Considering Substance Use in Dysfunctional Beliefs About Sleep: Validation of the DBAS-S

Objective Alcohol and cannabis are commonly used as sleep aids that inadvertently perpetuate symptoms of insomnia over time, but substance use is not a standard component of sleep assessments. This pre-registered study aimed to expand the validity and clinical utility of the Dysfunctional Beliefs about Sleep Scale (DBAS) by incorporating substance use expectancies for sleep.Method After using cognitive interviewing to improve the content validity of items (N = 24, age M = 33.3y, 83% female), iterative factor analyses and tests of external validation were used to refine and psychometrically evaluate the revised scale (full sample N = 966, age M = 42.6, 60% female, 65% with at least mild insomnia).Results The final set of 11 items was best represented by two factors (9 sleep and 2 substance use items; r = 0.15). The sleep subscale was strongly correlated with the DBAS-16 (r = 0.95, p < .001) and demonstrated measurement invariance across those who did (n = 309) and did not (n = 318) drink alcohol. The substance use subscale (β = 0.16, p < .001) accounted for a small but significant amount of unique variance in insomnia severity (sleep subscale β = 0.59, p < .001) and demonstrated concurrent and prospective associations with alcohol/cannabis use. Surprisingly, the single cannabis belief item was more strongly associated with cannabis use than longer validated scales of cannabis sleep expectancies.Conclusions The substance use version of the DBAS (DBAS-S) is a brief, reliable, and valid measure of dysfunctional sleep beliefs and substance use expectancies that will improve clinical practice by flagging substance use as a potential contributor to insomnia for clinicians.