ASSOCIATION BETWEEN CIRCADIAN RHYTHMS, SLEEP, AND DEPRESSION: A BIDIRECTIONAL MENDELIAN RANDOMISATION STUDY IN THE UK BIOBANK

Valentina Paz, Dylan M Williams, Marcus Richards, Bettina Tassino, Ana Silva, Victoria Garfield
doi: https://doi.org/10.1101/2024.12.08.24318680

Abstract

Increasing evidence links circadian and sleep disruptions with depression, though whether they are causally related remains unclear. This study used Mendelian randomization (MR) with exposures instrumented using genome-wide genetic variants and data from the UK Biobank (n=408 480; mean age=56.9 years) to explore causal associations between circadian and sleep traits, and depression. We conducted bidirectional MR analyses between morningness, sleep duration, insomnia symptoms, and depression, and secondary unidirectional analyses with circadian and sleep traits as exposures and mixed anxiety-depressive symptoms and anxiety disorders as outcomes. Using inverse-variance weighting and sensitivity analyses for horizontal pleiotropy (MR-Egger and Weighted Median Estimator), we found suggestive associations between morningness and lower odds of depression (diagnosis and treatment: OR=0.89, 95%CI=0.80, 0.98), and depression and shorter sleep duration (β=-4.67 minutes, 95%CI=-7.44; -1.89). Depression showed an effect on insomnia (any symptoms: OR=1.22, 95%CI=1.11; 1.34; frequent symptoms: OR=1.30, 95%CI=1.19; 1.42), while frequent insomnia was associated with increased odds of depression (diagnosis: OR=1.14, 95%CI=1.10; 1.19; diagnosis and treatment: OR=1.10, 95%CI=1.06; 1.14). Insomnia also had an effect on mixed anxiety-depressive symptoms (any symptoms: OR=1.19, 95%CI=1.08, 1.32; frequent symptoms: OR=1.20, 95%CI=1.17, 1.23), but no associations were found between the exposures and anxiety disorders. These findings shed light on the causal links between the circadian system and depression.

Introduction

The circadian system controls the endogenous rhythmic activity of numerous physiological functions in humans, the sleep-wake cycle being the most evident circadian rhythm [1]. Sleep is a complex behaviour characterised by multiple dimensions. Some of the most important dimensions of sleep are timing, the placement of sleep within the 24-hour day; duration, defined as the total amount of sleep obtained over 24 hours [2]; and quality, understood as the self-satisfaction with sleep [3]. Circadian disruption, considered the consequence of a temporal challenge or other perturbations in the circadian system [4], is a critical factor in numerous circadian and sleep disorders and has been linked to several adverse health outcomes, including cardiovascular, metabolic, and mental disorders [5].

Depression, a prevalent mental disorder and the leading cause of years of life lost due to disability worldwide, is characterized by persistent low mood, anhedonia, and high comorbidity with anxiety disorders [6]. This disorder is often associated with circadian and sleep disturbances [7–9]. These disturbances usually persist across different phases of the disorder and contribute to increased treatment resistance [8]. Moreover, sleep problems are recognized as a vegetative sign of depression [10] and one of its most common residual symptoms [11]. The association between depression and circadian preferences, the preference to time daily activities including sleep used as an estimate of circadian phase, has been extensively reported in cross-sectional studies, which indicate that evening types are more prone to depression [12–14]. Epidemiological research further suggests that both short and long sleep durations are associated with depression [15], and short sleep often persists even after the remission of emotional symptoms [16]. Insomnia —characterized by disturbances in sleep continuity, including early morning awakenings, difficulty falling asleep and staying asleep, and non-restorative or poor-quality sleep— has been historically considered an epiphenomenon of depression. However, recent longitudinal studies suggest that insomnia is a risk factor for the development of the disorder [17,18]. However, the causal nature of these associations remains elusive [19].

Mendelian randomisation (MR), which uses genetic variants to help assess causality, could be used to address these questions [20]. Regarding the association between circadian preferences and depression, some studies found that the risk of depression is lower in morning individuals [21,22]. Moreover, recent work has revealed that short sleep duration has an effect on the risk of depression [23] and bidirectional associations between sleep duration and depression using MR [24]. Studies also reported evidence of an effect of insomnia symptoms on depressive symptoms [25] and risk of depression [23,26–28] and bidirectional associations between insomnia and depression [29–33]. However, other studies have not found causal associations between circadian and sleep traits and depression, such as duration [34], morningness [23,35] or insomnia [36].

Considering that this field of study is still in development and that results reported to date are ambiguous, it is crucial to investigate whether the association between circadian preferences and sleep patterns and depression might be causal, and if so, to determine the directionality of this association. Furthermore, considering the comorbidity between depression and anxiety [6], this study aims to elucidate the role of anxiety in these associations. Specifically, the present study examines whether circadian and sleep phenotypes are causally associated with depression.

Materials and Methods

Sample

Study participants were from the UK Biobank (UKB) cohort, described in detail elsewhere [37]. The UKB is a large-scale biomedical database containing phenotypic and genetic data available for over 500 000 participants from the general United Kingdom population aged between 40 and 69 at recruitment (2006 to 2010). We used data on up to 408 480 white British individuals (genetically defined) from the UKB with quality-controlled genotype and phenotype data available. The mean age of participants was 56.9 ± 8.0 years, 54.0% were female, and 19.9% were in the most deprived socioeconomic quintile (Table 2 for full sample characteristics).

View this table:
Table 1. Summary of genetic instruments.
View this table:
Table 2. Sample characteristics.

Study design

This study used a pseudo two-sample MR design to estimate bidirectional associations between circadian and sleep traits, and depression and anxiety. Thus, we used a combination of genetic association estimates that we derived from individual-level UKB data and GWAS summary statistics, some of which analysed only UKB data, and others that analysed UKB alongside other datasets. This study followed the STROBE-MR reporting guidelines [38].

Genotyping and quality control

487 409 UKB participants were genotyped using one of two customised genome-wide arrays imputed to a combination of the UK10K, 1000 Genomes Phase 3, and the Haplotype Reference Consortium (HRC) reference panels, which resulted in data on 93 095 623 autosomal variants [39]. We then applied additional variant-level quality control (QC) and excluded genetic variants with Fisher’s exact test <0.3, minor allele frequency (MAF) <1%, and a missing call rate of ≥5%.

Participants were excluded if they had excessive or minimal heterozygosity, more than 10 putative third-degree relatives as per the kinship matrix, no consent to extract DNA, sex mismatches between self-reported and genetic sex, missing QC information and non-European ancestry (based on self-reported ancestry and the similarity with their “genetic ancestry” [40], as per a principal component analysis of their genotype) (see Lu et al. 2002 [41] for the differences between race, ethnicity, and ancestry).

Selection of genetic instruments

Morningness was instrumented using independent genetic variants (at LD cut-off r2≤0.3) from a GWAS by Jones et al. (2019) [42]. The variants were discovered in 697 828 European participants using data from the UKB cohort and summary statistics from a GWAS performed with 23andMe data (we used only UKB summary statistics). In the UKB cohort, circadian preferences was assessed based on the question “Do you consider yourself to be?” administered at baseline, with possible responses “More evening than morning” and “Definitely evening” coded as controls and “Definitely morning” and “More morning than evening” coded as cases (“Do not know” and “Prefer not to answer” responses were coded as missing). Of the respondents, 23.9% identified as “Definitely morning”, 32.2% as “More morning than evening”, 25.6% as “More evening than morning”, 8.0% as “Definitely evening”, and 10.3% as “Do not know”. Of the 351 variants discovered, we used 148.

Sleep duration was instrumented using independent genetic variants (at LD cut-off r2<0.05 and 1 Mb) discovered in a 2019 GWAS of 446 118 UKB European participants [34], which explained 0.69% of the variance in sleep duration. These variants were associated with the question “About how many hours sleep do you get in every 24 h? (please include naps)” administered at baseline, with responses in hourly increments (authors excluded responses >3H or <18h and, “Do not know” and “Prefer not to answer” responses were coded as missing). The authors reported a mean self-reported habitual sleep duration of 7.2 hours. Of the 78 variants discovered, we used 74.

