Data Imputation Improves Sedentary Behavior and Physical Activity Estimates in Low Wear Time Accelerometer Data

2017 ◽  
Vol 49 (5S) ◽  
pp. 1051
Author(s):  
David R. Paul ◽  
Brenda M. Vincent ◽  
Chantal A. Vella ◽  
Philip W. Scruggs ◽  
Ryan P. McGrath
Keyword(s):  
Physical Activity ◽  
Sedentary Behavior ◽  
Accelerometer Data ◽  
Data Imputation ◽  
Wear Time ◽  
Low Wear

scholarly journals Objectively Measured Sedentary Behavior, Physical Activity, and Cardiometabolic Risk in Hispanic Youth: Hispanic Community Health Study/Study of Latino Youth

2018 ◽  
Vol 103 (9) ◽  
pp. 3289-3298 ◽  
Author(s):  
Garrett Strizich ◽  
Robert C Kaplan ◽  
Daniela Sotres-Alvarez ◽  
Keith M Diaz ◽  
Amber L Daigre ◽  
...  
Keyword(s):  
Physical Activity ◽  
Sedentary Behavior ◽  
Cardiometabolic Risk ◽  
Vigorous Physical Activity ◽  
Latino Youth ◽  
Health Study ◽  
Accelerometer Data ◽  
Linear Regression Models ◽  
Plasminogen Activator Inhibitor 1 ◽  
Wear Time

Abstract Context Time spent in moderate-to-vigorous physical activity (MVPA), but not in sedentary behavior (SB), is related to cardiometabolic risk among non-Hispanic white youth. Objective Examine associations of SB and MVPA with cardiometabolic risk factors among Hispanic/Latino youth. Design Cross-sectional analysis. Setting Four US communities. Participants Hispanic/Latino youth (N = 1,426) ages 8 to 16 years. Measurements Associations of MVPA and SB, measured using 7-day accelerometer data (independent variables), with markers of glucose and lipid metabolism, inflammation, and endothelial function (dependent variables), were assessed in multivariable linear regression models while adjusting for sociodemographic characteristics and accelerometer wear time. Additional models controlled for obesity measures. Results SB comprised a mean (SD) of 75% (13%) of accelerometer wear time; mean (SD) time of MVPA was 35 min/d (22 min/d). Deleterious levels of high-density lipoprotein–cholesterol (HDL-C), triglycerides, insulin resistance, C-reactive protein, and plasminogen activator inhibitor-1 were associated with lower levels of MVPA and higher levels of SB (all P < 0.05). Associations of MVPA with log-transformed triglyceride concentrations (β per 15-min/d increment, −0.039; SE, 0.018; P = 0.037) and SB with HDL-C (β per 30-min/d increment, −0.63; SE, 0.26; P = 0.018), but not those with other markers, remained significant after adjusting for MVPA or SB and further adjustment for body mass index and waist circumference. Higher SB tertiles were associated with lower soluble receptor for advanced glycation end products in fully adjusted models (P for trend = 0.037). Conclusions Physiological precursors of diabetes and cardiovascular disease were associated with MVPA and SB among US Hispanic/Latino youth, a group that bears a disproportionate burden of metabolic disorders.

The Impact of Low Accelerometer Wear Time on the Estimates and Application of Sedentary Behavior and Physical Activity Data in Adults

2017 ◽  
Vol 14 (12) ◽  
pp. 919-924 ◽  
Author(s):  
Ryan McGrath ◽  
Chantal A. Vella ◽  
Philip W. Scruggs ◽  
Mark D. Peterson ◽  
Christopher J. Williams ◽  
...  
Keyword(s):  
Physical Activity ◽  
Sedentary Behavior ◽  
Body Fat ◽  
Regression Models ◽  
Body Fat Percentage ◽  
Accelerometer Data ◽  
Reference Dataset ◽  
Fat Percentage ◽  
Physical Activity Guidelines ◽  
Wear Time

Background:This investigation sought to determine how accelerometer wear (1) biased estimates of sedentary behavior (SB) and physical activity (PA), (2) affected misclassifications for meeting the Physical Activity Guidelines for Americans, and (3) impacted the results of regression models examining the association between moderate to vigorous physical activity (MVPA) and a clinically relevant health outcome.Methods:A total of 100 participants [age: 20.6 (7.9) y] wore an ActiGraph GT3X+ accelerometer for 15.9 (1.6) hours per day (reference dataset) on the hip. The BOD POD was used to determine body fat percentage. A data removal technique was applied to the reference dataset to create individual datasets with wear time ranging from 15 to 10 hours per day for SB and each intensity of PA.Results:Underestimations of SB and each intensity of PA increased as accelerometer wear time decreased by up to 167.2 minutes per day. These underestimations resulted in Physical Activity Guidelines for Americans misclassification rates of up to 42.9%. The regression models for the association between MVPA and body fat percentage demonstrated changes in the estimates for each wear-time adherence level when compared to the model using the reference MVPA data.Conclusions:Increasing accelerometer wear improves daily estimates of SB and PA, thereby also improving the precision of statistical inferences that are made from accelerometer data.

