0478 The Impact of Non-pharmacological Insomnia Therapy on Mood and Sleep in Morning and Evening Chronotypes in Older Adults

Abstract

Abstract Introduction As individuals age, the circadian-driven timing of their sleep shifts to an earlier hour. Whether such a shift moderates the effectiveness of insomnia treatment on sleep disturbances and mood in older adults is unknown. Methods We tested the hypothesis that circadian preference moderates improvements in mood and insomnia symptoms following a non-pharmacological insomnia treatment. Older adults (N=111, age=69±6.4 years, female=65%) with insomnia (Insomnia Severity Index (ISI) score >10) received a 6-session treatment regimen. Circadian preference was measured at baseline with the Composite Scale of Morningness (CSM mean score=41.2±7.3, median=42). Chronotypes were classified based on a median split of CSM scores. Depression, (Geriatric Depression Scale, GDS), insomnia severity (ISI), cognitive arousal (Glasgow Content of Thoughts Inventory, GCTI), and time in bed (TIB), total sleep time (TST), and sleep efficiency (TST /TIB=SE) from sleep diaries were collected pre- and post-treatment. Tests of proportion were used to characterize differences in demographic variables between chronotypes. Ranked correlation tests were used to test associations between circadian preference and variables of interest at pre- and post-treatment. T-tests with unequal variance were used to examine whether treatment outcomes differed between chronotypes. Results In this study, 58% of females and 38% of males were later chronotypes (p=0.04). Later chronotype was associated with greater pre-treatment TIB (p=0.01), and earlier chronotype was associated with higher post-treatment cognitive arousal (p=0.04). Later chronotypes had greater reduction in depression symptoms (Cohen’s d=0.43, p=0.04), cognitive arousal (d=0.39, p=0.05), and a trend for greater reduction in TIB (d=0.37, p=0.07). Earlier chronotypes had a greater increase in TST (d=0.42, p=0.04). However, both chronotypes saw equivalent changes in SE (d=0.12, p=0.58). Conclusion Later chronotypes had a greater reduction in TIB, while earlier chronotypes demonstrated a significant increase in TST due to insomnia treatment, and both chronotypes had improved SE. These patterns suggest that treatment equivalently improves the consolidation of sleep, but the mechanism of this treatment effect differs by chronotype. That is, a moderating effect of circadian preference on the mechanism of improved sleep consolidation by insomnia treatment in older adults. These results suggest greater attention to age-related changes in chronotype in insomnia treatment. Support (If Any) NIMHR01MH101468-01; MIRECC at the VAPAHCS