BACKGROUND:

The association between objectively measured sleep and cognition among community-dwelling elderly persons remains understudied. This observational, cross-sectional analysis examined this association.

METHODS:

Results are from 2932 women (mean age 83.5 years) in the Study of Osteoporotic Fractures between 2002 and 2004. Cognitive function was measured by Mini-Mental State Examination (MMSE) and Trail Making B Test (Trails B). Cognitive impairment was defined as MMSE < 26 or Trails B > 278 seconds. Sleep parameters measured objectively using actigraphy included total sleep time, sleep efficiency, sleep latency, wake after sleep onset (WASO), and total nap time.

RESULTS:

There were 305 women (10.6%) with MMSE < 26 and 257 women (9.3%) with Trails B > 278 seconds. Compared with women with sleep efficiency > or = 70%, those with <70% had a higher risk of cognitive impairment (MMSE < 26 multivariate odds ratio [MOR] = 1.61; 95% confidence interval [CI], 1.20-2.16; Trails B > 278 MOR = 1.96; 95% CI, 1.43-2.67). Higher sleep latency was associated with higher risk of cognitive impairment (per half hour: MMSE < 26 MOR = 1.23; 95% CI, 1.13-1.33; Trails B > 278 MOR = 1.13; 95% CI, 1.04-1.24), as was higher WASO (per half hour: MMSE < 26 MOR = 1.15; 95% CI, 1.06-1.23; Trails B > 278 MOR = 1.24; 95% CI, 1.15-1.34). Women who napped > or = 2 hours per day had a higher risk (MMSE < 26 MOR = 1.42; 95% CI, 1.05-1.93; Trails B > 278 MOR = 1.74; 95% CI, 1.26-2.40). There was no significant relationship for total sleep time.

CONCLUSION:

Objectively measured disturbed sleep was consistently related to poorer cognition, whereas total sleep time was not. This finding may suggest that it is disturbance of sleep rather than quantity that affects cognition.