Comparison of sleep parameters from wrist-worn ActiGraph and Actiwatch devices

Liu, Fangyu; Schrack, Jennifer A.; Wanigatunga, Sarah; Rabinowitz, Jill A.; He, Linchen; Wanigatunga, Amal A.; Zipunnikov, Vadim; Simonsick, Eleanor M.; Ferrucci, Luigi; Spira, Adam P.

doi:10.1093/sleep/zsad155

Cited by 5 publications

(3 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, for the Actigraph, triaxial accelerometer data were downloaded via a custom dock and aggregated into 1-min epochs for sleep-wake scoring using the Actilife implementation of the Cole-Kripke algorithm. 22,23 1-min epochs were scored from “Lights off” to “Lights on” and subsequently up-sampled to 30s epochs to match the resolution of the PSG data.…”

Section: Methodsmentioning

confidence: 99%

“…While the use of a research-grade actigraph like the GT9X provides acceptable sleep tracking as an adjunct to physical activity monitoring in research settings, the incremental value of having raw accelerometry data without HR sensors may be overestimated. Further, as recently demonstrated, 23 the equivalence of sleep measurements from research-grade actigraphy cannot be assumed.…”

Section: -Stage Sleep Classification Performancementioning

confidence: 96%

See 1 more Smart Citation

A Data Driven Approach for Choosing a Wearable Sleep Tracker

Ong,

Golkashani,

Ghorbani

et al. 2023

Preprint

View full text Add to dashboard Cite

Goal and AimsTo evaluate the performance of 6 wearable devices across 4 device classes (research-grade EEG-based headband, research-grade actigraphy, high-end consumer tracker, low-cost consumer tracker) over 3 age-groups (young: 18-30y, middle-aged: 31-50y and older adults: 51-70y).Focus TechnologyDreem 3 headband, Actigraph GT9X, Oura ring Gen3 running the latest sleep staging algorithm (OSSA 2.0), Fitbit Sense, Xiaomi Mi Band 7, Axtro Fit3.Reference TechnologyIn-lab polysomnography (PSG) with consensus sleep scoring.Sample60 participants (26 males) across 3 age groups (young: N=21, middle-aged: N=23 and older adults: N=16).DesignParticipants slept overnight in a sleep laboratory from their habitual sleep time to wake time, wearing 5 devices concurrently.Core AnalyticsDiscrepancy and epoch-by-epoch analyses for sleep/wake (2-stage) and sleep-stage (4-stage; wake/light/deep/REM) classification (devices vs. PSG). Mixed model ANOVAs for comparisons of biases across devices (within-subject), and age and sex (between-subjects).Core OutcomesThe EEG-based Dreem headband outperformed the other wearables in terms of 2-stage (kappa = .76) and 4-stage (kappa = .76-.86) classification but was not tolerated by at least 25% of participants. This was followed by the high-end, validated consumer trackers: Oura (2-stage kappa = .64, 4-stage kappa = .55-.70) and Fitbit (2-stage kappa = .58, 4-stage kappa = .45-.60). Next was the accelerometry-based research-grade Actigraph which only provided 2-stage classification (kappa = .47), and finally the low-cost consumer trackers which had very low kappa values overall (2-stage kappa < .31, 4-stage kappa < .33).Important Additional OutcomesProportional biases were driven by nights with poorer sleep (i.e., longer sleep onset latencies [SOL] and wake after sleep onset [WASO]). For those nights with sleep efficiency ≥85%, the large majority of sleep measure estimates from Dreem, Oura, Fitbit and Actigraph were within clinically acceptable limits of 30 mins. Biases for total sleep time [TST] and WASO were also largest in older participants who tended to have poorer sleep.Core ConclusionThe Dreem band is recommended for highest accuracy sleep tracking, but it has price, comfort and ease of use trade-offs. The high-end consumer sleep trackers (Oura, Fitbit) balance classification accuracy with cost, comfort and ease of use and are recommended for large-scale population studies where sleep is mostly normal. The low-cost trackers, despite poor wake detection could have some utility for logging time in bed.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: -Stage Sleep Classification Performancementioning

confidence: 96%

A Data Driven Approach for Choosing a Wearable Sleep Tracker

Ong,

Golkashani,

Ghorbani

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…But comparative studies of valid and reliable devices using longer epochs (30-60s) as in long-term data collection, are limited. In longitudinal recordings over multiple days the distinction of motionless alert from daytime sleep is required and necessitates calibrated actigraphic movement-based sensitivities and sleep algorithms (25, 26).…”

Section: Introductionmentioning

confidence: 99%

Behaviour-based movement cut-off points in 3-year old children comparing wrist- with hip-worn actigraphs MW8 and GT3X

Jansson,

Westlander,

Sandlund

et al. 2024

Preprint

View full text Add to dashboard Cite

Introduction: Behaviour-based physical intensities have not undergone rigorous calibration in long-term recordings of 3-year-old childrens sleep/activity patterns. This study aimed at (i) calibrating activity counts of motor behaviour measured simultaneously with MotionWatch 8 (MW8) and ActiGraph (GT3X) in 3-year-old children, (ii) documenting movement intensities in 30s-epochs at wrist/hip positions, and (iii) evaluating the accuracy of cut-off agreements between different behavioural activities. Methods: Thirty 3-year-old children of the NorthPop cohort performed six directed behavioural activities individually, each for 8-10 minutes while wearing two pairs of devices at hip and wrist position. Directly observed naturally-occurring behaviours included: watching cartoons, recumbent story listening, sit and handcraft, floor play with toys, engaging in a walk and a sprinting game. Receiver-Operating-Curve classification was applied to determine activity count thresholds and to assign context-guided, physical activity composite classes. Results: Activity counts of MW8 and GT3X pairs of wrist-worn (r = 0.94) and hip-worn (r = 0.79) devices correlated significantly (p < 0.001). Activity counts at hip position were significantly lower compared to those at the wrist position (p < 0.001), irrespective of device type. Sprinting, floorball/walk and floorplay assigned as "Physically Mobile" classes achieved outstanding accuracy (AUC >0.9) and two sedentary and a motionless activities assigned into "Physically Stationary" classes achieved excellent accuracy (AUC >0.8). Conclusion: This study provides useful cut-offs for physical activity levels of preschool children using two different devices. Contextual information of behaviour is advantageous over intensity classifications only, because interventions reallocate time among behaviours, which allows to establish dose-response relationships between behavioural changes and health outcomes. Our comparative calibration is one step forward to inform behaviour-based public health guidelines for 3-year-old children.

show abstract