Abstract
INTRODUCTION: Objective sleep disturbances, including short and fragmented sleep, are observed in neurodegenerative diseases. However, subjective sleep disturbances are inconsistently reported. Improved understanding of objective and subjective sleep estimation is needed to tailor sleep interventions. METHODS: Baseline subjective habitual sleep was compared to objective sleep measured by actigraphy averaged over 8 weeks in 20 patients with mild cognitive impairment (MCI) or dementia due to Alzheimer's disease (AD) or Lewy body disease (LBD) and 20 healthy older adults. Discrepancies between objective and subjective sleep parameters were used to predict cohort membership (AD, LBD, or control). RESULTS: Participants with AD and LBD estimated lower sleep disturbance than actigraphy. Subjective sleep quality was poorest in LBD and highest in AD. Subjective sleep and subjective-objective sleep discrepancy discriminated between cohorts with 80% accuracy. DISCUSSION: Subjective and objective sleep differ and both should be measured in MCI and dementia. Sleep discrepancy may have diagnostic utility.