1072 Dim Light Melatonin Onset and Mood Symptoms: Circadian Timing in a Clinically-Enriched Sample of Young Adults

Abstract Introduction Theoretical and empirical work supports circadian rhythm disruption in the pathogenesis of mood disorders. However, dim light melatonin onset (DLMO), the gold standard measure of circadian timing, has been understudied in relation to mood symptoms. This study evaluated DLMO timing and stability in a clinically-enriched young adult sample and tested associations between circadian timing indices and mood disorder symptoms. Methods Young adults (n=124; mean age = 21.9) with (n=70) and without (n=54) mood disorders completed a 20-day ecological momentary assessment (EMA) study. Participants reported hypomanic and depressive symptoms prior to beginning the EMA study, and thrice-daily via EMA. An at-home controlled DLMO procedure was completed on days 1, 10, and 20. Participants provided 10 saliva samples 30 minutes apart starting five hours prior to habitual bedtime while wearing light-attenuating goggles. DLMO stability was quantified using intraclass correlation coefficients (ICC) and within-person variability. Linear mixed models tested associations between DLMO, symptom severity, and the temporal pattern of mood symptoms. Results Mean DLMO for the sample was 22.01 h ± 1.44 h (SD), demonstrating substantial stability (ICC = 0.75; 95% CI = 0.60–0.83). The mood disorder group exhibited a significantly earlier DLMO than controls (b=–0.51, SE=0.24, p=.04), but there were no differences in DLMO stability by diagnostic group. Hypomanic, but not depressive, symptoms were associated with later DLMO both cross-sectionally (B = 0.20, SE = 0.09, p = .03) and prospectively (B=.24, SE=.09, p=.01). Time-of-day analyses revealed distinct diurnal patterns for hypomanic, but not depressive symptoms; individuals with earlier DLMOs exhibited peaks in hypomanic symptoms around 1400 h, whereas those with later DLMOs reported peak symptom levels in the evening. Conclusion Circadian phase was stable over short periods in young adults, with earlier timing observed in those with mood disorders. Hypomanic symptoms were linked to later circadian timing and showed a distinct daily pattern. Incorporating objective measures of endogenous circadian rhythms may improve understanding of mood disorder episodes and inform targeted interventions. Support (if any) National Aeronautics and Space Administration grants NNX14AN49G and 80NSSC20K0243 and NIH grant R01DK117488 to Namni Goel; NIH grants R01MH77908, R01MH102310, and R01MH126911 to Lauren Alloy.