Elsevier

Journal of Psychiatric Research

Volume 103, August 2018, Pages 156-160
Journal of Psychiatric Research

Prospective study of chronotype and incident depression among middle- and older-aged women in the Nurses’ Health Study II

https://doi.org/10.1016/j.jpsychires.2018.05.022Get rights and content

Abstract

Background

Prior cross-sectional studies have suggested that being a late chronotype is associated with depression and depressive symptoms, but prospective data are lacking.

Methods

We examined the association between chronotype and incident depression (defined as self-reported physician/clinician-diagnosed depression or antidepressant medication use) in 32,470 female participants of the Nurses’ Health Study II cohort who self-reported their chronotype (early, intermediate or late) and were free of depression at baseline in 2009 (average age: 55 yrs). Women updated their depression status on biennial questionnaires in 2011 and 2013. We used multivariable (MV)-adjusted Cox proportional hazards models to estimate hazard ratios (HR) and 95% confidence intervals (95%CI) for incident depression across chronotype categories (i.e., early, intermediate, and late chronotypes).

Results

Across a follow-up period of 4 years, we observed 2,581 cases of incident depression in this cohort. Compared to intermediate chronotypes, early chronotypes had a modestly lower risk of depression after MV adjustment (MVHR = 0.88, 95%CI = 0.81–0.96), whereas late chronotypes had a similar risk of 1.06 (95%CI = 0.93–1.20); the overall trend across chronotype categories was statistically significant (ptrend<0.01). Results were similar when we restricted analyses to women who reported average sleep durations (7–8 h/day) and no history of rotating night shift work at baseline.

Conclusions

Our results suggest that chronotype may influence the risk of depression in middle-to older-aged women. Additional studies are needed to confirm these findings and examine roles of both environmental and genetic factors to further our understanding of the role of chronotype in the etiology of mood disorders.

Introduction

The past decade has provided increasing evidence for links between the circadian system and mood, including data linking non-visual photoreception pathways in rodents to impaired mood (LeGates et al., 2012), and the beneficial role of timed light exposure therapy and sleep deprivation on mood, likely by resetting circadian rhythms (McClung, 2007, 2013, 2015; Wirz-Justice, 2003). Circadian rhythms regulate physiology and behavior – from gene expression and immune function to sleep and cognition – and the disturbance of those rhythms has also been put forward as one of the correlates of mood disorders (Wirz-Justice, 2006). The circadian system entrains to the light/dark cycle of the 24 h day (Duffy and Wright, 2005; Pittendrigh, 1981; Roenneberg and Merrow, 2007) and thereby gives rise to an individual's chronotype, with genetic variation being at least in part contributing to inter-individual differences in chronotype (Roenneberg et al., 2007a). This phenotype has also been shown to be associated with e.g. the period length of circadian clock gene expression in human fibroblasts (Brown et al., 2008). And while the implications of the circadian system and circadian photoreception pathways with mood seem established, it is unclear to what extent inter-individual differences in chronotype per se are linked to depression.

Previous work has mainly relied on cross-sectional data sets, with limited information on potential confounding factors. One of the largest datasets, FINRISK, included behavioral information on an established scale of Morningness/Eveningness (Horne and Ostberg, 1976) that correlates well with other proxies for chronotype, such as sleep timing (Kitamura et al., 2014) or dim light melatonin onset (Kantermann et al., 2015; Megdal and Schernhammer, 2007; the key circadian phase marker; Arendt, 2006); the FINRISK analysis showed that morningness was associated with lower levels of depressive symptoms, and lower odds of diagnosed depression and anti-depressant medication use (Konttinen et al., 2014; Merikanto et al., 2013, 2015). Studies around the world, in adult and adolescent populations have reported similar findings (Alvaro et al., 2014; Chan et al., 2014; de Souza and Hidalgo, 2014; Hidalgo et al., 2009; Jeong et al., 2015; Kitamura et al., 2010; Levandovski et al., 2011; Pabst et al., 2009), independent of whether they used sleep timing or morningness/eveningness as a proxy for chronotype. Patients with major depressive disorder have been reported to be more likely to be late chronotypes, as compared to individuals suffering from dysthymia or anxiety disorders. Those results suggest that depression might also alter an individual's chronotype (Antypa et al., 2016; Lemoine et al., 2013; Müller et al., 2015; Selvi et al., 2010), or possibly that changes in chronotype is related to depression severity. Indeed, recent findings in adolescents show no significant differences in chronotype between healthy controls and patients with remitted depression (Keller et al., 2017). Prospective analyses are crucial to better understand the directionality of the association of chronotype with depression, and the present study addresses this gap. We hypothesized that late chronotypes would be at higher risk and early chronotypes would be at lower risk for incident depression as compared to intermediate chronotypes.

Section snippets

Methods

The Nurses' Health Study II (NHSII) is a large, prospective cohort study of women's health, which started in 1989 when 116,434 registered US nurses (aged 25–42 years) responded to a baseline questionnaire. Biennial follow-up questionnaires have since been mailed to obtain updated information on medical history, lifestyle factors, and newly diagnosed diseases. Follow-up rates are high with approximately 90% participation at each two-year cycle (Schernhammer et al., 2011). This study was approved

Statistical analyses

We defined definite morning types as ‘early chronotypes’, definite evening types as ‘late chronotypes’, and others as ‘intermediate chronotypes’, and used Cox proportional hazard models to calculate hazard ratios (HR) and 95% confidence intervals (95%CI) across the three chronotype categories. Intermediate chronotypes served as the referent group in all analyses. We included the following covariates in multivariable (MV) adjusted models: menopausal status (pre/post-menopausal), marital status

Results

Across four years of follow-up, we observed 2,518 cases of incident depression. Table 1 shows the baseline characteristics of women free of depression and depressive symptoms at baseline in 2009. Late chronotypes were less likely to be married, and more likely to live alone, to be smokers, and more frequently reported extreme sleep durations.