scholarly journals Effects of Varying Epoch Lengths, Wear Time Algorithms, and Activity Cut-Points on Estimates of Child Sedentary Behavior and Physical Activity from Accelerometer Data

PLoS ONE ◽  
2016 ◽  
Vol 11 (3) ◽  
pp. e0150534 ◽  
Author(s):  
Jorge A. Banda ◽  
K. Farish Haydel ◽  
Tania Davila ◽  
Manisha Desai ◽  
Susan Bryson ◽  
...  
Keyword(s):  
Physical Activity ◽  
Sedentary Behavior ◽  
Accelerometer Data ◽  
Cut Points ◽  
Wear Time

Single Versus Multiple Monitoring Periods for Accelerometer-Measured Physical Activity in Medial Knee Osteoarthritis and Asymptomatic Controls

2020 ◽  
Vol 3 (1) ◽  
pp. 29-38
Author(s):  
Kerry E. Costello ◽  
Janie L. Astephen Wilson ◽  
Cheryl L. Hubley-Kozey
Keyword(s):  
Physical Activity ◽  
Knee Osteoarthritis ◽  
Sedentary Behavior ◽  
Vigorous Physical Activity ◽  
Accelerometer Data ◽  
Group Level ◽  
Single Session ◽  
Limits Of Agreement ◽  
Wear Time ◽  
Single Versus

Purpose: 1) To compare group-level physical activity calculated from a single versus multiple non-consecutive, one-week accelerometer monitoring periods in individuals with medial-compartment knee osteoarthritis and asymptomatic controls; and 2) to examine agreement among these estimates of physical activity at the individual-level. Methods: Accelerometer data from 38 individuals with knee osteoarthritis and 47 asymptomatic individuals was collected during three non-consecutive monitoring periods over one year. General linear models examined the effects of number of sessions averaged (one, two, or three) and group on light and moderate-to-vigorous intensity physical activity, step count, and sedentary behavior. Bland Altman analyses examined agreement between one-, two-, and/or three-session averages. Results: There were no sessions by group interactions. There was a main effect of sessions for sedentary behavior that was borderline significant when expressed as percent wear time. Limits of agreement indicated that two-session average versus single-session metrics could differ by ±50 minutes for light physical activity, ±20 minutes for moderate-to-vigorous physical activity, and ±2100 steps per day. Conclusions: These data suggest that objective physical activity monitoring practices might differ between clinical research, where group data are compared, and clinical decision making, where individual data are compared. Good estimates of group level differences in step count, light, and moderate-to-vigorous physical activity were found using a single session of accelerometer data, but a single session of sedentary behavior data should take wear time into account. The large limits of agreement indicate that multiple sessions may be needed to compare these metrics among or within individuals.

Abstract P248: Objectively Measured Physical Activity and Sedentary Behavior Change During a Lifestyle Intervention in Overweight Adults: Does Season Matter?

Circulation ◽  
2016 ◽  
Vol 133 (suppl_1) ◽  
Author(s):  
Bonny Rockette-Wagner ◽  
Andrea M Kriska ◽  
Qianheng Ma ◽  
Susan M Sereika ◽  
Christopher C Imes ◽  
...  
Keyword(s):  
Physical Activity ◽  
Sedentary Behavior ◽  
Lifestyle Intervention ◽  
Disease Risk ◽  
Lifestyle Interventions ◽  
Accelerometer Data ◽  
Wear Time ◽  
Step Counts ◽  
Valid Assessment ◽  
Objectively Measured

