Previous cross‐sectional studies have observed alterations in activity rhythms in dementia patients but the direction of causation is unclear. We determined whether circadian activity rhythms measured in community‐dwelling older women are prospectively associated with incident dementia or mild cognitive impairment (MCI).

Activity rhythm data were collected from 1,282 healthy community‐dwelling women from the Study of Osteoporotic Fractures (SOF) cohort (mean age 83 years) with wrist actigraphy for a minimum of three 24‐hour periods. Each participant completed a neuropsychological test battery and had clinical cognitive status (dementia, MCI, normal) adjudicated by an expert panel approximately 5 years later. All analyses were adjusted for demographics, body mass index (BMI), functional status, depression, medications, alcohol, caffeine, smoking, health status, and comorbidities.

After 4.9 years of follow‐up, 195 (15%) women had developed dementia and 302 (24%) had developed MCI. Older women with decreased activity rhythms had a higher likelihood of developing dementia or MCI when comparing those in the lowest quartiles of amplitude (odds ratio [OR] = 1.57; 95% CI, 1.09–2.25) or rhythm robustness (OR = 1.57; 95% CI, 1.10–2.26) to women in the highest quartiles. An increased risk of dementia or MCI (OR = 1.83; 95% CI, 1.29–2.61) was found for women whose timing of peak activity occurred later in the day (after 3:51 PM) when compared to those with average timing (1:34 PM–3:51 PM).

Older, healthy women with decreased circadian activity rhythm amplitude and robustness, and delayed rhythms have increased odds of developing dementia and MCI. If confirmed, future studies should examine whether interventions (physical activity, bright light exposure) that influence activity rhythms will reduce the risk of cognitive deterioration in the elderly. ANN NEUROL 2011