Browse

Purpose: Examine the effect of cutpoint selection on physical activity (PA) metrics calculated from young children’s accelerometer data and on the proportion of children meeting PA guidelines. Methods: A total of 262 children (3.6 ± 1.4 years, 126 males) wore ActiGraph wGT3X-BT accelerometers on their right hip for 7 days, 24 hr/day. Ten cutpoint sets were applied to the sample categorized by age, based on populations of the original cutpoint calibration studies using ActiLife software. Resulting sedentary behavior, light PA, moderate to vigorous PA, and total PA were compared using repeated-measures analysis of variance. Proportion of children meeting age-appropriate PA guidelines based on each cutpoint set was assessed using Cochran’s q tests. Results: Children wore the accelerometer for an average of 7.6 ± 1.2 days for an average of 11.9 ± 1.2 hr/day. Significant differences in time spent in each intensity were found across all cutpoints except for sedentary, and total PA for three comparisons (Trost vs. Butte Vertical Axis [VA], Pate vs. Puyau, and Costa VA vs. Evenson) and moderate to vigorous PA for four comparisons (Trost vs. Pate, Trost vs. Pate and Pfeiffer, Pate vs. Pate and Pfeiffer, and van Cauwenberghe vs. Evenson). When examined within age-appropriate groups, all sets of cutpoints resulted in significant differences across all intensities and in the number of children meeting PA guidelines. Conclusion: Choice of cutpoints applied to data from young children significantly affects times calculated for different movement intensities, which in turn impacts the proportion of children meeting guidelines. Thus, comparisons of movement intensities should not be made across studies using different sets of cutpoints.

The Physical Activity Vital Sign (PAVS) is a two-question assessment used to estimate habitual moderate to vigorous aerobic physical activity (MVPA). Previous studies have shown active adults cannot estimate the physical activity intensity properly. The initial purpose was to investigate the criterion validity of the PAVS for quantifying habitual MVPA in young adults meeting weekly MVPA guidelines (n = 140; 21 ± 3 years). A previously validated PiezoRx waist-worn accelerometer served as the criterion measure (wear time, 6.7 ± 0.6 days). All participants completed the PAVS once before wearing the PiezoRx. Standardized activity monitor validation procedures were followed. The PAVS (201 ± 142 min/week) underestimated (p < .001) MVPA compared to the PiezoRx (381 ± 155 min/week). To correct for this large error, the sample was divided into calibration model development (n = 70; 21 ± 3 years) and criterion validation (n = 70; 21 ± 3 years) groups. The PAVS score, age, gender, and body mass index outcomes from the development group were used to construct a multiple linear regression model-based calibrated PAVS (cPAVS) equation. In the validation group, the cPAVS was similar (p = .113; 352 ± 23 min/week) compared to accelerometry. Equivalence testing demonstrated the cPAVS, but not the PAVS, was equivalent to the PiezoRx. Despite achieving most statistical criteria, the PAVS and cPAVS still had high degrees of variability, preventing their use on an individual level. Alternative strategies are needed for the PAVS in an active young adult population. These results caution using the PAVS in active young adults and identify a case where obvious variabilities in accuracy conflict with statistically congruent results.

ActiGraph counts are commonly used for characterizing physical activity intensity and energy expenditure and are among the most well-studied accelerometer metrics. Researchers have recently replicated the counts processing method using a mechanical setup, now allowing users to generate counts from raw acceleration data. Purpose: The purpose of this study was to compare ActiGraph-generated counts to open-source counts and assess the impact on free-living physical activity levels derived from cut points, machine learning, and two-regression models. Methods: Children (n = 488, 13.0 ± 1.1 years of age) wore an ActiGraph wGT3X-BT on their right hip for 7 days during waking hours. ActiGraph counts and counts generated from raw acceleration data were compared at the epoch-level and as overall means. Seven methods were used to classify overall and epoch-level activity intensity. Outcomes were compared using weighted kappa, correlations, mean absolute deviation, and two one-sided equivalence testing. Results: All outcomes were statistically equivalent between ActiGraph and open-source counts; weighted kappa was ≥.971 and epoch-level correlations were ≥.992, indicating very high agreement. Bland–Altman plots indicated differences increased with activity intensity, but overall differences between ActiGraph and open-source counts were minimal (e.g., epoch-level mean absolute difference of 23.9 vector magnitude counts per minute). Regardless of classification model, average differences translated to 1.4–2.6 min/day for moderate- to vigorous-intensity physical activity. Conclusion: Open-source counts may be used to enhance comparability of future studies, streamline data analysis, and enable researchers to use existing developed models with alternative accelerometer brands. Future studies should verify the performance of open-source counts for other outcomes, like sleep.

