Association of Polygenic Liabilities for Major Depression, Bipolar Disorder, and Schizophrenia With Risk for Depression in the Danish Population

Katherine L Musliner; Preben B Mortensen; John J McGrath; Nis P Suppli; David M Hougaard; Jonas Bybjerg-Grauholm; Marie Bækvad-Hansen; Ole Andreassen; Carsten B Pedersen; Marianne G Pedersen; Ole Mors; Merete Nordentoft; Anders D Børglum; Thomas Werge; Esben Agerbo

doi:10.1001/jamapsychiatry.2018.4166

. 2019 Jan 30;76(5):516–525. doi: 10.1001/jamapsychiatry.2018.4166

Association of Polygenic Liabilities for Major Depression, Bipolar Disorder, and Schizophrenia With Risk for Depression in the Danish Population

Katherine L Musliner ^1,^2,^✉, Preben B Mortensen ^1,^2,³, John J McGrath ^2,^4,⁵, Nis P Suppli ^1,⁶, David M Hougaard ^1,⁷, Jonas Bybjerg-Grauholm ^1,⁷, Marie Bækvad-Hansen ^1,⁷, Ole Andreassen ^8,⁹, Carsten B Pedersen ^1,^2,³, Marianne G Pedersen ^1,^2,³, Ole Mors ^1,¹⁰, Merete Nordentoft ^1,⁶, Anders D Børglum ^1,¹¹, Thomas Werge ^1,^12,¹³, Esben Agerbo ^2,³, for the Bipolar Disorder Working Group of the Psychiatric Genomics Consortium

¹iPSYCH, The Lundbeck Foundation Initiative for Integrated Psychiatric Research, Denmark

²National Centre for Register-based Research, Department of Economics and Business Economics, Aarhus University, Aarhus, Denmark

³Center for Integrated Register-based Research at Aarhus University, Aarhus, Denmark

⁴Queensland Brain Institute, The University of Queensland, St Lucia, Queensland, Australia

⁵Queensland Centre for Mental Health Research, The Park Centre for Mental Health, Wacol, Queensland, Australia

⁶Mental Health Centre Copenhagen, Capital Region of Denmark, Copenhagen University Hospital, Copenhagen, Denmark

⁷Department for Congenital Disorders, Statens Serum Institut, Copenhagen, Denmark

⁸The Norwegian Centre for Mental Disorders Research, Division of Mental Health and Addiction, University of Oslo, Oslo, Norway

⁹Oslo University Hospital, Oslo, Norway

¹⁰Psychosis Research Unit, Aarhus University Hospital, Risskov, Denmark

¹¹Centre for Integrative Sequencing, Department of Biomedicine and iSEQ, Aarhus University, Aarhus, Denmark

¹²Department of Clinical Medicine, University of Copenhagen, Copenhagen, Denmark

¹³Institute of Biological Psychiatry, Copenhagen Mental Health Services, Copenhagen, Denmark

Group Information: The members of the Bipolar Disorder Working Group of the Psychiatric Genomics Consortium are listed at the end of this article.

^✉

Corresponding Author: Katherine L. Musliner, PhD, Fuglesangs Alle 26, Building R, 8210 Aarhus V, Denmark (klm@econ.au.dk).

Accepted for Publication: October 17, 2018.

Published Online: January 30, 2019. doi:10.1001/jamapsychiatry.2018.4166

Author Contributions: Dr Musliner had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Musliner, Hougaard, Borglum, Agerbo.

Acquisition, analysis, or interpretation of data: Musliner, Bo Mortensen, McGrath, Suppli, Bybjerg-Grauholm, Baekvad-Hansen, Andreassen, C. Pedersen, M. Pedersen, Mors, Nordentoft, Borglum, Werge, Agerbo.

Drafting of the manuscript: Musliner.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Musliner, Agerbo.

Obtained funding: Bo Mortensen, Hougaard, Mors, Nordentoft, Borglum, Werge.

Administrative, technical, or material support: Bybjerg-Grauholm, Baekvad-Hansen, Andreassen, C. Pedersen, Mors, Borglum, Werge, Agerbo.

Supervision: Bo Mortensen, Suppli, Mors, Nordentoft, Borglum, Agerbo.

