Response to Rifkin

To the Editor: We agree with Dr. Rifkin that our results do not preclude the possibility that the two treatments differ on posttreatment outcomes. The study was not powered to test equivalence or noninferiority. To have 80% power to detect equivalence in posttreatment remission within 10%, we would have needed 280 subjects per treatment condition (N=560). In the article, we discuss the “similar” and “comparable” pattern of results for cognitive-behavioral therapy for seasonal affective disorder (CBT-SAD) and light therapy, but we were careful not to call the treatments equally effective at posttreatment. We presented 95% CIs for differences between CBT-SAD and light therapy in the article to enable the reader to see what cut points for equivalence and noninferiority would lead to significant results.

Our trial was instead powered to test for superiority of one treatment over the other on winter depression recurrences in the 2 years following treatment of the index episode. We recently reported that CBT-SAD was associated with significantly fewer recurrences and less severe symptoms than light therapy 2 winters after study treatment (1).