Any insomnia symptoms were instrumented using 29 independent genetic variants from a previously published large-scale GWAS [25]. These variants were discovered in 453 379 UKB European participants, explaining 1.0% of the variance, based on the question “Do you have trouble falling asleep at night or do you wake up in the middle of the night?” administered at baseline, with possible responses “Never/rarely” defined as controls and “Sometimes” and “Usually” defined as cases (“Prefer not to answer” responses were coded as missing). If the participants asked for help, they were shown the message “if this varies a lot, answer this question in relation to the last 4 weeks”. Of the UKB respondents, 23.9% responded “Never/rarely” and 76.1% “Sometimes” or “Usually”. The authors also performed another GWAS for frequent insomnia symptoms not used here, as we decided to use Jansen et al.’s (2019) GWAS due to how cases and controls were defined [32].

Frequent insomnia symptoms were instrumented using independent genetic variants (at LD cut-off r2≤0.6 and >250 kb) discovered in the GWAS of Jansen et al. (2019) [32]. These variants were discovered in 1 331 010 European participants from the UKB and 23andMe cohorts combined and explained 2.6% of the variance. The lead genetic variants showed concordant effects in both samples. For UKB participants, the same question used in the GWAS by Lane et al. (2019) was employed. Responses “Never/rarely” and “Sometimes” were defined as controls and “Usually” as cases. For 23andMe participants, insomnia cases were defined as those who affirmed being told by a doctor, diagnosed or treated for insomnia, diagnosed with sleep disturbances, routinely having trouble getting to sleep at night, or ever taking prescription sleep aids. Controls were defined as those who did not provide a positive or uncertain response (e.g. "I don’t know"; "I am not sure") to any of the issues listed above, nor were diagnosed with Narcolepsy, Sleep apnea, Restless leg syndrome, Post-traumatic stress disorder, Autism, or Asperger. The prevalence of insomnia was 28.3% in the UKB sample, 30.5% in 23andMe, and 29.9% in the combined sample. Of the 248 variants discovered, we used 215.

Depression was instrumented using independent genetic variants discovered in two large-scale GWAS [43,44] to increase the number of valid variants following criteria from our recent review [45]. Wray et al. (2018) [44] conducted a genome-wide meta-analysis of 480 359 Europeans from seven different cohorts (including UKB), explaining 1.9% of the variance (at LD cut-off r2<0.1). Cases (28.2%) were defined as those with an MDD diagnosis established using structured diagnostic instruments from assessments by trained interviewers, self-reported symptoms, diagnosis or treatment, or medical records. Controls (71.8%) were screened for the absence of lifetime Major Depressive Disorder (MDD) and randomly selected from the population. Variants from Howard et al. (2019) [43] were discovered in a genome-wide meta-analysis of 807 553 Europeans from three large studies (LD cut-off r2<0.1 and 3 Mb). The analysis included the 23andMe_307k cohort [46], the UKB cohort [47], and the PGC_139k cohort [44] (we used only UKB summary statistics). The depression phenotype included probable MDD based on self-reported depressive symptoms with associated impairment, MDD identified from hospital admission records and self-reported help-seeking for problems with nerves, anxiety, tension, or depression. Controls were 69.5% of the total sample, while cases were 30.5%. Of the 146 variants discovered, we used 39 variants.

Following guidance from our recent review of selection of sleep genetic instruments [45], we selected genetic instruments for morningness, sleep duration, and insomnia symptoms (Table 1 and Supplementary Table 1). Genetic instruments for depression were selected following the same criteria. In particular, we excluded correlated variants using linkage disequilibrium (LD) clumping with r2<0.01 within 250kb. We only included genome-wide significant (p<5x10-8) variants. We excluded weak instruments, as assessed by the F-statistics (F>10 indicates that substantial weak instrument bias is unlikely [48]) using the Cragg-Donald formula (F=(n−k−1/k) (R2/1−R2)) [49] (k = number of instruments) or the ‘t-statistic’ summary-level method when the R2 was unknown [45]. We also harmonised the genetic variants between the exposure GWAS and our outcome sample by aligning effect alleles. We did not exclude palindromic variants because we mostly used a single sample. Finally, we calculated the potential bias due to overlapping samples. This bias implies overestimating the strongest variant in the data under analysis when using genetic variants discovered in the analytical sample [50]. We used an online calculator available at https://sb452.shinyapps.io/overlap/ [51] assuming 100% sample overlap. We estimated that the biases were small for all outcomes analysed (absolute value of biases ≤0.001) with type-1 error rate = 0.05 (Supplementary Table 2).

Circadian and sleep measures

We selected our morningness, sleep duration, and insomnia symptoms outcomes using phenotypes similar to our exposures. For morningness, we used the same questions and categories as Jones et al. (2019) [42]. For sleep duration, we used the same question as Dashti et al. (2019) [34] with responses in hourly increments. Based on our sleep duration distribution, we excluded extreme responses of <2H or >12h. For any and frequent insomnia symptoms, we used the same UKB question and categories as Lane et al. (2019) [25] and Jansen et al. (2019) [32], respectively.

Depression measures

For depression, we pragmatically decided to use two phenotypes: “diagnosis” (primary outcome for depression) and “diagnosis or treatment” (secondary outcome for depression). The diagnosis category was composed of participants who had answered yes to the question “Have you been diagnosed with one or more of the following mental health problems by a professional, even if you don’t have it currently?: depression” plus participants with primary or secondary MDD (mild, moderate, severe or recurrent without psychotic symptoms) identified from hospital inpatient records.

The diagnosis and treatment category was composed of participants in the diagnosis category plus those who have ever taken substances (prescribed medication for at least two weeks, unprescribed medication more than once or drugs or alcohol more than once) or undertaken certain activities more than once (therapies such as psychotherapy, counselling, group therapy or cognitive-behavioural therapy or other therapeutic activities such as mindfulness, yoga or art classes) to cope with symptoms of depression (prolonged feelings of sadness and/or anhedonia). We included this secondary outcome to increase the proportion of cases with the aim of replicating our results. However, this outcome is less precise in assessing depression compared to the primary outcome, as it encompasses a broad spectrum of therapeutic strategies to manage depressive symptoms.

Anxiety measures

Considering the comorbidity between depression and anxiety [6] and the fact that the genetic variants discovered for depression by Howard et al. (2019) [43] were based on a phenotype that included self-reported help-seeking for problems with nerves, anxiety, tension, or depression, we also analysed two outcomes related to anxiety. We did not conduct bidirectional analyses with anxiety since the SNPs discovered to date have not been replicated at the genome-wide level of significance [52].

The mixed anxiety-depressive symptoms outcome was composed of participants who had answered yes to at least one of the following questions: “Have you ever seen a general practitioner (GP) for nerves, anxiety, tension or depression?” and “Have you ever seen a psychiatrist for nerves, anxiety, tension or depression?”.

The anxiety disorders outcome was composed of participants who had answered yes to the question, “Have you been diagnosed with one or more of the following mental health problems by a professional, even if you don’t have it currently?: “anxiety, nerves or generalised anxiety disorder”. This category also included participants with primary or secondary “Generalised anxiety disorder” identified from hospital inpatient records.

