Integration of Report-Based Methods to Enhance the Interpretation of Monitor-Based Research: Results From the Free-Living Activity Study for Health Project

Click name to view affiliation

Nicholas R. Lamoureux Iowa State University, Ames, IA, USA

Search for other papers by Nicholas R. Lamoureux in
Current site
Google Scholar
PubMed
Close
*
,
Paul R. Hibbing Children’s Mercy Kansas City, Kansas City, MO, USA

Search for other papers by Paul R. Hibbing in
Current site
Google Scholar
PubMed
Close
*
,
Charles Matthews National Cancer Institute, Rockville, MD, USA

Search for other papers by Charles Matthews in
Current site
Google Scholar
PubMed
Close
*
, and
Gregory J. Welk Iowa State University, Ames, IA, USA

Search for other papers by Gregory J. Welk in
Current site
Google Scholar
PubMed
Close
Open access

Accelerometry-based monitors are commonly utilized to evaluate physical activity behavior, but the lack of contextual information limits the interpretability and value of the data. Integration of report-based with monitor-based data allows the complementary strengths of the two approaches to be used to triangulate information and to create a more complete picture of free-living physical behavior. This investigation utilizes data collected from the Free-Living Activity Study for Health to test the feasibility of annotating monitor data with contextual information from the Activities Completed Over Time in 24-hr (ACT24) previous-day recall. The evaluation includes data from 134 adults who completed the 24-hr free-living monitoring protocol and retrospective 24-hr recall. Analyses focused on the relative agreement of energy expenditure estimates between ACT24 and two monitor-based methods (ActiGraph and SenseWear Armband). Daily energy expenditure estimates from ACT24 were equivalent to the reference device-based estimate. Minute-level agreement of energy expenditure between ACT24 and device-based methods was moderate and was similar to the agreement between two different monitor-based methods. This minute-level agreement between ACT24 and device-based methods demonstrates the feasibility and utility of integrating self-report with accelerometer data to provide richer context on the monitored behaviors. This type of integration offers promise for advancing the assessment of physical behavior by aiding in data interpretation and providing opportunities to improve physical activity assessment methods under free-living conditions.

The accurate assessment of physical behavior is critical for advancing research on physical activity (PA) and sedentary behavior (SB). While accelerometers are widely utilized due to their ability to provide objective information of human movement, they provide limited contextual information (e.g., about specific types of activities or their purpose/domains). Fully integrating accelerometer data with self-report data has been recommended to address this limitation (Troiano et al., 2012); however, there are few examples of it being employed in practice.

There are many applications where context from report-based data can aid in interpretation. For example, surveillance research has demonstrated changes in PA levels both over the course of the lifespan (e.g., Farooq et al., 2020) and across generations (e.g., Canizares & Badley, 2018); however, report-based measures are critical for identifying the nature of these changes or differences. The limitations of report-based measures are well known and researchers have commented on specific issues with social desirability (Adams et al., 2005), the lack of reliability (Prince et al., 2008), and issues with standardized implementation. Despite the limitations, report-based measures remain by far the most practical way to differentiate among different domains of PA (Brownson et al., 2005).

Beyond the ability to classify domains of PA, linked self-report data provides opportunities to utilize information, such as environmental or social context, perceived activity intensity, and the intentions underlying activity patterns to advance understanding of PA patterns and their determinants. Linking self-report data also allows for activities typically not identified by accelerometers, such as cycling or weight training, to be detected. Capturing and coding complex activities under free-living conditions is important for addressing underlying errors in current assessment methods.

By sequentially accounting for each activity completed over the course of the day, the Activities Completed over Time in 24 Hours (ACT24; Matthews et al., 2018) previous-day recall allows data to be extracted in a “timestamped” format which can then be directly integrated with device-based methods to provide valuable contextual information. The present study capitalizes on the temporally linked data in the Free-Living Activity Study for Health (FLASH) project (Hibbing et al., 2021) to directly explore this potential integration. The primary purpose is to compare minute-level energy expenditure (EE) data from ACT24 with a device-based reference to evaluate temporal agreement. A secondary purpose is to explore the potential to contextualize data using examples of research questions that can be addressed when methods are integrated.

Method

Participants and Procedures

Data were collected as part of the ongoing FLASH project. Briefly, participants in the FLASH are recruited to wear several activity monitors for 24-hr of free-living activity. Prior to the assessment period, participants are oriented to the ACT24 and provided an overview of each of the monitors to be worn. The monitors are then worn for all nonwater-based activities (e.g., swimming, showering) from midnight to midnight on the day of evaluation. The following morning, participants completed an ACT24 recall, recording each activity completed during the assessment from memory. The present analyses use a subset of measures that were collected from the full cohort of original participants in FLASH (n = 134).

Measures

Activities Completed Over Time in 24-hr

The ACT24 is an internet-based recall, adapted from interviewer administered recalls to be self-administered (Matthews et al., 2018). Inputting each activity involves selecting from one of 175 specific activities (e.g., showering or bathing) from 14 broad categories (e.g., personal care), providing a start and end time rounded to the nearest 5-min unit of time, and answering follow-up questions to provide additional contextual details, such as posture or setting. Once all details for an activity are provided, it is added to the timeline and the process is repeated until the full 24 hr has been accounted for. Each activity entered is then linked to the compendium of physical activities (Ainsworth et al., 2011) to provide an estimate of EE in metabolic equivalents (METs). Sedentary behaviors were classified as sitting/lying behaviors in the waking day that required lower EE, typically ≤1.50 METs. Since accelerometer-based methods are not able to distinguish between SB and sleep, reported sleep time was combined with SB to allow for reasonable comparisons of time use with SenseWear Armband (SWA) and Sojourn (SOJ). Light PA (LPA) was classified as non-SB <3 METs, and moderate to vigorous PA (MVPA) was classified as ≥3.00 METs. The ACT24 has been shown to provide accurate estimates of EE, SB, and PA (Matthews et al., 2018) at the full-day level.

SenseWear Armband

The SWA (no longer marketed, BodyMedia Inc., Pittsburgh, PA) is utilized as a reference measure in FLASH due to the high agreement with estimates of EE determined by doubly labeled water (Calabro et al., 2013; Johannsen et al., 2010). Worn on the nondominant triceps, SWA includes accelerometer, galvanic skin response, and heat and heat flux sensors to estimate EE. After wear, data are downloaded using SenseWear Professional software (no longer marketed; BodyMedia Inc., Pittsburgh, PA) and saved to 60-second epoch files for analysis. The SB was classified as ≤1.50 METs, LPA was classified as >1.50 METs but <3.0 METs, and MVPA was classified as ≥3.00 METs.

