Psychometric comparisons of the standard and abbreviated DBAS-10 versions of the dysfunctional beliefs and attitudes about sleep questionnaire

Objective: To evaluate the psychometric properties of the DBAS-10, a recently proposed abbreviated version of the Dysfunctional Beliefs and Attitudes About Sleep Scale (DBAS). Population: Two hundred and eleven (69 normal sleepers; 142 insomnia suffers) middle-aged and older adults (age 40-79 years) drawn from two separate cohorts of research volunteers. Method: Volunteers in the first cohort (69 normal sleepers; 69 insomnia sufferers) completed the full DBAS on one occasion. Volunteers in the second cohort (73 insomnia sufferers) completed the full DBAS prior to treatment and at multiple subsequent time points to assess treatment-related changes. A series of statistical tests were conducted with one or both cohorts to investigate the comparability of the DBAS-10 and full DBAS, the internal consistency of each instrument, the factor structure of the DBAS-10, and the validity of this instrument. Results: Statistical findings showed that the DBAS-10 correlated highly with the full DBAS, had respectable internal consistency, effectively discriminated normal sleepers from insomnia sufferers, and detected cognitive changes resulting specifically from CBT intervention. Although factor analysis empirically identified three conceptually meaningful DBAS-10 subscales, the subscale structure varied somewhat from previous factor analytic findings with this instrument. Conclusions: The DBAS-10 generally appears to have very acceptable psychometric properties although subscales previously proposed for this instrument may vary across research populations. Nonetheless, results encourage the use of this instrument in studies concerned with the nature and treatment of sleep-disruptive cognitions. (C) 2001 Elsevier Science B. V. All rights reserved.