Statistical analyses

i. Main analyses

We performed logistic regression between each of the genetic variants and our outcomes (for sleep duration as an outcome we performed linear regressions), adjusting for ten principal components to minimise issues of residual confounding by population stratification (for the analyses of depression to morningness, insomnia, and sleep duration, we corroborated that when adjusting for anxiety disorders the results remained the same). Inverse-variance weighting (IVW) MR was implemented. The IVW, also known as ‘conventional MR’ estimates the effect of an exposure (e.g., depression) on a given outcome (e.g., insomnia) by taking an average of the genetic variants’ ratio of variant-outcome (SNP→Y) to variant-exposure (SNP→X) association, which is calculated using the principles of a fixed-effects meta-analysis [53]. We also implemented standard sensitivity analyses, including MR-Egger and the Weighted Median Estimator (WME). MR-Egger regression yields an intercept term to denote the presence or absence of unbalanced horizontal pleiotropy and its estimand accounts for this pleiotropy [54], while the WME can give more robust estimates when up to 50% of the genetic variants have invalid weights. Consistency in effect sizes and overlap of 95% confidence intervals (CI) were examined [55]. Results are expressed as odds ratios, which should be interpreted as odds of having the outcome for 1-log-odd of increase in the exposure. Only for sleep duration as an outcome, results are expressed as unstandardised beta coefficients, which should be interpreted as differences in the outcome for every 1-unit increase in the exposure. Analyses were performed using PLINK 2.0 and the “MendelianRandomization” package [56] for R studio version 2022.02.0.

ii. Additional analyses

Following the same approach as for our main analyses, we tested the association between our circadian and sleep exposures and mixed anxiety-depressive symptoms, and anxiety disorders as outcomes.

iii. Testing of MR assumptions

a. Associations between the genetic instrument and exposure instrumented (GWAS robust): variants instrumented have been robustly associated with our phenotypes in recent large-scale GWAS.

b. No association between genetic instruments and the outcome other than via the exposure under study (no horizontal pleiotropy): we implemented MR-Egger and WME sensitivity analyses to test this assumption.

c. No confounding of the SNP-outcome associations (e.g. confounding by population stratification, assortative mating or dynastic effects): we regressed several covariates on our main instruments using a Bonferroni multiple testing correction of 0.05/nSNPs. The list of covariates we selected was based on recent literature [57–59] and included age, sex (female, male), years of full-time education, deprivation (Townsend deprivation quintiles), smoking status (never or previous vs. current smoker), alcohol status (never or previous vs current consumer), body mass index (BMI) (kg/m2) and, physical activity (days of moderate activity for more than 10 minutes).

Results

Main MR results

Associations between morningness and depression

No associations were found between genetically predicted morningness and diagnosis of depression (OR=0.96, 95%CI=0.87; 1.06) (Figure 1a, Supplementary Tables 3 & 7). Associations were found between genetically predicted morningness and lower odds of having a diagnosis or treatment of depression only with the WME (OR=0.89, 95%CI=0.80; 0.98) (Figure 1b, Supplementary Tables 3 & 7). No associations were found between genetically predicted depression and morningness (OR=1.01, 95%CI=0.90; 1.14) (Figure 1c, Supplementary Tables 3 & 7).

Figure 1.
Figure 1.

Associations between morningness and depression, including sensitivity analyses. a) Morningness to depression (diagnosis). b) Morningness to depression (diagnosis and treatment). c) Depression to morningness. IVW, inverse-variance weighting; WME, weighted median estimator; 95% CI, 95% confidence interval. Odds ratio: odds of having the outcome for 1-log-odd of increase in the exposure.

Associations between sleep duration and depression

No associations were found between genetically predicted sleep duration and diagnosis of depression (OR=0.97, 95%CI=0.80; 1.18) (Figure 2a, Supplementary Tables 4 & 7) or diagnosis or treatment of depression (OR=0.94, 95%CI=0.80; 1.10) (Figure 2b, Supplementary Tables 4 & 7). Results showed an association between genetically predicted depression and shorter sleep duration only with the WME (β=-4.67 minutes, 95%CI=-7.44; -1.89) (Figure 2c, Supplementary Tables 4 & 7).

Figure 2.
Figure 2.

Associations between sleep duration and depression, including sensitivity analyses. a) Sleep duration to depression (diagnosis). b) Sleep duration to depression (diagnosis and treatment). c) Depression to sleep duration. IVW, inverse-variance weighting; WME, weighted median estimator; 95% CI, 95% confidence interval. Odds ratio: odds of having the outcome for 1-log-odd of increase in the exposure. Beta (unstandardised beta coefficient): differences in the outcome for every 1-unit increase in the exposure.

Associations between insomnia symptoms and depression

No associations were found between genetically predicted any insomnia symptoms and diagnosis of depression (OR=0.94, 95%CI=0.80; 1.10) or diagnosis or treatment of depression (OR=0.96, 95%CI=0.83; 1.10) (Figure 3a & 3b, Supplementary Tables 5 & 7). Genetically predicted depression was associated with higher odds of having any insomnia symptoms (OR=1.22, 95%CI=1.11; 1.34). MR-Egger indicated no unbalanced horizontal pleiotropy (MR-Egger intercept P-values > 0.05). The MR-Egger slope was not directionally consistent with the IVW estimate. However, the WME estimate was consistent in terms of direction and size (Figure 3c, Supplementary Tables 5 & 7).

Figure 3.
Figure 3.

Associations between insomnia and depression, including sensitivity analyses. a) Any insomnia symptoms to depression (diagnosis). b) Any insomnia symptoms to depression (diagnosis and treatment). c) Depression to any insomnia symptoms. d) Frequent insomnia symptoms to depression (diagnosis). e) Frequent insomnia symptoms to depression (diagnosis and treatment). f) Depression to frequent insomnia symptoms. IVW, inverse-variance weighting; WME, weighted median estimator; 95% CI, 95% confidence interval. Odds ratio: odds of having the outcome for 1-log-odd of increase in the exposure.

Genetically predicted frequent insomnia symptoms were associated with higher odds of having a diagnosis of depression (OR=1.14, 95%CI=1.10; 1.19) (Figure 3d, Supplementary Tables 6 & 7) and diagnosis or treatment of depression (OR=1.10, 95%CI=1.06; 1.14) (Figure 3e, Supplementary Tables 6 & 7), and genetically predicted depression was associated with higher odds of having frequent insomnia symptoms (OR=1.30, 95%CI=1.19; 1.42) (Figure 3f, Supplementary Tables 6 & 7). Both MR-Egger and WME approaches indicated no unbalanced horizontal pleiotropy (MR-Egger intercept P-values ≥ 0.05) and were consistent in terms of direction and size (although the “diagnosis or treatment” result from the MR-Egger was not directionally consistent).

Additional MR results

Associations between morningness, sleep duration, insomnia and anxiety outcomes (mixed anxiety-depressive symptoms and anxiety disorders)

No associations were found between genetically predicted morningness (OR=0.97, 95%CI=0.91; 1.03) or genetically predicted sleep duration (OR=0.87, 95%CI=0.75; 1.00) and mixed anxiety-depressive symptoms (Figure 4a & Figure 4b, Supplementary Table 3-6). Results showed that genetically predicted any insomnia symptoms were associated with higher odds of having mixed anxiety-depressive symptoms (OR=1.19, 95%CI=1.08; 1.32). Both MR-Egger and WME approaches indicated no unbalanced horizontal pleiotropy (MR-Egger intercept P-values > 0.05). The MR-Egger slope was not directionally consistent with the IVW estimate, but the WME estimate was (Figure 4c, Supplementary Table 3-6). Genetically predicted frequent insomnia symptoms were associated with higher odds of having mixed anxiety-depressive symptoms (OR=1.20, 95%CI=1.17; 1.23). The MR-Egger approach was consistent in terms of direction. However, the intercept p-value was 0.007; thus, we performed a leave-one-out analysis [60]. The analysis did not suggest undue influence by single outliers or systemic bias (Supplementary Table 8). WME estimate was consistent in terms of direction and size (Figure 4d, Supplementary Table 3-6). No associations were found between genetically predicted morningness (OR=0.92, 95%CI=0.78; 1.09) (Figure 4e, Supplementary Table 3-6), sleep duration (OR=1.09, 95%CI=0.80; 1.47) (Figure 4f, Supplementary Table 3-6), any insomnia symptoms (OR=1.03, 95%CI=0.82; 1.29) (Figure 4g, Supplementary Table 3-6) or frequent insomnia symptoms (OR=1.02, 95%CI=0.95; 1.10) (Figure 4h, Supplementary Table 3-6) with anxiety disorders.

Figure 4.
Figure 4.