Sojourn

The Sojourn method (Lyden et al., 2014) was used to estimate EE using data from an ActiGraph wGT3X-BT (ActiGraph LLC, Pensacola, FL) worn on the hip. The monitor is initialized to collect data at 100 Hz, and data are then downloaded and stored in 1-second epochs. The Sojourn methods, applied using open source R code (Sojourn version 1.1.0; Hibbing et al., 2019), first identifies distinct activity bouts with no gaps between them, then estimates EE for each bout (rather than the traditional approach of estimating EE for each epoch). Due to the unstructured and highly variable nature of free-living activity, this method has been shown to better estimate active and SB than traditional regression-based methods (Hibbing, Ellingson, et al., 2018; Lyden et al., 2014). Within each bout, EE is assumed to be constant, meaning the value can be assigned for each individual 1-second epoch in the bout. These values were then averaged each minute to obtain minute-by-minute EE in the same epoch length as the SWA. The SB was classified as ≤1.50 METs, LPA was classified as >1.5 and <3.0 METs, and MVPA was classified as ≥3.00 METs.

Statistical Analysis

Descriptive statistics of EE and time spent in SB, LPA, and MVPA were calculated to compare group-level estimates across the three measures. Activity intensity classification agreement was determined as the proportion of minutes that the comparison measures (ACT24 and SOJ) agreed with the reference measure (SWA) for each activity intensity. Pearson product correlations and mean absolute percent error (MAPE – calculated as [abs(SWA − comparison)/SWA] × 100) were calculated between EE estimates for each pair of measures for both 24 hr and minute-by-minute (calculated as the average of each participant’s individual agreement) levels of analysis. For ACT24–SOJ comparisons, SOJ was used as the reference measure. Agreement was evaluated at both the day- and minute-level using paired t tests, with minute-level agreement being calculated as the overall average of each participant’s individual minute-level agreement. Since a nonsignificant t test only indicates “not significantly different” rather than “the same,” when paired t test results were nonsignificant, equivalency was tested using procedures outlined by Dixon et al. (2018), with a ±10% zone of equivalence.

Exploratory Contextual Analysis

The ACT24 responses were used to provide examples of the value of additional contextual detail. Data were stratified by self-reported activity type to examine where MAPE was highest or lowest (i.e., to explore free-living sources of error). Agreement was examined using contextual information, such as location and social context to further explore potential sources of EE variability among measures. The contextual analyses will highlight the value of integrated data for the assessment of physical behavior.

Results

Agreement

Complete data (ActiGraph, SWA, and ACT24) were available from 134 participants. Table 1 shows the sample characteristics. Participants were predominantly young, White non-Hispanic, active individuals with favorable health self-perceptions. Since recruitment primarily occurred by convenience sampling on a university campus, most participants reported having completed at least some college, with 43% holding at least an undergraduate degree. Participants with more than 300 min of nonwear time for SWA or SOJ were excluded. Included participants had 1,353.8 ± 151.7 min of SOJ wear time and 1,400 ± 52.2 min of SWA wear time.

Table 1

Participant Characteristics

TotalMale (n = 54)Female (n = 80)
Age (years)26.1 ± 10.0927.7 ± 10.225 ± 9.95
Height173.07 ± 19.54180.91 ± 7.68167.78 ± 6.58
Weight77.59 ± 19.987.48 ± 22.2170.91 ± 14.85
BMI25.79 ± 5.6226.68 ± 6.2125.2 ± 5.13
BF%25.14 ± 9.1818.7 ± 7.829.46 ± 7.34
Handedness91.8% right handed88.9% R93.4% R

Note. BF% = body fat percentage; BMI = body mass index.

At the day level, estimates of total daily EE were similar for ACT24 and SWA (and equivalent at 10% level), but lower for SOJ (significantly different at p < .0001 compared with both ACT24 and SWA; see Table 2). Compared with either device-based method, ACT24 yielded higher estimates of SB time and lower estimates of LPA time. Compared with SWA, SOJ estimates of LPA were higher with correspondingly lower estimates of MVPA.

Table 2

Description of 24-hr EE and PA Time by Intensity

TotalMaleFemale
EE (MET·hr)
 ACT2437.8 ± 5.2238.63 ± 5.4737.22 ± 5.00
 SWA37.1 ± 7.2838.16 ± 7.2736.36 ± 7.24
 SOJ34.2 ± 4.1634.3 ± 3.7834.19 ± 4.42
SB (min)
 ACT241,133.6 ± 141.531,127.61 ± 161.251,137.64 ± 127.43
 SWA1,001.5 ± 152.12983.07 ± 158.751,014.01 ± 147.17
 SOJ1,021.4 ± 163.511,016.04 ± 211.671,025.08 ± 122.16
LPA (min)
 ACT24189 ± 127.53186.15 ± 139.85190.93 ± 119.36
 SWA264.9 ± 100.07267.44 ± 100.04263.18 ± 100.68
 SOJ346.3 ± 166.54349.35 ± 211.47344.29 ± 129.16
MVPA (min)
 ACT24110.9  ± 97.74116.30 ± 87.68107.29 ± 104.37
 SWA134.1 ± 84.08146.76 ± 86.76125.47 ± 81.66
 SOJ72.2 ± 40.3174.61 ± 36.0870.64 ± 43.08

Note. SWA = SenseWear Armband; SOJ = Sojourn analysis method using waist-worn ActiGraph; MET = metabolic equivalents; SB = sedentary behavior; ACT24 = Activities Completed Over Time in 24-hr; PA = physical activity; EE = energy expenditure.

Estimates of daily EE between SWA–ACT24 and SWA–SOJ were weakly correlated (.19 and .21, respectively), while correlation between SOJ–ACT24 estimates were more than twofold greater (.52). Day-level EE estimates were similar, with MAPE ranging from 13.2% to 18.1% for pairwise comparisons. Agreement for comparisons including the survey-based method were similar to the agreement between the two device-based assessments.

Correlations of MET estimates at the minute level (calculated as the mean of minute-by-minute correlations for each participant) ranged from .25 to .37. The minute-level correlations between ACT24 and either device-based method was higher than correlations between the two device-based methods (.31 and .37 vs. .25). Absolute differences in minute-level MET estimates between ACT24 and the two device-based methods were similar, though slightly higher than absolute agreement between SWA and SOJ (calculated as the average of each participant’s average minute-by-minute MAPE; see Table 3). All minute-level comparisons of EE were significantly different (p < .0001).

