Change in circadian preference predicts sustained treatment outcomes in patients with unipolar depression and evening preference

Eveningness is associated with worse outcomes in depression. It remained unclear if eveningness could be altered with chronobiological therapy and whether such a change would predict long-term outcomes of depression. Data from a randomized controlled trial of 5-week adjunctive bright light therapy with a gradual advance protocol conducted in 91 adult patients with nonseasonal unipolar depression and eveningness (Morningness-Eveningness Questionnaire, score ≤ 41) was examined. "Change of eveningness" was defined by Morningness-Eveningness Questionnaire score over 41 at posttreatment week 5 and "persistent change of eveningness" was defined as maintenance of Morningness-Eveningness Questionnaire score > 41 throughout the follow-up period from week 5 to posttreatment 5 months. Thirty-three participants (36%) had change of eveningness at week 5. Generalized estimating equations models showed that a change of eveningness at week 5 predicted a 2-fold increase in remission of depression over the 5-month follow up (odds ratio = 2.61 95% confidence interval 1.20-5.71, = .016). Twenty-five participants (75.7%) had a persistent change and were more likely to achieve a remission of depression over the 5-month follow up (odds ratio = 3.18, 95% confidence interval: 1.35-7.50, = .008). One-third of the patients with depression changed their evening-preference after 5-week of chronotherapeutic treatment, and such change predicted a higher likelihood of depression remission over 5 months of follow-up. Registry: Chinese Clinical Trial Registry; Name: Adjunctive light treatment in major depressive disorder patients with evening chronotype-A randomized controlled trial; URL: https://www.chictr.org.cn/showprojen.aspx?proj=11672; Identifier: ChiCTR-IOR-15006937. Chan JWY, Chan NY, Li SX, et al. Change in circadian preference predicts sustained treatment outcomes in patients with unipolar depression and evening preference. . 2022;18(2):523-531.