Comparative efficacy of digital CBT-I versus stepped-care CBT-I to reduce comorbid depression

Document Type

Conference Proceeding

Publication Date


Publication Title



Introduction: Depression is commonly comorbid with insomnia and can be effectively targeted with digital cognitive behavioral therapy for insomnia (dCBT-I). Digital delivery of CBT-I is advantageous because it is highly accessible and requires minimal clinical resources; however, tradeoffs include the loss of clinician support and the ability to personalize treatment. A stepped-care model can optimize care by starting with a least resource intensive intervention (step 1: dCBT-I) and stepping-up non-remitters to specialized treatment (step 2: face-to-face CBT-I). This study tested the efficacy of a stepped-care approach to target co-morbid depression. Methods: 261 individuals with insomnia (DSM-5 diagnostic criteria) were randomized into two conditions at step 1: dCBT-I (N=104), or an online sleep education control (N=157). Participants in the dCBT-I condition who did not show remission for insomnia (ISI>9) were further randomized to either face-to-face CBT-I (N=23) or sleep education (N=32). Depression (Quick Inventory of Depressive Symptomatology) was assessed at baseline, post-step 1, and post-step 2. Results: Those who received stepped-care (dCBT-I to face-to-face CBT-I) achieved the same improvements in depression (pre-treatment QIDS: 6.9 ± 2.1 SD; post-treatment QIDS: 4.0 ± 3.2 SD) compared to those who remitted following only dCBT-I (pre-treatment QIDS: 7.2 ± 2.2 SD; post-treatment QIDS: 4.2 ± 2.2 SD). Furthermore, depression remission (QIDS ≤ 5) in the face-to-face CBT-I condition at step 2 (82.6%) was twice as high compared to the control condition at step 2 (40.6%), indicating that the steppedcare condition produced higher rates of depression remission compared to dCBT-I alone. Conclusion: Preliminary evidence from this study provide suggest that a stepped-care approach that adds face-to-face CBT-I for non-remitters to dCBT-I may produce greater improvement in co-morbid depression than dCBT-I alone.




Suppl 1

First Page


Last Page


This document is currently not available here.