Associations between any insomnia, frequent insomnia, morningness, sleep duration and anxiety outcomes, including sensitivity analyses. a) Morningness to mixed anxiety-depressive symptoms. b) Sleep duration to mixed anxiety-depressive symptoms. c) Any insomnia symptoms to mixed anxiety-depressive symptoms. d) Frequent insomnia symptoms to mixed anxiety-depressive symptoms. e) Morningness to anxiety disorders. f) Sleep duration to anxiety disorders. g) Any insomnia symptoms to anxiety disorders. h) Frequent insomnia symptoms to anxiety disorders. IVW, inverse-variance weighting; WME, weighted median estimator; 95% CI, 95% confidence interval. Odds ratio: odds of having the outcome for 1-log-odd of increase in the exposure.

Testing MR Assumption III

After Bonferroni corrections, we observed that several genetic variants were associated with some covariates (especially with education and BMI) (Supplementary Table 9).

Discussion

The present study investigated the association between several circadian and sleep phenotypes and depression using a comprehensive Mendelian randomisation design. We found a causal bidirectional association between depression and insomnia, and suggestive evidence of an effect of morningness on lower odds of depression, and of depression on shorter sleep duration. We also demonstrated that insomnia confers an effect on mixed anxiety-depressive symptoms, while no associations were found between the circadian and sleep traits studied and anxiety disorders. Contrary to the predominant view of circadian rhythm disruptions only as a secondary consequence of depression, this study follows current evidence suggesting a more influential, causal role of these traits on this condition [18].

Our main finding demonstrated bidirectional associations between depression and insomnia following previous cohort [11,61] and MR studies [29–33]. Notably, we found that frequent insomnia, but not any insomnia symptoms, confers an effect on depression. Nowadays, it is widely accepted that insomnia is a risk factor for the development of depression [11], as confirmed by meta-analyses of longitudinal studies [62,63]. In the UKB, a recent machine learning analysis found that one of the best predictors of depression are insomnia symptoms [64]. However, despite advances in the field, the pathway from insomnia to depression is still unclear. Riemann et al. (2012) [65] point towards rapid eye movement (REM) sleep instability. The authors suggest that REM fragmentation in individuals with insomnia impedes correct overnight emotional regulation processes and, when it becomes frequent, heightens responses to negative stimuli and facilitates the development of depression. This theory is supported by experimental studies showing that induced sleep deprivation alters the processing of emotional information [66,67]. Other studies have proposed that insomnia and sleep deprivation may lead to dysregulations in the stress and inflammatory systems, brain neurotrophic factors and monoaminergic, glutamatergic and orexinergic neurotransmission, contributing to mood dysregulations [68].

We also found that depression confers an effect on any and frequent insomnia symptoms. Insomnia has long been recognised as a vegetative sign of depression [10]. Previous observational [16,61] and MR studies [29–33] have found that depression increases the risk of subsequent insomnia, which is a common complaint in this population [69]. Notably, in contrast to some previous MR studies [29,30], we found that depression exerts a larger effect on insomnia than vice-versa. Interestingly, research has revealed shared genetic and neurobiological mechanisms among these conditions [29,30,70], leading to the proposal that the aetiology of insomnia is even more closely associated with biological processes found in psychiatric disorders, such as depression, rather than those involved in regulating sleep-wake cycles [71].

Our results also indicate suggestive evidence of an association between morningness and lower odds of depression but no evidence of an effect of genetically predicted depression on morningness. Numerous cross-sectional studies have reported that morning types are less prone to depressive symptoms, diagnosed depression and antidepressant use than evening types [12]. Accordingly, a randomised clinical trial demonstrated that shifting sleep timing to early hours in evening types leads to lower depressive symptoms [72]. Further, Jones et al. (2019) [42] reported a negative genetic correlation between morningness and diagnosis of depression. However, the directionality of this association is still unclear [73]. Some MR studies have found robust [21,22] and suggestive evidence [33,42] that morningness decreases the risk of depression. However, other MR studies have not found causal associations between morningness and depression [23,35]. A recent study has suggested causality from circadian phase disturbances to mood symptoms using transfer entropy, a non-linear causal inference method [74]. Possible mechanisms for the association between diurnal preferences and depression include better alignment with typical work-rest schedules and higher daily light exposure [75,76]. Moreover, a recent cross-sectional analysis and genome-wide association study have proposed that sleep inertia drives the association of evening types with psychiatric disorders such as depression [77]. Our findings support the well-documented association between eveningness and depression [12–14], and suggest that, at least in the context of modern social schedules, being a morning type may be more beneficial for mental health.

A suggestive effect of depression on shorter sleep duration was also found, while no associations were found between genetically predicted sleep duration and depression. In line with our results, Sun et al. (2022) [33] observed a suggestive inverse association of depression with sleep duration using MR. Moreover, a three-year prospective study found that older adults with incident depression were more likely to shift from an adequate sleep duration to shorter sleep duration [78]. Additionally, both genetic [34] and non-genetic [58] studies have revealed associations between sleep duration and depression. A more thorough examination of this association is warranted since prior studies have identified U-shaped associations between sleep duration and depression [34,79], which could not be assessed in this work due to the potential for biased results from nonlinear MR methods [80].

Additional results showed that any and frequent insomnia symptoms were associated with higher odds of having mixed anxiety-depressive symptoms, while no associations were found with the other circadian and sleep phenotypes studied. These results support our main findings; insomnia is the trait more strongly associated with depression (even when it is measured mixed with anxiety symptoms). Notably, both any and frequent insomnia symptoms exert an influence on mixed anxiety-depressive symptoms, while only frequent symptoms confer an effect on depression. We speculate that this could be attributed to the outcome being more nuanced, allowing for the effect to be observed even with mild insomnia symptoms. Moreover, we did not find evidence of an effect of circadian and sleep traits on anxiety disorders, implying a unique association with depression that is not replicated when the outcome includes only anxiety. There is considerable literature documenting circadian and sleep problems in individuals with anxiety disorders [81–83]. However, most of these studies were observational in nature, precluding causal inference. In the case of insomnia, a meta-analysis of longitudinal studies indicated that it increased the risk of anxiety disorders [62], which is consistent with a recent MR study [84]. We did not observe this association, which may be due to discrepancies in how anxiety was measured. The lack of association between morningness and anxiety follows previous literature, including MR studies [12,85], as well as the association between sleep duration and anxiety disorders [84]. Considering that anxiety still lacks robust instruments to perform MR analyses, future studies should assess its effect on circadian and sleep traits.

Despite substantial observational evidence linking depression with circadian and sleep traits [12–15,17,18], consistent causal associations have been difficult to establish. It is thus noteworthy that we found bidirectional causal associations between depression and insomnia symptoms. These associations are robust even though they rely on an estimation of insomnia symptoms based solely on one self-reported question, and they are specific as they were confirmed for insomnia symptoms but not for morningness or sleep duration. Therefore, it is important to note that a persistent challenge in the chronobiological field is the distinction between traits; for example, early morning awakening, a classic feature of insomnia [86] (as well as depression [87]), might be misinterpreted as a morning preference. Future research should explore whether distinct components of insomnia are causally related to depression and performing in-depth explorations into the mechanisms driving these associations.

Finally, the three core MR assumptions were corroborated to validate our findings. We instrumented the best genetic variants available (assumption I), implemented sensitivity analyses to check horizontal pleiotropy (assumption II), and performed regressions between our genetic instruments and unobserved confounders (assumption III). Regarding this last assumption, we found that some of the variants instrumented were associated with common confounders. These associations warrant further examination, as they could be related to the traits being polygenic in nature and/or may constitute vertical rather than horizontal pleiotropy. Future research shall explore whether any of the SNPs are vertically pleiotropic by conducting MR mediation analyses [88].

