0278 Bidirectional Associations Between Gait Speed and Multidimensional Sleep Health: Evidence from the MIDUS Longitudinal Study

Abstract Introduction Gait speed and sleep health each reflect key aspects of functional and physiological aging. Although both are predictors of morbidity and mortality, little is known about their longitudinal, bidirectional relationship across adulthood. Slower gait speed may indicate underlying pain, inflammation, or neuromuscular decline that disrupts sleep, while poor sleep can impair physical performance over time. In this study, we examined the reciprocal association between gait speed and multidimensional sleep health among middle-aged and older adults in the US. Methods We used two waves of longitudinal data from the Midlife in the United States Study (MIDUS; N = 687; ages 34–81 at baseline). Gait speed was assessed using a two-trial 50-foot timed walk, with outcome in meters-per-second. Sleep health was measured using the five-component RuSATED index (regularity, satisfaction, alertness, efficiency, duration), and a composite score created by summing five dichotomized sleep problems, where higher scores reflect poorer sleep health. We estimated cross-lagged panel models to evaluate reciprocal associations between gait speed and sleep health over 9-year period. Models sequentially adjusted for sociodemographic covariates (age, sex, race, education, marital status) and then additional health covariates (BMI, depressive symptoms, chronic conditions). Results In sociodemographic-adjusted models, greater sleep problems at baseline predicted slower gait speed at follow-up (β = -0.018, CI: -0.034, -0.002). Baseline higher gait speed also predicted fewer sleep problems nine years later (β = -0.38, CI: -0.75, -0.01). After full adjustment for sociodemographic and health covariates, the association from sleep to gait speed remained significant (β = -0.016, CI: -0.032, -0.001), whereas the reverse association from gait speed to later sleep problems was attenuated and no longer significant (β = -0.22, CI: -0.60, 0.16). In sex-stratified analyses, the association from sleep to gait speed remained significant only in men (β = -0.027, CI: -0.051, -0.003), but not in women. Conclusion Findings indicate that sleep problems may be a stronger and more consistent predictor of later gait decline than gait speed is of subsequent sleep problems. These results suggest that sleep health represents a potentially modifiable early risk factor for mobility deterioration in midlife and older adulthood, particularly among men. Support (if any)