Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Comparison of Genetic Liability for Sleep Traits Among Individuals With Bipolar Disorder I or II and Control Participants

Comparison of Genetic Liability for Sleep Traits Among Individuals With Bipolar Disorder I or II... Research JAMA Psychiatry | Original Investigation Comparison of Genetic Liability for Sleep Traits Among Individuals With Bipolar Disorder I or II and Control Participants Katie J. S. Lewis, PhD; Alexander Richards, PhD; Robert Karlsson, PhD; Ganna Leonenko, PhD; Samuel E. Jones, PhD; Hannah J. Jones, PhD; Katherine Gordon-Smith, PhD; Liz Forty, PhD; Valentina Escott-Price, PhD; Michael J. Owen, FRCPsych, PhD; Michael N. Weedon, PhD; Lisa Jones, PhD; Nick Craddock, FRCPsych, PhD; Ian Jones, MRCPsych, PhD; Mikael Landén, MD, PhD; Michael C. O’Donovan, FRCPsych, PhD; Arianna Di Florio, MD, PhD Editorial page 235 IMPORTANCE Insomnia, hypersomnia, and an evening chronotype are common in individuals Supplemental content with bipolar disorder (BD), but whether this reflects shared genetic liability is unclear. Stratifying by BD subtypes could elucidate this association and inform sleep and BD research. OBJECTIVE To assess whether polygenic risk scores (PRSs) for sleep traits are associated with BD subtypes I and II. DESIGN, SETTING, AND PARTICIPANTS This case-control study was conducted in the United Kingdom and Sweden with participants with BD and control participants. Multinomial regression was used to assess whether PRSs for insomnia, daytime sleepiness, sleep duration, and chronotype are associated with BD subtypes compared with control participants. Affected individuals were recruited from the Bipolar Disorder Research Network. Control participants were recruited from the 1958 British Birth Cohort and the UK Blood Service. Analyses were repeated in an independent Swedish sample from August 2018 to July 2019. All participants were of European ancestry. EXPOSURES Standardized PRSs derived using alleles from genome-wide association studies of insomnia, sleep duration, daytime sleepiness, and chronotype. These were adjusted for the first 10 population principal components, genotyping platforms, and sex. MAIN OUTCOMES AND MEASURES Association of PRSs with BD subtypes, determined by semistructured psychiatric interview and case notes. RESULTS The main analysis included 4672 participants with BD (3132 female participants [67.0%]; 3404 with BD-I [72.9%]) and 5714 control participants (2812 female participants [49.2%]). Insomnia PRS was associated with increased risk of BD-II (relative risk [RR], 1.14 −5 [95% CI, 1.07-1.21]; P =8.26×10 ) but not BD-I (RR, 0.98 [95% CI, 0.94-1.03]; P = .409) relative to control participants. Sleep-duration PRS was associated with BD-I (RR, 1.10 −5 [95% CI, 1.06-1.15]; P =1.13×10 ) but not BD-II (RR, 0.99 [95% CI, 0.93-1.06]; P = .818). Associations between (1) insomnia PRS and BD-II and (2) sleep-duration PRS and BD-I were replicated in the Swedish sample of 4366 individuals with BD (2697 female participants [61.8%]; 2627 with BD-I [60.2%]) and 6091 control participants (3767 female participants [61.8%]). Chronotype and daytime-sleepiness PRS were not associated with BD subtypes. CONCLUSIONS AND RELEVANCE Per this analysis, BD subtypes differ in genetic liability to insomnia and hypersomnia, providing further evidence that the distinction between BD-I and BD-II has genetic validity. This distinction will be crucial in selecting participants for future research on the role of sleep disturbance in BD. Author Affiliations: Author affiliations are listed at the end of this article. Corresponding Author: Arianna Di Florio, MD, PhD, Medical Research Council Centre for Neuropsychiatric Genetics and Genomics, Institute of Psychological Medicine and Clinical Neurosciences, Cardiff University, Hadyn Ellis Building, Maindy Road, JAMA Psychiatry. 2020;77(3):303-310. doi:10.1001/jamapsychiatry.2019.4079 Cathays, Cardiff, CF24 4HQ, United Published online November 21, 2019. Kingdom (diflorioa@cardiff.ac.uk). (Reprinted) 303 Downloaded From: https://jamanetwork.com/ by a Deepdyve User on 07/07/2021 Research Original Investigation Comparison of Genetic Liability for Sleep Traits in Individuals With Bipolar Disorder and Controls ipolar disorder (BD) and sleep have often been linked. First, reduced sleep duration, a symptom of manic epi- Key Points B sodes, has been implicated as a prodrome and trigger Question Does genetic liability to insomnia, hypersomnia, and 1-5 of mania. Second, insomnia (difficulty initiating and main- chronotype differentiate subtypes of bipolar disorder? taining sleep ) and hypersomnia (prolonged sleep duration or Findings In this case-control study of 4672 participants with excessive daytime sleepiness ) are commonly reported symp- bipolar disorder and 5714 control participants, individuals with 7-9 toms of bipolar depression that persist in the interepisode bipolar disorder I had significantly greater genetic liability to longer 9-18 period and are associated with significant distress and sleep duration, whereas individuals with bipolar disorder II had impairment. Third, there is evidence that individuals with significantly greater genetic liability to insomnia; these findings BD display greater evening preference for sleep (ie, an eve- were replicated in an independent sample. Individuals with bipolar 19-21 subtypes did not differ in genetic liability to morning or evening ning chronotype) than healthy control participants. chronotype. At present, sleep interventions for individuals with BD pri- 22-24 marily focus on reducing insomnia and stabilizing circa- Meaning Associations between polygenic liability to insomnia and dian rhythms. Understanding the association between sleep hypersomnia and clinical strata within bipolar disorder are shown in this study for the first time, to our knowledge. and BD is important and could inform clinical interventions. 26-31 Recent genome-wide association studies (GWAS) pro- vide an opportunity to examine the association between sleep and BD at the genomic level. Using summary-level data, some voluntary groups, such as Bipolar UK). Participants were studies have demonstrated a positive genetic correlation be- excluded if they had affective illness experienced solely in 28,30 tween BD and sleep duration. Other studies, however, have response to alcohol or substance misuse or secondary to found no significant genetic correlations between BD, chro- medical illness or medication use. 26,28 notype, and insomnia. These analyses have used summary- Participants provided written informed consent. The study level data and therefore may have been limited by a lack of in- had ethical approval from the West Midlands Multi-Centre Re- dividual bipolar phenotypic and genotypic data. In particular, search Ethics Committee, in addition to local research and de- associations between BD subtypes (ie, type 1 [BD-I] and type velopment approval by UK National Health Service Trusts and 2 [BD-II]) and sleep traits have been neglected, despite evi- Health Boards. dence of heterogeneity between BD subtypes in genetics, ill- 32-36 ness course, clinical features, and etiologies. There is also Control Participants evidence that individuals with BD subtypes differ in sensitiv- Control participants aged 18 years or older were recruited via ity to sleep loss and rates of hypersomnia and insomnia dur- the UK Blood Service and the 1958 Birth Cohort. Characteris- ing depressive episodes. tics and recruitment of this sample has been described We therefore aimed to determine whether genetic liabil- previously. All control participants reported their race as ity for insomnia, hypersomnia, and chronotype differs in BD-I white. and BD-II. Given a lack of evidence on whether these sleep traits differ between individuals with BD-I or BD-II in the interepi- Measures sode period, we had no prior hypothesis about which sleep Individuals with BD were assessed using the Schedules for traits, if any, would be associated with BD subtypes. Clinical Assessment in Neuropsychiatry interview, admin- To test the associations between sleep and BD pheno- istered by trained research psychologists or psychiatrists in the types, we adopted the polygenic risk score (PRS) method to es- research team (A.D.F., L.F., K.G.-S., L.J., N.C., and I.J.). Infor- timate the burden of risk alleles associated with 4 sleep- mation from this interview was combined with clinical case associated phenotypes (insomnia, sleep duration, excessive note data to make lifetime best-estimate DSM-IV diagnoses. daytime sleepiness, and chronotype) in people with BD-I or Measures taken to increase reliability of distinguishing BD sub- 38,39 BD-II and control participants. In secondary analyses, we types are outlined in eAppendix 1 in the Supplement. Inter- conducted a 2-sample mendelian randomization (MR) study rater reliability for differentiating between a best-estimate life- to test whether the data were consistent with a causal asso- time DSM-IV diagnosis of BD-I and BD-II was found to be good ciation between sleep and BD phenotypes. (κ, 0.85). Discovery Data Sets for Sleep Traits The discovery data sets were GWAS summary statistics for Method 27 30 31 insomnia, sleep duration, daytime sleepiness, and Sample Recruitment chronotype conducted in participants recruited to the UK Biobank. Sleep phenotypes were assessed using touch- Individuals With BD Participants with BD were recruited within the United King- screen questions. To assess insomnia symptoms, participants dom by the Bipolar Disorder Research Network (bdrn.org). were asked, “Do you have trouble falling asleep at night, or do All participants reported their race as white, were genetically you wake up in the middle of the night?” with the possible re- unrelated, were 18 years or older, and had been recruited sponses “never/rarely,” “sometimes,” “usually,” and “prefer not systematically (eg, via community mental health teams) or to answer.” The insomnia GWAS was conducted in 236 163 par- nonsystematically (eg, via websites, radio advertisements, or ticipants who answered “usually” (affected individuals) or 304 JAMA Psychiatry March 2020 Volume 77, Number 3 (Reprinted) jamapsychiatry.com Downloaded From: https://jamanetwork.com/ by a Deepdyve User on 07/07/2021 Comparison of Genetic Liability for Sleep Traits in Individuals With Bipolar Disorder and Controls Original Investigation Research “never/rarely” (control participants). We chose to use this rela- ing the reference group from control participants to partici- tively extreme GWAS rather than a larger GWAS conducted by pants with BD-I or BD-II, and then by using logistic regressions the same authors (ie, comparing responses of never or rarely that corrected for age, sex, and 10 population principal compo- with sometimes or usually for the question on insomnia symp- nents. Results are reported at theP value thresholds that showed toms) because (1) we considered this to better approximate the most significant results with a false-discovery rate correc- meaningful insomnia; (2) it identified a larger number of ge- tion applied (using the Benjamini and Hochberg approach). nome-wide significant loci (28 vs 9 in the larger GWAS), sug- gesting that, despite the smaller sample size, a clearer distinc- MR Analyses tion in phenotype offered better power; and (3) the authors used In cases in which we observed significant associations between results of the extremes GWAS, not the larger GWAS, in valida- sleep phenotype PRS and BD subtypes, we conducted follow-up tion analyses. The sleep-duration GWAS was conducted in 2-sample MR studies to test whether sleep phenotypes (expo- 448 609 participants who were asked, “About how many hours sures) were potentially causally related to BD subtypes (the out- sleep do you get in every 24 hours? (Please include naps.)” Re- come). Mendelian randomization is a causal inference method sponses were in 1-hour increments and were analyzed as a con- that uses genetic variants as instrumental variables for an expo- tinuous variable. The daytime sleepiness GWAS was con- sure of interest. It relies on 3 assumptions: (1) genetic variants ducted in 452 071 participants and assessed using the question, must be strongly associated with the exposure, (2) genetic vari- “How likely are you to doze off or fall asleep during the day- ants should not be associated with confounders of the exposure- time when you don’t mean to? (eg, when working, reading, outcome relationship, and (3) genetic variants should only be as- or driving),” with responses “never/rarely,” “sometimes,” sociated with the outcome through the exposure in question. “often,” or “all the time” analyzed on a scale of 1 to 4 points. We