Objective: To identify mediators of the beneficial effect of cognitive–behavioral therapy (CBT) on fatigue severity in chronically fatigued patients with Type 1 diabetes. Method: We performed secondary analyses of a randomized controlled trial testing the efficacy of CBT. Primary outcome was fatigue severity assessed with the Checklist Individual Strength, subscale fatigue severity. We used multiple mediation analysis to determine potential mediators of the treatment effect. Proposed mediators were symptom focusing, self-efficacy concerning fatigue and pain, perceived activity, sleep disturbances, confidence in diabetes self-care, diabetes distress and discrepancy regarding social support. Actigraphy was used to assess the level of physical activity. The analysis was repeated with depressive symptoms as potential mediator to test whether this caused a change in the other fatigue-related mediators. Results: The effect of CBT on fatigue severity was partly mediated by a change in symptom focusing (−1.39, 95% CI [−3.32, −0.19]), fear avoidance (−1.10, 95% CI [−2.49, −0.22]), self-efficacy concerning fatigue (−1.95, 95% CI [−4.51, −0.40]), and perceived physical activity (−2.44, 95% CI [−4.53, −1.07]). Depressive symptoms were also a mediator (−1.22, 95% CI [−2.56, −0.38]), but the aforementioned fatigue-perpetuating factors still explained part of the treatment effect. Conclusions: Changes in cognitions about fatigue and activity, and a change in depressive symptoms partly mediated the treatment effect of CBT on fatigue severity in Type 1 diabetes. The mediators found are similar to those explaining the effects of CBT in other chronic diseases. This provides knowledge for the development of more effective and efficient interventions for fatigue.