laboratory or free-living protocol. Lab participants wore and removed a wrist-worn GT9X on 3-5 occasions during a 3-hour directly observed activity protocol. The 2-day free-living protocol used an independent temperature sensor and self-report as the reference to determine if wrist and hip-worn GT9X accurately determined wear (i.e., sensitivity) and non-wear (i.e., specificity). Free-living estimates of wear/non-wear were also compared to Troiano 2007 and Choi 2012 wear/non-wear algorithms. In lab, sensitivity and specificity of the wrist-worn GT9X in detecting total minutes of wear-on and off was 93% and 49%, respectively. The GT9X detected wear-off more often than wear-on, but with a greater margin of error (4.8 ± 11.6 vs. 1.4 ± 1.4 min). In the free-living protocol, wrist and hip-worn GT9X's yielded sensitivity and specificity of 72 and 90% and 84 and 92%, respectively. GT9X estimations had inferior sensitivity but superior specificity to Troiano 2007 and Choi 2012 algorithms. Due to inaccuracies, it may not be advisable to singularly use the proximity-sensor-based wear-time detection method to detect wear-time.

Supplemental digital content is available in the text.

Studies using wearable sensors to measure posture, physical activity (PA), and sedentary behavior typically use a single sensor worn on the ankle, thigh, wrist, or hip. Although the use of single sensors may be convenient, using multiple sensors is becoming more practical as sensors miniaturize. Purpose We evaluated the effect of single-site versus multisite motion sensing at seven body locations (both ankles, wrists, hips, and dominant thigh) on the detection of physical behavior recognition using a machine learning algorithm. We also explored the effect of using orientation versus orientation-invariant features on performance. Methods Performance (F1 score) of PA and posture recognition was evaluated using leave-one-subject-out cross-validation on a 42-participant data set containing 22 physical activities with three postures (lying, sitting, and upright). Results Posture and PA recognition models using two sensors had higher F1 scores (posture, 0.89 ± 0.06; PA, 0.53 ± 0.08) than did models using a single sensor (posture, 0.78 ± 0.11; PA, 0.43 ± 0.03). Models using two nonwrist sensors for posture recognition (F1 score, 0.93 ± 0.03) outperformed two-sensor models including one or two wrist sensors (F1 score, 0.85 ± 0.06). However, two-sensor models for PA recognition with at least one wrist sensor (F1 score, 0.60 ± 0.05) outperformed other two-sensor models (F1 score, 0.47 ± 0.02). Both posture and PA recognition F1 scores improved with more sensors (up to seven; 0.99 for posture and 0.70 for PA), but with diminishing performance returns. Models performed best when including orientation-based features. Conclusions Researchers measuring posture should consider multisite sensing using at least two nonwrist sensors, and researchers measuring PA should consider multisite sensing using at least one wrist sensor and one nonwrist sensor. Including orientation-based features improved both posture and PA recognition.