The impact of sedative-hypnotic use on sleep symptoms in elderly nursing home residents*

Abstract

OBJECTIVE: To determine the frequency of sleep-related complaints among institutionalized elderly subjects and to assess the relationship between perceived sleep quality and the use of sedative-hypnotic agents and other psychoactive medications. METHODS: In 12 nursing homes in Massachusetts, we conducted observational, cross-sectional, and longitudinal studies of 145 institutionalized elderly subjects (average age, 83.0 years; age range, 65 to 105 years). We recorded the patients' demographic characteristics and all medication use (both scheduled and as needed) during a 1-month baseline period. A research assistant who was blinded to diagnoses and medication use performed detailed neuropsychologic testing and administered a series of standardized questions concerning difficulty sleeping, early morning awakening, and time spent awake in bed. Medication use and patient assessments were repeated after a 6-month interval. RESULTS: One or more sleep-related complaints were present at baseline in 94 (65%) of the residents studied. Using logistic regression to adjust for potential confounding, we found no relationship in the baseline month between use of sedative-hypnotic agents and the presence or absence of sleep complaints. After 6 months of follow-up, 27 (19%) of the residents had decreased their use of sedative-hypnotic agents and 23 (16%) had increased their use. However, there was no relationship between decreased use of sedative-hypnotic agents and worsened sleep (p > 0.20) or between their increased use and improved sleep reports (p > 0.10). Improvement in functional status was significantly associated with improved sleep at follow-up (p < 0.005). CONCLUSIONS: Sleep complaints occur in the majority of institutionalized elderly persons. Neither cross-sectional nor longitudinal analyses showed a relationship between patterns of sedative-hypnotic use and the presence, absence, or change in sleep complaints.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 56 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.