Insomnia disorder is the second most prevalent mental disorder and the most prevalent sleep disorder. Cognitive Behavioral Therapy for Insomnia (CBT-I) is considered the treatment of choice with well-documented effects. Nevertheless, a significant proportion of patients fail to respond, and an even larger proportion fail to remit from the condition. In addition, very little is known about the effects of CBT-I's separate components or about what moderates and mediates their effect. Gaining knowledge about components, predictors, and mediators could be one route for optimizing and tailoring CBT-I and ultimately enhancing outcomes.

The overall aim of this thesis was to advance our theoretical and clinical knowledge about CBT-I by exploring Cognitive Therapy (CT) and Behavior Therapy's (BT) comparative efficacy and their potential moderators and mediators.

To pursue the study aims, one large randomized controlled trial was performed that involved 219 individuals with insomnia disorder randomized to CT, BT, or a waitlist control group. Study 1 examined CT and BT's comparative efficacy against a waitlist control on a broad range of outcomes. Study 2 examined theoretically derived constructs from both therapy models, and insomnia-associated correlates as potential predictors and moderators of outcome for the two therapies. Study 3 examined theoretically driven process variables from the cognitive model as mediators of outcome in both CT and BT.

Study I showed that both therapies outperformed the waitlist and turned out as comparably effective treatments on the majority of outcomes. BT was associated with significantly more adverse events, whereas CT received significantly more minutes of telephone support.

Study II showed that early morning waketime and bedtime variability moderated the effect of both CT and BT. Those experiencing lower early morning waketime and bedtime variability achieved greater insomnia severity reductions in CT. In contrast, those experiencing greater early morning waketime and bedtime variability achieved larger insomnia severity reductions in BT. The findings also showed that greater insomnia severity, waketime after sleep onset, and lower sleep efficiency at baseline predicted greater insomnia severity at posttreatment.

Study III provided evidence that reductions in dysfunctional beliefs and monitoring for sleep during treatment acted as drivers of the reduction in insomnia severity in CT. The results also indicated that reductions in safety behaviors and dysfunctional beliefs mediated reductions in insomnia severity in BT, although not as clear as the drivers of change for CT since they were also reciprocally predicted by reductions in insomnia severity.

Study I indicate that CT and BT achieve similar effects and that both therapies are effective as standalone therapies for insomnia disorder. Study II provided evidence that the two therapies in CBT-I can depend on different patient characteristics at baseline to be effective. The results from study II thus suggest that the therapies in CBT-I could be tailored based on patient's characteristics before treatment to optimize outcomes. Study III provided support for the role of cognitive processes as important routes to remediate insomnia and underscore the value of assessing and targeting dysfunctional beliefs, monitoring, and safety behaviors to achieve reductions in insomnia severity and emphasize the importance of these concepts in understanding insomnia.