0307 Correlations Between Subjective and Objective Estimations of Sleep Quantity and Quality from Wearable Devices
Joseph Cuellar,Prayag Gordy,Jason Jameson,Rachel Markwald,Evan Chinoy
DOI: https://doi.org/10.1093/sleep/zsae067.0307
IF: 6.313
2024-04-20
SLEEP
Abstract:Abstract Introduction Beyond sleep quantity, another important factor for health and performance is sleep quality. While sleep quantity is reliably measured with objective methods like actigraphy, sleep quality is typically only assessed via subjective report, so it is still debated whether it can be estimated objectively. Many new-generation wearable devices have been found to equal or outperform actigraphy in sleep/wake measurement, but whether their estimates of sleep quantity or quality also correlate with subjective sleep estimates/ratings has yet to be established. Therefore, we explored whether subjective estimates/ratings of sleep (total sleep time [TST] and sleep quality) were correlated with objective estimates provided by commercial wearable devices. Methods In total, 33 healthy young adults participated in week-long home sleep studies. In Study 1, participants (n=21) habitually slept at night using a device set including Philips Respironics Actiwatch 2, Fatigue Science ReadiBand, Fitbit Inspire HR, Polar Vantage V Titan, and Oura Ring (OR). In Study 2, participants (n=12) with habitual irregular sleep schedules (i.e., some naps/daytime sleep) used a device set including Actiwatch 2, Garmin Fenix 7, and OR. For all sleep episodes, participants wore their devices and completed sleep diaries. Pearson correlations were calculated between devices and diaries for TST, and for sleep quality scores (0–100) from the OR algorithm compared with diary, Likert-type sleep quality ratings (1–5). P-values <.05 were considered statistically significant. Results Correlations between subjective and objective TST were significant for all devices. Likewise, OR sleep quality scores significantly correlated with subjective sleep quality ratings. Conclusion This analysis provides initial evidence for relationships between objective estimates of sleep quantity and quality from commercial wearable devices and subjective estimates/ratings from sleep diaries in healthy adults. While TST had previously been known to correlate between actigraphy and diaries, this had not been extensively tested in commercial wearables. Additionally, we found these relationships extended to OR’s sleep quality score. Such proprietary “sleep quality” scores are often the primary sleep-tracking metric displayed from new-generation wearables; however, scores are proprietary and without actigraphy equivalents. More testing is needed to explore whether these relationships extend to other populations/settings. Support (if any) Office of Naval Research, Code 34
neurosciences,clinical neurology