Relationships of psychiatric disorders with sleep duration in an adult general population sample.

The objective was to explore psychiatric disorders as potential predictors of sleep duration. A cross-sectional survey study with a probability sample of residents of a northern German area was carried out. There were 4075 study participants, aged 18--64 years, with a participation rate of 70.2%. Face-to-face in-home computer-aided interviews (Composite International Diagnostic Interview) provided diagnoses of nicotine and alcohol dependence, alcohol abuse, depressive, anxiety and somatoform disorders according to the Diagnostic and Statistical Manual of the American Psychiatric Association (DSM-IV), and included questions about sleep duration. Results show that subjects with a short sleep duration of 5h or less had significantly increased odds ratios (OR) for current nicotine dependence (OR 1.9,; confidence interval, CI, 1.2--2.9), alcohol dependence (OR 2.6, CI 1.2--5.6), depressive disorder (OR 3.0, CI 1.7--5.4) or anxiety disorder (OR 2.1, CI 1.3--3.4) compared to individuals who never had the respective psychiatric disorder after adjustment for sex, age, and school education in a multinomial regression analysis. The conclusion is drawn that current nicotine or alcohol dependence, depressive, and anxiety disorders may add to short sleep duration in this random adult general population sample.