Data from ActiGraph accelerometers have long been imported into ActiLife software, where the company’s proprietary “activity counts” were generated in order to understand physical behavior metrics. In 2022, ActiGraph released an open-source method to generate activity counts from any raw, triaxial accelerometer data using Python, which has been translated into RStudio packages. However, it is unclear if outcomes are comparable when generated in ActiLife and RStudio. Therefore, the authors’ technical note systematically compared activity counts and related physical behavior metrics generated from ActiGraph accelerometer data using ActiLife or available packages in RStudio and provides example code to ease implementation of such analyses in RStudio. In addition to comparing triaxial activity counts, physical behavior outputs (sleep, sedentary behavior, light-intensity physical activity, and moderate- to vigorous-intensity physical activity) were compared using multiple nonwear algorithms, epochs, cut points, sleep scoring algorithms, and accelerometer placement sites. Activity counts and physical behavior outcomes were largely the same between ActiLife and the tested packages in RStudio. However, peculiarities in the application of nonwear algorithms to the first and last portions of a data file (that occurred on partial, first or last days of data collection), differences in rounding, and handling of counts values on the borderline of activity intensities resulted in small but inconsequential differences in some files. The hope is that researchers and both hardware and software manufacturers continue to push efforts toward transparency in data analysis and interpretation, which will enhance comparability across devices and studies and help to advance fields examining links between physical behavior and health.
Browse
Comparability of 24-hr Activity Cycle Outputs From ActiGraph Counts Generated in ActiLife and RStudio
Alexander H.K. Montoye, Kimberly A. Clevenger, Benjamin D. Boudreaux, and Michael D. Schmidt
Pre- Versus Postmeal Sedentary Duration—Impact on Postprandial Glucose in Older Adults With Overweight or Obesity
Elizabeth Chun, Irina Gaynanova, Edward L. Melanson, and Kate Lyden
Introduction : Reducing sedentary time is associated with improved postprandial glucose regulation. However, it is not known if the timing of sedentary behavior (i.e., pre- vs. postmeal) differentially impacts postprandial glucose in older adults with overweight or obesity. Methods : In this secondary analysis, older adults (≥65 years) with overweight and obesity (body mass index ≥ 25 kg/m2) wore a continuous glucose monitor and a sedentary behavior monitor continuously in their real-world environments for four consecutive days on four separate occasions. Throughout each 4-day measurement period, participants followed a standardized eucaloric diet and recorded mealtimes in a diary. Glucose, sedentary behavior, and meal intake data were fused using sensor and diary timestamps. Mixed-effect linear regression models were used to evaluate the impact of sedentary timing relative to meal intake. Results : Premeal sedentary time was significantly associated with both the increase from premeal glucose to the postmeal peak (ΔG) and the percent of premeal glucose increase that was recovered 1-hr postmeal glucose peak (%Baseline Recovery; p < .05), with higher levels of premeal sedentary time leading to both a larger ΔG and a smaller %Baseline Recovery. Postmeal sedentary time was significantly associated with the time from meal intake to glucose peak (ΔT; p < .05), with higher levels of postmeal sedentary time leading to a longer time to peak. Conclusions : Pre- versus postmeal sedentary behavior differentially impacts postprandial glucose response in older adults with overweight or obesity, suggesting that the timing of sedentary behavior reductions might play an influential role on long-term glycemic control.