Table 3

Agreement of MET Estimates Between SWA, SOJ, and ACT24

SWA–ACT24SOJ–ACT24SWA–SOJ
Correlation
 Day level.19.52.21
 Minute level.31 ± .21.37 ± .19.25 ± .29
MAPE (%)
 Day level18.13 ± 14.3813.16 ± 11.6716.55 ± 12.35
 Minute level40.38 ± 14.3843.04 ± 12.2538.96 ± 14.50

Note. Minute-level agreement is calculated as the group mean of all participant level minute-by-minute agreement averages. For calculations of MAPE, the method listed first is used as the reference for percentage calculations. SWA = SenseWear Armband; SOJ = Sojourn analysis method using waist-worn ActiGraph; MET = metabolic equivalents; ACT24 = Activities Completed Over Time in 24-hr; MAPE = mean absolute percentage error.

Although methods resulted in substantial differences in absolute minute-by-minute MET estimates, agreement for intensity (i.e., time in SB, LPA, and MVPA) was generally good. Agreement was highest for SWA–ACT24 SB time. When compared with classification agreement between the two device-based methods, ACT24 had higher agreement with SWA for the classification of SB and MVPA, and lower for LPA (Table 4).

Table 4

Agreement of Activity Intensity Classification for Each Minute, Using SWA as Criterion Method for Determining Intensity

ACT24 intensitySOJ intensity
SWA intensitySBLPAMVPASBLPAMVPA
SB0.8420.1040.0510.7560.2120.030
LPA0.6870.1930.1110.6340.3050.061
MVPA0.5920.1930.2060.5090.3090.182

Note. SWA = SenseWear Armband; SOJ = Sojourn analysis method using waist-worn ActiGraph; SB = sedentary behavior; ACT24 = Activities Completed Over Time in 24-hr; MVPA = moderate to vigorous physical activity; LPA = light physical activity.

Contextualized Agreement

Table 5 shows results of our exploratory analysis of activities leading to discrepancies in device-assessed PA and how context of activity can aid in understanding variability in activity and potential sources of error. The top panel lists commonly reported activities with high levels of agreement between SOJ and SWA, such as sleeping (minute-by-minute MAPE = 24.9%) or computer work (33.6%). The bottom panel lists other activities with very low levels of agreement (such as jogging or cycling which had minute-by-minute MAPE of 100% or higher). In addition to summarizing agreement via MAPE, Table 5 also shows the percentage of minutes where EE estimates from SOJ were within 10% of SWA, ranging from 44.8% to 0.5%. This highlights the variability and error cancellation during evaluations of agreement at high resolution. Generally, SB and LPA associated with common lifestyle activities (e.g., preparing food or household cleaning) demonstrated higher agreement, while higher intensity activities, such as exercise (e.g., jogging or walking) corresponded to greater disagreement between SOJ and SWA. Although activity intensity is acknowledged to dramatically influence MAPE, more nuanced aspects of activity also appear to be important when comparing devices. For example, both computer work and driving are characterized as SB with EE estimates near 1.5 METs; however, minute-by-minute MAPE for driving is nearly 50% greater (47% vs. 31.9%), highlighting the importance of context to understanding sources of error.

Table 5

Agreement Between SWA and SOJ by Reported Activity (Minimum Five Reported Bouts)

Highest agreement
ActivityBouts reportedMean durationMean SWA METMean SOJ METMAPEProportion SOJ ± 10% SWA
Child care (dress, bath, and feed)829.8751.821.7223.4 ± 12.944.8
Sleeping or in bed193319.221.141.1224.9 ± 18.536.9
Cannot remember914.441.401.4625.3 ± 14.340.4
Playing computer or electronic games2895.861.531.2126.5 ± 14.727.1
Volunteer work9140.331.521.2126.7 ± 13.133.9
Attending service787.571.441.2229.4 ± 17.516.1
Computer or electronic games679.171.401.3631.3 ± 12.811.9
Showering or bathing6916.521.741.2532.2 ± 18.819.6
Reading (books, papers, and magazines)3962.381.721.2732.3 ± 21.521.9
Taking a nap3175.101.521.1333.5 ± 24.923.7
Lowest agreement
ActivityBouts reportedMean durationMean SWA METMean SOJ METMAPEProportion SOJ ± 10% SWA
Running or jogging2134.524.056.69252.9 ± 253.611.8
Cardio machines632.002.916.65212.1 ± 212.90.5
Stretching or flexibility exercises1134.551.974.19201.9 ± 274.929.2
Walking for exercise2229.822.894.53157.8 ± 127.67.8
Walking for transportation45411.262.133.22117.3 ± 12512.8
Bicycling for transportation4611.542.472.65103.1 ± 101.19.9
Walking the dog813.751.682.16100.8 ± 157.921.7
Standing—much walking in work area16135.131.842.51100.5 ± 133.220.1
Bicycling or exercise bike1229.172.102.6594.4 ± 806.6
Food shopping2822.962.172.2176 ± 89.215.4

Note. SWA = SenseWear Armband; SOJ = Sojourn analysis method using waist-worn ActiGraph; MET = metabolic equivalents; MAPE = mean absolute percentage error.

Contextual information related to location and social elements provided additional insights about variability in physical behavior. For example, the intensity during periods of self-reported walking for exercise varied based on who an individual was walking with and where the walk occurred. Analysis showed that during periods of reported walking for exercise, individuals had higher EE when walking with a spouse or partner than when walking alone (3.83 ± 0.47 vs. 2.74 ± 2.12 MET, p < .05), and that walking reported to have occurred in parks or trails was higher intensity than walks performed in fitness centers or other locations (3.76 ± 1.51 compared with 2.33 ± 2.64 and 2.15 ± 1.66 METs, respectively).

Discussion

This study documented the level of minute-by-minute and day-level agreement between EE estimates from the self-report-based ACT24 and two device-based methods. The exploratory analysis also demonstrated the value of temporal integration of monitor-based data with ACT24 data. Previous research has documented that the ACT24 has good agreement with device-based assessments at the full-day level (Matthews et al., 2018). The present study demonstrates that the utility of ACT24 extends to higher resolutions. While MAPE values are much higher at finer resolution, this is expected as coarser estimates allow periods of overestimation and underestimation to cancel out.

Although ACT24 was not designed for reporting activity at minute-by-minute resolution (Hibbing et al., 2021), it nevertheless achieved similar minute-level agreement with the reference measure (SWA) as SOJ did. Thus, ACT24 may be a suitable method of adding important contextual information to assessments of free-living activity. The ACT24 showed the greatest convergence with monitor methods for periods of SB, while agreement was much lower for active periods (ACT24 periods of non-SB agreed with 34.4% of SWA periods of non-SB), these limitations were similarly present when evaluating agreement between SOJ and SWA (30.3%). For some reported activities, such as household chores, weight training, or cycling, high agreement was observed between different device-based estimates but lower agreement between device and self-report estimates. This demonstrates the challenges of capturing activities where intensity is influenced by external resistance or isolated limb movements, as well as activities with potential differences between absolute and relative intensity.