used genome-wide significant single-nucleotide polymor- The chronotype GWAS consisted of 403 195 individuals phisms as genetic instruments for the sleep phenotypes. who answered the question “Do you consider yourself to Instrument-exposure effects were taken from the sleep-trait be….?” Those who answered “definitely a ‘morning’ person” GWAS summary statistics, and instrument-outcome effects were or “more a ‘morning’ than ‘evening’ person” were coded as taken from BD-I and BD-II GWAS summary statistics. Four MR affected individuals, and those who answered “more an methods were used to assess relationships between sleep phe- ‘evening’ than a ‘morning’ person” or “definitely an ‘evening’ notypes and BD subtypes: the inverse variance weighted, 50 51 49 person” coded as control participants. Hence positive effect weighted median, weighted mode, and MR Egger regres- sizes from this GWAS indicate a morning chronotype, whereas sion methods. To test for evidence of pleiotropy, we examined negative effect sizes indicate an evening chronotype. the intercept of MR Egger regressions and the Cochran Q and 52,53 Rücker Q statistics. Data pruning, harmonization, and analy- ses were conducted in R version 3.33 using the “TwoSampleMR” Polygenic Risk Scores Full details on genotyping, quality control, and imputation are package. in eAppendix 1 and the eFigure in the Supplement. We gener- ated polygenic risk scores (PRSs) using PLINK version 1.9 in Replication Sample PRSice. Imputed genotypes were clumped for linkage dis- We sought to replicate the study findings using Swedish equilibrium (window, 500 kb; r = 0.20), and single- individuals with BD (n = 4366) and control participants nucleotide polymorphisms most significantly associated with (n = 6091) recruited via the St Göran Bipolar project and sleep traits were retained. Clumping resulted in retaining the Swedish National Quality Register for Bipolar Affective 56,57 92 085, 92 096, and 91 950 single-nucleotide polymor- Disorder (BipoläR). Full details of the samples, genotyp- phisms for daytime sleepiness, sleep duration, and insom- ing, quality control, and imputation are in eAppendix 2 in nia, respectively. After clumping, PRSs for sleep traits were gen- the Supplement. erated using PRSice at P value thresholds (P ) P < 1.00, P ≤ .50, P ≤.20, P ≤.10, P ≤ .05, P ≤ .01, and P ≤ .001 and con- verted to z scores. This range of P value thresholds was cho- Results sen in the absence of an independent sample that indicated which PRS P value threshold explained the most variance in Sample Description each of the respective sleep phenotypes. Among the individuals with BD, 3132 were female (67.0%), with a median age of 46 (range, 18-89) years. A total of 3404 par- Statistical Analysis ticipants met criteria for BD-I, and 1268 met criteria for BD-II. Data analysis was conducted in R version 3.33 (R Foundation Among control participants, 2812 of 5714 (49.2%) were fe- for Statistical Computing). We performed multinomial logistic male. The Swedish sample consisted of 6091 control partici- regression analyses examining associations between PRS for the pants (3767 female participants [61.8%]) and 4366 affected in- aforementioned sleep traits (at the range of P value thresholds dividuals (2697 female participants [61.8%]), of whom 2627 described) and individuals with BD subtypes (BD-I or BD-II) vs met criteria for BD-I and 1739 met criteria for BD-II. control participants. All analyses were adjusted for sex and 10 population principal components. In sensitivity analyses, we Correlations Between PRSs for Sleep Traits performed direct comparisons between the BD subtype groups Across all PRS P value thresholds, insomnia PRSs were nega- by first performing the same multinomial regressions but chang- tively associated with sleep-duration PRSs (r range, −0.17 to jamapsychiatry.com (Reprinted) JAMA Psychiatry March 2020 Volume 77, Number 3 305 Downloaded From: https://jamanetwork.com/ by a Deepdyve User on 07/07/2021 Research Original Investigation Comparison of Genetic Liability for Sleep Traits in Individuals With Bipolar Disorder and Controls Figure. Relative Risk Ratios for Individuals With Bipolar Subtypes vs Control Participants A Insomnia B Sleep duration 1.3 1.3 1.2 1.2 1.1 1.1 1.0 1.0 0.9 0.9 0.8 0.8 Individuals Individuals Individuals Individuals With Bipolar I With Bipolar II With Bipolar I With Bipolar II C Daytime sleepiness D Morningness 1.3 1.3 Polygenic risk score P value threshold .001 .01 .05 .10 .20 .50 1.00 1.2 1.2 1.1 1.1 1.0 1.0 0.9 0.9 0.8 0.8 Individuals Individuals Individuals Individuals With Bipolar I With Bipolar II With Bipolar I With Bipolar II Relative risk of insomnia (A), sleep duration (B), daytime sleepiness (C), and morningness (D) for patients with bipolar subtypes compared with control participants, as anticipated based on polygenic risk scores. Error bars indicate 95% CIs. −4 −0.30; P <1×10 ) and positively associated with daytime- PRS for Sleep Traits by BD Subtypes −4 sleepiness PRSs (r range, 0.04-0.10; P = .007 to P <1×10 ). Results at the most significant PRS P value thresholds (with Sleep-duration PRSs were negatively associated with daytime- corrected P values) are summarized. Results at other PRS sleepiness PRSs (r range, −0.03 to −0.00), but these associa- P value thresholds (P ) are provided in eAppendix 1 and tions were not significant across all thresholds (range, P = .028- eTables 18 and 19 in the Supplement. .916). Morningness PRSs were not significantly associated with PRS for insomnia, sleep duration, or daytime sleepiness. Insomnia PRS Multinomial regressions comparing individuals with BD PRSs for Sleep Traits: Case-Control Analyses subtypes to control participants revealed that insomnia PRS Logistic regression comparing individuals with BD with con- was significantly associated with a decreased risk of BD-I at trol participants revealed that, across all PRS P value thresh- aPRS P of P < 1.00 and P ≤ .50, but significant associations olds, case status was significantly associated with PRSs for sleep were not seen at other P value thresholds (eTable 6 in the duration (odds ratio [OR], 1.07 [95% CI, 1.03-1.12]; Supplement). At all P value thresholds, insomnia PRS was −4 −4 P =5.52×10 ; P =5.52×10 with adjustment for false- significantly associated with BD-II (relative risk [RR], 1.14 −5 discovery rate [PRS P < 1.00]) and daytime sleepiness (OR, 1.10 [95% CI, 1.07-1.21]; P =8.26×10 , P = .001 with false- −6 −5 [95% CI, 1.06-1.15]; P =2.31×10 ; P =1.05×10 with adjust- discovery rate adjustment [PRS P ≤ .001]). Results at all P T T ment for false-discovery rate [PRS P ≤ .01]) and negatively are shown in the Figure, A. In direct tests, insomnia PRS was associated with morning chronotype (OR, 0.91 [95% CI, 0.88- significantly associated with BD-II compared with BD-I (RR, −5 −5 −5 −4 0.95]; P =1.86×10 ; P =6.26×10 with adjustment for false- 1.16 [95% CI, 1.08-1.24]; P =1.39×10 ; P =1.95×10 with discovery rate [PRS P ≤ .05]) but not significantly associated false-discovery rate adjustment; OR, 1.14 [95% CI, 1.07- −5 with insomnia (OR, 0.98 [95% CI, 0.94-1.02]; P = .39 [PRS 1.22]; P =6.81×10 [PRS P ≤ .001]; eTables 10-11 in the P < 1.00]; eTables 2-5 in the Supplement). Supplement). 306 JAMA Psychiatry March 2020 Volume 77, Number 3 (Reprinted) jamapsychiatry.com Downloaded From: https://jamanetwork.com/ by a Deepdyve User on 07/07/2021 Relative Risk Relative Risk Relative Risk Relative Risk Comparison of Genetic Liability for Sleep Traits in Individuals With Bipolar Disorder and Controls Original Investigation Research Table. Results of 2-Sample Mendelian Randomization Studies a b Insomnia in Bipolar Disorder II Sleep Duration in Bipolar Disorder I Mendelian Randomization log(Odds Ratio) or Q log(Odds Ratio) or Q Method Statistic SE or df P Value Statistic SE or df P Value Inverse variance weighted 0.256 0.149 .087 0.396 0.209 .059 Weighted median 0.355 0.183 .052 0.245 0.219 .264 Weighted mode 0.410 0.300 .180 0.005 0.361 .990 MR Egger 0.577 0.498 .255 -0.274 0.757 .719 d −8 Rücker Q 53.33 35 .024 126.92 54 8.33 × 10 d −8 Cochran Q 54.03 36 .027 128.91 55 7.18 × 10 a c 37 Single-nucleotide polymorphisms. Log(odds ratio) and SEs are presented. b d 56 Single-nucleotide polymorphisms. Q statistics and df are presented. ent from zero for analyses of BD-I and BD-II , the Cochran Q Sleep-Duration PRS At all PRS P value thresholds, multinomial regression compar- and Rücker Q statistics indicated significant heterogeneity in ing individuals with BD subtypes with control participants re- effect estimates (insomnia: Rücker Q = 53.33; P = .024; Coch- vealed that sleep-duration PRS was associated with BD-I (RR, ran Q = 54.03; P = .027; sleep duration: Rücker Q = 126.92; 36 54 −5 −4 −8 −8 1.10 [95% CI, 1.06-1.15]; P =1.13×10 ; P =1.07×10 with P =8.33×10 ; Cochran Q = 128.91; P =7.18×10 ; Table), false-discovery rate adjustment [PRS P < 1.00];eTable7inthe possibly because of horizontal pleiotropy. Supplement). Associations between sleep-duration PRS and BD-II were not significant at any PRS P value threshold (eTable 7 Replication Sample in the Supplement). Results at all P value thresholds are shown In the Swedish sample, insomnia PRS was significantly asso- in the Figure, B. Direct comparisons between the subgroups ciated with BD-II (RR, 1.07 [95% CI, 1.01-1.13]; P = .013 [PRS with BD-I and BD-II revealed that sleep-duration PRS was sig- P ≤ .001]) compared with control participants, whereas the nificantly associated with BD-I (RR, 1.11 [95% CI, 1.04-1.19]; association with BD-I was not significant. Sleep-duration PRS −3 −3 P =1.69×10 ; P =4.74×10 with false-discovery rate ad- was associated with a significant increased relative risk of BD-I justment; OR, 1.11 [95% CI, 1.04-1.19]; P = .002 [PRS P <1.00]; compared with control participants (RR, 1.11 [95% CI, 1.06- −5 eTables 12-13 in the Supplement). 1.16]; P =1.72×10 [PRS P < 1.00]). The association be- tween sleep-duration PRS and BD-II was marginally signifi- Daytime-Sleepiness PRS cant (RR, 1.06 [95% CI, 1.00-1.12]; P = .042 [PRS P <1.00]). Compared with the control group, daytime-sleepiness PRS was associated with BD-I and BD-II at all PRS P value thresholds (except P < .001; eTable 8 in the Supplement). Results at all Discussion PRS P are shown in the Figure, C. Direct comparisons be- tween BD subtypes were not significant after correction for Bipolar disorder is heterogeneous in symptom presentation and multiple testing (eTable 14-15 in the Supplement). most likely in the mechanisms that underlie these presenta- tions. Genetics can help refine diagnostic groups that share Chronotype PRS similar etiologies. In this study, we provide what is to our Polygenic risk score for morningness was associated with a re- knowledge the first evidence that genetic liability to insom- duced relative risk of BD-I in affected individuals compared nia and longer sleep duration differs according to BD sub- with the control participants (RR, 0.90 [95% CI, 0.86-0.95]; type. −5 −4 P =1.06×10 ; P =1.11×10 with false-discovery rate ad- Genetic liability to insomnia as indexed by PRS was asso- justment [PRS P ≤ .50] at all PRS P value thresholds except ciated with increased relative risk of BD-II compared with P less than .001. In individuals with BD compared with con- control participants and those with BD-I. The stronger asso- trol participants, morningness PRS was associated with a re- ciation between insomnia PRS and BD-II may explain nonsig- duced risk of BD-II, but this finding was not significant across nificant genetic correlations between BD and insomnia in 26,28 most PRS P value thresholds (eTable 9 in the Supplement). Re- previous research, because individuals with BD-II are usu- sults at all PRSs are shown in the Figure. Direct comparisons ally underrepresented in BD GWAS (eg, only 11% in a recent between BD subtypes were not significant (eTable 16-17 in the study ). Future research should explore possible reasons for Supplement). this association. Hypersomnia in BD populations has remained relatively underresearched, but researchers have recently called for in- MR Analyses Across all MR methods, we did not find evidence of a poten- creased efforts to understand its underlying biology and role 16,60 tial causal relationship between insomnia and BD-II or sleep in BD. This is because of its high prevalence and recur- 7,37 duration and BD-I. However, the direction of effect was con- rence across bipolar depressive episodes in addition to high 17,61 sistent when assessing the effect of insomnia with BD-II (Table). interepisode prevalence and association with relapse. We Although MR Egger intercepts were not significantly differ- used sleep-duration and daytime-sleepiness PRS