Strengths and limitations

The study’s strengths include using an MR approach, which reduces bias due to confounding and reverse causality, a large sample size that ensures sufficient power, carefully selecting genetic instruments [45], and assessing several circadian, sleep and depression phenotypes. However, limitations should also be acknowledged. First, given that UKB participants are healthier than the general population, the findings may not apply to individuals with significant comorbidities [89]. Second, the sample was restricted to white Europeans, as the genetic variants were discovered in this population; future work should examine these genetic associations in diverse racial and ethnic groups for generalisation and to promote health equity. Third, volunteers were 40-69 years old; thus, our results cannot be generalised to other age groups. Fourth, most of our exposures were binary, which has some limitations [90]. Fifth, the circadian and sleep phenotypes were assessed using self-reported questions, emphasising the need for replication using objective measures such as accelerometer-derived data when possible. Sixth, we found that several genetic variants were linked to confounders, and we cannot rule out the possibility of other confounding factors. Last, depression is a highly heterogeneous condition [91,92], underscoring the need to examine its subtypes (e.g., typical and atypical; mild, moderate and severe; episodic, recurrent and persistent).

Conclusions

In conclusion, this MR study offers novel evidence of causal associations between circadian rhythms, sleep, and depression through the analysis of multiple phenotypes. Notably, the findings demonstrate bidirectional causal associations between depression and insomnia, as well as suggestive evidence linking morningness with lower odds of depression and an effect of depression on sleep deprivation. These findings provide insights into the association between the circadian system and depression, enhancing our understanding of the etiopathogenic processes involved in this condition. Future studies should focus on depressive subtypes and other circadian and sleep traits, and replicate these findings using other datasets and methods.

Data Availability

The UK Biobank data are publicly available to all bona fide researchers at https://www.ukbiobank.ac.uk.

Conflict of Interest

The authors declare that they have no conflict of interest.

Ethics approval

UKB participants had given written informed consent and ethical approval for the study was granted by the North West Haydock Research Ethics Committee of the UK’s Health Research Authority.

Acknowledgements

This research has been conducted using the UK Biobank Resource under Application Number 68232. This work uses data provided by patients and collected by the NHS as part of their care and support. We thank all UKB researchers and volunteers. We thank Albert Henry for his suggestions on analyses. VP was supported by Programa de Desarrollo de las Ciencias Básicas (PEDECIBA, MEC-UdelaR, Uruguay), Agencia Nacional de Investigación e Innovación (ANII, Uruguay) [grant number MOV_CA_2020_1_163153], Comisión Sectorial de Investigación Científica (CSIC, UdelaR, Uruguay) and, Comisión Académica de Posgrados (CAP, UdelaR, Uruguay). MR is funded by the UK Medical Research Council (grants MC_UU_00019/1 and 3). BT and AS were supported by Comisión Sectorial de Investigación Científica (CSIC, UdelaR, Uruguay) [grant number 883158]. VG was supported by the Professor David Matthews Non-Clinical Fellowship from the Diabetes Research and Wellness Foundation (United Kingdom) [grant number SCA/01/NCF/22].