For activities that are traditionally captured accurately by accelerometers, the integration also shows that the inherent variability within a single bout of free-living activity may not be reliably captured by self-report methods. For example, ACT24 codes all bouts of “walking for transportation” as 3.5 METs, while values from both SWA (2.17 ± 1.4) and SOJ (3.42 ± 2.66) vary more widely. Though the purpose of this evaluation is to determine if ACT24 can effectively be utilized to add contextual information to device-based assessments of activity, it is important to acknowledge that perfect agreement of EE is likely not possible. However, this is likely due to limitations within the methods themselves, rather than the specific measures used. For example, it is difficult to tease out whether the lack of agreement is due to intensity misclassification or problems associated with synchronizing reported clock time on ACT24 and computer clock time for the devices.

Matching data from PA recalls and accelerometers has previously been suggested as an important step in capturing the most complete picture of participant PA by including context and perception of activity alongside movement data (Troiano et al., 2012). While accelerometers may be useful for objectively quantifying activity intensity, ACT24 makes it possible to examine allocations into different activity domains. The ability to distinguish activities coded as leisure PA or occupational PA can be important in understanding behavior or evaluating interventions (Saint-Maurice et al., 2021). Similarly, distinguishing forms of SB is also important for understanding how to change these patterns. For example, allocations from the ACT24 can help to differentiate computer time from television time or work time (Matthews et al., 2021). While devices can capture the amount of SB (and associated breaks), the context is needed to explain and interpret decisions involved in human behavior.

The use of ecological momentary assessment methods have been previously advanced to capture subjective assessments of behaviors, such as environmental and social contexts, emotional state, and other details that are not easily inferred from accelerometers (Dunton, 2017). While there are some clear advantages of ecological momentary assessment, the present study demonstrates that ACT24 can provide detailed contextualized information on the full 24-hr activity cycle instead of being limited to isolated experiences within the day. While FLASH is not a representative sample and specific activities may not occur frequently enough to draw generalizable conclusions, the exploratory analysis reveals the potential to further extend on the domain-specific effects of activity and capture important contextual details that influence behavior, such as locations or social interactions. This type of insight provided by temporally linked context data can then be used to better understand changes in patterns of behavior or to design more effective behavior change interventions.

Linking survey and monitor data also offers opportunities to improve other methods of assessing PA. The temporally linked information about activity type provides a direct way to identify activities that may not be well detected with monitor data (e.g., biking or weightlifting) or to detect places where monitor methods do not agree with each other. In this case, the report-based data provide a way to triangulate the outcomes and more effectively interpret the data obtained from the monitor. Research has documented the challenges of differentiating SB from periods of nonwear (Oliver et al., 2011), as well as distinguishing between light and moderate activities (Welch et al., 2013). This has contributed to differences in recommendations of nonwear thresholds (Vanhelst et al., 2019), and varying recommendations for data analysis methods (e.g., Kim et al., 2012; Montoye, Begum, et al., 2017; Troiano et al., 2014). Research to date has also not been able to overcome the fundamental limitations of accelerometers for capturing swimming or cycling, nor the inability to accurately capture the metabolic costs of resistance training, or carrying heavy objects. Finally, studies have consistently observed that monitors perform worse when evaluating activities under free-living conditions (Dutta et al., 2018; Montoye, Conger, et al., 2017; Pavey et al., 2017). The contention here is that a more detailed understanding of agreement and error can be gained when participant reported activity is directly integrated with monitor-based data. For example, overall EE agreement between methods was poor for LPA but a more detailed look shows that for many common light activities, such as washing dishes or laundry, the two methods demonstrate relatively good agreement. Thus, the weaker associations for overall LPA may be driven by poor agreement for a few specific activities (e.g., shopping), and added context can help point researchers toward these specific activities to help refine activity assessment methods and algorithms.

The incorporation of ACT24 in the open-access FLASH data set provides investigators with opportunities to evaluate new algorithms or to compare existing algorithms for evaluating specific behaviors. Previous studies have demonstrated the value of integrated sensors, such as global positioning systems (Allahbakhshi et al., 2020), gyroscopes (Hibbing, Lamunion, et al., 2018), or heart rate monitors (Tapia et al., 2007) to more effectively assess common lifestyle activities. Validation studies have typically relied on direct observation (Keadle et al., 2014; Lyden et al., 2017; Montoye et al., 2020) or wearable cameras (Doherty et al., 2013; Harms et al., 2019) to capture the context of free-living behaviors. While these methods are robust for small-scale studies, the burden and cost likely preclude the use of this method for adding context to large-scale activity assessment efforts. The integration of the ACT24 presents an efficient method for capturing the context of data collected with monitor-based methods over a full 24-hr period. A previous paper documented the utility of the FLASH repository (Hibbing et al., 2021), but the present study demonstrates the specific value of the linked ACT24 data for PA and SB research.

Strengths and Limitations

To the best of our knowledge, this investigation is the first to temporally link report-based data with device-based methods of PA assessment at a minute-by-minute level. Previous research has linked monitor-based data to calibrate recall instruments (Saint-Maurice et al., 2017; Welk et al., 2017); however, studies to date have not directly linked report-based data to study or evaluate accelerometer assessments. A strength of the study was the evaluation of data over a full 24-hr cycle, which provides a more comprehensive evaluation of lifestyle patterns (Rosenberger et al., 2019). A methodological strength is the inclusion of data from two different monitor-based methods as this allowed for rich pairwise comparisons among the methods as well as a demonstration of how the context from the ACT24 can provide insights about lack of agreement in monitor-based methods.

The study documented the advantages of this methodological approach; however, there are also some limitations with the analysis. The study utilized a sample of mostly young, healthy, and educated participants, and included only a single day of monitoring. Thus, there are limitations with generalizability of the conclusions. However, the value of this type of contextual information when evaluating activity patterns is really the major conclusion from the paper. The FLASH project is an ongoing project so the sample of available data will grow over time. However, other researchers could also replicate the approach by including the ACT24 or another report-based tool to enhance the interpretation of monitor-based data.