Introduction : Instrumented gait mat systems have been regarded as one of the gold standard methods for measuring spatiotemporal gait parameters. However, their portable walkways confine walking to a restricted area and limit the number of gait cycles collected. Wearable inertial sensors are a potential alternative that allow more natural walking behavior and have fewer space restrictions. The objective of this pilot study was to establish the concurrent validity of body-worn sensors against the portable walkway system in older children. Methods : Twenty-one participants (10 males) 7–17 years old performed 2-min walk tests at a self-selected and fast pace in a 25-m-long hallway, while wearing three inertial sensors. Data collection were synchronized between devices and the portions of the walk when subjects passed on the walkway were used to compare gait speed, stride length, gait cycle duration, cadence, and double support time. Regression models and Bland–Altman analysis were completed to determine agreement between systems for the selected gait parameters. Results : Gait speed, cadence, gait cycle duration, and stride length as measured by inertial sensors demonstrated strong agreement overall. Double support time was found to have lower validity due to a combined bias of age, height, weight, and walking pace. Conclusion : These results support the validity of wearable inertial sensors in measuring gait speed, cadence, gait cycle duration, and stride length in children 7 years old and above during a 2-min walking test. Future studies are warranted with a broader age range to thoroughly represent the pediatric population.

Introduction: Patterns of physical activity (PA) and time in bed (TIB) across the 24-hr cycle have important implications for many health outcomes; therefore, wearable accelerometers are often implemented in behavioral research to measure free-living PA and TIB. Two accelerometers, the activPAL and Actiwatch, are common accelerometers for measuring PA (activPAL) and TIB (Actiwatch), respectively. Both accelerometers have the capacity to measure TIB, but the degree to which these accelerometers agree is not clear. Therefore, this study compared estimates of TIB between activPAL and the Actiwatch accelerometers. Methods: Participants (mean ± SD_age = 39.8 ± 7.6 years) with overweight or obesity (N = 83) wore an activPAL and Actiwatch continuously for 7 days, 24 hr per day. TIB was assessed using manufacturer-specific algorithms. Repeated-measures mixed-effect models and Bland–Altman plots were used to compare the activPAL and Actiwatch TIB estimates. Results: Statistical differences between TIB assessed by activPAL versus Actiwatch (p < .001) were observed. There was not a significant interaction between accelerometer and day of wear (p = .87). The difference in TIB between accelerometers ranged from −72.9 ± 15.7 min (Day 7) to −98.6 ± 14.5 min (Day 3), with the Actiwatch consistently estimating longer TIB compared with the activPAL. Conclusion: Data generated by the activPAL and Actiwatch accelerometers resulted in divergent estimates of TIB. Future studies should continue to explore the validity of activity monitoring accelerometers for estimating TIB.