as proxies for jamapsychiatry.com (Reprinted) JAMA Psychiatry March 2020 Volume 77, Number 3 307 Downloaded From: https://jamanetwork.com/ by a Deepdyve User on 07/07/2021 Research Original Investigation Comparison of Genetic Liability for Sleep Traits in Individuals With Bipolar Disorder and Controls genetic liability to hypersomnia. Sleep-duration PRS was as- views and case notes with high interrater reliability. We were sociated with increased relative risk of BD-I but not BD-II and therefore able to explore genetic associations using individual- was significantly more strongly associated with BD-I than BD-II level genetic data, which provided more granularity than sum- in a direct comparison. In contrast, daytime-sleepiness PRS was mary statistics. In addition, we were able to replicate the re- not significantly associated with BD subtypes. Daytime- sults for insomnia PRS and BD-II and sleep-duration PRS and sleepiness PRS may be a proxy for insomnia, in that daytime BD-I in an independent sample. sleepiness can be induced by insomnia, and we observed sig- nificant positive correlations between insomnia and daytime- Limitations sleepiness PRS. These results support existing research on the This study has several limitations. First, potential recruit- 16,60 importance of hypersomnia in individuals with BD (and ment bias in our BD sample may have reduced its representa- 37 66 BD-I in particular ) and provide further evidence that hyper- tiveness and influenced the results. Second, we were un- 63,64 somnia is not a unitary construct. able to adjust for additional variables (eg, age, education), The results of the MR analyses do not support potentially because these were unavailable for control participants. Third, causal relationships between insomnia and BD-II or sleep du- the index of hypersomnia is imprecise because the available ration and BD-I. However, we observed significant heteroge- GWAS summary statistics measured sleep duration as total neity in the genetic instruments, thereby violating the third hours slept ; previous work suggests that hypersomnia is 17,64,67 assumption of MR and potentially biasing the results. There- better characterized by total time in bed. Fourth, there fore, while insomnia and sleep duration could be useful clini- is evidence that 5% to 17% of patients with BD-II convert to 68,69 cal stratifiers, there is currently insufficient evidence to sup- BD-I, which could have resulted in some individuals port a causal inference. Further research is needed to elucidate with BD-II being misclassified in this sample. However, this the biological mechanisms underpinning the genetic associa- would have reduced power to observe differences between the tion between BD-I and longer sleep duration. 2 subtypes rather than resulted in positive results we observe for insomnia and sleep duration. Finally, variants associated Implications with insomnia or hypersomnia at ages 40 to 69 years (the age Clinical and biological heterogeneity, combined with a classi- of the UK Biobank sample ) may differ from those associ- fication that is not grounded in biology, are obstacles to ad- ated in childhood or early adulthood. This may have in- vancing BD research. We provide some evidence of heteroge- creased our type-2 error rate, because most patients with BD neity in genetic propensity to some sleep traits within BD experience the first onset of impairing symptoms in adoles- (specifically insomnia and sleep duration), highlighting dif- cence or early adulthood. Genetic risk for insomnia or hy- ferences in the way some sleep-associated genetic factors are persomnia that manifests during or prior to early adulthood associated with BD subtypes. This adds to previously pub- may be more strongly associated with BD than those associ- lished work on stratification in BD and work suggesting that ated with midlife insomnia. These results should be repli- 32-34 different factors may influence the 2 conditions. cated using future sleep trait GWAS in younger samples of suf- These results suggest that clinical trials of sleep interven- ficient size for PRS analysis. tions should stratify participants by clinical subtype and ge- netic liability to insomnia or hypersomnia. Future work should explore which factors drive the differences in genetic liability Conclusions for insomnia/sleep duration between BD subtypes. To our knowledge, this is the first study to explore whether genetic liability for sleep traits is associated with clinical strata Strengths This study was conducted on the world’s largest single co- of individuals with BD. Future work should explore potential hort of BD with genotypic and phenotypic data. Phenotypic mechanisms underlying differences between the BD sub- data were collected using face-to-face semistructured inter- types in genetic liability for sleep traits. ARTICLE INFORMATION Biostatistics, Karolinska Institutet, Stockholm, University, Cardiff, United Kingdom (Owen, Sweden (Lewis, Karlsson, Landén); Genetics of Craddock, I. Jones, O’Donovan, Di Florio); Institute Accepted for Publication: September 29, 2019. Complex Traits, University of Exeter Medical of Neuroscience and Physiology, Sahlgenska Published Online: November 21, 2019. School, Exeter, United Kingdom (S. E. Jones, Academy at the Gothenburg University, doi:10.1001/jamapsychiatry.2019.4079 Weedon); Centre for Academic Mental Health, Gothenburg, Sweden (Landén). Open Access: This is an open access article Population Health Sciences, Bristol Medical School, Author Contributions: Dr Di Florio had full access distributed under the terms of the CC-BY License. University of Bristol, Bristol, United Kingdom to all the data in the study and takes responsibility © 2019 Lewis KJS et al. JAMA Psychiatry. (H. J. Jones); Medical Research Council Integrative for the integrity of the data and the accuracy of the Epidemiology Unit, University of Bristol, Bristol, Author Affiliations: Medical Research Council data analysis. United Kingdom (H. J. Jones); National Institute of Centre for Neuropsychiatric Genetics and Concept and design: Lewis, H. Jones, Health Research Biomedical Research Centre at Genomics, Institute of Psychological Medicine and Gordon-Smith, L. Jones, Craddock, I. Jones, University Hospitals Bristol NHS Foundation Trust Clinical Neurosciences, School of Medicine, Cardiff O'Donovan, Di Florio. and the University of Bristol, Bristol, United University, Cardiff, United Kingdom (Lewis, Acquisition, analysis, or interpretation of data: Lewis, Kingdom (H. J. Jones); Department of Richards, Leonenko, Forty, Escott-Price, Owen, Richards,Karlsson,Leonenko,S.Jones,Gordon-Smith, Psychological Medicine, University of Worcester, Craddock, I. Jones, O’Donovan, Di Florio); Forty,Escott-Price,Owen,Weedon,L.Jones,Craddock, Worcester, United Kingdom (Gordon-Smith, Department of Medical Epidemiology and I. Jones, Landén, O'Donovan, Di Florio. L. Jones); National Centre for Mental Health, Cardiff 308 JAMA Psychiatry March 2020 Volume 77, Number 3 (Reprinted) jamapsychiatry.com Downloaded From: https://jamanetwork.com/ by a Deepdyve User on 07/07/2021 Comparison of Genetic Liability for Sleep Traits in Individuals With Bipolar Disorder and Controls Original Investigation Research Drafting of the manuscript: Lewis, Richards, Padiñas, PhD, Cardiff University, for helpful a systematic review and meta-analysis. Sleep Med H. Jones, Gordon-Smith, L. Jones, O'Donovan, comments on the manuscript. He was not Rev. 2015;20:46-58. doi:10.1016/j.smrv.2014.06.006 Di Florio. compensated for his contribution. 16. Kaplan KA, Williams R. Hypersomnia: an Critical revision of the manuscript for important overlooked, but not overestimated, sleep intellectual content: Lewis, Karlsson, Leonenko, REFERENCES disturbance in bipolar disorder. Evid Based Ment S. Jones, H. Jones, Gordon-Smith, Forty, 1. Wehr TA, Sack DA, Rosenthal NE. Sleep Health. 2017;20(2):59. doi:10.1136/eb-2016-102433 Escott-Price, Owen, Weedon, L. Jones, Craddock, reduction as a final common pathway in the genesis 17. Kaplan KA, Gruber J, Eidelman P, Talbot LS, I. Jones, Landén, O'Donovan, Di Florio. of mania. Am J Psychiatry. 1987;144(2):201-204. Harvey AG. Hypersomnia in inter-episode bipolar Statistical analysis: Lewis, Richards, Karlsson, 2. Wehr TA. Sleep loss: a preventable cause of disorder: does it have prognostic significance? Leonenko, S. Jones, H. Jones, Escott-Price, mania and other excited states. J Clin Psychiatry. J Affect Disord. 2011;132(3):438-444. doi:10.1016/j. O'Donovan. 1989;50(12)(suppl):8-16. jad.2011.03.013 Obtained funding: Owen, L. Jones, I. Jones, Landén, Di Florio. 3. Lewis KS, Gordon-Smith K, Forty L, et al. Sleep 18. Ritter PS, Marx C, Lewtschenko N, et al. Administrative, technical, or material support: loss as a trigger of mood episodes in bipolar The characteristics of sleep in patients with Richards, S. Jones, Gordon-Smith, Forty, Landén. disorder: individual differences based on diagnostic manifest bipolar disorder, subjects at high risk of Supervision: Gordon-Smith, Forty, Owen, Weedon, subtype and gender. Br J Psychiatry. 2017;211(3): developing the disease and healthy controls. L. Jones, Craddock, I. Jones, O'Donovan, Di Florio. 169-174. doi:10.1192/bjp.bp.117.202259 J Neural Transm (Vienna). 2012;119(10):1173-1184. doi:10.1007/s00702-012-0883-y Conflict of Interest Disclosures: Dr Di Florio 4. Leibenluft E, Albert PS, Rosenthal NE, Wehr TA. reported receiving a Wellcome Trust grant outside Relationship between sleep and mood in patients 19. Kim KL, Weissman AB, Puzia ME, et al. the submitted work. Dr Owen reported grants from with rapid-cycling bipolar disorder. Psychiatry Res. Circadian phase preference in pediatric bipolar Takeda Pharmaceuticals outside the submitted 1996;63(2-3):161-168. doi:10.1016/0165-1781(96) disorder. J Clin Med. 2014;3(1):255-266. doi:10. work. Dr Landén reports grants from Stanley Center 02854-5 3390/jcm3010255 for Psychiatric Research, Broad Institute (via a grant 5. Perlman CA, Johnson SL, Mellman TA. 20. Melo MCA, Abreu RLC, Linhares Neto VB, from Stanley Medical Research Institute), and The prospective impact of sleep duration on de Bruin PFC, de Bruin VMS. Chronotype and grants from the National Institute for Mental Health depression and mania. Bipolar Disord. 2006;8(3): circadian rhythm in bipolar disorder: a systematic (grant MH077139) during the conduct of the study. 271-274. doi:10.1111/j.1399-5618.2006.00330.x review. Sleep Med Rev. 2017;34:46-58. doi:10.1016/ Dr O'Donovan reported grants from the Medical j.smrv.2016.06.007 6. American Psychiatric Association. Diagnostic Research Council of the United Kingdom and grants and Statistical Manual of Mental Disorders. 5th ed. 21. Takaesu Y. Circadian rhythm in bipolar disorder: from the National Institute of Mental Health during Arlington, VA: American Psychiatric Publishing; 2013. a review of the literature. Psychiatry Clin Neurosci. the conduct of the study and grants from Takeda 2018;72(9):673-682. doi:10.1111/pcn.12688 outside the submitted work. No other disclosures 7. Kaplan KA, Harvey AG. Hypersomnia across were reported. mood disorders: a review and synthesis. Sleep Med 22. Sheaves B, Freeman D, Isham L, et al. Rev. 2009;13(4):275-285. doi:10.1016/j.smrv.2008. Stabilising sleep for patients admitted at acute crisis Funding/Support: This study was funded by a 09.001 to a psychiatric hospital (OWLS): an assessor-blind National Alliance for Research on Schizophrenia & pilot randomised controlled trial. Psychol Med. Depression Young Investigator Grant awarded 8. Forty L, Smith D, Jones L, et al. Clinical 2018;48(10):1694-1704. doi:10.1017/ by the Brain & Behavior Research Foundation differences between bipolar and unipolar S0033291717003191 (Dr Di Florio). The Bipolar Disorder Research depression. Br J Psychiatry. 2008;192(5):388-389. Network was funded by the Wellcome Trust (grant doi:10.1192/bjp.bp.107.045294 23. Kaplan KA, Harvey AG. Behavioral treatment of 078901) and Stanley Medical Research Institute. insomnia in bipolar disorder. Am J Psychiatry. 2013; 9. Harvey AG, Talbot LS, Gershon A. Sleep Additional support was provided under Medical 170(7):716-720. doi:10.1176/appi.ajp.2013.12050708 disturbance in bipolar disorder across the lifespan. Research Council Centre (grant G0800509) and Clin Psychol (New York). 2009;16(2):256-277. 24. Kaplan KA, Talavera DC, Harvey AG. Rise and Program Grants (grant G0801418). This study doi:10.1111/j.1468-2850.2009.01164.x shine: a treatment experiment testing a morning makes use of genome-wide association data routine to decrease subjective sleep inertia in 10. Kanady JC, Soehnera AM, Harvey AG. generated by the Wellcome Trust Case-Control insomnia and bipolar disorder. Behav Res Ther. A retrospective examination of sleep disturbance Consortium 2. The National Centre for Mental 2018;111(111):106-112. doi:10.1016/j.brat.2018.10.009 across the course of bipolar disorder. J Sleep Disord Health is a collaboration between Cardiff, Swansea, Ther. 2015;4(2):318-328. 