The role of dysfunctional beliefs about sleep in cognitive behaviour therapy for insomnia

While a good evidence-base exists for the efficacy of cognitive behavioural therapy for insomnia (CBT-I), the most common cognitive intervention involves restructuring dysfunctional beliefs about sleep which are thought to be important in the aetiology and/or maintenance of insomnia. A literature review revealed that the importance of this component of CBT-I had not been investigated directly and the evidence was not consistent and only provided weak evidence for such cognitive restructuring. The review also exposed flaws in the instrument used to assess such cognitions, the Dysfunctional Beliefs about Sleep Scale (DBAS: Morin, 1993, 1994) and its variants. The psychometric properties of the DBAS and three variants (DBAS-10, DBAS-SF, DBAS-16) and the relationship between DBAS scores and sleep were assessed in community (n = 224) and treatment-seeking insomniac (n = 180) sample. The psychometric properties of all versions of the DBAS were not optimal and only weak relationships were found between DBAS scores and sleep measures. DBAS scores of good and poor sleepers were found to differ with large effect size. While DBAS scores decreased in a sample of people with insomnia (n = 185) over the course of CBT-I, the relationships between changes in DBAS scores and improvements in sleep were not strong. Using the combined evidence, it was concluded that there was no advantage of any of the versions of the DBAS over the original scale and that theme 2 scores of the DBAS-ii were the best measure of dysfunctional beliefs about sleep. It was also concluded that there was only weak evidence for such beliefs being involved in the aetiology and/or maintenance of insomnia, and that a more parsimonious explanation of the results was that these beliefs were a consequence of poor sleep. Better evidence is required about the importance of modifying dysfunctional cognitions about sleep in CBT for insomnia (CBT-I). It is possible that the inclusion of this component is unimportant and distracts the therapists' and clients' attention away from components which are more central to the efficacy of the approach. Ultimately, studies are required which directly compare outcomes using a CBT package including cognitive restructuring and the same package with the cognitive restructuring omitted or which compare a package involving cognitive restructuring alone with a suitable comparison group. However, these studies need a better measure of dysfunctional beliefs about sleep. Suggestions for improvements in the measure and for future research were made.