Insomnia is a heterogeneous disorder, for which cognitive behavioural therapy for insomnia (CBT-I) is the recommended first-line treatment. However, CBT-I is not effective for all patients. It is important to identify pre-treatment predictors and moderators of treatment-response to CBT-I and its variants to guide precision insomnia management approaches. This systematic review (PROSPERO: CRD42023476013) aimed to synthesise evidence on pre-treatment characteristics that predict or moderate insomnia outcomes following CBT-I. Peer-reviewed papers that reported such analyses from randomised and non-randomised studies of adults with insomnia were eligible. Four databases and citations of included reports were searched until April 2025. From 7,325 unique records, 1,494 full-text records were screened, of which 103 reports were included. Among 195 unique predictors/moderators identified from 1,208 results, 92.3% were infrequently studied and/or rarely demonstrated statistical significance as predictors of insomnia improvement. The most consistent predictors of improvement included shorter insomnia duration, more severe insomnia, longer objective sleep duration, positive attitudes towards treatment, lower depression severity, and the absence of comorbid mental disorders, fatigue and pain. Further high-quality research using sufficiently powered samples and theoretically-justified moderators is needed to establish characteristics related to differential outcomes between cognitive behavioural interventions.
Smith et al. (Sun,) studied this question.