25. Frank E, Kupfer DJ, Thase ME, et al. Two-year and Bangor Universities and is funded by the Welsh outcomes for interpersonal and social rhythm government through Health and Care Research 11. Harvey AG, Schmidt DA, Scarnà A, Semler CN, therapy in individuals with bipolar I disorder. Arch Wales. Additional support was received from the Goodwin GM. Sleep-related functioning in Gen Psychiatry. 2005;62(9):996-1004. doi:10. Medical Research Council (grant MR/M005070/1 euthymic patients with bipolar disorder, patients 1001/archpsyc.62.9.996 [Drs S.E. Jones and Weedon]). Funding support for with insomnia, and subjects without sleep the Swedish Bipolar Collection was provided by the problems. Am J Psychiatry. 2005;162(1):50-57. 26. Lane JM, Liang J, Vlasac I, et al. Genome-wide Stanley Center for Psychiatric Research, Broad doi:10.1176/appi.ajp.162.1.50 association analyses of sleep disturbance traits Institute from a grant from Stanley Medical identify new loci and highlight shared genetics with 12. Geoffroy PA, Scott J, Boudebesse C, et al. Sleep Research Institute, the Swedish Research Council neuropsychiatric and metabolic traits. Nat Genet. in patients with remitted bipolar disorders: (grant 2018-02653), and the Swedish Federal 2017;49(2):274-281. doi:10.1038/ng.3749 a meta-analysis of actigraphy studies. Acta Government under the ALF agreement (grant ALF Psychiatr Scand. 2015;131(2):89-99. doi:10.1111/acps. 27. Lane JM, Jones SE, Dashti HS, et al; HUNT All In 20170019). 12367 Sleep. Biological and clinical insights from genetics Role of the Funder/Sponsor: The funding sources of insomnia symptoms. Nat Genet. 2019;51(3): 13. Plante DT, Winkelman JW. Sleep disturbance in had no role in the design and conduct of the study; 387-393. doi:10.1038/s41588-019-0361-7 bipolar disorder: therapeutic implications. Am J collection, management, analysis, and Psychiatry. 2008;165(7):830-843. doi:10.1176/appi. 28. Jansen PR, Watanabe K, Stringer S, et al; interpretation of the data; preparation, review, or ajp.2008.08010077 23andMe Research Team. Genome-wide analysis of approval of the manuscript; and decision to submit insomnia in 1,331,010 individuals identifies new risk 14. Harvey AG, Soehner AM, Kaplan KA, et al. the manuscript for publication. loci and functional pathways. Nat Genet. 2019;51 Treating insomnia improves mood state, sleep, and Meeting Presentation: This paper was presented (3):394-403. doi:10.1038/s41588-018-0333-3 functioning in bipolar disorder: a pilot randomized at the British Sleep Society Scientific Meeting; controlled trial. J Consult Clin Psychol. 2015;83(3): 29. Jones SE, Tyrrell J, Wood AR, et al. November 21, 2019; Birmingham, United Kingdom. 564-577. doi:10.1037/a0038655 Genome-wide association analyses in 128,266 Additional Contributions: We are grateful to all individuals identifies new morningness and sleep 15. Ng TH, Chung KF, Ho FYY, Yeung WF, Yung KP, participants in the Bipolar Disorder Research duration loci. PLoS Genet. 2016;12(8):e1006125. Lam TH. Sleep-wake disturbance in interepisode Network and the Swedish Bipolar Collection, as well doi:10.1371/journal.pgen.1006125 bipolar disorder and high-risk individuals: as all control participants who gave their time to this research. We are also grateful to Antonio jamapsychiatry.com (Reprinted) JAMA Psychiatry March 2020 Volume 77, Number 3 309 Downloaded From: https://jamanetwork.com/ by a Deepdyve User on 07/07/2021 Research Original Investigation Comparison of Genetic Liability for Sleep Traits in Individuals With Bipolar Disorder and Controls 30. Dashti HS, Jones SE, Wood AR, et al. GWAS in middle and old age. PLoS Med. 2015;12(3):e1001779. risk factors in a cohort of 6086 patients. PLoS One. 446,118 European adults identifies 78 genetic loci doi:10.1371/journal.pmed.1001779 2014;9(4):e94097. doi:10.1371/journal.pone. for self-reported 2 habitual sleep duration 0094097 44. Chang CC, Chow CC, Tellier LCAM, Vattikuti S, supported by accelerometer-derived estimates. Purcell SM, Lee JJ. Second-generation PLINK: rising 58. Craddock N, O’Donovan MC, Owen MJ. bioRxiv. https://www.biorxiv.org/content/biorxiv/ to the challenge of larger and richer datasets. Psychosis genetics: modeling the relationship early/2018/04/19/274977.full.pdf. Published 2018. Gigascience. 2015;4(1):7. doi:10.1186/s13742-015- between schizophrenia, bipolar disorder, and mixed doi:10.1101/274977 0047-8 (or “schizoaffective”) psychoses. Schizophr Bull. 31. Wang H, Lane JM, Jones SE, et al. Genome-wide 2009;35(3):482-490. doi:10.1093/schbul/sbp020 45. Euesden J, Lewis CM, O’Reilly PF. PRSice: association analysis of excessive daytime sleepiness polygenic risk score software. Bioinformatics. 2015; 59. Sklar P, Ripke S, Scott L, et al. Large-scale identifies 42 loci that suggest phenotypic 31(9):1466-1468. doi:10.1093/bioinformatics/btu848 genome-wide association analysis of bipolar subgroups. bioRxiv. https://www.biorxiv.org/ disorder identifies a new susceptibility locus near 46. Benjamini Y, Hochberg Y. Controlling the false content/10.1101/454561v1.full. Published 2018. ODZ4. 2011;43(10):977-983. doi:10.1038/ng.943 discovery rate: a practical and powerful approach to Accessed 8, 2019. doi:10.1101/454561 multiple testing. J R Stat Soc B. 1995;57(1):289-300. 60. Plante DT. Hypersomnia in mood disorders: 32. Baek JH, Park DY, Choi J, et al. Differences doi:10.1111/j.2517-6161.1995.tb02031.x a rapidly changing landscape. Curr Sleep Med Rep. between bipolar I and bipolar II disorders in clinical 2015;1(2):122-130. doi:10.1007/s40675-015-0017-9 47. Zheng J, Baird D, Borges M-C, et al. Recent features, comorbidity, and family history. J Affect developments in mendelian randomization studies. 61. Kaplan KA, McGlinchey EL, Soehner A, et al. Disord. 2011;131(1-3):59-67. doi:10.1016/j.jad.2010. Curr Epidemiol Rep. 2017;4(4):330-345. doi:10. Hypersomnia subtypes, sleep and relapse in bipolar 11.020 1007/s40471-017-0128-6 disorder. Psychol Med. 2015;45(8):1751-1763. 33. Caseras X, Murphy K, Lawrence NS, et al. doi:10.1017/S0033291714002918 48. Stahl E, Forstner A, McQuillin A, et al. Emotion regulation deficits in euthymic bipolar I Genome-wide association study identifies 30 loci 62. American Academy of Sleep Medicine. versus bipolar II disorder: a functional and associated with bipolar disorder. bioRxiv. https:// The International Classification of Sleep Disorders: diffusion-tensor imaging study. Bipolar Disord. www.biorxiv.org/content/10.1101/173062v. Published Diagnostic and Coding Manual. 2nd ed. Rochester, 2015;17(5):461-470. doi:10.1111/bdi.12292 2017. Accessed November 8, 2019. doi:10.1101/ MN: American Academy of Sleep Medicine; 2005. 34. Caseras X, Lawrence NS, Murphy K, Wise RG, 63. Ohayon MM, Dauvilliers Y, Reynolds CF III. Phillips ML. Ventral striatum activity in response to 49. Bowden J, Davey Smith G, Burgess S. Operational definitions and algorithms for reward: differences between bipolar I and II Mendelian randomization with invalid instruments: excessive sleepiness in the general population: disorders. Am J Psychiatry. 2013;170(5):533-541. effect estimation and bias detection through Egger implications for DSM-5 nosology. Arch Gen Psychiatry. doi:10.1176/appi.ajp.2012.12020169 regression. Int J Epidemiol. 2015;44(2):512-525. 2012;69(1):71-79. doi:10.1001/archgenpsychiatry. 35. Jones L, Metcalf A, Gordon-Smith K, et al. doi:10.1093/ije/dyv080 2011.1240 Gambling problems in bipolar disorder in the UK: 50. Bowden J, Davey Smith G, Haycock PC, 64. Nofzinger EA, Thase ME, Reynolds CF III, et al. prevalence and distribution. Br J Psychiatry. 2015; Burgess S. Consistent estimation in mendelian Hypersomnia in bipolar depression: a comparison 207(4):328-333. doi:10.1192/bjp.bp.114.154286 randomization with some invalid instruments using with narcolepsy using the multiple sleep latency 36. Charney AW, Ruderfer DM, Stahl EA, et al. a weighted median estimator. Genet Epidemiol. test. Am J Psychiatry. 1991;148(9):1177-1181. doi:10. Evidence for genetic heterogeneity between clinical 2016;40(4):304-314. doi:10.1002/gepi.21965 1176/ajp.148.9.1177 subtypes of bipolar disorder. Transl Psychiatry. 51. Hartwig FP, Davey Smith G, Bowden J. Robust 65. Allardyce J, Leonenko G, Hamshere M, et al. 2017;7(1):e993. doi:10.1038/tp.2016.242 inference in summary data Mendelian Association between schizophrenia-related 37. Steinan MK, Scott J, Lagerberg TV, et al. Sleep randomization via the zero modal pleiotropy polygenic liability and the occurrence and level of problems in bipolar disorders: more than just assumption. Int J Epidemiol. 2017;46(6):1985-1998. mood-incongruent psychotic symptoms in bipolar insomnia. Acta Psychiatr Scand. 2016;133(5):368-377. doi:10.1093/ije/dyx102 disorder. JAMA Psychiatry. 2018;75(1):28-35. doi:10. doi:10.1111/acps.12523 1001/jamapsychiatry.2017.3485 52. Bowden J, Del Greco M F, Minelli C, 38. Wray NR, Lee SH, Mehta D, Vinkhuyzen AAE, Davey Smith G, Sheehan N, Thompson J. 66. Munafò MR, Tilling K, Taylor AE, Evans DM, Dudbridge F, Middeldorp CM. Research review: A framework for the investigation of pleiotropy in Davey Smith G. Collider scope: when selection bias Polygenic methods and their application to two-sample summary data Mendelian can substantially influence observed associations. psychiatric traits. J Child Psychol Psychiatry. 2014; randomization. Stat Med. 2017;36(11):1783-1802. Int J Epidemiol. 2018;47(1):226-235. doi:10.1093/ 55(10):1068-1087. doi:10.1111/jcpp.12295 doi:10.1002/sim.7221 ije/dyx206 39. Dudbridge F. Power and predictive accuracy of 53. Greco M FD, Minelli C, Sheehan NA, 67. Billiard M, Dolenc L, Aldaz C, Ondze B, polygenic risk scores. PLoS Genet. 2013;9(3): Thompson JR. Detecting pleiotropy in Mendelian Besset A. Hypersomnia associated with mood e1003348. doi:10.1371/journal.pgen.1003348 randomisation studies with summary data and a disorders: a new perspective. J Psychosom Res. 40. Wellcome Trust Case Control Consortium. continuous outcome. Stat Med. 2015;34(21): 1994;38(1)(suppl 1):41-47. doi:10.1016/0022-3999 Genome-wide association study of 14,000 cases of 2926-2940. doi:10.1002/sim.6522 (94)90134-1 seven common diseases and 3,000 shared 54. Hemani G, Zheng J, Elsworth B, et al. 68. Coryell W, Endicott J, Maser JD, Keller MB, controls. Nature. 2007;447(7145):661-678. doi:10. The MR-Base platform supports systematic causal Leon AC, Akiskal HS. Long-term stability of polarity 1038/nature05911 inference across the human phenome. Elife. 2018; distinctions in the affective disorders. Am J Psychiatry. 41. Wing JK, Babor T, Brugha T, et al; Schedules for pii:e34408. doi:10.7554/eLife.34408 1995;152(3):385-390. doi:10.1176/ajp.152.3.385 Clinical Assessment in Neuropsychiatry. SCAN: 55. Jakobsson J, Zetterberg H, Blennow K, 69. Alloy LB, Urošević S, Abramson LY, et al. schedules for clinical assessment in Johan Ekman C, Johansson AGM, Landén M. Progression along the bipolar spectrum: neuropsychiatry. Arch Gen Psychiatry. 1990;47(6): Altered concentrations of amyloid precursor a longitudinal study of predictors of conversion 589-593. doi:10.1001/archpsyc.1990. protein metabolites in the cerebrospinal fluid of from bipolar spectrum conditions to bipolar I and II patients with bipolar disorder. disorders. J Abnorm Psychol. 2012;121(1):16-27. 42. Jones SE, Lane JM, Wood AR, et al. Neuropsychopharmacology. 2013;38(4):664-672. doi:10.1037/a0023973 Genome-wide association analyses of chronotype doi:10.1038/npp.2012.231 70. Tondo L, Lepri B, Cruz N, Baldessarini RJ. Age in 697,828 individuals provides insights into 56. Karanti A, Bobeck C, Osterman M, et al. Gender at onset in 3014 Sardinian bipolar and major circadian rhythms. Nat Commun. 2019;10(1):343. differences in the treatment of patients with bipolar depressive disorder patients. Acta Psychiatr Scand. doi:10.1038/s41467-018-08259-7 disorder: a study of 7354 patients. J Affect Disord. 2010;121(6):446-452. doi:10.1111/j.1600-0447.2009. 43. Sudlow C, Gallacher J, Allen N, et al. UK 2015;174:303-309. doi:10.1016/j.jad.2014.11.058 01523.x biobank: an open access resource for identifying 57. Tidemalm D, Haglund A, Karanti A, Landén M, the causes of a wide range of complex diseases of Runeson B. Attempted suicide in bipolar disorder: 310 JAMA Psychiatry March 2020 Volume 77, Number 3 (Reprinted) jamapsychiatry.com Downloaded From: https://jamanetwork.com/ by a Deepdyve User on 07/07/2021 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png JAMA Psychiatry Unpaywall

Loading next page...