Conclusions

Overall, daily EE estimates from the self-report based ACT24 demonstrate equivalency within a 10% tolerance to EE estimates from SWA, which is used as a criterion within FLASH. The strong associations are consistent with previous studies showing good agreement between ACT24 and criterion EE estimates from doubly labeled water (Matthews et al., 2018). Results are also supported by a cross-cultural study, which demonstrated the utility of previous day recall instruments, such as ACT24 for examining daily patterns of PA and SB (Matthews et al., 2019). However, this is the first study to test integration using more discrete data exports from the ACT24. At a minute-by-minute resolution, ACT24 demonstrates high agreement with SWA for identification of periods of SB, and reduced, although still comparable, agreement with evaluations from another established device-based method for estimates of EE and classification of activity intensity.

Although the methods may be variable with respect to EE estimates, integration provides an opportunity to add missing context to device-based methods of PA assessment. These contextual details, such as PA domain, intention, setting, or social setting can be used to address more complex epidemiological research questions, enhance understanding of PA determinants, and better evaluate interventions designed to increase PA or reduce SB.

Acknowledgment

FLASH is an ongoing project involving the collection of PA monitor data that are then made available through an open-access repository to promote and enable data sharing and collaboration to improve the assessment of physical behaviors.

References

  • Adams, S.A., Matthews, C.E., Ebbeling, C.B., Moore, C.G., Cunningham, J.E., Fulton, J., & Hebert, J.R. (2005). The effect of social desirability and social approval on self-reports of physical activity. American Journal of Epidemiology, 161(4), 389398. https://doi.org/10.1093/aje/kwi054

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ainsworth, B.E., Haskell, W.L., Herrmann, S.D., Meckes, N., Bassett, D.R., Jr., Tudor-Locke, C., Greer, J.L., Vezina, J., Whitt-Glover, M.C., & Leon, A.S. (2011). 2011 Compendium of Physical Activities: A second update of codes and MET values. Medicine & Science in Sports & Exercise, 43(8), 15751581. https://doi.org/10.1249/MSS.0b013e31821ece12

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Allahbakhshi, H., Conrow, L., Naimi, B., & Weibel, R. (2020). Using accelerometer and GPS data for real-life physical activity type detection. Sensors, 20(3), 588. https://doi.org/10.3390/s20030588

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Brownson, R.C., Boehmer, T.K., & Luke, D.A. (2005). Declining rates of physical activity in the United States: What are the contributors? Annual Review of Public Health, 26(1), 421443. https://doi.org/10.1146/annurev.publhealth.26.021304.144437

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Calabro, M.A., Stewart, J.M., & Welk, G.J. (2013). Validation of pattern-recognition monitors in children using doubly labeled water. Medicine & Science in Sports & Exercise, 45(7), 13131322.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Canizares, M., & Badley, E.M. (2018). Generational differences in patterns of physical activities over time in the Canadian population: An age-period-cohort analysis. BMC Public Health, 18(1), 111. https://doi.org/10.1186/s12889-018-5189-z

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dixon, P.M., Saint-Maurice, P.F., Kim, Y., Hibbing, P., Bai, Y., & Welk, G.J. (2018). A primer on the use of equivalence testing for evaluating measurement agreement. Medicine & Science in Sports & Exercise, 50(4), 837. https://doi.org/10.1249/MSS.0000000000001481

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Doherty, A.R., Kelly, P., Kerr, J., Marshall, S., Oliver, M., Badland, H., Hamilton, A., & Foster, C. (2013). Using wearable cameras to categorise type and context of accelerometer-identified episodes of physical activity. The International Journal of Behavioral Nutrition and Physical Activity, 10(1), 22. https://doi.org/10.1186/1479-5868-10-22

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dunton, G.F. (2017). Ecological momentary assessment in physical activity research. Exercise and Sport Sciences Reviews, 45(1), 48. https://doi.org/10.1249/JES.0000000000000092

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dutta, A., Ma, O., Toledo, M., Pregonero, A.F., Ainsworth, B.E., Buman, M.P., & Bliss, D.W. (2018). Identifying free-living physical activities using lab-based models with wearable accelerometers. Sensors, 18(11), 3893. https://doi.org/10.3390/s18113893

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Farooq, A., Martin, A., Janssen, X., Wilson, M.G., Gibson, A.M., Hughes, A., & Reilly, J.J. (2020). Longitudinal changes in moderate-to-vigorous-intensity physical activity in children and adolescents: A systematic review and meta-analysis. Obesity Reviews, 21(1), e12953. https://doi.org/10.1111/obr.12953

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Harms, T., Gershuny, J., Doherty, A., Thomas, E., Milton, K., & Foster, C. (2019). A validation study of the Eurostat harmonised European time use study (HETUS) diary using wearable technology. BMC Public Health, 19(2), 19.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hibbing, P.R., Ellingson, L.D., Dixon, P.M., & Welk, G.J. (2018). Adapted sojourn models to estimate activity intensity in youth: A suite of tools. Medicine & Science in Sports & Exercise, 50(4), 846854. https://doi.org/10.1249/MSS.0000000000001486

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hibbing, P.R., Lamoureux, N.R., Matthews, C.E., & Welk, G.J. (2021). Protocol and data description: The free-living activity study for health. Journal for the Measurement of Physical Behaviour, 4(3), 197204. https://doi.org/10.1123/jmpb.2020-0052

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hibbing, P.R., Lamunion, S.R., Kaplan, A.S., & Crouter, S.E. (2018). Estimating energy expenditure with ActiGraph GT9X Inertial Measurement Unit. Medicine & Science in Sports & Exercise, 50(5), 10931102. https://doi.org/10.1249/MSS.0000000000001532

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hibbing, P.R., Lyden, K., & Schwabacher, I.J. (2019). Sojourn: Apply Sojourn methods for processing ActiGraph accelerometer data. https://cran.r-project.org/package = Sojourn

    • Search Google Scholar
    • Export Citation
  • Johannsen, D.L., Calabro, M.A., Stewart, J., Franke, W., Rood, J.C., & Welk, G.J. (2010). Accuracy of armband monitors for measuring daily energy expenditure in healthy adults. Medicine & Science in Sports & Exercise, 42(11), 21342140. https://doi.org/10.1249/MSS.0b013e3181e0b3ff

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Keadle, S.K., Lyden, K., Hickey, A., Ray, E.L., Fowke, J.H., Freedson, P.S., & Matthews, C.E. (2014). Validation of a previous day recall for measuring the location and purpose of active and sedentary behaviors compared to direct observation. The International Journal of Behavioral Nutrition and Physical Activity, 11(1), 12. https://doi.org/10.1186/1479-5868-11-12

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kim, Y., Beets, M.W., & Welk, G.J. (2012). Everything you wanted to know about selecting the “right” Actigraph accelerometer cut-points for youth, but. . .: A systematic review. Journal of Science and Medicine in Sport, 15(4), 311321. https://doi.org/10.1016/j.jsams.2011.12.001

