Prediction of outcome of bright light treatment in patients with seasonal affective disorder: Discarding the early response, confirming a higher atypical balance, and uncovering a higher body mass index at baseline as predictors of endpoint outcome

• Published in: Journal of affective disorders 2017-11, Vol.222, p.126-132
• Main Authors: Dimitrova, Tzvetelina D., Reeves, Gloria M., Snitker, Soren, Lapidus, Manana, Sleemi, Aamar R., Balis, Theodora G., Manalai, Partam, Tariq, Muhammad M., Cabassa, Johanna A., Karim, Naila N., Johnson, Mary A., Langenberg, Patricia, Rohan, Kelly J., Miller, Michael, Stiller, John W., Postolache, Teodor T.
• Format: Article
• Language: English
• Subjects: Adult; Body Mass Index; Cross-Over Studies; Depression; Female; Humans; Light therapy; Male; Middle Aged; Mood disorder; Obesity; Phototherapy - methods; Seasonal affective disorder; Seasonal Affective Disorder - physiopathology; Seasonal Affective Disorder - therapy; Treatment Outcome
• ISSN: 0165-0327; 1573-2517
• DOI: 10.1016/j.jad.2017.06.038

We tested the hypothesis that the early improvement in mood after the first hour of bright light treatment compared to control dim-red light would predict the outcome at six weeks of bright light treatment for depressed mood in patients with Seasonal Affective Disorder (SAD). We also analyzed the value of Body Mass Index (BMI) and atypical symptoms of depression at baseline in predicting treatment outcome. Seventy-eight adult participants were enrolled. The first treatment was controlled crossover, with randomized order, and included one hour of active bright light treatment and one hour of control dim-red light, with one-hour washout. Depression was measured on the Structured Interview Guide for the Hamilton Rating Scale for Depression-SAD version (SIGH-SAD). The predictive association of depression scores changes after the first session. BMI and atypical score balance with treatment outcomes at endpoint were assessed using multivariable linear and logistic regressions. No significant prediction by changes in depression scores after the first session was found. However, higher atypical balance scores and BMI positively predicted treatment outcome. Absence of a control intervention for the six-weeks of treatment (only the first session in the laboratory was controlled). Exclusion of patients with comorbid substance abuse, suicidality and bipolar I disorder, and patients on antidepressant medications, reducing the generalizability of the study. Prediction of outcome by early response to light treatment was not replicated, and the previously reported prediction of baseline atypical balance was confirmed. BMI, a parameter routinely calculated in primary care, was identified as a novel predictor, and calls for replication and then exploration of possible mediating mechanisms. •Mood change after first session of light did not predict end-point outcome.•Atypical balance was confirmed as a positive predictor of outcome.•BMI was a strong positive predictor of outcome.