Introduction: Lifestyle interventions, with physical activity (PA) as a key component, are important to reducing cardiometabolic disease risk. In our work in both efficacy and effectiveness clinical trials, we have shown that season has a significant impact on subjectively determined moderate-vigorous (MV) PA levels, both at baseline and during the intervention. However, the effect of season in these lifestyle interventions has not yet been examined utilizing objective measurements of PA so that time spent in all PA intensities and sedentary behavior(SB) can be quantified. Hypothesis: Our hypothesis is that PA would increase and SB would decrease due to the intervention but that season would have an additional effect on both. Methods: We enrolled 150 overweight/obese adults (51.1±10.2 y; 79% Caucasian; 91% female) in a 12-month lifestyle intervention for weight loss that provided regular feedback to participants on diet and PA goal achievement. Six cohorts were recruited from 2012-2014. The PA goal was to achieve and maintain 150 minutes/week of MVPA. ActiGraph GT3x accelerometers, worn on the waist, were used to assess average daily step counts and time spent in PA and SB. Accelerometer recordings with 10 hours/day of wear time on ≥4 days were considered a valid assessment of typical PA and SB. Changes in activity variables at 6 months and 12 months were examined using linear mixed models. We also examined the season (winter, spring, summer, or autumn) when the intervention was implemented and if this affected changes in PA and SB. Results: Baseline accelerometer data were valid for 149 participants. Mean (SD) baseline values were 6132 (1873) steps counts/day, 11 (11) MVPA min/day, 245 (64) light intensity (L)PA min/day, 635 (85) SB min/day. Season was significantly related to step counts, LPA, MVPA, and SB with significantly lower PA and higher SB in the winter (p<0.05). Changes in LPA and SB were not significant over the entire follow-up (p>0.05). When adjusted (for monitor wear time/day and season) mean (SD) increases in step counts from baseline were 1128 (208) and 742(209) steps/day at 6 and 12 months, respectively (both p<0.0001). For MVPA adjusted mean (SD) increases from baseline were 7(1) and 6(1) min/day at 6 and 12 months, respectively (both p<0.0001). Conclusions: Relevant improvements in steps counts and MVPA were recorded at 6 and 12 months. This was true even after controlling for the effect of differences in the season of implementation. When considering the effect of lifestyle interventions on activity, future studies should consider the effect of seasonal changes on PA levels.

Abstract P138: Association between Cardiovascular Fitness and Accelerometer-Derived Physical Activity and Sedentary Time in the General Population

Circulation ◽  
2013 ◽  
Vol 127 (suppl_12) ◽  
Author(s):  
Jacquelyn Kulinski ◽  
Amit Khera ◽  
Colby Ayers ◽  
Sandeep Das ◽  
James de Lemos ◽  
...  
Keyword(s):  
Physical Activity ◽  
Sedentary Behavior ◽  
Sedentary Time ◽  
Cardiovascular Fitness ◽  
Inverse Association ◽  
Accelerometer Data ◽  
Vigorous Activity ◽  
Vo2 Max ◽  
Wear Time ◽  
A Minor

Objectives: Prior literature suggests that sedentary behavior may represent a risk factor that is independent of physical activity. The mechanism for this association remains poorly understood. We hypothesized that sedentary behavior might be inversely associated with cardiorespiratory fitness, independent of moderate and vigorous activity. Methods: We included 2,223 participants (ages 12-49 years, 47% female) without known heart disease who had both cardiovascular fitness testing and at least one valid day of accelerometer data from NHANES 2003-2004. From accelerometer data, we quantified bouts of exercise (at least 8 of 10 minutes above previously defined thresholds corresponding to moderate- or vigorous-intensity activity) as mean minutes of activity bouts per day for each participant. Sedentary time was defined as <100 counts per minute of wear time in mean minutes per day. VO2 max estimates were derived from a sub-maximal exercise treadmill test using measured heart rate responses to known levels of exercise workloads. Multivariable-adjusted linear regression analyses were performed with fitness as the dependent variable. Models were stratified by gender and adjusted for age, BMI, mean wear time and included both sedentary and moderate-vigorous activity time. Results: We observed that moderate and vigorous exercises levels were positively correlated to VO2 max (R=0.237, p<0.0001). Sedentary time was found to be inversely associated with VO2 max (R=-0.07, p=0.002). With each additional minute of moderate-vigorous activity, VO2 max increased by 0.05 ml/kg/min (p=0.001, men) and 0.08 ml/kg/min (p=0.004, women). For each additional minute of sedentary time, VO2 decreased by 0.01 ml/kg/min in both men (p=0.026) and women (p=0.001). Conclusions: After adjustment for moderate and vigorous activity, sedentary behavior appears to have a minor, inverse association with fitness. These findings suggest that the risk related to sedentary behavior might be mediated, in part, through lower fitness levels.

scholarly journals Capturing the Pattern of Physical Activity and Sedentary Behavior: Exposure Variation Analysis of Accelerometer Data

