Abstract

STUDY OBJECTIVES: We investigated the prevalence and correlates of sleep complaints in a large well-characterized population. DESIGN: The Atherosclerosis Risk in Communities Study is a prospective population-based study of cardiovascular disease. Using this well-characterized cohort, we undertook a cross-sectional epidemiologic analysis of correlates of 3 sleep complaints that are commonly included in insomnia definitions. We hypothesized that different sleep complaints included under the rubric of "insomnia" would have different correlates. Using multivariate regression analysis we predicted the likelihood of endorsing the symptoms of difficulty falling asleep, difficulty staying asleep, and nonrestorative sleep by age, sex, alcohol intake, smoking, diabetes, heart disease, menopausal status, hypnotic use, hypertension, depression, education level, income, body mass index, respiratory symptoms, and pulmonary function. SETTING: North American communities. PARTICIPANTS: 13,563 participants aged 47 to 69 years. INTERVENTIONS: None. MEASUREMENTS AND RESULTS: The prevalences of sleep complaints were 22%, 39%, and 35% for difficulty falling asleep, difficulty staying asleep, and nonrestorative sleep, respectively. In contrast to previous reports, we found that Black race was associated with reduced risk of sleep complaints. We also found that increasing age was associated with difficulty staying asleep but not with difficulty falling asleep or nonrestorative sleep. Medical illness, depression, lower socioeconomic status, and unhealthy behaviors were associated with increased risk of sleep complaints, which varied by risk factor. CONCLUSIONS: In a well-characterized population-based study, specific sleep complaints have differing covariates. It is likely that difficulty falling asleep and difficulty staying asleep have different causes and outcomes. Sleep difficulties do not appear to be associated with black race per se.