A Self-Paced Walk Test for Individual Calibration of Heart Rate to Energy Expenditure
Kate Westgate, Tomas I. Gonzales, Stefanie Hollidge, Tim Lindsay, Nick Wareham, and Søren Brage
Introduction: Estimating free-living physical activity (PA) with continuous heart rate (HR) monitoring is challenging due to individual variation in the relationship between HR and energy expenditure. This variation can be captured through individual calibration with graded exercise tests, but structured tests with prescribed load require medical screening and are not always feasible in population settings. We present and evaluate an individual calibration method using HR response to a less demanding self-paced walk test. Methods: Six hundred and forty-three participants from the Fenland Study (Cambridgeshire, the United Kingdom) completed a 200-m self-paced walk test, a treadmill test, and 1 week of continuous HR and accelerometry monitoring. Mixed-effects regression was used to derive a walk test calibration model from HR response to the walk using treadmill-based parameters as criterion. Free-living PA estimates from the calibration model were compared with treadmill-calibrated and non-exercise-calibrated estimates. Results: Walk calibration captured 57% of the variance in the HR–energy expenditure relationship determined by the treadmill test. Applying walk calibration to data from free-living yielded similar PA estimates to those using treadmill calibration (52.7 vs. 52.0 kJ·kg−1·day−1; mean difference: 0.7 kJ·kg−1·day−1, 95% confidence interval [−0.0, 1.5]) and high correlation (r = .89). Individual differences were observed (root mean square error: 10.0 kJ·kg−1·day−1; 95% limits of agreement: −20.6, 19.1 kJ·kg−1·day−1). Walk calibration improved precision by 29% compared with nonexercise group calibration (root mean square error: 14.0 kJ·kg−1·day−1; 95% limits of agreement: −30.4, 24.5 kJ·kg−1·day−1). Conclusions: A 200-m self-paced walk test captures between-individual variation in the HR–energy expenditure relationship and facilitates estimation of free-living PA in population settings.
The KID Study (Kids Interacting With Dogs): Piloting a Novel Approach for Measuring Dog-Facilitated Youth Physical Activity
Colleen J. Chase, Sarah Burkart, and Katie Potter
Background: Two-thirds of children in the United States do not meet the National Physical Activity Guidelines, leaving a majority at higher risk for negative health outcomes. Novel, effective children’s physical activity (PA) interventions are urgently needed. Dog-facilitated PA (e.g., dog walking and active play) is a promising intervention target, as dogs support many of the known correlates of children’s PA. There is a need for accurate methods of quantifying dog-facilitated PA. Purpose: The study purpose was to determine the feasibility and acceptability of a novel method for quantifying the volume and intensity of dog-facilitated PA among dog-owning children. Methods: Children and their dog(s) wore ActiGraph accelerometers with a Bluetooth proximity feature for 7 days. Additionally, parents logged child PA with the family dog(s). Total minutes of dog-facilitated PA and percentage of overall daily moderate to vigorous PA performed with the dog were calculated. Results: Twelve children (mean age = 7.8 ± 2.9 years) participated. There was high feasibility, with 100% retention, valid device data (at least 4 days ≥6-hr wear time), and completion of daily parent log and questionnaire packets. On average, dog-facilitated PA contributed 22.9% (9.2 min) and 15.1% (7.3 min) of the overall daily moderate to vigorous PA for children according to Bluetooth proximity data and parent report, respectively. Conclusions: This pilot study demonstrated the feasibility of utilizing an accelerometer with a proximity feature to quantify dog-facilitated PA. Future research should use this protocol with a larger, more diverse sample to determine whether dog-facilitated PA contributes a clinically significant amount toward overall PA in dog-owning youth.
Comparison of Sleep and Physical Activity Metrics From Wrist-Worn ActiGraph wGT3X-BT and GT9X Accelerometers During Free-Living in Adults
Duncan S. Buchan
Background: ActiGraph accelerometers can monitor sleep and physical activity (PA) during free-living, but there is a need to confirm agreement in outcomes between different models. Methods: Sleep and PA metrics from two ActiGraphs were compared after participants (N = 30) wore a GT9X and wGT3X-BT on their nondominant wrist for 7 days during free-living. PA metrics including total steps, counts, average acceleration—Euclidean Norm Minus One (ENMO) and Mean Amplitude Deviation, intensity gradient, the minimum acceleration value of the most active 10 and 30 min (M10, M30), time spent in activity intensities from vector magnitude (VM) counts, and ENMO cut points and sleep metrics (sleep period time window, sleep duration, sleep onset, and waking time) were compared. Results: Excellent agreement was evident for average acceleration-Mean Amplitude Deviation, counts, total steps, M10, and light PA (VM counts) with good agreement evident from the remaining PA metrics apart from moderate–vigorous PA (VM counts) which demonstrated moderate agreement. Mean bias for all PA metrics were low, as were the limits of agreement for the intensity gradient, average acceleration-Mean Amplitude Deviation, and inactive time (ENMO and VM counts). The limits of agreement for all other PA metrics were >10%. Excellent agreement, low mean bias, and narrow limits of agreement were evident for all sleep metrics. All sleep and PA metrics demonstrated equivalence (equivalence zone of ≤10%) apart from moderate–vigorous PA (ENMO) which needed an equivalence zone of 16%. Conclusions: Equivalent estimates of almost all PA and sleep metrics are provided from the GT9X and wGT3X-BT worn on the nondominant wrist.