Introduction: Current best practice for objective measurement of sedentary behavior and moderate-to-vigorous intensity physical activity (MVPA) requires two separate devices. This study assessed concurrent agreement between the ActiGraph GT3X and the activPAL3 micro for measuring MVPA to determine if activPAL can accurately measure MVPA in addition to its known capacity to measure sedentary behavior. Methods: Forty participants from two studies, including pregnant women (n = 20) and desk workers (n = 20), provided objective measurement of MVPA from waist-worn ActiGraph GT3X and thigh-worn activPAL micro3. MVPA from the GT3X was compared with MVPA from the activPAL using metabolic equivalents of task (MET)- and step-based data across three epochs. Intraclass correlation coefficient and Bland–Altman analyses, overall and by study sample, compared MVPA minutes per day across methods. Results: Mean estimates of activPAL MVPA ranged from 22.7 to 35.2 (MET based) and 19.7 to 25.8 (step based) minutes per day, compared with 31.4 min/day (GT3X). MET-based MVPA had high agreement with GT3X, intraclass correlation coefficient ranging from .831 to .875. Bland–Altman analyses revealed minimal bias between 15- and 30-s MET-based MVPA and GT3X MVPA (−3.77 to 8.63 min/day, p > .10) but with wide limits of agreement (greater than ±27 min). Step-based MVPA had moderate to high agreement (intraclass correlation coefficient: .681–.810), but consistently underestimated GT3X MVPA (bias: 5.62–11.74 min/day, p < .02). For all methods, activPAL appears to better estimate GT3X at lower quantities of MVPA. Results were similar when repeated separately by pregnant women and desk workers. Conclusion: activPAL can measure MVPA in addition to sedentary behavior, providing an option for concurrent, single device monitoring. MET-based MVPA using 30-s activPAL epochs provided the best estimate of GT3X MVPA in pregnant women and desk workers.

Purpose: To determine accuracy of activPAL Technologies’ CREA algorithm to assess bedtime, wake time, and sleep time. Methods: As part of a larger study, 104 participants recorded nightly sleep logs (LOGs) and wore the activPAL accelerometer at the thigh and ActiGraph accelerometer at the hip for 24 hr/day, for seven consecutive days. For sleep LOGs, participants recorded nightly bed and daily wake times. Previously validated ActiGraph, proprietary activPAL, and the Winkler sleep algorithm were used to compute sleep variables. Eighty-seven participants provided 2+ days of valid data. Pearson correlations, paired samples t tests, and equivalency tests were used to examine relationships and differences between methods (activPAL vs. ActiGraph, activPAL vs. LOG, and activPAL vs. Winkler algorithm). Results: For screened data, moderately high to high correlations but significant mean differences were found between activPAL versus ActiGraph for bedtime (t ₈₆ = −6.80, p ≤ .01, r = .84), wake time (t ₈₆ = 4.80, p ≤ .01, r = .93), and sleep time (t ₈₆ = 7.99, p ≤ .01, r = .88). activPAL versus LOG comparisons also yielded significant mean differences and moderately high to high correlations for bedtime (t ₈₆ = −4.68, p ≤ .01, r = .82), wake time (t ₈₆ = 8.14, p ≤ .01, r = .93), and sleep time (t ₈₆ = 8.60, p ≤ .01, r = .72). Equivalency testing revealed that equivalency could not be claimed between activPAL versus LOG or activPAL versus ActiGraph comparisons, though the activPAL and Winkler algorithm were equivalent. Conclusion: The activPAL algorithm overestimated sleep time by detecting earlier bedtimes and later wake times. Because of the significant differences between algorithms, bedtime, wake time, and sleep time are not interchangeable between methods.

Purpose: The purpose of this study was to evaluate the validity of the step count and Active Minutes features of the Garmin Vivofit Jr. 2 consumer activity monitor. Methods: Participants included 35 students (age 8–11) enrolled in an after-school physical activity (PA) and nutrition program. Participants wore an ActiGraph GT3x+ monitor on their waist and the Vivofit monitor on their wrist during the PA portion of the program. Data were collected across multiple sessions, resulting in 158 unique pairs of data. Pearson correlation, mean absolute percent error, and equivalence testing were performed to compare step count and minutes of activity (Vivofit Active Minutes vs ActiGraph moderate to vigorous PA) between the two monitors. Results: Moderate correlations were found between the monitors for steps (r = .65) and minutes (r = .43). Mean absolute percent error was 26% for steps and 43% for minutes, suggesting that there were high amounts of individual error. Equivalence testing showed significant agreement between the monitors for steps (p = .046), but not for minutes (p = .98). Conclusion: The Garmin Vivofit Jr. 2 shows acceptable validity for measurement of steps at a group level in a field-based setting, although the amount of individual variability must be considered. The Vivofit Jr. 2 was not valid for measurement of minutes of activity.