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lyden, K., Keadle, S.K., Staudenmayer, J., & Freedson, P.S. (2014). A method to estimate free-living active and sedentary behavior from an accelerometer. Medicine & Science in Sports & Exercise, 46(2), 386. https://doi.org/10.1249/MSS.0b013e3182a42a2d

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lyden, K., Keadle, S.K., Staudenmayer, J., & Freedson, P.S. (2017). The activPAL accurately classifies activity intensity categories in healthy adults. Medicine & Science in Sports & Exercise, 49(5), 1022. https://doi.org/10.1249/MSS.0000000000001177

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Matthews, C.E., Berrigan, D., Fischer, B., Gomersall, S.R., Hillreiner, A., Kim, Y., Leitzmann, M.F., Saint-Maurice, P., Olds, T.S., & Welk, G.J. (2019). Use of previous-day recalls of physical activity and sedentary behavior in epidemiologic studies: Results from four instruments. BMC Public Health, 19(S2), 478. https://doi.org/10.1186/s12889-019-6763-8

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Matthews, C.E., Carlson, S.A., Saint-Maurice, P.F., Patel, S., Salemo, E., Loftfield, E., Troiano, R.P., Fulton, J.E., Sampson, J.N., Tribby, C., Keadle, S., & Berrigan, D. (2021). Sedentary behavior in United States adults: Fall 2019. Medicine & Science in Sports & Exercise, 53(12), 25122519.

    • Search Google Scholar
    • Export Citation
  • Matthews, C.E., Keadle, S.K., Moore, S.C., Schoeller, D.S., Carroll, R.J., Troiano, R.P., & Sampson, J.N. (2018). Measurement of active and sedentary behavior in context of large epidemiologic studies. Medicine & Science in Sports & Exercise, 50(2), 266. https://doi.org/10.1249/MSS.0000000000001428

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Montoye, A.H., Begum, M., Henning, Z., & Pfeiffer, K.A. (2017). Comparison of linear and non-linear models for predicting energy expenditure from raw accelerometer data. Physiological Measurement, 38(2), 343. https://doi.org/10.1088/1361-6579/38/2/343

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Montoye, A.H., Clevenger, K.A., Pfeiffer, K.A., Nelson, M.B., Bock, J.M., Imboden, M.T., & Kaminsky, L.A. (2020). Development of cut-points for determining activity intensity from a wrist-worn ActiGraph accelerometer in free-living adults. Journal of Sports Sciences, 38(22), 25692578. https://doi.org/10.1080/02640414.2020.1794244

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Montoye, A.H., Conger, S.A., Connolly, C.P., Imboden, M.T., Nelson, M.B., Bock, J.M., & Kaminsky, L.A. (2017). Validation of accelerometer-based energy expenditure prediction models in structured and simulated free-living settings. Measurement in Physical Education and Exercise Science, 21(4), 223234. https://doi.org/10.1080/1091367X.2017.1337638

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Oliver, M., Badland, H.M., Schofield, G.M., & Shepherd, J. (2011). Identification of accelerometer nonwear time and sedentary behavior. Research Quarterly for Exercise and Sport, 82(4), 779783. https://doi.org/10.1080/02701367.2011.10599814

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Pavey, T.G., Gilson, N.D., Gomersall, S.R., Clark, B., & Trost, S.G. (2017). Field evaluation of a random forest activity classifier for wrist-worn accelerometer data. Journal of Science and Medicine in Sport, 20(1), 7580. https://doi.org/10.1016/j.jsams.2016.06.003

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Prince, S.A., Adamo, K.B., Hamel, M.E., Hardt, J., Gorber, S.C., & Tremblay, M. (2008). A comparison of direct versus self-report measures for assessing physical activity in adults: A systematic review. The International Journal of Behavioral Nutrition and Physical Activity, 5(1), 56. https://doi.org/10.1186/1479-5868-5-56

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rosenberger, M.E., Fulton, J.E., Buman, M.P., Troiano, R.P., Grandner, M.A., Buchner, D.M., & Haskell, W.L. (2019). The 24-hour activity cycle: A new paradigm for physical activity. Medicine & Science in Sports & Exercise, 51(3), 454464. https://doi.org/10.1249/MSS.0000000000001811

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Saint-Maurice, P.F., Berrigan, D., Whitfield, G.P., Watson, K.B., Patel, S., Loftfield, E., Sampson, J.N., Fulton, J.E., & Matthews, C.E. (2021). Amount, type, and timing of domain-specific moderate-to-vigorous physical activity among US adults. Journal of Physical Activity and Health, 18(Suppl. 1), S114S122. https://doi.org/10.1123/jpah.2021-0174

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Saint-Maurice, P.F., Kim, Y., Hibbing, P., Oh, A.Y., Perna, F.M., & Welk, G.J. (2017). Calibration and validation of the youth activity profile: The FLASHE study. American Journal of Preventive Medicine, 52(6), 880887. https://doi.org/10.1016/j.amepre.2016.12.010

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tapia, E.M., Intille, S.S., Haskell, W., Larson, K., Wright, J., King, A., & Friedman, R. (2007). Real-time recognition of physical activities and their intensities using wireless accelerometers and a heart rate monitor. Presented at the 2007 11th IEEE International Symposium on Wearable Computers.

    • Search Google Scholar
    • Export Citation
  • Troiano, R.P., Gabriel, K.K.P., Welk, G.J., Owen, N., & Sternfeld, B. (2012). Reported physical activity and sedentary behavior: Why do you ask? Journal of Physical Activity and Health, 9(Suppl. 1), S68S75. https://doi.org/10.1123/jpah.9.s1.s68

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Troiano, R.P., McClain, J.J., Brychta, R.J., & Chen, K.Y. (2014). Evolution of accelerometer methods for physical activity research. British Journal of Sports Medicine, 48(13), 10191023. https://doi.org/10.1136/bjsports-2014-093546

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Vanhelst, J., Vidal, F., Drumez, E., Béghin, L., Baudelet, J.-B., Coopman, S., & Gottrand, F. (2019). Comparison and validation of accelerometer wear time and non-wear time algorithms for assessing physical activity levels in children and adolescents. BMC Medical Research Methodology, 19(1), 72. https://doi.org/10.1186/s12874-019-0712-1

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Welch, W.A., Bassett, D.R., Thompson, D.L., Freedson, P.S., Staudenmayer, J.W., John, D., Steeves, J.A., Conger, S.A., Ceaser, T., & Howe, C.A. (2013). Classification accuracy of the wrist-worn GENEA accelerometer. Medicine & Science in Sports & Exercise, 45(10), 2012. https://doi.org/10.1249/MSS.0b013e3182965249

