Key points are not available for this paper at this time.
BACKGROUND: This study tested the hypothesis that behaviorally defined sleep disturbance among residents in a Special Care Unit (SCU) for Alzheimer's disease (AD) was related to the severity of their dementia. Previously, sleep laboratory studies have reported such relationships when sleep has been recorded polysomnographically over several nights. Observational studies of sleep have not shown such relationships, presumably because of the impression involved in determining sleep/wake state behaviorally. METHODS: Nightly sleep data based on observations made every two hours by nursing staff for a period of 13 to 18 months were examined for 47 AD residents with a mean age of 80.7 +/- 6.5 years. Level of dementia and functional capacity were assessed with the Mattis Dementia Rating Scale (DRS) and the Katz Activities of Daily Living Scale (ADL). RESULTS: Data indicated that these SCU residents experienced a moderately disturbed night of sleep an average of 24% +/- 10% of their nights in the facility and a severely disturbed night of sleep on 7% +/- 6% of those nights. More profound dementia was associated with more sleep disturbance; however, incapacity in ADLs, age, gender, and psychoactive medications were unrelated to such disturbances. CONCLUSIONS: The results indicate that patterns of relationships noted between laboratory-based measures of sleep and variables such as severity of dementia can be detected using behavioral observations of sleep, provided that the number of nights of observation are sufficiently large to offset the measurement error involved in their use.
Bliwise et al. (Wed,) studied this question.