Nocturnal sleep duration and cognitive impairment in a population-based study of older adults.

OBJECTIVE: We aimed to investigate the association between nocturnal sleep duration, changes in nocturnal sleep duration and cognitive impairment in older adults. METHODS: 4010 participants of a population-based cohort study provided information on nocturnal sleep duration at baseline (1991-1995) and at follow-up (2002/2003). 792 follow-up participants aged 70+ by 2006 participated in telephone-based cognitive assessments. Several cognitive tests were used including the telephone interview for cognitive status (TICS). Cognitive impairment was defined as <31 points on the TICS (13.0%) and as below this percentile on the other tests. Based on individual tests, a verbal memory score and a total score were constructed. Multivariable prevalence ratios (PRs) of cognitive impairment and 95% confidence intervals (95%CIs) were computed using Poisson regression. Analyses were restricted to those free of depression in 2002/2003 (n = 695). RESULTS: Sleeping or=9 h was positively, although imprecisely, associated with impairment of verbal memory (PR = 1.7, 95%CI = 1.0, 3.0), and less pronounced with the other cognitive measures. An increase in sleep duration from 7-8 h in 1992-1995 to >or=9 h 8.5 years later (versus sleeping 7-8 h at both time points) was associated with an increased prevalence of cognitive impairment according to the TICS (PR = 2.1, 95% = 1.0, 4.5) and the verbal memory score (PR = 2.0, 95%CI = 1.0, 3.8). CONCLUSIONS: Increases in sleep duration are associated with cognitive impairment. A biological explanation for this association is currently lacking. Increases in sleep duration could be a marker of cognitive deficits.

Diabetes and cognitive function in a population-based study of elderly women and men.

OBJECTIVES: The aim of this study was to examine the association between diabetes and cognitive function in the elderly. RESEARCH DESIGN AND METHODS: From January to December 2003, all 740 participants, aged 70 years or more, of an ongoing population-based cohort study were eligible for a telephone interview on cognitive function. Cognitive function was assessed using validated instruments, including the Telephone Interview of Cognitive Status (TICS) and the East Boston Memory Test (EBMT). Information on diabetes was available from prior questionnaires and was validated in 2002. We used multivariable logistic regression to estimate odds ratios (OR) of an impaired cognitive function (below 25th percentile) and their 95% confidence intervals (CI) adjusting for age, gender, smoking, alcohol consumption, body mass index (BMI), physical exercise, educational level, and depressive symptoms. RESULTS: Out of 473 participants interviewed (64.9%), 66 had diabetes (14.1%). The adjusted OR for diabetes and impaired cognitive function assessed by TICS was 2.3 (95% CI: 1.2-4.3). Diabetes was also associated with performance on delayed recall EBMT (adjusted OR=2.0; 95% CI: 1.0-4.1), but not immediate EBMT recall (adjusted OR=1.0; 95% CI: 0.5-2.1). The association between diabetes and cognitive function was a bit more pronounced in participants in whom diabetes was diagnosed 12 (median) or more years prior (adjusted OR with TICS=2.4; 95% CI: 1.0-5.8) and in those without antidiabetic treatment (age- and sex-adjusted OR=3.4; 95% CI: 1.7-6.5). CONCLUSION: Diabetes should be considered to be a risk factor for cognitive impairment in the elderly, which might be attenuated by antidiabetic treatment.