References

  1. [1].↵
    Aschoff J. A self-sustained oscillator with an inherent frequency underlies human 24-hour periodicity. 1965;148:7.
    OpenUrl
  2. [2].↵
    Buysse DJ. Sleep Health: Can We Define It? Does It Matter? Sleep 2014;37:9–17. doi:10.5665/sleep.3298.
    OpenUrlCrossRefPubMed
  3. [3].↵
    Nelson KL, Davis JE, Corbett CF. Sleep quality: An evolutionary concept analysis. Nurs Forum (Auckl) 2022;57:144–51. doi:10.1111/nuf.12659.
    OpenUrlCrossRefPubMed
  4. [4].↵
    Vetter C. Circadian disruption: What do we actually mean? Eur J Neurosci 2020;51:531–50. doi:10.1111/ejn.14255.
    OpenUrlCrossRefPubMed
  5. [5].↵
    Medic G, Wille M, Hemels M. Short- and long-term health consequences of sleep disruption. Nat Sci Sleep 2017;Volume 9:151–61. doi:10.2147/NSS.S134864.
    OpenUrlCrossRefPubMed
  6. [6].↵
    Saha S, Lim CCW, Cannon DL, Burton L, Bremner M, Cosgrove P, et al. Co-morbidity between mood and anxiety disorders: A systematic review and meta-analysis. Depress Anxiety 2021;38:286–306. doi:10.1002/da.23113.
    OpenUrlCrossRefPubMed
  7. [7].↵
    World Health Organization. Depression 2021. https://www.who.int/news-room/fact-sheets/detail/depression (accessed January 31, 2023).
  8. [8].↵
    Zaki NFW, Spence DW, BaHammam AS, Pandi-Perumal SR, Cardinali DP, Brown GM. Chronobiological theories of mood disorder. Eur Arch Psychiatry Clin Neurosci 2018;268:107–18. doi:10.1007/s00406-017-0835-5.
    OpenUrlCrossRef
  9. [9].↵
    Whiteford HA, Ferrari AJ, Degenhardt L, Feigin V, Vos T. The Global Burden of Mental, Neurological and Substance Use Disorders: An Analysis from the Global Burden of Disease Study 2010. PLOS ONE 2015;10:e0116820. doi:10.1371/journal.pone.0116820.
    OpenUrlCrossRefPubMed
  10. [10].↵
    American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders (5th ed.). Washington DC: American Psychiatric Association; 2013.
  11. [11].↵
    Franzen PL, Buysse DJ. Sleep disturbances and depression: risk relationships for subsequent depression and therapeutic implications. Dialogues Clin Neurosci 2008;10:473–81. doi:10.31887/DCNS.2008.10.4/plfranzen.
    OpenUrlCrossRefPubMed
  12. [12].↵
    Kivelä L, Papadopoulos MR, Antypa N. Chronotype and Psychiatric Disorders. Curr Sleep Med Rep 2018;4:94–103. doi:10.1007/s40675-018-0113-8.
    OpenUrlCrossRef
  13. [13].
    Hidalgo MP, Caumo W, Posser M, Coccaro SB, Camozzato AL, Chaves MLF. Relationship between depressive mood and chronotype in healthy subjects: Depressive mood and chronotype. Psychiatry Clin Neurosci 2009;63:283–90. doi:10.1111/j.1440-1819.2009.01965.x.
    OpenUrlCrossRefPubMed
  14. [14].↵
    Kitamura S, Hida A, Watanabe M, Enomoto M, Aritake-Okada S, Moriguchi Y, et al. Evening preference is related to the incidence of depressive states independent of sleep-wake conditions. Chronobiol Int 2010;27:1797–812. doi:10.3109/07420528.2010.516705.
    OpenUrlCrossRefPubMed
  15. [15].↵
    Zhai L, Zhang H, Zhang D. Sleep duration and depression among adults: a meta-analysis of prospective studies. Depress Anxiety 2015;32:664–70. doi:10.1002/da.22386.
    OpenUrlCrossRefPubMed
  16. [16].↵
    Van Mill JG, Hoogendijk WJG, Vogelzangs N, Van Dyck R, Penninx BWJH. Insomnia and Sleep Duration in a Large Cohort of Patients With Major Depressive Disorder and Anxiety Disorders. J Clin Psychiatry 2010;71:239–46. doi:10.4088/JCP.09m05218gry.
    OpenUrlCrossRefPubMedWeb of Science
  17. [17].↵
    Perlis ML, Posner D, Riemann D, Bastien CH, Teel J, Thase M. Insomnia. The Lancet 2022;400:1047–60. doi:10.1016/S0140-6736(22)00879-0.
    OpenUrlCrossRefPubMed
  18. [18].↵
    Fang H, Tu S, Sheng J, Shao A. Depression in sleep disturbance: A review on a bidirectional relationship, mechanisms and treatment. J Cell Mol Med 2019;23:2324–32. doi:10.1111/jcmm.14170.
    OpenUrlCrossRef
  19. [19].↵
    Köhler CA, Evangelou E, Stubbs B, Solmi M, Veronese N, Belbasis L, et al. Mapping risk factors for depression across the lifespan: An umbrella review of evidence from meta-analyses and Mendelian randomization studies. J Psychiatr Res 2018;103:189–207. doi:10.1016/j.jpsychires.2018.05.020.
    OpenUrlCrossRefPubMed
  20. [20].↵
    Smith GD, Ebrahim S. ‘Mendelian randomization’: can genetic epidemiology contribute to understanding environmental determinants of disease? Int J Epidemiol 2003;32:1–22. doi:10.1093/ije/dyg070.
    OpenUrlCrossRefPubMedWeb of Science
  21. [21].↵
    O’Loughlin J, Casanova F, Jones SE, Hagenaars SP, Beaumont RN, Freathy RM, et al. Using Mendelian Randomisation methods to understand whether diurnal preference is causally related to mental health. Mol Psychiatry 2021;26:6305–16. doi:10.1038/s41380-021-01157-3.
    OpenUrlCrossRef
  22. [22].↵
    Daghlas I, Lane JM, Saxena R, Vetter C. Genetically Proxied Diurnal Preference, Sleep Timing, and Risk of Major Depressive Disorder. JAMA Psychiatry 2021;78:903. doi:10.1001/jamapsychiatry.2021.0959.
    OpenUrlCrossRefPubMed
  23. [23].↵
    Qiu-Qiang Z, Wei-Wei Y, Shan-Shu H, Yi-Ran L. Mendelian randomization of individual sleep traits associated with major depressive disorder. J Affect Disord 2024;365:105–11. doi:10.1016/j.jad.2024.08.068.
    OpenUrlCrossRefPubMed
  24. [24].↵
    Austin-Zimmerman I, Levey DF, Giannakopoulou O, Deak JD, Galimberti M, Adhikari K, et al. Genome-wide association studies and cross-population meta-analyses investigating short and long sleep duration. Nat Commun 2023;14:6059. doi:10.1038/s41467-023-41249-y.
    OpenUrlCrossRefPubMed
  25. [25].↵
    Lane JM, Jones SE, Dashti HS, Wood AR, Aragam KG, van Hees VT, et al. Biological and clinical insights from genetics of insomnia symptoms. Nat Genet 2019;51:387–93. doi:10.1038/s41588-019-0361-7.
    OpenUrlCrossRefPubMed
  26. [26].↵
    Nassan M, Daghlas I, Winkelman JW, Dashti HS, International Suicide Genetics Consortium, Saxena R. Genetic evidence for a potential causal relationship between insomnia symptoms and suicidal behavior: a Mendelian randomization study. Neuropsychopharmacology 2022. doi:10.1038/s41386-022-01319-z.
    OpenUrlCrossRef
  27. [27].
    He M, Zhou J, Li X, Wang R. Investigating the causal effects of smoking, sleep, and BMI on major depressive disorder and bipolar disorder: a univariable and multivariable two-sample Mendelian randomization study. Front Psychiatry 2023;14:1206657. doi:10.3389/fpsyt.2023.1206657.
    OpenUrlCrossRefPubMed
  28. [28].↵
    Gibson MJ, Lawlor DA, Millard LAC. Identifying the potential causal role of insomnia symptoms on 11,409 health-related outcomes: a phenome-wide Mendelian randomisation analysis in UK Biobank. BMC Med 2023;21:128. doi:10.1186/s12916-023-02832-8.
    OpenUrlCrossRefPubMed
  29. [29].↵
    Baranova A, Cao H, Zhang F. Shared genetic liability and causal effects between major depressive disorder and insomnia. Hum Mol Genet 2022;31:1336–45. doi:10.1093/hmg/ddab328.
    OpenUrlCrossRefPubMed
  30. [30].↵
    Cai L, Bao Y, Fu X, Cao H, Baranova A, Zhang X, et al. Causal links between major depressive disorder and insomnia: A Mendelian randomisation study. Gene 2021;768:145271. doi:10.1016/j.gene.2020.145271.
    OpenUrlCrossRefPubMed
  31. [31].
    Zhou F, Guo Y, Wang Z, Liu S, Xu H. Assessing the causal associations of insomnia with depressive symptoms and subjective well-being: a bidirectional Mendelian randomization study. Sleep Med 2021;87:85–91. doi:10.1016/j.sleep.2021.08.025.
    OpenUrlCrossRefPubMed
  32. [32].↵
    Jansen PR, Watanabe K, Stringer S, Skene N, Bryois J, Hammerschlag AR, et al. Genome-wide analysis of insomnia in 1,331,010 individuals identifies new risk loci and functional pathways. Nat Genet 2019;51:394–403. doi:10.1038/s41588-018-0333-3.