 
/lp/unpaywall/comparison-of-genetic-liability-for-sleep-traits-among-individuals-fYqDohQMft

References (86)

Publisher
Unpaywall
ISSN
2168-622X
DOI
10.1001/jamapsychiatry.2019.4079
Publisher site
See Article on Publisher Site

Abstract

Research JAMA Psychiatry | Original Investigation Comparison of Genetic Liability for Sleep Traits Among Individuals With Bipolar Disorder I or II and Control Participants Katie J. S. Lewis, PhD; Alexander Richards, PhD; Robert Karlsson, PhD; Ganna Leonenko, PhD; Samuel E. Jones, PhD; Hannah J. Jones, PhD; Katherine Gordon-Smith, PhD; Liz Forty, PhD; Valentina Escott-Price, PhD; Michael J. Owen, FRCPsych, PhD; Michael N. Weedon, PhD; Lisa Jones, PhD; Nick Craddock, FRCPsych, PhD; Ian Jones, MRCPsych, PhD; Mikael Landén, MD, PhD; Michael C. O’Donovan, FRCPsych, PhD; Arianna Di Florio, MD, PhD Editorial page 235 IMPORTANCE Insomnia, hypersomnia, and an evening chronotype are common in individuals Supplemental content with bipolar disorder (BD), but whether this reflects shared genetic liability is unclear. Stratifying by BD subtypes could elucidate this association and inform sleep and BD research. OBJECTIVE To assess whether polygenic risk scores (PRSs) for sleep traits are associated with BD subtypes I and II. DESIGN, SETTING, AND PARTICIPANTS This case-control study was conducted in the United Kingdom and Sweden with participants with BD and control participants. Multinomial regression was used to assess whether PRSs for insomnia, daytime sleepiness, sleep duration, and chronotype are associated with BD subtypes compared with control participants. Affected individuals were recruited from the Bipolar Disorder Research Network. Control participants were recruited from the 1958 British Birth Cohort and the UK Blood Service. Analyses were repeated in an independent Swedish sample from August 2018 to July 2019. All participants were of European ancestry. EXPOSURES Standardized PRSs derived using alleles from genome-wide association studies of insomnia, sleep duration, daytime sleepiness, and chronotype. These were adjusted for the first 10 population principal components, genotyping platforms, and sex. MAIN OUTCOMES AND MEASURES Association of PRSs with BD subtypes, determined by semistructured psychiatric interview and case notes. RESULTS The main analysis included 4672 participants with BD (3132 female participants [67.0%]; 3404 with BD-I [72.9%]) and 5714 control participants (2812 female participants [49.2%]). Insomnia PRS was associated with increased risk of BD-II (relative risk [RR], 1.14 −5 [95% CI, 1.07-1.21]; P =8.26×10 ) but not BD-I (RR, 0.98 [95% CI, 0.94-1.03]; P = .409) relative to control participants. Sleep-duration PRS was associated with BD-I (RR, 1.10 −5 [95% CI, 1.06-1.15]; P =1.13×10 ) but not BD-II (RR, 0.99 [95% CI, 0.93-1.06]; P = .818). Associations between (1) insomnia PRS and BD-II and (2) sleep-duration PRS and BD-I were replicated in the Swedish sample of 4366 individuals with BD (2697 female participants [61.8%]; 2627 with BD-I [60.2%]) and 6091 control participants (3767 female participants [61.8%]). Chronotype and daytime-sleepiness PRS were not associated with BD subtypes. CONCLUSIONS AND RELEVANCE Per this analysis, BD subtypes differ in genetic liability to insomnia and hypersomnia, providing further evidence that the distinction between BD-I and BD-II has genetic validity. This distinction will be crucial in selecting participants for future research on the role of sleep disturbance in BD. Author Affiliations: Author affiliations are listed at the end of this article. Corresponding Author: Arianna Di Florio, MD, PhD, Medical Research Council Centre for Neuropsychiatric Genetics and Genomics, Institute of Psychological Medicine and Clinical Neurosciences, Cardiff University, Hadyn Ellis Building, Maindy Road, JAMA Psychiatry. 2020;77(3):303-310. doi:10.1001/jamapsychiatry.2019.4079 Cathays, Cardiff, CF24 4HQ, United Published online November 21, 2019. Kingdom (diflorioa@cardiff.ac.uk). (Reprinted) 303 Downloaded From: https://jamanetwork.com/ by a Deepdyve User on 07/07/2021 Research Original Investigation Comparison of Genetic Liability for Sleep Traits in Individuals With Bipolar Disorder and Controls ipolar disorder (BD) and sleep have often been linked. First, reduced sleep duration, a symptom of manic epi- Key Points B sodes, has been implicated as a prodrome and trigger Question Does genetic liability to insomnia, hypersomnia, and 1-5 of mania. Second, insomnia (difficulty initiating and main- chronotype differentiate subtypes of bipolar disorder? taining sleep ) and hypersomnia (prolonged sleep duration or Findings In this case-control study of 4672 participants with excessive daytime sleepiness ) are commonly reported symp- bipolar disorder and 5714 control participants, individuals with 7-9 toms of bipolar depression that persist in the interepisode bipolar disorder I had significantly greater genetic liability to longer 9-18 period and are associated with significant distress and sleep duration, whereas individuals with bipolar disorder II had impairment. Third, there is evidence that individuals with significantly greater genetic liability to insomnia; these findings BD display greater evening preference for sleep (ie, an eve- were replicated in an independent sample. Individuals with bipolar 19-21 subtypes did not differ in genetic liability to morning or evening ning chronotype) than healthy control participants. chronotype. At present, sleep interventions for individuals with BD pri- 22-24 marily focus on reducing insomnia and stabilizing circa- Meaning Associations between polygenic liability to insomnia and dian rhythms. Understanding the association between sleep hypersomnia and clinical strata within bipolar disorder are shown in this study for the first time, to our knowledge. and BD is important and could inform clinical interventions. 26-31 Recent genome-wide association studies (GWAS) pro- vide an opportunity to examine the association between sleep and BD at the genomic level. Using summary-level data, some voluntary groups, such as Bipolar UK). Participants were studies have demonstrated a positive genetic correlation be- excluded if they had affective illness experienced solely in 28,30 tween BD and sleep duration. Other studies, however, have response to alcohol or substance misuse or secondary to found no significant genetic correlations between BD, chro- medical illness or medication use. 26,28 notype, and insomnia. These analyses have used summary- Participants provided written informed consent. The study level data and therefore may have been limited by a lack of in- had ethical approval from the West Midlands Multi-Centre Re- dividual bipolar phenotypic and genotypic data. In particular, search Ethics Committee, in addition to local research and de- associations between BD subtypes (ie, type 1 [BD-I] and type velopment approval by UK National Health Service Trusts and 2 [BD-II]) and sleep traits have been neglected, despite evi- Health Boards. dence of heterogeneity between BD subtypes in genetics, ill- 32-36 ness course, clinical features, and etiologies. There is also Control Participants evidence that individuals with BD subtypes differ in sensitiv- Control participants aged 18 years or older were recruited via ity to sleep loss and rates of hypersomnia and insomnia dur- the UK Blood Service and the 1958 Birth Cohort. Characteris- ing depressive episodes. tics and recruitment of this sample has been described We therefore aimed to determine whether genetic liabil- previously. All control participants reported their race as ity for insomnia, hypersomnia, and chronotype differs in BD-I white. and BD-II. Given a lack of evidence on whether these sleep traits differ between individuals with BD-I or BD-II in the interepi- Measures sode period, we had no prior hypothesis about which sleep Individuals with BD were assessed using the Schedules for traits, if any, would be associated with BD subtypes. Clinical Assessment in Neuropsychiatry interview, admin- To test the associations between sleep and BD pheno- istered by trained research psychologists or psychiatrists in the types, we adopted the polygenic risk score (PRS) method to es- research team (A.D.F., L.F., K.G.-S., L.J., N.C., and I.J.). Infor- timate the burden of risk alleles associated with 4 sleep- mation from this interview was combined with clinical case associated phenotypes (insomnia, sleep duration, excessive note data to make lifetime best-estimate DSM-IV diagnoses. daytime sleepiness, and chronotype) in people with BD-I or Measures taken to increase reliability of distinguishing BD sub- 38,39 BD-II and control participants. In secondary analyses, we types are outlined in eAppendix 1 in the Supplement. Inter- conducted a 2-sample mendelian randomization (MR) study rater reliability for differentiating between a best-estimate life- to test whether the data were consistent with a causal asso- time DSM-IV diagnosis of BD-I and BD-II was found to be good ciation between sleep and BD phenotypes. (κ, 0.85). Discovery Data Sets for Sleep Traits The discovery data sets were GWAS summary statistics for Method 27 30 31 insomnia, sleep duration, daytime sleepiness, and Sample Recruitment chronotype conducted in participants recruited to the UK Biobank. Sleep phenotypes were assessed using touch- Individuals With BD Participants with BD were recruited within the United King- screen questions. To assess insomnia symptoms, participants dom by the Bipolar Disorder Research Network (bdrn.org). were asked, “Do you have trouble falling asleep at night, or do All participants reported their race as white, were genetically you wake up in the middle of the night?” with the possible re- unrelated, were 18 years or older, and had been recruited sponses “never/rarely,” “sometimes,” “usually,” and “prefer not systematically (eg, via community mental health teams) or to answer.” The insomnia GWAS was conducted in 236 163 par- nonsystematically (eg, via websites, radio advertisements, or ticipants who answered “usually” (affected individuals) or 304 JAMA Psychiatry March 2020 Volume 77, Number 3 (Reprinted) jamapsychiatry.com Downloaded From: https://jamanetwork.com/ by a Deepdyve User on 07/07/2021 Comparison of Genetic Liability for Sleep Traits in Individuals With Bipolar Disorder and Controls Original Investigation Research “never/rarely” (control participants). We chose to use this rela- ing the reference group from control participants to partici- tively extreme GWAS rather than a larger GWAS conducted by pants with BD-I or BD-II, and then by using logistic regressions the same authors (ie, comparing responses of never or rarely that corrected for age, sex, and 10 population principal compo- with sometimes or usually for the question on insomnia symp- nents. Results are reported at theP value thresholds that showed toms) because (1) we considered this to better approximate the most significant results with a false-discovery rate correc- meaningful insomnia; (2) it identified a larger number of ge- tion applied (using the Benjamini and Hochberg approach). nome-wide significant loci (28 vs 9 in the larger GWAS), sug- gesting that, despite the smaller sample size, a clearer distinc- MR Analyses tion in phenotype offered better power; and (3) the authors used In cases in which we observed significant associations between results of the extremes GWAS, not the larger GWAS, in valida- sleep phenotype PRS and BD subtypes, we conducted follow-up tion analyses. The sleep-duration GWAS was conducted in 2-sample MR studies to test whether sleep phenotypes (expo- 448 609 participants who were asked, “About how many hours sures) were potentially causally related to BD subtypes (the out- sleep do you get in every 24 hours? (Please include naps.)” Re- come). Mendelian randomization is a causal inference method sponses were in 1-hour increments and were analyzed as a con- that uses genetic variants as instrumental variables for an expo- tinuous variable. The daytime sleepiness GWAS was con- sure of interest. It relies on 3 assumptions: (1) genetic variants ducted in 452 071 participants and assessed using the question, must be strongly associated with the exposure, (2) genetic vari- “How likely are you to doze off or fall asleep during the day- ants should not be associated with confounders of the exposure- time when you don’t mean to? (eg, when working, reading, outcome relationship, and (3) genetic variants should only be as- or driving),” with responses “never/rarely,” “sometimes,” sociated with the outcome through the exposure in question. “often,” or “all the time” analyzed on a scale of 1 to 4 points. We used genome-wide significant