Cognition and 20-year subsequent sleep disturbances

Abstract

Introduction: There is a paucity of data exploring the extent that preclinical cognitive changes are predictive of subsequent sleep outcomes.

Methods: Logistic regression models were used to evaluate data from a cohort of 196 African American adults who had measures of cognitive function assessed at 2 time points during a 20-year period across the mid- to late-life transition. Cognitive testing included the Delayed Word Recall, the Digit Symbol Substitution, and the Word Fluency tests, which were summarized as a composite cognitive z-score. Sleep apnea was measured by in-home sleep apnea testing and sleep duration and quality were derived from 7-day wrist actigraphy at the end of the study period.

Results: A one standard deviation (SD) lower composite cognitive z-score at baseline was significantly associated with greater odds of low sleep efficiency (<85%) (odds ratio [OR] = 1.85, 95% confidence interval [CI] = 1.13, 3.04) and greater odds of increased wakefulness after sleep onset time (WASO; >60 minutes) (OR = 1.65, 95% CI = 1.05, 2.60) in adjusted models. A one SD faster rate of cognitive decline over the study period was significantly associated with greater odds of low sleep efficiency (OR = 1.68, 95% CI = 1.04, 2.73), greater odds of sleep fragmentation (>35%); (OR = 1.73, 95% CI = 1.05, 2.85), and greater odds of increased WASO (OR = 1.85, 95% CI = 1.15, 2.95) in adjusted models. Neither baseline cognitive z-score nor rate of cognitive decline was associated with sleep apnea or the total average sleep duration.

Conclusion: Cognition at baseline and change over time predicts sleep quality and may reflect common neural mechanisms and vulnerabilities.

Keywords: Cognition; Cognitive decline; Longitudinal analysis; Sleep; Sleep disturbances.