1109 DO SLEEP PARAMETERS MEDIATE THE ASSOCIATION BETWEEN CHRONOTYPE AND MENTAL HEALTH?. (28th April 2017)

• Record Type: Journal Article
• Authors: Smidt, AM; Latham, MD; Allen, NB
• Is Part Of: Sleep. Volume 40(2017)Supplement 1
• Page Start: A413
• Page End: A413
• Publication Date: 2017-04-28
• Subjects: Sleep -- Physiological aspects -- Periodicals; Sleep disorders -- Periodicals; Sommeil -- Aspect physiologique -- Périodiques; Sommeil, Troubles du -- Périodiques; Sleep disorders; Sleep -- Physiological aspects; Sleep -- physiological aspects; Sleep Wake Disorders; Psychophysiology; Electronic journals; Periodicals; 616.8498
• DOI: 10.1093/sleepj/zsx050.1108
• Languages: English
• ISSNs: 0161-8105

Abstract

Abstract: Introduction: Chronotype, specifically a preference for eveningness, is a well-established predictor of mental health problems, particularly depression and anxiety. However, research is lacking on whether specific sleep parameters may act as mediators of the correlation between chronotype and depression & anxiety. An understanding of how sleep parameters - including efficiency, sleep onset latency (SOL), total sleep time (TST), wake after sleep onset (WASO), and number of awakenings (NA) - may act as mediators of the correlation between chronotype and depression & anxiety would offer a basis for future longitudinal research on how chronotype specifically affects mental health outcomes. Methods: Fifty-four participants (age: M=19.58, SD=1.49; 82.7% female) used actigraph watches (Actigraph Corp.) for an average of 7 days to assess their sleep. Actigraphy data were informed by daily sleep diaries, and sleep parameters (including efficiency, SOL, TST, WASO, and NA) were calculated using ActiLife software. Self-report measures were used to assess chronotype (reduced Morningness-Eveningess Questionnaire; rMEQ) and depression & anxiety symptoms (Patient Health Questionnaire; PHQ). Two participants were excluded from analyses due to missing data. Results: rMEQ scores (range: 7–21) were significantly negatively correlated with depression symptoms (r = -.282, p = .042), such that greater preference for eveningness was associated with increased depressive symptoms. However, rMEQ scores were not significantly correlated with anxiety symptoms (r = -.160, ns). None of the 10 (one sleep index per model times two [depression and anxiety]) correlational mediation models predicting depression and anxiety from rMEQ scores were statistically significant. The effect sizes (Preacher & Kelley's kappa-squared) of the overall mediation models that included efficiency and TST predicting depression (0.02 and 0.05, respectively) and anxiety (0.02 and 0.04, respectively) are in the small-to-medium range. Conclusion: rMEQ scores correlated with mental health outcomes as expected. None of the correlational indirect effects models were significant, suggesting that the association between chronotype and mental health symptoms might not be mediated by sleep parameters. However, effect sizes of the mediation models of sleep efficiency and TST on depression and anxiety, respectively, point to effects that may bear further investigation with longitudinal designs with more statistical power. Support (If Any):