single-nucleotide polymor- The chronotype GWAS consisted of 403 195 individuals phisms as genetic instruments for the sleep phenotypes. who answered the question “Do you consider yourself to Instrument-exposure effects were taken from the sleep-trait be….?” Those who answered “definitely a ‘morning’ person” GWAS summary statistics, and instrument-outcome effects were or “more a ‘morning’ than ‘evening’ person” were coded as taken from BD-I and BD-II GWAS summary statistics. Four MR affected individuals, and those who answered “more an methods were used to assess relationships between sleep phe- ‘evening’ than a ‘morning’ person” or “definitely an ‘evening’ notypes and BD subtypes: the inverse variance weighted, 50 51 49 person” coded as control participants. Hence positive effect weighted median, weighted mode, and MR Egger regres- sizes from this GWAS indicate a morning chronotype, whereas sion methods. To test for evidence of pleiotropy, we examined negative effect sizes indicate an evening chronotype. the intercept of MR Egger regressions and the Cochran Q and 52,53 Rücker Q statistics. Data pruning, harmonization, and analy- ses were conducted in R version 3.33 using the “TwoSampleMR” Polygenic Risk Scores Full details on genotyping, quality control, and imputation are package. in eAppendix 1 and the eFigure in the Supplement. We gener- ated polygenic risk scores (PRSs) using PLINK version 1.9 in Replication Sample PRSice. Imputed genotypes were clumped for linkage dis- We sought to replicate the study findings using Swedish equilibrium (window, 500 kb; r = 0.20), and single- individuals with BD (n = 4366) and control participants nucleotide polymorphisms most significantly associated with (n = 6091) recruited via the St Göran Bipolar project and sleep traits were retained. Clumping resulted in retaining the Swedish National Quality Register for Bipolar Affective 56,57 92 085, 92 096, and 91 950 single-nucleotide polymor- Disorder (BipoläR). Full details of the samples, genotyp- phisms for daytime sleepiness, sleep duration, and insom- ing, quality control, and imputation are in eAppendix 2 in nia, respectively. After clumping, PRSs for sleep traits were gen- the Supplement. erated using PRSice at P value thresholds (P ) P < 1.00, P ≤ .50, P ≤.20, P ≤.10, P ≤ .05, P ≤ .01, and P ≤ .001 and con- verted to z scores. This range of P value thresholds was cho- Results sen in the absence of an independent sample that indicated which PRS P value threshold explained the most variance in Sample Description each of the respective sleep phenotypes. Among the individuals with BD, 3132 were female (67.0%), with a median age of 46 (range, 18-89) years. A total of 3404 par- Statistical Analysis ticipants met criteria for BD-I, and 1268 met criteria for BD-II. Data analysis was conducted in R version 3.33 (R Foundation Among control participants, 2812 of 5714 (49.2%) were fe- for Statistical Computing). We performed multinomial logistic male. The Swedish sample consisted of 6091 control partici- regression analyses examining associations between PRS for the pants (3767 female participants [61.8%]) and 4366 affected in- aforementioned sleep traits (at the range of P value thresholds dividuals (2697 female participants [61.8%]), of whom 2627 described) and individuals with BD subtypes (BD-I or BD-II) vs met criteria for BD-I and 1739 met criteria for BD-II. control participants. All analyses were adjusted for sex and 10 population principal components. In sensitivity analyses, we Correlations Between PRSs for Sleep Traits performed direct comparisons between the BD subtype groups Across all PRS P value thresholds, insomnia PRSs were nega- by first performing the same multinomial regressions but chang- tively associated with sleep-duration PRSs (r range, −0.17 to jamapsychiatry.com (Reprinted) JAMA Psychiatry March 2020 Volume 77, Number 3 305 Downloaded From: https://jamanetwork.com/ by a Deepdyve User on 07/07/2021 Research Original Investigation Comparison of Genetic Liability for Sleep Traits in Individuals With Bipolar Disorder and Controls Figure. Relative Risk Ratios for Individuals With Bipolar Subtypes vs Control Participants A Insomnia B Sleep duration 1.3 1.3 1.2 1.2 1.1 1.1 1.0 1.0 0.9 0.9 0.8 0.8 Individuals Individuals Individuals Individuals With Bipolar I With Bipolar II With Bipolar I With Bipolar II C Daytime sleepiness D Morningness 1.3 1.3 Polygenic risk score P value threshold .001 .01 .05 .10 .20 .50 1.00 1.2 1.2 1.1 1.1 1.0 1.0 0.9 0.9 0.8 0.8 Individuals Individuals Individuals Individuals With Bipolar I With Bipolar II With Bipolar I With Bipolar II Relative risk of insomnia (A), sleep duration (B), daytime sleepiness (C), and morningness (D) for patients with bipolar subtypes compared with control participants, as anticipated based on polygenic risk scores. Error bars indicate 95% CIs. −4 −0.30; P <1×10 ) and positively associated with daytime- PRS for Sleep Traits by BD Subtypes −4 sleepiness PRSs (r range, 0.04-0.10; P = .007 to P <1×10 ). Results at the most significant PRS P value thresholds (with Sleep-duration PRSs were negatively associated with daytime- corrected P values) are summarized. Results at other PRS sleepiness PRSs (r range, −0.03 to −0.00), but these associa- P value thresholds (P ) are provided in eAppendix 1 and tions were not significant across all thresholds (range, P = .028- eTables 18 and 19 in the Supplement. .916). Morningness PRSs were not significantly associated with PRS for insomnia, sleep duration, or daytime sleepiness. Insomnia PRS Multinomial regressions comparing individuals with BD PRSs for Sleep Traits: Case-Control Analyses subtypes to control participants revealed that insomnia PRS Logistic regression comparing individuals with BD with con- was significantly associated with a decreased risk of BD-I at trol participants revealed that, across all PRS P value thresh- aPRS P of P < 1.00 and P ≤ .50, but significant associations olds, case status was significantly associated with PRSs for sleep were not seen at other P value thresholds (eTable 6 in the duration (odds ratio [OR], 1.07 [95% CI, 1.03-1.12]; Supplement). At all P value thresholds, insomnia PRS was −4 −4 P =5.52×10 ; P =5.52×10 with adjustment for false- significantly associated with BD-II (relative risk [RR], 1.14 −5 discovery rate [PRS P < 1.00]) and daytime sleepiness (OR, 1.10 [95% CI, 1.07-1.21]; P =8.26×10 , P = .001 with false- −6 −5 [95% CI, 1.06-1.15]; P =2.31×10 ; P =1.05×10 with adjust- discovery rate adjustment [PRS P ≤ .001]). Results at all P T T ment for false-discovery rate [PRS P ≤ .01]) and negatively are shown in the Figure, A. In direct tests, insomnia PRS was associated with morning chronotype (OR, 0.91 [95% CI, 0.88- significantly associated with BD-II compared with BD-I (RR, −5 −5 −5 −4 0.95]; P =1.86×10 ; P =6.26×10 with adjustment for false- 1.16 [95% CI, 1.08-1.24]; P =1.39×10 ; P =1.95×10 with discovery rate [PRS P ≤ .05]) but not significantly associated false-discovery rate adjustment; OR, 1.14 [95% CI, 1.07- −5 with insomnia (OR, 0.98 [95% CI, 0.94-1.02]; P = .39 [PRS 1.22]; P =6.81×10 [PRS P ≤ .001]; eTables 10-11 in the P < 1.00]; eTables 2-5 in the Supplement). Supplement). 306 JAMA Psychiatry March 2020 Volume 77, Number 3 (Reprinted) jamapsychiatry.com Downloaded From: https://jamanetwork.com/ by a Deepdyve User on 07/07/2021 Relative Risk Relative Risk Relative Risk Relative Risk Comparison of Genetic Liability for Sleep Traits in Individuals With Bipolar Disorder and Controls Original Investigation Research Table. Results of 2-Sample Mendelian Randomization Studies a b Insomnia in Bipolar Disorder II Sleep Duration in Bipolar Disorder I Mendelian Randomization log(Odds Ratio) or Q log(Odds Ratio) or Q Method Statistic SE or df P Value Statistic SE or df P Value Inverse variance weighted 0.256 0.149 .087 0.396 0.209 .059 Weighted median 0.355 0.183 .052 0.245 0.219 .264 Weighted mode 0.410 0.300 .180 0.005 0.361 .990 MR Egger 0.577 0.498 .255 -0.274 0.757 .719 d −8 Rücker Q 53.33 35 .024 126.92 54 8.33 × 10 d −8 Cochran Q 54.03 36 .027 128.91 55 7.18 × 10 a c 37 Single-nucleotide polymorphisms. Log(odds ratio) and SEs are presented. b d 56 Single-nucleotide polymorphisms. Q statistics and df are presented. ent from zero for analyses of BD-I and BD-II , the Cochran Q Sleep-Duration PRS At all PRS P value thresholds, multinomial regression compar- and Rücker Q statistics indicated significant heterogeneity in ing individuals with BD subtypes with control participants re- effect estimates (insomnia: Rücker Q = 53.33; P = .024; Coch- vealed that sleep-duration PRS was associated with BD-I (RR, ran Q = 54.03; P = .027; sleep duration: Rücker Q = 126.92; 36 54 −5 −4 −8 −8 1.10 [95% CI, 1.06-1.15]; P =1.13×10 ; P =1.07×10 with P =8.33×10 ; Cochran Q = 128.91; P =7.18×10 ; Table), false-discovery rate adjustment [PRS P < 1.00];eTable7inthe possibly because of horizontal pleiotropy. Supplement). Associations between sleep-duration PRS and BD-II were not significant at any PRS P value threshold (eTable 7 Replication Sample in the Supplement). Results at all P value thresholds are shown In the Swedish sample, insomnia PRS was significantly asso- in the Figure, B. Direct comparisons between the subgroups ciated with BD-II (RR, 1.07 [95% CI, 1.01-1.13]; P = .013 [PRS with BD-I and BD-II revealed that sleep-duration PRS was sig- P ≤ .001]) compared with control participants, whereas the nificantly associated with BD-I (RR, 1.11 [95% CI, 1.04-1.19]; association with BD-I was not significant. Sleep-duration PRS −3 −3 P =1.69×10 ; P =4.74×10 with false-discovery rate ad- was associated with a significant increased relative risk of BD-I justment; OR, 1.11 [95% CI, 1.04-1.19]; P = .002 [PRS P <1.00]; compared with control participants (RR, 1.11 [95% CI, 1.06- −5 eTables 12-13 in the Supplement). 1.16]; P =1.72×10 [PRS P < 1.00]). The association be- tween sleep-duration PRS and BD-II was marginally signifi- Daytime-Sleepiness PRS cant (RR, 1.06 [95% CI, 1.00-1.12]; P = .042 [PRS P <1.00]). Compared with the control group, daytime-sleepiness PRS was associated with BD-I and BD-II at all PRS P value thresholds (except P < .001; eTable 8 in the Supplement). Results at all Discussion PRS P are shown in the Figure, C. Direct comparisons be- tween BD subtypes were not significant after correction for Bipolar disorder is heterogeneous in symptom presentation and multiple testing (eTable 14-15 in the Supplement). most likely in the mechanisms that underlie these presenta- tions. Genetics can help refine diagnostic groups that share Chronotype PRS similar etiologies. In this study, we provide what is to our Polygenic risk score for morningness was associated with a re- knowledge the first evidence that genetic liability to insom- duced relative risk of BD-I in affected individuals compared nia and longer sleep duration differs according to BD sub- with the control participants (RR, 0.90 [95% CI, 0.86-0.95]; type. −5 −4 P =1.06×10 ; P =1.11×10 with false-discovery rate ad- Genetic liability to insomnia as indexed by PRS was asso- justment [PRS P ≤ .50] at all PRS P value thresholds except ciated with increased relative risk of BD-II compared with P less than .001. In individuals with BD compared with con- control participants and those with BD-I. The stronger asso- trol participants, morningness PRS was associated with a re- ciation between insomnia PRS and BD-II may explain nonsig- duced risk of BD-II, but this finding was not significant across nificant genetic correlations between BD and insomnia in 26,28 most PRS P value thresholds (eTable 9 in the Supplement). Re- previous research, because individuals with BD-II are usu- sults at all PRSs are shown in the Figure. Direct comparisons ally underrepresented in BD GWAS (eg, only 11% in a recent between BD subtypes were not significant (eTable 16-17 in the study ). Future research should explore possible reasons for Supplement). this association. Hypersomnia in BD populations has remained relatively underresearched, but researchers have recently called for in- MR Analyses Across all MR methods, we did not find evidence of a poten- creased efforts to understand its underlying biology and role 16,60 tial causal relationship between insomnia and BD-II or sleep in BD. This is because of its high prevalence and recur- 7,37 duration and BD-I. However, the direction of effect was con- rence across bipolar depressive episodes in addition to high 17,61 sistent when assessing the effect of insomnia with BD-II (Table). interepisode prevalence and association with relapse. We Although MR Egger intercepts were not significantly differ- used sleep-duration and daytime-sleepiness PRS as proxies for