2014 ◽  
Vol 11 (3) ◽  
pp. 614-625 ◽  
Author(s):  
Leon Straker ◽  
Amity Campbell ◽  
Svend Erik Mathiassen ◽  
Rebecca Anne Abbott ◽  
Sharon Parry ◽  
...  
Keyword(s):  
Physical Activity ◽  
Sedentary Behavior ◽  
Public Health Research ◽  
Office Workers ◽  
Accelerometer Data ◽  
Variation Analysis ◽  
Occupational Groups ◽  
Time Pattern ◽  
Exposure Variation Analysis ◽  
First Time

Background:Capturing the complex time pattern of physical activity (PA) and sedentary behavior (SB) using accelerometry remains a challenge. Research from occupational health suggests exposure variation analysis (EVA) could provide a meaningful tool. This paper (1) explains the application of EVA to accelerometer data, (2) demonstrates how EVA thresholds and derivatives could be chosen and used to examine adherence to PA and SB guidelines, and (3) explores the validity of EVA outputs.Methods:EVA outputs are compared with accelerometer data from 4 individuals (Study 1a and1b) and 3 occupational groups (Study 2): seated workstation office workers (n = 8), standing workstation office workers (n = 8), and teachers (n = 8).Results:Line graphs and related EVA graphs highlight the use of EVA derivatives for examining compliance with guidelines. EVA derivatives of occupational groups confirm no difference in bouts of activity but clear differences as expected in extended bouts of SB and brief bursts of activity, thus providing evidence of construct validity.Conclusions:EVA offers a unique and comprehensive generic method that is able, for the first time, to capture the time pattern (both frequency and intensity) of PA and SB, which can be tailored for both occupational and public health research.

Abstract MP01: Association of Objectively-measured Physical Activity with Cardiovascular Risk in Mobility-limited Older Adults

Circulation ◽  
2014 ◽  
Vol 129 (suppl_1) ◽  
Author(s):  
Thomas W Buford ◽  
Don G Hire ◽  
Walter T Ambrosius ◽  
Stephen D Anton ◽  
Timothy S Church ◽  
...  
Keyword(s):  
Physical Activity ◽  
Older Adults ◽  
Sedentary Behavior ◽  
Risk Score ◽  
Activity Data ◽  
Mean Intensity ◽  
Wear Time ◽  
Low Intensity ◽  
Objectively Measured ◽  
The Impact

Introduction: In middle-aged adults, time spent being sedentary is associated with cardiovascular (CV) health risks independent of structured physical activity (PA). However, data are sparse regarding the impact of sedentary behavior on CV risk in older adults. The extent to which the absolute duration or intensity of daily PA reduces CV risk in older adults is also unknown. Objectives: Our objective was to examine the cross-sectional association between objectively-measured sedentary behavior and predicted CV risk among older adults in the Lifestyle Interventions and Independence for Elders (LIFE) study. The secondary objective was to evaluate associations between the duration/intensity of daily PA and predicted CV risk. Methods: LIFE is a randomized clinical trial to determine if regular PA prevents mobility disability among mobility-limited older adults. Activity data were collected by hip-worn accelerometer at baseline prior to participation in study interventions. Only participants with at least three days of accelerometry data (≥ 10 hrs wear time) were included. Unadjusted and adjusted linear regression was used to model the relationship between accelerometry measures and predicted 10-year Framingham risk of Hard Coronary Heart Disease (HCHD; i.e. myocardial infarction or coronary death). Adjusted models included demographic confounders (e.g. education, race, income) and health parameters (e.g. depression, cognition, arthritis) not in the risk score. Accelerometry cut-points were (in counts/min): sedentary behavior: < 100; low-intensity activity: 100-499; higher intensity activity: > 500. Results: Participants (n = 1170; 78.7 ± [SD] 5.3 years; 66.1% female) had a median HCHD risk of 10.3% (25 th -75 th %: 5.7-18.6). Over a mean accelerometer wear time of 8.1 ± 3.2 days, participants spent 77.0 ± 8.2% of their time sedentary. They also spent 16.6 ± 5.0% of their time in low-intensity PA and 6.4 ± 4.4% in higher-intensity PA. For all PA performed (> 100 counts/min), participants achieved a median of 393.4 (337.8-473.5) counts/min. In the unadjusted model, time spent sedentary (β = 2.41; 95% CI : 1.94, 2.89), in low-intensity PA (-2.56; -3.03, -2.08), and in higher-intensity PA (-1.60; -2.09, -1.11) were all associated with HCHD risk (all p’s < 0.001). These associations remained significant after adjustment. The mean intensity of daily PA was not significantly associated with HCHD risk in any model (p > 0.05). Conclusions: Daily time spent being sedentary is positively associated with predicted 10-year HCHD risk among mobility-limited older adults. Duration, but not mean intensity, of daily PA is inversely associated with HCHD risk score in this population.