Criterion Validity of Accelerometers in Determining Knee-Flexion Angles During Sitting in a Laboratory Setting
Yanlin Wu, Myles W. O’Brien, Alexander Peddle, W. Seth Daley, Beverly D. Schwartz, Derek S. Kimmerly, and Ryan J. Frayne
Introduction: Device-based monitors often classify all sedentary positions as the sitting posture, but sitting with bent or straight legs may exhibit unique physiological and biomechanical effects. The classifications of the specific nuances of sitting have not been understood. The purpose of this study was to validate a dual-monitor approach from a trimonitor configuration measuring knee-flexion angles compared to motion capture (criterion) during sitting in laboratory setting. Methods: Nineteen adults (12♀, 24 ± 4 years) wore three activPALs (torso, thigh, tibia) while 14 motion capture cameras simultaneously tracked 15 markers located on bony landmarks. Each participant completed a 45-s supine resting period and eight, 45-s seated trials at different knee flexion angles (15° increment between 0° and 105°, determined via goniometry), followed by 15 s of standing. Validity was assessed via Friedman’s test (adjusted p value = .006), mean absolute error, Bland–Altman analyses, equivalence testing, and intraclass correlation. Results: Compared to motion capture, the calculated angles from activPALs were not different during 15°–90° (all, p ≥ .009), underestimated at 105° (p = .002) and overestimated at 0°, as well as the supine position (both, p < .001). Knee angles between 15° and 105° exhibited a mean absolute error of ∼5°, but knee angles <15° exhibited larger degrees of error (∼10°). A proportional (β = −0.12, p < .001) bias was observed, but a fixed (0.5° ± 1.7°, p = .405) bias did not exist. In equivalence testing, the activPALs were statistically equivalent to motion capture across 30°–105°. Strong agreement between the activPALs and motion capture was observed (intraclass correlation = .97, p < .001). Conclusions: The usage of a three-activPAL configuration detecting seated knee-flexion angles in free-living conditions is promising.
Erratum. Context Matters: The Importance of Physical Activity Domains for Public Health
Journal for the Measurement of Physical Behaviour
Influence of Accelerometer Calibration on the Estimation of Objectively Measured Physical Activity: The Tromsø Study
Marc Weitz, Bente Morseth, Laila A. Hopstock, and Alexander Horsch
Accelerometers are increasingly used to observe human behavior such as physical activity under free-living conditions. An important prerequisite to obtain reliable results is the correct calibration of the sensors. However, accurate calibration is often neglected, leading to potentially biased results. Here, we demonstrate and quantify the effect of accelerometer miscalibration on the estimation of objectively measured physical activity under free-living conditions. The total volume of moderate to vigorous physical activity (MVPA) was significantly reduced after post hoc auto-calibration for uniaxial and triaxial count data, as well as for Euclidean Norm Minus One and mean amplitude deviation raw data. Weekly estimates of MVPA were reduced on average by 5.5, 9.2, 45.8, and 4.8 min, respectively, when compared to the original uncalibrated estimates. Our results indicate a general trend of overestimating physical activity when using factory-calibrated sensors. In particular, the accuracy of estimates derived from the Euclidean Norm Minus One feature suffered from uncalibrated sensors. For all modalities, the more uncalibrated the sensor was, the more MVPA was overestimated. This might especially affect studies with lower sample sizes.
