0161 Structural Changes in Mood Domains During Sleep Deprivation and Recovery

Abstract

Abstract Introduction Mood is known to vary as a function of sleep debt. Mood batteries often measure multiple, non-orthogonal domains. In the present study, the latent structure of the Automated Neuropsychological Assessment Metrics Mood (ANAM-mood) battery was assessed during five different states of alertness. We present a novel scheme to evaluate changes in the structure of such batteries across conditions. It was hypothesized that the structure would change, with the greatest changes manifest during sleep deprivation. Methods The present analyses were conducted on data from a prior sleep deprivation study (n = 25 healthy young adults). We used the identity property of MANOVA. Given correlation matrices Σi and Σj, structural invariance was defined as Σi Σj-1 ≈ I (Identity). MANOVA statistic m was the largest eigenvalue of Σi Σj-1. Resampling was used to generate m-distributions within and between the five different alertness conditions - Rested, Sleep Deprived (SD) days 1 and 2, and Recovery days 1 and 2. In addition, the relative number of mood dimensions was estimated by the odds (via resampling) of two dimensions explaining 80% of the variance of Σi compared to Σj. Results The pairwise MANOVA-statistic revealed significant structural variations across conditions (p< 0.001 for all comparisons). Significant differences included Rested vs SD1 and SD2. Recovery 1 showed the most pronounced differences to all conditions. Recovery 2 was like Recovery 1, but more like Rested than the prior day. The odds ratio of two-dimensions for Recovery 1 was much higher compared to all other conditions (OR: 2.4-245, p< 0.001). Conclusion Individuals respond differently to a mood survey across specific alertness conditions resulting in structural changes to the data. Wakefulness during recovery from severe sleep loss differs most significantly from all states, contrary to our hypothesis. This is important to consider when interpreting mood changes as measured by such batteries (including related Spielberger's State Anxiety Inventory and Beck Depression Inventory-II) during sleep studies. Follow up studies are needed to fully assess such effects and devise methods to account for them. Support (if any) This effort was fully supported by the Department of Defense Military Operational Medicine Research Program (MOMRP).