Classification and Processing of 24-Hour Wrist Accelerometer Data

2018 ◽  
Vol 1 (2) ◽  
pp. 51-59 ◽  
Author(s):  
Anna Pulakka ◽  
Eric J. Shiroma ◽  
Tamara B. Harris ◽  
Jaana Pentti ◽  
Jussi Vahtera ◽  
...  
Keyword(s):  
Physical Activity ◽  
Sedentary Time ◽  
Total Activity ◽  
Sleep Time ◽  
Accelerometer Data ◽  
Classification Methods ◽  
Wake Time ◽  
Wear Time ◽  
Aging Study ◽  
Activity Measures

Background: An important step in accelerometer data analysis is the classification of continuous, 24-hour data into sleep, wake, and non-wear time. We compared classification times and physical activity metrics across different data processing and classification methods.Methods: Participants (n = 576) from the Finnish Retirement and Aging Study (FIREA) wore an accelerometer on their non-dominant wrist for seven days and nights and filled in daily logs with sleep and waking times. Accelerometer data were first classified as sleep or wake time by log, and Tudor-Locke, Tracy, and ActiGraph algorithms. Then, wake periods were classified as wear or non-wear by log, Choi algorithm, and wear sensor. We compared time classification (sleep, wake, and wake wear time) as well as physical activity measures (total activity volume and sedentary time) across these classification methods.Results:M(SD) nightly sleep time was 467 (49) minutes by log and 419 (88), 522 (86), and 453 (74) minutes by Tudor-Locke, Tracy, and ActiGraph algorithms, respectively. Wake wear time did not differ substantially when comparing Choi algorithm and the log. The wear sensor did not work properly in about 29% of the participants. Daily sedentary time varied by 8–81 minutes after excluding sleep by different methods and by 1–18 minutes after excluding non-wear time by different methods. Total activity volume did not substantially differ across the methods.Conclusion: The differences in wear and sedentary time were larger than differences in total activity volume. Methods for defining sleep periods had larger impact on outcomes than methods for defining wear time.

scholarly journals Reactivity and reproducibility of accelerometer-based sedentary behavior and physical activity in two measurement periods: Results of a feasibility randomized-controlled study

2019 ◽  
Author(s):  
Antje Ullrich ◽  
Sophie Baumann ◽  
Lisa Voigt ◽  
Ulrich John ◽  
Sabina Ulbricht
Keyword(s):  
Physical Activity ◽  
Sedentary Behavior ◽  
Growth Models ◽  
Controlled Study ◽  
Latent Growth ◽  
Latent Growth Models ◽  
Wear Time ◽  
Measurement Reactivity ◽  
Intraclass Correlations ◽  
Registration Number

Abstract Background: The aims of the study were to investigate measurement reactivity in sedentary behavior (SB), physical activity (PA), and accelerometer wear time in two measurement periods, to examine the reproducibility of these outcomes and to quantify measurement reactivity as a confounder for the reproducibility of SB and PA data.Methods: A total of 136 participants (65% women, mean age = 54.6 years, study period 02/2015 to 08/2016) received 7-day accelerometry at baseline and after 12 months. Latent growth models were used to identify measurement reactivity in each period. Intraclass correlations (ICC) were calculated to examine the reproducibility using two-level mixed-effects linear regression analyses. Results: At both measurements periods, participants increased time spent in SB (b=2.3 min/d; b=3.8 min/d), reduced time spent in light PA (b=2.0 min/day; b=3.3 min/d), but did not change moderate-to-vigorous PA. Participants reduced accelerometer wear time (b=4.6 min/d) only at baseline. The ICC coefficients ranged from 0.42 (95% CI=0.29-0.57) for accelerometer wear time to 0.70 (95% CI=0.61-0.78) for moderate-to-vigorous PA. In none of the regression models, a reactivity indicator was identified as a confounder for the reproducibility of SB and PA data.Conclusions: The results show that measurement reactivity differentially influences SB and PA in two measurement periods. Although 7-day accelerometry seems to be a reproducible measure of SB and PA, our findings highlight the importance of accelerometer wear time as a crucial confounder when using accelerometry in monitoring SB and PA, planning interventions, and analyzing SB and PA data.Clinical trial registration number: NCT02990039 (retrospectively registered); December 12, 2016

Sign in / Sign up

Export Citation Format

Share Document