jamapsychiatry.com (Reprinted) JAMA Psychiatry March 2020 Volume 77, Number 3 307 Downloaded From: https://jamanetwork.com/ by a Deepdyve User on 07/07/2021 Research Original Investigation Comparison of Genetic Liability for Sleep Traits in Individuals With Bipolar Disorder and Controls genetic liability to hypersomnia. Sleep-duration PRS was as- views and case notes with high interrater reliability. We were sociated with increased relative risk of BD-I but not BD-II and therefore able to explore genetic associations using individual- was significantly more strongly associated with BD-I than BD-II level genetic data, which provided more granularity than sum- in a direct comparison. In contrast, daytime-sleepiness PRS was mary statistics. In addition, we were able to replicate the re- not significantly associated with BD subtypes. Daytime- sults for insomnia PRS and BD-II and sleep-duration PRS and sleepiness PRS may be a proxy for insomnia, in that daytime BD-I in an independent sample. sleepiness can be induced by insomnia, and we observed sig- nificant positive correlations between insomnia and daytime- Limitations sleepiness PRS. These results support existing research on the This study has several limitations. First, potential recruit- 16,60 importance of hypersomnia in individuals with BD (and ment bias in our BD sample may have reduced its representa- 37 66 BD-I in particular ) and provide further evidence that hyper- tiveness and influenced the results. Second, we were un- 63,64 somnia is not a unitary construct. able to adjust for additional variables (eg, age, education), The results of the MR analyses do not support potentially because these were unavailable for control participants. Third, causal relationships between insomnia and BD-II or sleep du- the index of hypersomnia is imprecise because the available ration and BD-I. However, we observed significant heteroge- GWAS summary statistics measured sleep duration as total neity in the genetic instruments, thereby violating the third hours slept ; previous work suggests that hypersomnia is 17,64,67 assumption of MR and potentially biasing the results. There- better characterized by total time in bed. Fourth, there fore, while insomnia and sleep duration could be useful clini- is evidence that 5% to 17% of patients with BD-II convert to 68,69 cal stratifiers, there is currently insufficient evidence to sup- BD-I, which could have resulted in some individuals port a causal inference. Further research is needed to elucidate with BD-II being misclassified in this sample. However, this the biological mechanisms underpinning the genetic associa- would have reduced power to observe differences between the tion between BD-I and longer sleep duration. 2 subtypes rather than resulted in positive results we observe for insomnia and sleep duration. Finally, variants associated Implications with insomnia or hypersomnia at ages 40 to 69 years (the age Clinical and biological heterogeneity, combined with a classi- of the UK Biobank sample ) may differ from those associ- fication that is not grounded in biology, are obstacles to ad- ated in childhood or early adulthood. This may have in- vancing BD research. We provide some evidence of heteroge- creased our type-2 error rate, because most patients with BD neity in genetic propensity to some sleep traits within BD experience the first onset of impairing symptoms in adoles- (specifically insomnia and sleep duration), highlighting dif- cence or early adulthood. Genetic risk for insomnia or hy- ferences in the way some sleep-associated genetic factors are persomnia that manifests during or prior to early adulthood associated with BD subtypes. This adds to previously pub- may be more strongly associated with BD than those associ- lished work on stratification in BD and work suggesting that ated with midlife insomnia. These results should be repli- 32-34 different factors may influence the 2 conditions. cated using future sleep trait GWAS in younger samples of suf- These results suggest that clinical trials of sleep interven- ficient size for PRS analysis. tions should stratify participants by clinical subtype and ge- netic liability to insomnia or hypersomnia. Future work should explore which factors drive the differences in genetic liability Conclusions for insomnia/sleep duration between BD subtypes. To our knowledge, this is the first study to explore whether genetic liability for sleep traits is associated with clinical strata Strengths This study was conducted on the world’s largest single co- of individuals with BD. Future work should explore potential hort of BD with genotypic and phenotypic data. Phenotypic mechanisms underlying differences between the BD sub- data were collected using face-to-face semistructured inter- types in genetic liability for sleep traits. ARTICLE INFORMATION Biostatistics, Karolinska Institutet, Stockholm, University, Cardiff, United Kingdom (Owen, Sweden (Lewis, Karlsson, Landén); Genetics of Craddock, I. Jones, O’Donovan, Di Florio); Institute Accepted for Publication: September 29, 2019. Complex Traits, University of Exeter Medical of Neuroscience and Physiology, Sahlgenska Published Online: November 21, 2019. School, Exeter, United Kingdom (S. E. Jones, Academy at the Gothenburg University, doi:10.1001/jamapsychiatry.2019.4079 Weedon); Centre for Academic Mental Health, Gothenburg, Sweden (Landén). Open Access: This is an open access article Population Health Sciences, Bristol Medical School, Author Contributions: Dr Di Florio had full access distributed under the terms of the CC-BY License. University of Bristol, Bristol, United Kingdom to all the data in the study and takes responsibility © 2019 Lewis KJS et al. JAMA Psychiatry. (H. J. Jones); Medical Research Council Integrative for the integrity of the data and the accuracy of the Epidemiology Unit, University of Bristol, Bristol, Author Affiliations: Medical Research Council data analysis. United Kingdom (H. J. Jones); National Institute of Centre for Neuropsychiatric Genetics and Concept and design: Lewis, H. Jones, Health Research Biomedical Research Centre at Genomics, Institute of Psychological Medicine and Gordon-Smith, L. Jones, Craddock, I. Jones, University Hospitals Bristol NHS Foundation Trust Clinical Neurosciences, School of Medicine, Cardiff O'Donovan, Di Florio. and the University of Bristol, Bristol, United University, Cardiff, United Kingdom (Lewis, Acquisition, analysis, or interpretation of data: Lewis, Kingdom (H. J. Jones); Department of Richards, Leonenko, Forty, Escott-Price, Owen, Richards,Karlsson,Leonenko,S.Jones,Gordon-Smith, Psychological Medicine, University of Worcester, Craddock, I. Jones, O’Donovan, Di Florio); Forty,Escott-Price,Owen,Weedon,L.Jones,Craddock, Worcester, United Kingdom (Gordon-Smith, Department of Medical Epidemiology and I. Jones, Landén, O'Donovan, Di Florio. L. Jones); National Centre for Mental Health, Cardiff 308 JAMA Psychiatry March 2020 Volume 77, Number 3 (Reprinted) jamapsychiatry.com Downloaded From: https://jamanetwork.com/ by a Deepdyve User on 07/07/2021 Comparison of Genetic Liability for Sleep Traits in Individuals With Bipolar Disorder and Controls Original Investigation Research Drafting of the manuscript: Lewis, Richards, Padiñas, PhD, Cardiff University, for helpful a systematic review and meta-analysis. Sleep Med H. Jones, Gordon-Smith, L. Jones, O'Donovan, comments on the manuscript. He was not Rev. 2015;20:46-58. doi:10.1016/j.smrv.2014.06.006 Di Florio. compensated for his contribution. 16. Kaplan KA, Williams R. Hypersomnia: an Critical revision of the manuscript for important overlooked, but not overestimated, sleep intellectual content: Lewis, Karlsson, Leonenko, REFERENCES disturbance in bipolar disorder. Evid Based Ment S. Jones, H. Jones, Gordon-Smith, Forty, 1. Wehr TA, Sack DA, Rosenthal NE. Sleep Health. 2017;20(2):59. doi:10.1136/eb-2016-102433 Escott-Price, Owen, Weedon, L. Jones, Craddock, reduction as a final common pathway in the genesis 17. Kaplan KA, Gruber J, Eidelman P, Talbot LS, I. Jones, Landén, O'Donovan, Di Florio. of mania. Am J Psychiatry. 1987;144(2):201-204. Harvey AG. Hypersomnia in inter-episode bipolar Statistical analysis: Lewis, Richards, Karlsson, 2. Wehr TA. Sleep loss: a preventable cause of disorder: does it have prognostic significance? Leonenko, S. Jones, H. Jones, Escott-Price, mania and other excited states. J Clin Psychiatry. J Affect Disord. 2011;132(3):438-444. doi:10.1016/j. O'Donovan. 1989;50(12)(suppl):8-16. jad.2011.03.013 Obtained funding: Owen, L. Jones, I. Jones, Landén, Di Florio. 3. Lewis KS, Gordon-Smith K, Forty L, et al. Sleep 18. Ritter PS, Marx C, Lewtschenko N, et al. Administrative, technical, or material support: loss as a trigger of mood episodes in bipolar The characteristics of sleep in patients with Richards, S. Jones, Gordon-Smith, Forty, Landén. disorder: individual differences based on diagnostic manifest bipolar disorder, subjects at high risk of Supervision: Gordon-Smith, Forty, Owen, Weedon, subtype and gender. Br J Psychiatry. 2017;211(3): developing the disease and healthy controls. L. Jones, Craddock, I. Jones, O'Donovan, Di Florio. 169-174. doi:10.1192/bjp.bp.117.202259 J Neural Transm (Vienna). 2012;119(10):1173-1184. doi:10.1007/s00702-012-0883-y Conflict of Interest Disclosures: Dr Di Florio 4. Leibenluft E, Albert PS, Rosenthal NE, Wehr TA. reported receiving a Wellcome Trust grant outside Relationship between sleep and mood in patients 19. Kim KL, Weissman AB, Puzia ME, et al. the submitted work. Dr Owen reported grants from with rapid-cycling bipolar disorder. Psychiatry Res. Circadian phase preference in pediatric bipolar Takeda Pharmaceuticals outside the submitted 1996;63(2-3):161-168. doi:10.1016/0165-1781(96) disorder. J Clin Med. 2014;3(1):255-266. doi:10. work. Dr Landén reports grants from Stanley Center 02854-5 3390/jcm3010255 for Psychiatric Research, Broad Institute (via a grant 5. Perlman CA, Johnson SL, Mellman TA. 20. Melo MCA, Abreu RLC, Linhares Neto VB, from Stanley Medical Research Institute), and The prospective impact of sleep duration on de Bruin PFC, de Bruin VMS. Chronotype and grants from the National Institute for Mental Health depression and mania. Bipolar Disord. 2006;8(3): circadian rhythm in bipolar disorder: a systematic (grant MH077139) during the conduct of the study. 271-274. doi:10.1111/j.1399-5618.2006.00330.x review. Sleep Med Rev. 2017;34:46-58. doi:10.1016/ Dr O'Donovan reported grants from the Medical j.smrv.2016.06.007 6. American Psychiatric Association. Diagnostic Research Council of the United Kingdom and grants and Statistical Manual of Mental Disorders. 5th ed. 21. Takaesu Y. Circadian rhythm in bipolar disorder: from the National Institute of Mental Health during Arlington, VA: American Psychiatric Publishing; 2013. a review of the literature. Psychiatry Clin Neurosci. the conduct of the study and grants from Takeda 2018;72(9):673-682. doi:10.1111/pcn.12688 outside the submitted work. No other disclosures 7. Kaplan KA, Harvey AG. Hypersomnia across were reported. mood disorders: a review and synthesis. Sleep Med 22. Sheaves B, Freeman D, Isham L, et al. Rev. 2009;13(4):275-285. doi:10.1016/j.smrv.2008. Stabilising sleep for patients admitted at acute crisis Funding/Support: This study was funded by a 09.001 to a psychiatric hospital (OWLS): an assessor-blind National Alliance for Research on Schizophrenia & pilot randomised controlled trial. Psychol Med. Depression Young Investigator Grant awarded 8. Forty L, Smith D, Jones L, et al. Clinical 2018;48(10):1694-1704. doi:10.1017/ by the Brain & Behavior Research Foundation differences between bipolar and unipolar S0033291717003191 (Dr Di Florio). The Bipolar Disorder Research depression. Br J Psychiatry. 2008;192(5):388-389. Network was funded by the Wellcome Trust (grant doi:10.1192/bjp.bp.107.045294 23. Kaplan KA, Harvey AG. Behavioral treatment of 078901) and Stanley Medical Research Institute. insomnia in bipolar disorder. Am J Psychiatry. 2013; 9. Harvey AG, Talbot LS, Gershon A. Sleep Additional support was provided under Medical 170(7):716-720. doi:10.1176/appi.ajp.2013.12050708 disturbance in bipolar disorder across the lifespan. Research Council Centre (grant G0800509) and Clin Psychol (New York). 2009;16(2):256-277. 24. Kaplan KA, Talavera DC, Harvey AG. Rise and Program Grants (grant G0801418). This study doi:10.1111/j.1468-2850.2009.01164.x shine: a treatment experiment testing a morning makes use of genome-wide association data routine to decrease subjective sleep inertia in 10. Kanady JC, Soehnera AM, Harvey AG. generated by the Wellcome Trust Case-Control insomnia and bipolar disorder. Behav Res Ther. A retrospective examination of sleep disturbance Consortium 2. The National Centre for Mental 2018;111(111):106-112. doi:10.1016/j.brat.2018.10.009 across the course of bipolar disorder. J Sleep Disord Health is a collaboration between Cardiff, Swansea, Ther. 2015;4(2):318-328. 