Self-reported sleep health is more than a combination of multiple objective sleep measures.

OBJECTIVES

The self-reported Ru-SATED questionnaire is a psychometrically validated measure of multidimensional sleep health. Ru-SATED is increasingly examined in relation to health outcomes; however, its relationship to behavioral sleep is unknown. This study describes the combinations of bed-measured sleep health features that are associated with the Ru-SATED sleep health score.

METHODS

Sleep Number bed users (N = 2674, Mage = 52.4 ± 16.2, 55.2% women) completed the Ru-SATED 4.0 multidimensional sleep health scale and contributed past-month smart bed data (7-28 days). Associations between Ru-SATED and bed features (duration, efficiency, regularity, timing) were modeled using: (1) Spearman correlations; (2) multivariable regression with bed features included additively; (3) polynomial regression with complex interactions among bed features combined with novel graphical techniques; and (4) non-parametric tree-structured regression.

RESULTS

Individual bed measures had weak correlations with Ru-SATED (r = 0.04-0.21; R2-adjusted = 0.002-0.040). Models incorporating multiple bed features explained larger proportions of variance in Ru-SATED (Additive regression R2-adjusted = 0.078; polynomial regression R2-adjusted = 0.084; tree-structured regression R2-adjusted = 0.090). Across models, bed-measured duration (ΔR2 = 0.015-0.018), regularity (ΔR2 = 0.011-0.019), and efficiency (ΔR2 = 0.010-0.015) made the largest contributions to Ru-SATED, followed by timing (ΔR2 = 0.001-0.003). Individuals with the best Ru-SATED score had good bed-measured regularity, efficiency, duration, and timing, while those with the worst Ru-SATED had some combination of poor regularity, duration, efficiency, and/or timing.

CONCLUSIONS

Ru-SATED was associated with combinations of behavioral bed measures, above any single bed measure. Bed-measured duration, regularity, and efficiency made the strongest contributions. Findings further validate the multidimensional sleep health framework and may assist with presenting clinically relevant information regarding patient sleep health patterns and goals.