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Welk, G.J., Beyler, N.K., Kim, Y., & Matthews, C.E. (2017). Calibration of self-report measures of physical activity and sedentary behavior. Medicine & Science in Sports & Exercise, 49(7), 1473. https://doi.org/10.1249/MSS.0000000000001237

    • Crossref
    • Search Google Scholar
    • Export Citation

  • Collapse
  • Expand
  • Adams, S.A., Matthews, C.E., Ebbeling, C.B., Moore, C.G., Cunningham, J.E., Fulton, J., & Hebert, J.R. (2005). The effect of social desirability and social approval on self-reports of physical activity. American Journal of Epidemiology, 161(4), 389398. https://doi.org/10.1093/aje/kwi054

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ainsworth, B.E., Haskell, W.L., Herrmann, S.D., Meckes, N., Bassett, D.R., Jr., Tudor-Locke, C., Greer, J.L., Vezina, J., Whitt-Glover, M.C., & Leon, A.S. (2011). 2011 Compendium of Physical Activities: A second update of codes and MET values. Medicine & Science in Sports & Exercise, 43(8), 15751581. https://doi.org/10.1249/MSS.0b013e31821ece12

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Allahbakhshi, H., Conrow, L., Naimi, B., & Weibel, R. (2020). Using accelerometer and GPS data for real-life physical activity type detection. Sensors, 20(3), 588. https://doi.org/10.3390/s20030588

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Brownson, R.C., Boehmer, T.K., & Luke, D.A. (2005). Declining rates of physical activity in the United States: What are the contributors? Annual Review of Public Health, 26(1), 421443. https://doi.org/10.1146/annurev.publhealth.26.021304.144437

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Calabro, M.A., Stewart, J.M., & Welk, G.J. (2013). Validation of pattern-recognition monitors in children using doubly labeled water. Medicine & Science in Sports & Exercise, 45(7), 13131322.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Canizares, M., & Badley, E.M. (2018). Generational differences in patterns of physical activities over time in the Canadian population: An age-period-cohort analysis. BMC Public Health, 18(1), 111. https://doi.org/10.1186/s12889-018-5189-z

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dixon, P.M., Saint-Maurice, P.F., Kim, Y., Hibbing, P., Bai, Y., & Welk, G.J. (2018). A primer on the use of equivalence testing for evaluating measurement agreement. Medicine & Science in Sports & Exercise, 50(4), 837. https://doi.org/10.1249/MSS.0000000000001481

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Doherty, A.R., Kelly, P., Kerr, J., Marshall, S., Oliver, M., Badland, H., Hamilton, A., & Foster, C. (2013). Using wearable cameras to categorise type and context of accelerometer-identified episodes of physical activity. The International Journal of Behavioral Nutrition and Physical Activity, 10(1), 22. https://doi.org/10.1186/1479-5868-10-22

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dunton, G.F. (2017). Ecological momentary assessment in physical activity research. Exercise and Sport Sciences Reviews, 45(1), 48. https://doi.org/10.1249/JES.0000000000000092

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dutta, A., Ma, O., Toledo, M., Pregonero, A.F., Ainsworth, B.E., Buman, M.P., & Bliss, D.W. (2018). Identifying free-living physical activities using lab-based models with wearable accelerometers. Sensors, 18(11), 3893. https://doi.org/10.3390/s18113893

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Farooq, A., Martin, A., Janssen, X., Wilson, M.G., Gibson, A.M., Hughes, A., & Reilly, J.J. (2020). Longitudinal changes in moderate-to-vigorous-intensity physical activity in children and adolescents: A systematic review and meta-analysis. Obesity Reviews, 21(1), e12953. https://doi.org/10.1111/obr.12953

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Harms, T., Gershuny, J., Doherty, A., Thomas, E., Milton, K., & Foster, C. (2019). A validation study of the Eurostat harmonised European time use study (HETUS) diary using wearable technology. BMC Public Health, 19(2), 19.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hibbing, P.R., Ellingson, L.D., Dixon, P.M., & Welk, G.J. (2018). Adapted sojourn models to estimate activity intensity in youth: A suite of tools. Medicine & Science in Sports & Exercise, 50(4), 846854. https://doi.org/10.1249/MSS.0000000000001486

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hibbing, P.R., Lamoureux, N.R., Matthews, C.E., & Welk, G.J. (2021). Protocol and data description: The free-living activity study for health. Journal for the Measurement of Physical Behaviour, 4(3), 197204. https://doi.org/10.1123/jmpb.2020-0052

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hibbing, P.R., Lamunion, S.R., Kaplan, A.S., & Crouter, S.E. (2018). Estimating energy expenditure with ActiGraph GT9X Inertial Measurement Unit. Medicine & Science in Sports & Exercise, 50(5), 10931102. https://doi.org/10.1249/MSS.0000000000001532

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hibbing, P.R., Lyden, K., & Schwabacher, I.J. (2019). Sojourn: Apply Sojourn methods for processing ActiGraph accelerometer data. https://cran.r-project.org/package = Sojourn

    • Search Google Scholar
    • Export Citation
  • Johannsen, D.L., Calabro, M.A., Stewart, J., Franke, W., Rood, J.C., & Welk, G.J. (2010). Accuracy of armband monitors for measuring daily energy expenditure in healthy adults. Medicine & Science in Sports & Exercise, 42(11), 21342140. https://doi.org/10.1249/MSS.0b013e3181e0b3ff

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Keadle, S.K., Lyden, K., Hickey, A., Ray, E.L., Fowke, J.H., Freedson, P.S., & Matthews, C.E. (2014). Validation of a previous day recall for measuring the location and purpose of active and sedentary behaviors compared to direct observation. The International Journal of Behavioral Nutrition and Physical Activity, 11(1), 12. https://doi.org/10.1186/1479-5868-11-12

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kim, Y., Beets, M.W., & Welk, G.J. (2012). Everything you wanted to know about selecting the “right” Actigraph accelerometer cut-points for youth, but. . .: A systematic review. Journal of Science and Medicine in Sport, 15(4), 311321. https://doi.org/10.1016/j.jsams.2011.12.001

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lyden, K., Keadle, S.K., Staudenmayer, J., & Freedson, P.S. (2014). A method to estimate free-living active and sedentary behavior from an accelerometer. Medicine & Science in Sports & Exercise, 46(2), 386. https://doi.org/10.1249/MSS.0b013e3182a42a2d

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lyden, K., Keadle, S.K., Staudenmayer, J., & Freedson, P.S. (2017). The activPAL accurately classifies activity intensity categories in healthy adults. Medicine & Science in Sports & Exercise, 49(5), 1022. https://doi.org/10.1249/MSS.0000000000001177

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Matthews, C.E., Berrigan, D., Fischer, B., Gomersall, S.R., Hillreiner, A., Kim, Y., Leitzmann, M.F., Saint-Maurice, P., Olds, T.S., & Welk, G.J. (2019). Use of previous-day recalls of physical activity and sedentary behavior in epidemiologic studies: Results from four instruments. BMC Public Health, 19(S2), 478. https://doi.org/10.1186/s12889-019-6763-8

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Matthews, C.E., Carlson, S.A., Saint-Maurice, P.F., Patel, S., Salemo, E., Loftfield, E., Troiano, R.P., Fulton, J.E., Sampson, J.N., Tribby, C., Keadle, S., & Berrigan, D. (2021). Sedentary behavior in United States adults: Fall 2019. Medicine & Science in Sports & Exercise, 53(12), 25122519.

