The relationship between insomnia and mortality among community-dwelling older women

OBJECTIVES: To investigate the relationship between insomnia and 6-year survival among older, community-dwelling, white women. DESIGN: Longitudinal study. SETTING: The sample was selected randomly from a 20-census tract area of Baltimore, Maryland. PARTICIPANTS: In 1984, 778 women aged 65 years and older participated in the baseline survey; 613 were reinterviewed in 1985; 596 were reinterviewed in 1986. MEASUREMENTS: Insomnia was measured as self-report of at least one of three sleep complaints: trouble initiating sleep, waking at night and trouble falling back to sleep, and waking too early. Cox proportional hazards models were used to-assess the relationship between insomnia, as reported at the most recent of three interviews, and mortality. RESULTS: Twenty-nine percent of the sample reported trouble initiating sleep, 22.4% reported waking at night and having trouble falling back to sleep, and 26.6% reported waking too early. Of the sample, 43.2% had insomnia at the baseline interview; 23.2% percent of respondents with insomnia at baseline died during the follow-up period compared with 22.0% of those without insomnia at baseline. Recent insomnia was not associated significantly with mortality in either crude analyses (Hazards Ratio (HR) = 1.11, 95% CI, 0.78-1.60) or analyses adjusted for age, prescription medication use, functional limitations, self-rated health, and number of chronic conditions (HR = 0.74, 95% CI, 0.50-1.09). CONCLUSION: Our results, combined with the results from previous studies, suggest strongly that insomnia is not an indicator of approaching mortality in older community-dwelling populations.