UC Davis

OBJECTIVES: Insomnia is a common problem that affects a significant portion of the population. Dysfunctional beliefs and attitudes about sleep contribute to developing and maintaining insomnia. We developed a Brazilian-Portuguese version of the Dysfunctional Beliefs and Attitudes about Sleep Scale (DBAS-16) and conducted a comprehensive psychometric evaluation using latent variable and psychometric network frameworks. METHOD: Participants (N = 1,3851,386) were between 18 and 59 years old, with and without insomnia complaints. We used Confirmatory Factor Analysis (CFA) with dynamic fit indices to evaluate the original DBAS-16 structure and tested longitudinal invariance (14 days) and invariance across good and bad sleepers. We applied Unique Variable Analysis (UVA) and Exploratory Graph Analysis (EGA) to a derivation subsample (n = 693), with the remaining half reserved for cross-validation. RESULTS: The original structure was replicated with moderate fit. We found support for longitudinal but not cross-group metric invariance. UVA identified three redundant items for exclusion (1. Need 8 h of sleep, 3. Consequences of insomnia on health, and 15. Medication as a solution). EGA identified two highly stable dimensions, replicated in the cross-validation subsample. CONCLUSION: The EGA model fit significantly better than the theoretical model, endorsing an alternative DBAS dimensionality. After excluding locally dependent variables, two dimensions may better represent dysfunctional beliefs and attitudes about sleep.