    • Search Google Scholar
    • Export Citation
  • Matthews, C.E., Keadle, S.K., Moore, S.C., Schoeller, D.S., Carroll, R.J., Troiano, R.P., & Sampson, J.N. (2018). Measurement of active and sedentary behavior in context of large epidemiologic studies. Medicine & Science in Sports & Exercise, 50(2), 266. https://doi.org/10.1249/MSS.0000000000001428

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Montoye, A.H., Begum, M., Henning, Z., & Pfeiffer, K.A. (2017). Comparison of linear and non-linear models for predicting energy expenditure from raw accelerometer data. Physiological Measurement, 38(2), 343. https://doi.org/10.1088/1361-6579/38/2/343

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Montoye, A.H., Clevenger, K.A., Pfeiffer, K.A., Nelson, M.B., Bock, J.M., Imboden, M.T., & Kaminsky, L.A. (2020). Development of cut-points for determining activity intensity from a wrist-worn ActiGraph accelerometer in free-living adults. Journal of Sports Sciences, 38(22), 25692578. https://doi.org/10.1080/02640414.2020.1794244

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Montoye, A.H., Conger, S.A., Connolly, C.P., Imboden, M.T., Nelson, M.B., Bock, J.M., & Kaminsky, L.A. (2017). Validation of accelerometer-based energy expenditure prediction models in structured and simulated free-living settings. Measurement in Physical Education and Exercise Science, 21(4), 223234. https://doi.org/10.1080/1091367X.2017.1337638

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Oliver, M., Badland, H.M., Schofield, G.M., & Shepherd, J. (2011). Identification of accelerometer nonwear time and sedentary behavior. Research Quarterly for Exercise and Sport, 82(4), 779783. https://doi.org/10.1080/02701367.2011.10599814

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Pavey, T.G., Gilson, N.D., Gomersall, S.R., Clark, B., & Trost, S.G. (2017). Field evaluation of a random forest activity classifier for wrist-worn accelerometer data. Journal of Science and Medicine in Sport, 20(1), 7580. https://doi.org/10.1016/j.jsams.2016.06.003

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Prince, S.A., Adamo, K.B., Hamel, M.E., Hardt, J., Gorber, S.C., & Tremblay, M. (2008). A comparison of direct versus self-report measures for assessing physical activity in adults: A systematic review. The International Journal of Behavioral Nutrition and Physical Activity, 5(1), 56. https://doi.org/10.1186/1479-5868-5-56

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rosenberger, M.E., Fulton, J.E., Buman, M.P., Troiano, R.P., Grandner, M.A., Buchner, D.M., & Haskell, W.L. (2019). The 24-hour activity cycle: A new paradigm for physical activity. Medicine & Science in Sports & Exercise, 51(3), 454464. https://doi.org/10.1249/MSS.0000000000001811

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Saint-Maurice, P.F., Berrigan, D., Whitfield, G.P., Watson, K.B., Patel, S., Loftfield, E., Sampson, J.N., Fulton, J.E., & Matthews, C.E. (2021). Amount, type, and timing of domain-specific moderate-to-vigorous physical activity among US adults. Journal of Physical Activity and Health, 18(Suppl. 1), S114S122. https://doi.org/10.1123/jpah.2021-0174

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Saint-Maurice, P.F., Kim, Y., Hibbing, P., Oh, A.Y., Perna, F.M., & Welk, G.J. (2017). Calibration and validation of the youth activity profile: The FLASHE study. American Journal of Preventive Medicine, 52(6), 880887. https://doi.org/10.1016/j.amepre.2016.12.010

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tapia, E.M., Intille, S.S., Haskell, W., Larson, K., Wright, J., King, A., & Friedman, R. (2007). Real-time recognition of physical activities and their intensities using wireless accelerometers and a heart rate monitor. Presented at the 2007 11th IEEE International Symposium on Wearable Computers.

    • Search Google Scholar
    • Export Citation
  • Troiano, R.P., Gabriel, K.K.P., Welk, G.J., Owen, N., & Sternfeld, B. (2012). Reported physical activity and sedentary behavior: Why do you ask? Journal of Physical Activity and Health, 9(Suppl. 1), S68S75. https://doi.org/10.1123/jpah.9.s1.s68

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Troiano, R.P., McClain, J.J., Brychta, R.J., & Chen, K.Y. (2014). Evolution of accelerometer methods for physical activity research. British Journal of Sports Medicine, 48(13), 10191023. https://doi.org/10.1136/bjsports-2014-093546

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Vanhelst, J., Vidal, F., Drumez, E., Béghin, L., Baudelet, J.-B., Coopman, S., & Gottrand, F. (2019). Comparison and validation of accelerometer wear time and non-wear time algorithms for assessing physical activity levels in children and adolescents. BMC Medical Research Methodology, 19(1), 72. https://doi.org/10.1186/s12874-019-0712-1

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Welch, W.A., Bassett, D.R., Thompson, D.L., Freedson, P.S., Staudenmayer, J.W., John, D., Steeves, J.A., Conger, S.A., Ceaser, T., & Howe, C.A. (2013). Classification accuracy of the wrist-worn GENEA accelerometer. Medicine & Science in Sports & Exercise, 45(10), 2012. https://doi.org/10.1249/MSS.0b013e3182965249

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Welk, G.J., Beyler, N.K., Kim, Y., & Matthews, C.E. (2017). Calibration of self-report measures of physical activity and sedentary behavior. Medicine & Science in Sports & Exercise, 49(7), 1473. https://doi.org/10.1249/MSS.0000000000001237

    • Crossref
    • Search Google Scholar
    • Export Citation
All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 3431 601 26
PDF Downloads 311 92 4