DOI: 10.1002/alz.077991 ISSN: 1552-5260

Application of a digital ambulatory protocol to assess the associations of level and consistency of sleep with daily cognitive performance and cognitive variability

Carol A. Derby, Qi Gao, Mindy J. Katz, Linying Ji, Orfeu M. Buxton, Cuiling Wang
  • Psychiatry and Mental health
  • Cellular and Molecular Neuroscience
  • Geriatrics and Gerontology
  • Neurology (clinical)
  • Developmental Neuroscience
  • Health Policy
  • Epidemiology

Abstract

Background

Most prior work regarding sleep and cognition in older adults has not examined the relation of sleep health to daily cognitive performance measured in real world settings. Fewer studies have examined the extent to which both level and consistency of sleep are related to variability in daily cognitive performance. We applied daily ambulatory assessment methods to examine whether associations of objective and daily subjective measures of sleep quality and duration were related to mean and day‐to‐day variability in cognitive performance.

Method

Adults (N = 261) age ≥ 70 years (mean 77.2 ± 4.7) in the Einstein Aging Study sleep project wore an actigraphy watch and completed smart‐phone cognitive assessments over two weeks. All participants were free of dementia (47% Non‐Hispanic White, 40% Non‐Hispanic Black, 13% Other race/ethnicity; mean years education 15 ± 3.6; 67% female). Brief (4‐5 minute) cognitive tests were completed 6 times/day to assess processing speed (Symbol Match Test‐SMT), Visual short term memory binding (Color Shape Test‐CST) and spatial working memory (Grid Memory Test‐GMT). Wake after sleep onset (WASO), night sleep duration (hours), and daytime napping (minutes) were estimated. All measures were summarized as mean over the 2‐weeks and day‐to‐day variability (SD). Linear regression was used to examine associations of mean and variability of sleep with mean and variability (SD) of cognitive performance, adjusting for age, sex, race, and education.

Result

Both mean and SD of sleep were associated with mean cognitive performance: Higher mean and greater SD WASO were associated with worse mean SMT and CST (all p <0.04). Greater SD duration was associated with worse mean GMT (p = 0.003) and greater SD napping was associated with worse SMT (p = 0.04). Sleep was also associated with daily variability in cognition. Greater mean napping was associated with greater SD GMT (p = 0.009), while SD WASO, SD duration and SD napping were each associated with greater variability in GMT (all p <0.01).

Conclusion

Both level and consistency of optimal sleep health are related to both average levels of cognitive functioning and day‐to‐day variability in cognitive function. Consistency of sleep health may be particularly important for minimizing variability in spatial working memory.

More from our Archive