Are personality dimensions associated with sleep length in a large nationally representative sample?

OBJECTIVES

The purpose of this hypothesis-generating study was to determine whether personality domains and specific personality traits are uniquely associated with sleep duration using data obtained from the National Comorbidity Survey (NCS).

METHODS

Using trained interviewers, we administered to the 5877 noninstitutionalized adults living in the United States the Composite International Diagnostic Interview (World Health Organization. Composite International Diagnostic Interview. Geneva, Switzerland: WHO, 1990) to assess for any DSM-III-R psychiatric diagnoses, and they completed self-report measures of personality and sleep. This was a secondary data analysis using information from a large existing public use data set (NCS-part II). The NCS-part II was an epidemiologic survey based on a stratified multistage area probability method with a response rate of 82.4%.

RESULTS

Using a multivariate logistic regression technique, we found significant and positive associations between short sleep (defined as <or=6 hours of sleep per 24-hour period) and self-criticism (odds ratio [OR] = 1.36), the presence of a medical condition (OR = 1.35), neuroticism (OR = 1.30), and the use of sedating medication in the past 12 months (OR = 1.26). Significant and positive associations were found for long sleep (defined as >or=9 hours per 24 hour period) and a diagnosis of dysthymia (OR = 1.52), the use of a sedating medication in the past 12 months (OR = 1.52), emotional reliance on another person (OR = 1.37), employment status (OR = 1.31), and marital status (OR = 1.20).

CONCLUSIONS

Findings suggest that personality, even after controlling for psychiatric and medical conditions, is associated with sleep length and may be an additional factor to consider when assessing any individual patient.