25. Frank E, Kupfer DJ, Thase ME, et al. Two-year and Bangor Universities and is funded by the Welsh outcomes for interpersonal and social rhythm government through Health and Care Research 11. Harvey AG, Schmidt DA, Scarnà A, Semler CN, therapy in individuals with bipolar I disorder. Arch Wales. Additional support was received from the Goodwin GM. Sleep-related functioning in Gen Psychiatry. 2005;62(9):996-1004. doi:10. Medical Research Council (grant MR/M005070/1 euthymic patients with bipolar disorder, patients 1001/archpsyc.62.9.996 [Drs S.E. Jones and Weedon]). Funding support for with insomnia, and subjects without sleep the Swedish Bipolar Collection was provided by the problems. Am J Psychiatry. 2005;162(1):50-57. 26. Lane JM, Liang J, Vlasac I, et al. Genome-wide Stanley Center for Psychiatric Research, Broad doi:10.1176/appi.ajp.162.1.50 association analyses of sleep disturbance traits Institute from a grant from Stanley Medical identify new loci and highlight shared genetics with 12. Geoffroy PA, Scott J, Boudebesse C, et al. Sleep Research Institute, the Swedish Research Council neuropsychiatric and metabolic traits. Nat Genet. in patients with remitted bipolar disorders: (grant 2018-02653), and the Swedish Federal 2017;49(2):274-281. doi:10.1038/ng.3749 a meta-analysis of actigraphy studies. Acta Government under the ALF agreement (grant ALF Psychiatr Scand. 2015;131(2):89-99. doi:10.1111/acps. 27. Lane JM, Jones SE, Dashti HS, et al; HUNT All In 20170019). 12367 Sleep. Biological and clinical insights from genetics Role of the Funder/Sponsor: The funding sources of insomnia symptoms. Nat Genet. 2019;51(3): 13. Plante DT, Winkelman JW. Sleep disturbance in had no role in the design and conduct of the study; 387-393. doi:10.1038/s41588-019-0361-7 bipolar disorder: therapeutic implications. Am J collection, management, analysis, and Psychiatry. 2008;165(7):830-843. doi:10.1176/appi. 28. Jansen PR, Watanabe K, Stringer S, et al; interpretation of the data; preparation, review, or ajp.2008.08010077 23andMe Research Team. Genome-wide analysis of approval of the manuscript; and decision to submit insomnia in 1,331,010 individuals identifies new risk 14. Harvey AG, Soehner AM, Kaplan KA, et al. the manuscript for publication. loci and functional pathways. Nat Genet. 2019;51 Treating insomnia improves mood state, sleep, and Meeting Presentation: This paper was presented (3):394-403. doi:10.1038/s41588-018-0333-3 functioning in bipolar disorder: a pilot randomized at the British Sleep Society Scientific Meeting; controlled trial. J Consult Clin Psychol. 2015;83(3): 29. Jones SE, Tyrrell J, Wood AR, et al. November 21, 2019; Birmingham, United Kingdom. 564-577. doi:10.1037/a0038655 Genome-wide association analyses in 128,266 Additional Contributions: We are grateful to all individuals identifies new morningness and sleep 15. Ng TH, Chung KF, Ho FYY, Yeung WF, Yung KP, participants in the Bipolar Disorder Research duration loci. PLoS Genet. 2016;12(8):e1006125. Lam TH. Sleep-wake disturbance in interepisode Network and the Swedish Bipolar Collection, as well doi:10.1371/journal.pgen.1006125 bipolar disorder and high-risk individuals: as all control participants who gave their time to this research. We are also grateful to Antonio jamapsychiatry.com (Reprinted) JAMA Psychiatry March 2020 Volume 77, Number 3 309 Downloaded From: https://jamanetwork.com/ by a Deepdyve User on 07/07/2021 Research Original Investigation Comparison of Genetic Liability for Sleep Traits in Individuals With Bipolar Disorder and Controls 30. Dashti HS, Jones SE, Wood AR, et al. GWAS in middle and old age. PLoS Med. 2015;12(3):e1001779. risk factors in a cohort of 6086 patients. PLoS One. 446,118 European adults identifies 78 genetic loci doi:10.1371/journal.pmed.1001779 2014;9(4):e94097. doi:10.1371/journal.pone. for self-reported 2 habitual sleep duration 0094097 44. Chang CC, Chow CC, Tellier LCAM, Vattikuti S, supported by accelerometer-derived estimates. Purcell SM, Lee JJ. Second-generation PLINK: rising 58. Craddock N, O’Donovan MC, Owen MJ. bioRxiv. https://www.biorxiv.org/content/biorxiv/ to the challenge of larger and richer datasets. Psychosis genetics: modeling the relationship early/2018/04/19/274977.full.pdf. Published 2018. Gigascience. 2015;4(1):7. doi:10.1186/s13742-015- between schizophrenia, bipolar disorder, and mixed doi:10.1101/274977 0047-8 (or “schizoaffective”) psychoses. Schizophr Bull. 31. Wang H, Lane JM, Jones SE, et al. Genome-wide 2009;35(3):482-490. doi:10.1093/schbul/sbp020 45. Euesden J, Lewis CM, O’Reilly PF. PRSice: association analysis of excessive daytime sleepiness polygenic risk score software. Bioinformatics. 2015; 59. Sklar P, Ripke S, Scott L, et al. Large-scale identifies 42 loci that suggest phenotypic 31(9):1466-1468. doi:10.1093/bioinformatics/btu848 genome-wide association analysis of bipolar subgroups. bioRxiv. https://www.biorxiv.org/ disorder identifies a new susceptibility locus near 46. Benjamini Y, Hochberg Y. Controlling the false content/10.1101/454561v1.full. Published 2018. ODZ4. 2011;43(10):977-983. doi:10.1038/ng.943 discovery rate: a practical and powerful approach to Accessed 8, 2019. doi:10.1101/454561 multiple testing. J R Stat Soc B. 1995;57(1):289-300. 60. Plante DT. Hypersomnia in mood disorders: 32. Baek JH, Park DY, Choi J, et al. Differences doi:10.1111/j.2517-6161.1995.tb02031.x a rapidly changing landscape. Curr Sleep Med Rep. between bipolar I and bipolar II disorders in clinical 2015;1(2):122-130. doi:10.1007/s40675-015-0017-9 47. Zheng J, Baird D, Borges M-C, et al. Recent features, comorbidity, and family history. J Affect developments in mendelian randomization studies. 61. Kaplan KA, McGlinchey EL, Soehner A, et al. Disord. 2011;131(1-3):59-67. doi:10.1016/j.jad.2010. Curr Epidemiol Rep. 2017;4(4):330-345. doi:10. Hypersomnia subtypes, sleep and relapse in bipolar 11.020 1007/s40471-017-0128-6 disorder. Psychol Med. 2015;45(8):1751-1763. 33. Caseras X, Murphy K, Lawrence NS, et al. doi:10.1017/S0033291714002918 48. Stahl E, Forstner A, McQuillin A, et al. Emotion regulation deficits in euthymic bipolar I Genome-wide association study identifies 30 loci 62. American Academy of Sleep Medicine. versus bipolar II disorder: a functional and associated with bipolar disorder. bioRxiv. https:// The International Classification of Sleep Disorders: diffusion-tensor imaging study. Bipolar Disord. www.biorxiv.org/content/10.1101/173062v. Published Diagnostic and Coding Manual. 2nd ed. Rochester, 2015;17(5):461-470. doi:10.1111/bdi.12292 2017. Accessed November 8, 2019. doi:10.1101/ MN: American Academy of Sleep Medicine; 2005. 34. Caseras X, Lawrence NS, Murphy K, Wise RG, 63. Ohayon MM, Dauvilliers Y, Reynolds CF III. Phillips ML. Ventral striatum activity in response to 49. Bowden J, Davey Smith G, Burgess S. Operational definitions and algorithms for reward: differences between bipolar I and II Mendelian randomization with invalid instruments: excessive sleepiness in the general population: disorders. Am J Psychiatry. 2013;170(5):533-541. effect estimation and bias detection through Egger implications for DSM-5 nosology. Arch Gen Psychiatry. doi:10.1176/appi.ajp.2012.12020169 regression. Int J Epidemiol. 2015;44(2):512-525. 2012;69(1):71-79. doi:10.1001/archgenpsychiatry. 35. Jones L, Metcalf A, Gordon-Smith K, et al. doi:10.1093/ije/dyv080 2011.1240 Gambling problems in bipolar disorder in the UK: 50. Bowden J, Davey Smith G, Haycock PC, 64. Nofzinger EA, Thase ME, Reynolds CF III, et al. prevalence and distribution. Br J Psychiatry. 2015; Burgess S. Consistent estimation in mendelian Hypersomnia in bipolar depression: a comparison 207(4):328-333. doi:10.1192/bjp.bp.114.154286 randomization with some invalid instruments using with narcolepsy using the multiple sleep latency 36. Charney AW, Ruderfer DM, Stahl EA, et al. a weighted median estimator. Genet Epidemiol. test. Am J Psychiatry. 1991;148(9):1177-1181. doi:10. Evidence for genetic heterogeneity between clinical 2016;40(4):304-314. doi:10.1002/gepi.21965 1176/ajp.148.9.1177 subtypes of bipolar disorder. Transl Psychiatry. 51. Hartwig FP, Davey Smith G, Bowden J. Robust 65. Allardyce J, Leonenko G, Hamshere M, et al. 2017;7(1):e993. doi:10.1038/tp.2016.242 inference in summary data Mendelian Association between schizophrenia-related 37. Steinan MK, Scott J, Lagerberg TV, et al. Sleep randomization via the zero modal pleiotropy polygenic liability and the occurrence and level of problems in bipolar disorders: more than just assumption. Int J Epidemiol. 2017;46(6):1985-1998. mood-incongruent psychotic symptoms in bipolar insomnia. Acta Psychiatr Scand. 2016;133(5):368-377. doi:10.1093/ije/dyx102 disorder. JAMA Psychiatry. 2018;75(1):28-35. doi:10. doi:10.1111/acps.12523 1001/jamapsychiatry.2017.3485 52. Bowden J, Del Greco M F, Minelli C, 38. Wray NR, Lee SH, Mehta D, Vinkhuyzen AAE, Davey Smith G, Sheehan N, Thompson J. 66. Munafò MR, Tilling K, Taylor AE, Evans DM, Dudbridge F, Middeldorp CM. Research review: A framework for the investigation of pleiotropy in Davey Smith G. Collider scope: when selection bias Polygenic methods and their application to two-sample summary data Mendelian can substantially influence observed associations. psychiatric traits. J Child Psychol Psychiatry. 2014; randomization. Stat Med. 2017;36(11):1783-1802. Int J Epidemiol. 2018;47(1):226-235. doi:10.1093/ 55(10):1068-1087. doi:10.1111/jcpp.12295 doi:10.1002/sim.7221 ije/dyx206 39. Dudbridge F. Power and predictive accuracy of 53. Greco M FD, Minelli C, Sheehan NA, 67. Billiard M, Dolenc L, Aldaz C, Ondze B, polygenic risk scores. PLoS Genet. 2013;9(3): Thompson JR. Detecting pleiotropy in Mendelian Besset A. Hypersomnia associated with mood e1003348. doi:10.1371/journal.pgen.1003348 randomisation studies with summary data and a disorders: a new perspective. J Psychosom Res. 40. Wellcome Trust Case Control Consortium. continuous outcome. Stat Med. 2015;34(21): 1994;38(1)(suppl 1):41-47. doi:10.1016/0022-3999 Genome-wide association study of 14,000 cases of 2926-2940. doi:10.1002/sim.6522 (94)90134-1 seven common diseases and 3,000 shared 54. Hemani G, Zheng J, Elsworth B, et al. 68. Coryell W, Endicott J, Maser JD, Keller MB, controls. Nature. 2007;447(7145):661-678. doi:10. The MR-Base platform supports systematic causal Leon AC, Akiskal HS. Long-term stability of polarity 1038/nature05911 inference across the human phenome. Elife. 2018; distinctions in the affective disorders. Am J Psychiatry. 41. Wing JK, Babor T, Brugha T, et al; Schedules for pii:e34408. doi:10.7554/eLife.34408 1995;152(3):385-390. doi:10.1176/ajp.152.3.385 Clinical Assessment in Neuropsychiatry. SCAN: 55. Jakobsson J, Zetterberg H, Blennow K, 69. Alloy LB, Urošević S, Abramson LY, et al. schedules for clinical assessment in Johan Ekman C, Johansson AGM, Landén M. Progression along the bipolar spectrum: neuropsychiatry. Arch Gen Psychiatry. 1990;47(6): Altered concentrations of amyloid precursor a longitudinal study of predictors of conversion 589-593. doi:10.1001/archpsyc.1990. protein metabolites in the cerebrospinal fluid of from bipolar spectrum conditions to bipolar I and II patients with bipolar disorder. disorders. J Abnorm Psychol. 2012;121(1):16-27. 42. Jones SE, Lane JM, Wood AR, et al. Neuropsychopharmacology. 2013;38(4):664-672. doi:10.1037/a0023973 Genome-wide association analyses of chronotype doi:10.1038/npp.2012.231 70. Tondo L, Lepri B, Cruz N, Baldessarini RJ. Age in 697,828 individuals provides insights into 56. Karanti A, Bobeck C, Osterman M, et al. Gender at onset in 3014 Sardinian bipolar and major circadian rhythms. Nat Commun. 2019;10(1):343. differences in the treatment of patients with bipolar depressive disorder patients. Acta Psychiatr Scand. doi:10.1038/s41467-018-08259-7 disorder: a study of 7354 patients. J Affect Disord. 2010;121(6):446-452. doi:10.1111/j.1600-0447.2009. 43. Sudlow C, Gallacher J, Allen N, et al. UK 2015;174:303-309. doi:10.1016/j.jad.2014.11.058 01523.x biobank: an open access resource for identifying 57. Tidemalm D, Haglund A, Karanti A, Landén M, the causes of a wide range of complex diseases of Runeson B. Attempted suicide in bipolar disorder: 310 JAMA Psychiatry March 2020 Volume 77, Number 3 (Reprinted) jamapsychiatry.com Downloaded From: https://jamanetwork.com/ by a Deepdyve User on 07/07/2021

Journal

JAMA PsychiatryUnpaywall

Published: Mar 1, 2020

There are no references for this article.