Appetitive Symptoms Differentially Predict Treatment Response to Fluoxetine, Light, and Placebo in Nonseasonal Major Depression

Objective: We previously reported that morning bright light therapy is efficacious in adults with nonseasonal major depressive disorder (MDD), both on its own and in combination with fluoxetine. Given that appetitive symptoms predict response to bright light therapy in seasonal depression, we examined, in this secondary analysis, whether the same held true in these nonseasonal MDD patients.

Methods: Data were collected from October 7, 2009, to March 11, 2014. One hundred twenty-two patients who met DSM-IV-TR criteria for MDD without a seasonal pattern were randomly assigned to light monotherapy, fluoxetine, combination light and fluoxetine, or double-placebo (inactivated negative ion generator plus placebo pill). Multiple regression assessed the percentage change in Montgomery-Asberg Depression Rating Scale (MADRS) scores based on treatment condition, appetitive symptom score at baseline (sum of 4 items on the Structured Interview Guide for the Hamilton Depression Rating Scale, Seasonal Affective Disorders version), and the condition-by-appetitive score interaction. Sex was considered as a possible moderator of these effects.

Results: The overall regression model predicting treatment response was highly significant (P < .001), and the treatment condition-by-appetitive score interaction was a strong predictor of MADRS change scores (t = 2.65, P = .009). For individuals in the placebo group, more appetitive symptoms at baseline predicted less decrease in MADRS scores at 8 weeks (r = -0.37; large effect size). In contrast, for individuals in the active treatment groups, more appetitive symptoms at baseline predicted more of a decrease in depression scores at 8 weeks (fluoxetine group r = +0.23, medium effect size; light therapy group r = +0.11, small effect size; combination group r = +0.32, medium to large effect size). No moderation effect of sex was found.

Conclusions: More severe appetitive symptoms at baseline predicted treatment response differentially across the 4 treatment groups. Contrary to prior findings in seasonal depression, this association was not robust for MDD patients receiving light therapy alone, although it was stronger in patients receiving fluoxetine with or without light. As the group sample sizes were modest, the current findings should be considered as preliminary only.

Trial Registration: ClinicalTrials.gov identifier: NCT00958204