    OpenUrlCrossRefPubMed
  33. [33].↵
    Sun X, Liu B, Liu S, Wu DJH, Wang J, Qian Y, et al. Sleep disturbance and psychiatric disorders: a bidirectional Mendelian randomisation study. Epidemiol Psychiatr Sci 2022;31:e26. doi:10.1017/S2045796021000810.
    OpenUrlCrossRefPubMed
  34. [34].↵
    Dashti HS, Jones SE, Wood AR, Lane JM, van Hees VT, Wang H, et al. Genome-wide association study identifies genetic loci for self-reported habitual sleep duration supported by accelerometer-derived estimates. Nat Commun 2019;10:1100. doi:10.1038/s41467-019-08917-4.
    OpenUrlCrossRefPubMed
  35. [35].↵
    Hu Y, Shmygelska A, Tran D, Eriksson N, Tung JY, Hinds DA. GWAS of 89,283 individuals identifies genetic variants associated with self-reporting of being a morning person. Nat Commun 2016;7:10448. doi:10.1038/ncomms10448.
    OpenUrlCrossRefPubMed
  36. [36].↵
    Gao X, Meng L-X, Ma K-L, Liang J, Wang H, Gao Q, et al. The bidirectional causal relationships of insomnia with five major psychiatric disorders: A Mendelian randomization study. Eur Psychiatry 2019;60:79–85. doi:10.1016/j.eurpsy.2019.05.004.
    OpenUrlCrossRefPubMed
  37. [37].↵
    Sudlow C, Gallacher J, Allen N, Beral V, Burton P, Danesh J, et al. UK Biobank: An Open Access Resource for Identifying the Causes of a Wide Range of Complex Diseases of Middle and Old Age. PLOS Med 2015;12:e1001779. doi:10.1371/journal.pmed.1001779.
    OpenUrlCrossRefPubMed
  38. [38].↵
    Skrivankova VW, Richmond RC, Woolf BAR, Yarmolinsky J, Davies NM, Swanson SA, et al. Strengthening the Reporting of Observational Studies in Epidemiology Using Mendelian Randomization: The STROBE-MR Statement. JAMA 2021;326:1614. doi:10.1001/jama.2021.18236.
    OpenUrlCrossRefPubMed
  39. [39].↵
    Bycroft C, Freeman C, Petkova D, Band G, Elliott LT, Sharp K, et al. The UK Biobank resource with deep phenotyping and genomic data. Nature 2018;562:203–9. doi:10.1038/s41586-018-0579-z.
    OpenUrlCrossRefPubMed
  40. [40].↵
    Constantinescu A-E, Mitchell RE, Zheng J, Bull CJ, Timpson NJ, Amulic B, et al. A framework for research into continental ancestry groups of the UK Biobank. Hum Genomics 2022;16:3. doi:10.1186/s40246-022-00380-5.
    OpenUrlCrossRefPubMed
  41. [41].↵
    Lu C, Ahmed R, Lamri A, Anand SS. Use of race, ethnicity, and ancestry data in health research. PLOS Glob Public Health 2022;2:e0001060. doi:10.1371/journal.pgph.0001060.
    OpenUrlCrossRefPubMed
  42. [42].↵
    Jones SE, Lane JM, Wood AR, van Hees VT, Tyrrell J, Beaumont RN, et al. Genome-wide association analyses of chronotype in 697,828 individuals provides insights into circadian rhythms. Nat Commun 2019;10:343. doi:10.1038/s41467-018-08259-7.
    OpenUrlCrossRefPubMed
  43. [43].↵
    Howard DM, Adams MJ, Clarke T-K, Hafferty JD, Gibson J, Shirali M, et al. Genome-wide meta-analysis of depression identifies 102 independent variants and highlights the importance of the prefrontal brain regions. Nat Neurosci 2019;22:343–52. doi:10.1038/s41593-018-0326-7.
    OpenUrlCrossRefPubMed
  44. [44].↵
    Wray NR, Ripke S, Mattheisen M, Trzaskowski M, Byrne EM, Abdellaoui A, et al. Genome-wide association analyses identify 44 risk variants and refine the genetic architecture of major depression. Nat Genet 2018;50:668–81. doi:10.1038/s41588-018-0090-3.
    OpenUrlCrossRefPubMed
  45. [45].↵
    Paz V, Dashti HS, Burgess S, Garfield V. Selection of genetic instruments in Mendelian randomisation studies of sleep traits. Sleep Med 2023;112:342–51. doi:10.1016/j.sleep.2023.10.036.
    OpenUrlCrossRefPubMed
  46. [46].↵
    Hyde CL, Nagle MW, Tian C, Chen X, Paciga SA, Wendland JR, et al. Identification of 15 genetic loci associated with risk of major depression in individuals of European descent. Nat Genet 2016;48:1031–6. doi:10.1038/ng.3623.
    OpenUrlCrossRefPubMed
  47. [47].↵
    Howard DM, Adams MJ, Shirali M, Clarke T-K, Marioni RE, Davies G, et al. Genome-wide association study of depression phenotypes in UK Biobank identifies variants in excitatory synaptic pathways. Nat Commun 2018;9:1470. doi:10.1038/s41467-018-03819-3.
    OpenUrlCrossRefPubMed
  48. [48].↵
    Burgess S, Thompson SG, CRP CHD Genetics Collaboration. Avoiding bias from weak instruments in Mendelian randomization studies. Int J Epidemiol 2011;40:755–64. doi:10.1093/ije/dyr036.
    OpenUrlCrossRefPubMedWeb of Science
  49. [49].↵
    Burgess S, Thompson SG. Mendelian Randomization: Methods for Causal Inference Using Genetic Variants. CRC Press; 2021.
  50. [50].↵
    Taylor AE, Davies NM, Ware JJ, VanderWeele T, Smith GD, Munafò MR. Mendelian randomization in health research: Using appropriate genetic variants and avoiding biased estimates. Econ Hum Biol 2014;13:99–106. doi:10.1016/j.ehb.2013.12.002.
    OpenUrlCrossRefPubMed
  51. [51].↵
    Burgess S, Davies NM, Thompson SG. Bias due to participant overlap in two-sample Mendelian randomization. Genet Epidemiol 2016;40:597–608. doi:10.1002/gepi.21998.
    OpenUrlCrossRefPubMed
  52. [52].↵
    Van Der Walt K, Campbell M, Stein DJ, Dalvie S. Systematic review of genome-wide association studies of anxiety disorders and neuroticism. World J Biol Psychiatry 2023;24:280–91. doi:10.1080/15622975.2022.2099970.
    OpenUrlCrossRefPubMed
  53. [53].↵
    Burgess S, Bowden J. Integrating summarized data from multiple genetic variants in Mendelian randomization: bias and coverage properties of inverse-variance weighted methods. ArXiv151204486 Stat 2015.
  54. [54].↵
    Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. Int J Epidemiol 2015;44:512–25. doi:10.1093/ije/dyv080.
    OpenUrlCrossRefPubMed
  55. [55].↵
    Bowden J, Smith GD, Haycock PC, Burgess S. Consistent Estimation in Mendelian Randomization with Some Invalid Instruments Using a Weighted Median Estimator. Genet Epidemiol 2016;40:304–14. doi:10.1002/gepi.21965.
    OpenUrlCrossRefPubMed
  56. [56].↵
    Yavorska OO, Burgess S. MendelianRandomization: an R package for performing Mendelian randomization analyses using summarized data. Int J Epidemiol 2017;46:1734–9. doi:10.1093/ije/dyx034.
    OpenUrlCrossRefPubMed
  57. [57].↵
    Merikanto I, Kronholm E, Peltonen M, Laatikainen T, Vartiainen E, Partonen T. Circadian preference links to depression in general adult population. J Affect Disord 2015;188:143–8. doi:10.1016/j.jad.2015.08.061.
    OpenUrlCrossRefPubMed
  58. [58].↵
    Sun Y, Shi L, Bao Y, Sun Y, Shi J, Lu L. The bidirectional relationship between sleep duration and depression in community-dwelling middle-aged and elderly individuals: evidence from a longitudinal study. Sleep Med 2018;52:221–9. doi:10.1016/j.sleep.2018.03.011.
    OpenUrlCrossRefPubMed
  59. [59].↵
    Taylor DJ, Lichstein KL, Durrence HH, Reidel BW, Bush AJ. Epidemiology of Insomnia, Depression, and Anxiety. Sleep 2005;28: 1457–64. doi:10.1093/sleep/28.11.1457.
    OpenUrlCrossRefPubMedWeb of Science
  60. [60].↵
    Burgess S, Thompson SG. Interpreting findings from Mendelian randomization using the MR-Egger method. Eur J Epidemiol 2017;32:377–89. doi:10.1007/s10654-017-0255-x.
    OpenUrlCrossRef
  61. [61].↵
    Roberts RE, Duong HT. Depression and insomnia among adolescents: A prospective perspective. J Affect Disord 2013;148:66–71. doi:10.1016/j.jad.2012.11.049.
    OpenUrlCrossRefPubMed
  62. [62].↵
    Hertenstein E, Feige B, Gmeiner T, Kienzler C, Spiegelhalder K, Johann A, et al. Insomnia as a predictor of mental disorders: A systematic review and meta-analysis. Sleep Med Rev 2019;43:96–105. doi:10.1016/j.smrv.2018.10.006.
    OpenUrlCrossRefPubMed
  63. [63].↵
    Li L, Wu C, Gan Y, Qu X, Lu Z. Insomnia and the risk of depression: a meta-analysis of prospective cohort studies. BMC Psychiatry 2016;16:375. doi:10.1186/s12888-016-1075-3.