In age-adjusted models, early chronotypes had a significantly reduced risk of depression as compared to intermediate types (Table 2, HR = 0.86,

Discussion

In this study of more than 30,000 middle-aged and older women, chronotype was associated with incident depression, even after accounting for potential confounders of the association. Sensitivity analyses excluding women with potentially disturbed circadian rhythms, such as shift workers or women who had a history of extreme sleep durations, similarly showed that early chronotypes were at lower risk of newly-occurring depression, as compared to intermediate. Thus, our findings indicate that

Funding

This work was supported by Center for Disease Control and Prevention/The National Institute for Occupational Safety and Health grants 5R01OH009803 and 5R21OH011052. The Nurses’ Health Study II is supported by the National Cancer Institute grant UM1CA176726.

Conflicts of interest

None.

References (54)

  • T. Roenneberg et al.

    Epidemiology of the human circadian clock

    Sleep Med. Rev.

    (2007)
  • T. Roenneberg et al.

    The human circadian clock entrains to sun time

    Curr. Biol.

    (2007)
  • E.R. Stothard et al.

    Circadian entrainment to the natural light-dark cycle across seasons and the weekend

    Curr. Biol.

    (2017)
  • K.V. Allebrandt et al.

    The search for circadian clock components in humans: new perspectives for association studies

    Brazilian Journal of Medical and Biological Research = Revista Brasileira de Pesquisas Medicas E Biologicas

    (2008)
  • N. Antypa et al.

    Chronotype associations with depression and anxiety disorders in a large cohort study

    Depress. Anxiety

    (2016)
  • J. Arendt

    Melatonin and human rhythms

    Chronobiol. Int.

    (2006)
  • K.A. Bertrand et al.

    Determinants of plasma 25-hydroxyvitamin D and development of prediction models in three US cohorts

    Br. J. Nutr.

    (2012)
  • D.M. Berwick et al.

    Performance of a five-item mental health screening test

    Med. Care

    (1991)
  • S.A. Brown et al.

    Molecular insights into human behavior

    Proc. Natl. Acad. Sci.

    (2008)
  • J.W.Y. Chan et al.

    Eveningness and insomnia: independent risk factors of nonremission in major depressive disorder

    Sleep

    (2014)
  • C.M. de Souza et al.

    Midpoint of sleep on school days is associated with depression among adolescents

    Chronobiol. Int.

    (2014)
  • J.F. Duffy et al.

    Entrainment of the human circadian system by light

    J. Biol. Rhythm.

    (2005)
  • D.J. Gottlieb et al.

    Genome-wide association of sleep and circadian phenotypes

    BMC Med. Genet.

    (2007)
  • A.C. Heath et al.

    Evidence for genetic influences on sleep disturbance and sleep pattern in twins

    Sleep

    (1990)
  • M.P. Hidalgo et al.

    Relationship between depressive mood and chronotype in healthy subjects

    Psychiatr. Clin. Neurosci.

    (2009)
  • J.A. Horne et al.

    A self-assessment questionnaire to determine morningness-eveningness in human circadian rhythms

    Int. J. Chronobiol.

    (1976)
  • T. Kantermann et al.

    Comparing the morningness-eveningness questionnaire and munich ChronoType questionnaire to the dim light melatonin onset

    J. Biol. Rhythm.

    (2015)
  • Cited by (38)

    • Adaptation and Pilot Study of a Behavioral Intervention Targeting Morning Activation Deficits in Dementia Caregivers: Scheduling Activity and Monitoring Mornings (SAMM)

      2022, American Journal of Geriatric Psychiatry
      Citation Excerpt :

      As opposed to morning behavioral disengagement (as is often found in depression), a subset of the population identify as “morning types” (i.e., people who habitually prefer and engage in morning/earlier timed activity patterns). A prospective study found that being a “morning type” was protective against major depression, while incidence rates did not differ between “intermediate” and “evening types.”2 Our recent research examined these associations in older dementia caregivers, based in part on their high vulerability to depression symptoms and disorders.3

    • Sleep disorders and non-sleep circadian disorders predict depression: A systematic review and meta-analysis of longitudinal studies

      2022, Neuroscience and Biobehavioral Reviews
      Citation Excerpt :

      With regard to chronotype, only 3 eligible cohort studies were retrieved, with one focusing on nurses and another two on children or adolescents. They all used the morningness/eveningness scale to identify chronotype (Chen et al., 2021a; Haraden et al., 2017; Vetter et al., 2018). Meal time or eating patterns was assessed in an eligible study (Chen et al., 2021a).

    • Evaluating the timing of differences in activity related to depression symptoms across adulthood in the United States

      2021, Journal of Affective Disorders
      Citation Excerpt :

      Conclusions: Our findings confirm prior observations that highlight the relevance of morning activity to mood. Specifically, past research indicates that being a “morning type” is protective against depression (Vetter et al., 2018), and that people with depression often report difficulty “getting going” in the morning (Trotti, 2017). Strengths of our study include the use of objective accelerometer measures, which are not subject to affective or recall biases, and that our study sample was large and designed to be representative of people in the United States.

    View all citing articles on Scopus
    View full text