    OpenUrlCrossRefPubMed
  64. [64].↵
    Lyall LM, Sangha N, Zhu X, Lyall DM, Ward J, Strawbridge RJ, et al. Subjective and objective sleep and circadian parameters as predictors of depression-related outcomes: A machine learning approach in UK Biobank. J Affect Disord 2023;335:83–94. doi:10.1016/j.jad.2023.04.138.
    OpenUrlCrossRefPubMed
  65. [65].↵
    Riemann D, Spiegelhalder K, Nissen C, Hirscher V, Baglioni C, Feige B. REM Sleep Instability – A New Pathway for Insomnia? Pharmacopsychiatry 2012:s-0031-1299721. doi:10.1055/s-0031-1299721.
    OpenUrlCrossRef
  66. [66].↵
    Franzen PL, Siegle GJ, Buysse DJ. Relationships between affect, vigilance, and sleepiness following sleep deprivation. J Sleep Res 2008;17:34–41. doi:10.1111/j.1365-2869.2008.00635.x.
    OpenUrlCrossRefPubMedWeb of Science
  67. [67].↵
    Franzen PL, Buysse DJ, Dahl RE, Thompson W, Siegle GJ. Sleep deprivation alters pupillary reactivity to emotional stimuli in healthy young adults. Biol Psychol 2009;80:300–5. doi:10.1016/j.biopsycho.2008.10.010.
    OpenUrlCrossRefPubMedWeb of Science
  68. [68].↵
    Palagini L, Bastien CH, Marazziti D, Ellis JG, Riemann D. The key role of insomnia and sleep loss in the dysregulation of multiple systems involved in mood disorders: A proposed model. J Sleep Res 2019;28:e12841. doi:10.1111/jsr.12841.
    OpenUrlCrossRefPubMed
  69. [69].↵
    Yates WR, Mitchell J, Rush AJ, Trivedi MH, Wisniewski SR, Warden D, et al. Clinical features of depressed outpatients with and without co-occurring general medical conditions in STAR*D. Gen Hosp Psychiatry 2004;26:421–9. doi:10.1016/j.genhosppsych.2004.06.008.
    OpenUrlCrossRefPubMedWeb of Science
  70. [70].↵
    Bagherzadeh-Azbari S, Khazaie H, Zarei M, Spiegelhalder K, Walter M, Leerssen J, et al. Neuroimaging insights into the link between depression and Insomnia: A systematic review. J Affect Disord 2019;258:133–43. doi:10.1016/j.jad.2019.07.089.
    OpenUrlCrossRefPubMed
  71. [71].↵
    Lane JM, Qian J, Mignot E, Redline S, Scheer FAJL, Saxena R. Genetics of circadian rhythms and sleep in human health and disease. Nat Rev Genet 2023;24:4–20. doi:10.1038/s41576-022-00519-z.
    OpenUrlCrossRef
  72. [72].↵
    Facer-Childs ER, Middleton B, Skene DJ, Bagshaw AP. Resetting the late timing of ‘night owls’ has a positive impact on mental health and performance. Sleep Med 2019;60:236–47. doi:10.1016/j.sleep.2019.05.001.
    OpenUrlCrossRefPubMed
  73. [73].↵
    Bauducco S, Richardson C, Gradisar M. Chronotype, circadian rhythms and mood. Curr Opin Psychol 2020;34: 77-83. doi:10.1016/j.copsyc.2019.09.002.
    OpenUrlCrossRefPubMed
  74. [74].↵
    Song YM, Jeong J, De Los Reyes AA, Lim D, Cho C-H, Yeom JW, et al. Causal dynamics of sleep, circadian rhythm, and mood symptoms in patients with major depression and bipolar disorder: insights from longitudinal wearable device data. eBioMedicine 2024;103:105094. doi:10.1016/j.ebiom.2024.105094.
    OpenUrlCrossRefPubMed
  75. [75].↵
    LeGates TA, Fernandez DC, Hattar S. Light as a central modulator of circadian rhythms, sleep and affect. Nat Rev Neurosci 2014;15:443–54. doi:10.1038/nrn3743.
    OpenUrlCrossRefPubMed
  76. [76].↵
    Baron KG, Reid KJ. Circadian misalignment and health. Int Rev Psychiatry 2014;26:139–54. doi:10.3109/09540261.2014.911149.
    OpenUrlCrossRefPubMed
  77. [77].↵
    Burns A, Sinnott-Armstrong N, Daghlas I, Windred D, Phillips A, Cain S, et al. Sleep inertia, not chronotype, is a marker of circadian misalignment and a risk factor for psychiatric disorders: genetic and epidemiological evidence. Sleep Med 2024;115:S64. doi:10.1016/j.sleep.2023.11.210.
    OpenUrlCrossRef
  78. [78].↵
    Chen H-C, Chou P. Predictors of Change in Self-Reported Sleep Duration in Community-Dwelling Older Adults: The Shih-Pai Sleep Study, Taiwan. Sci Rep 2017;7:4729. doi:10.1038/s41598-017-04932-x.
    OpenUrlCrossRefPubMed
  79. [79].↵
    Dong L, Xie Y, Zou X. Association between sleep duration and depression in US adults: A cross-sectional study. J Affect Disord 2022;296:183–8. doi:10.1016/j.jad.2021.09.075.
    OpenUrlCrossRefPubMed
  80. [80].↵
    Wade KH, Hamilton FW, Carslake D, Sattar N, Davey Smith G, Timpson NJ. Challenges in undertaking nonlinear Mendelian randomization. Obesity 2023;31:2887–90. doi:10.1002/oby.23927.
    OpenUrlCrossRef
  81. [81].↵
    Mellman TA. Sleep and Anxiety Disorders. Psychiatr Clin North Am 2006;29:1047–58. doi:10.1016/j.psc.2006.08.005.
    OpenUrlCrossRefPubMedWeb of Science
  82. [82].
    Uhde TW, Cortese BM, Vedeniapin A. Anxiety and sleep problems: Emerging concepts and theoretical treatment implications. Curr Psychiatry Rep 2009;11:269–76. doi:10.1007/s11920-009-0039-4.
    OpenUrlCrossRefPubMed
  83. [83].↵
    Cox RC, Olatunji BO. A systematic review of sleep disturbance in anxiety and related disorders. J Anxiety Disord 2016;37:104–29. doi:10.1016/j.janxdis.2015.12.001.
    OpenUrlCrossRef
  84. [84].↵
    Zhou F, Li S, Xu H. Insomnia, sleep duration, and risk of anxiety: A two-sample Mendelian randomization study. J Psychiatr Res 2022;155:219–25. doi:10.1016/j.jpsychires.2022.08.012.
    OpenUrlCrossRefPubMed
  85. [85].↵
    O’Loughlin J, Casanova F, Jones SE, Hagenaars SP, Beaumont RN, Freathy RM, et al. Using Mendelian Randomisation methods to understand whether diurnal preference is causally related to mental health. Mol Psychiatry 2021;26:6305–16. doi:10.1038/s41380-021-01157-3.
    OpenUrlCrossRef
  86. [86].↵
    Riemann D, Nissen C, Palagini L, Otte A, Perlis ML, Spiegelhalder K. The neurobiology, investigation, and treatment of chronic insomnia. Lancet Neurol 2015;14:547–58. doi:10.1016/S1474-4422(15)00021-6.
    OpenUrlCrossRefPubMed
  87. [87].↵
    Richard IH, Lyness JM. An Overview of Depression. In: Jests DV, Friedman JH, editors. Psychiatry Neurol., Totowa, NJ: Humana Press; 2006, p. 33–41. doi:10.1007/978-1-59259-960-8_4.
    OpenUrlCrossRef
  88. [88].↵
    Carter AR, Sanderson E, Hammerton G, Richmond RC, Davey Smith G, Heron J, et al. Mendelian randomisation for mediation analysis: current methods and challenges for implementation. Eur J Epidemiol 2021;36:465–78. doi:10.1007/s10654-021-00757-1.
    OpenUrlCrossRefPubMed
  89. [89].↵
    Fry A, Littlejohns TJ, Sudlow C, Doherty N, Adamska L, Sprosen T, et al. Comparison of Sociodemographic and Health-Related Characteristics of UK Biobank Participants With Those of the General Population. Am J Epidemiol 2017;186:1026–34. doi:10.1093/aje/kwx246.
    OpenUrlCrossRefPubMed
  90. [90].↵
    Burgess S, Labrecque JA. Mendelian randomization with a binary exposure variable: interpretation and presentation of causal estimates. Eur J Epidemiol 2018;33:947–52. doi:10.1007/s10654-018-0424-6.
    OpenUrlCrossRefPubMed
  91. [91].↵
    Cai N, Choi KW, Fried EI. Reviewing the genetics of heterogeneity in depression: operationalizations, manifestations and etiologies. Hum Mol Genet 2020;29:R10–8. doi:10.1093/hmg/ddaa115.
    OpenUrlCrossRefPubMed
  92. [92].↵
    Fried E. Moving forward: how depression heterogeneity hinders progress in treatment and research. Expert Rev Neurother 2017;17:423–5. doi:10.1080/14737175.2017.1307737
    OpenUrlCrossRefPubMed
Back to top
Posted December 10, 2024.
Download PDF

Supplementary Material

Data/Code
Email
ASSOCIATION BETWEEN CIRCADIAN RHYTHMS, SLEEP, AND DEPRESSION: A BIDIRECTIONAL MENDELIAN RANDOMISATION STUDY IN THE UK BIOBANK
Valentina Paz, Dylan M Williams, Marcus Richards, Bettina Tassino, Ana Silva, Victoria Garfield
medRxiv 2024.12.08.24318680; doi: https://doi.org/10.1101/2024.12.08.24318680
Twitter logo Facebook logo LinkedIn logo Mendeley logo
Citation Tools

Subject Area