Group Information: The Bipolar Disorder Working Group of the Psychiatric Genomics Consortium members are Eli A. Stahl, Icahn School of Medicine at Mount Sinai and Broad Institute; Gerome Breen, King’s College London; Andreas J. Forstner, University of Basel, University of Bonn, Institute of Medical Genetics and Pathology, and University Hospital Basel; Andrew McQuillin, University College London; Stephan Ripke, Stanley Center for Psychiatric Research, Broad Institute, Charité-Universitätsmedizin, and Massachusetts General Hospital; Vassily Trubetskoy, Charité-Universitätsmedizin; Manuel Mattheisen, Aarhus University, Karolinska Institutet, University Hospital Würzburg, and The Lundbeck Foundation Initiative for Integrative Psychiatric Research; Yunpeng Wang, Mental Health Centre Sct. Hans and University of Oslo; Jonathan R I Coleman, King’s College London; Héléna A. Gaspar, King’s College London; Christiaan A. de Leeuw, Universiteit Amsterdam; Stacy Steinberg, Amgen; Jennifer M. Whitehead Pavlides, The University of Queensland; Maciej Trzaskowski, University of Queensland; Tune H. Pers, Broad Institute and Boston Children’s Hospital; Peter A. Holmans, Cardiff University; Liam Abbott, Broad Institute; Esben Agerbo, The Lundbeck Foundation Initiative for Integrative Psychiatric Research and Aarhus University; Huda Akil, University of Michigan; Diego Albani, Istituto Di Ricerche Farmacologiche Mario Negri; Ney Alliey-Rodriguez, University of Chicago; Thomas D. Als, Aarhus University and the Lundbeck Foundation Initiative for Integrative Psychiatric Research; Adebayo Anjorin, Berkshire Healthcare NHS Foundation Trust; Verneri Antilla, Massachusetts General Hospital; Swapnil Awasthi, Charité-Universitätsmedizin; Judith A. Badner, Rush University Medical Center; Marie Bækvad-Hansen, the Lundbeck Foundation Initiative for Integrative Psychiatric Research and Statens Serum Institut; Jack D. Barchas, Weill Cornell Medical College; Nicholas Bass, University College London; Michael Bauer, University Hospital Carl Gustav Carus; Richard Belliveau, Broad Institute; Sarah E. Bergen, Karolinska Institutet; Carsten Bøcker Pedersen, the Lundbeck Foundation Initiative for Integrative Psychiatric Research and Aarhus University; Erlend Bøen, Diakonhjemmet Hospital; Marco Boks, UMC Utrecht Hersencentrum Rudolf Magnus; James Boocock, University of California Los Angeles; Monika Budde, University Hospital, LMU Munich; William Bunney, University of California, Irvine; Margit Burmeister, University of Michigan; Jonas Bybjerg-Grauholm, the Lundbeck Foundation Initiative for Integrative Psychiatric Research and Statens Serum Institut; William Byerley, University of California San Francisco; Miquel Casas, Instituto de Salud Carlos III, Biomedical Network Research Centre on Mental Health, Hospital Universitari Vall d´Hebron, Universitat Autònoma de Barcelona, and Universitat Autònoma de Barcelona; Felecia Cerrato, Broad Institute; Pablo Cervantes, McGill University Health Center; Kimberly Chambert, Broad Institute; Alexander W. Charney, Icahn School of Medicine at Mount Sinai; Danfeng Chen, Broad Institute; Claire Churchhouse, Broad Institute and Massachusetts General Hospital; Toni-Kim Clarke, University of Edinburgh; William Coryell, University of Iowa Hospitals and Clinics; David W. Craig, USC; Cristiana Cruceanu, McGill University Health Center and Max Planck Institute of Psychiatry; David Curtis, Queen Mary University of London and University College London; Piotr M. Czerski, Poznan University of Medical Sciences; Anders M. Dale, University of California San Diego; Simone de Jong, King’s College London; Franziska Degenhardt, University of Bonn; Jurgen Del-Favero, University of Antwerp; J. Raymond DePaulo, Johns Hopkins University School of Medicine; Srdjan Djurovic, Oslo University Hospital Ullevål and University of Bergen; Amanda L. Dobbyn, Icahn School of Medicine at Mount Sinai; Ashley Dumont, Broad Institute; Torbjørn Elvsåshagen, Oslo University Hospital; Valentina Escott-Price, Cardiff University; Chun Chieh Fan, University of California San Diego; Sascha B Fischer, University of Basel and University Hospital Basel; Matthew Flickinger, University of Michigan; Tatiana M. Foroud, Indiana University; Liz Forty, Cardiff University; Josef Frank, Heidelberg University; Christine Fraser, Cardiff University; Nelson B. Freimer, University of California Los Angeles; Louise Frisén, Karolinska University Hospital; Katrin Gade, University Hospital, LMU Munich and University Medical Center Göttingen; Diane Gage, Broad Institute; Julie Garnham, Dalhousie University; Claudia Giambartolomei, University of California Los Angeles; Marianne Giørtz Pedersen, the Lundbeck Foundation Initiative for Integrative Psychiatric Research and Aarhus University; Jaqueline Goldstein, Broad Institute; Scott D. Gordon, QIMR Berghofer Medical Research Institute; Katherine Gordon-Smith, University of Worcester; Elaine K Green, Plymouth University Peninsula Schools of Medicine and Dentistry; Melissa J. Green, University of New South Wales; Tiffany A. Greenwood, University of California San Diego; Jakob Grove, Aarhus University and the Lundbeck Foundation Initiative for Integrative Psychiatric Research; Weihua Guan, University of Minnesota System; José Guzman Parra, University Regional Hospital, Biomedicine Institute, IBIMA; Marian L Hamshere, Cardiff University; Martin Hautzinger, Eberhard Karls Universität Tübingen; Urs Heilbronner, University Hospital, LMU Munich; Stefan Herms, University of Basel, University of Bonn, and University Hospital Basel; Maria Hipolito, Howard University Hospital; Per Hoffmann, University of Basel, University of Bonn, and University Hospital Basel; Dominic Holland, University of California San Diego, La Jolla; Laura Huckins, Icahn School of Medicine at Mount Sinai; Stéphane Jamain, Université Paris Est; Jessica S. Johnson, Icahn School of Medicine at Mount Sinai; Anders Juréus, Karolinska Institutet; Radhika Kandaswamy, King’s College London; Robert Karlsson, Karolinska Institutet; James L. Kennedy, Campbell Family Mental Health Research Institute and University of Toronto; Sarah Kittel-Schneider, University Hospital Frankfurt; Sarah V. Knott, (University of Worcester; James A. Knowles, SUNY Downstate Medical Center College of Medicine; Manolis Kogevinas, Center for Research in Environmental Epidemiology, Barcelona; Anna C. Koller, University of Bonn; Ralph Kupka, VU medisch centrum, Amsterdam; Catharina Lavebratt, Karolinska University Hospital; Jacob Lawrence, North East London NHS Foundation Trust; William B. Lawson, Howard University Hospital; Markus Leber, University Hospital Cologne; Phil H Lee, Broad Institute and Massachusetts General Hospital; Shawn E. Levy, HudsonAlpha Institute for Biotechnology; Jun Z. Li, University of Michigan; Chunyu Liu, University of Illinois at Chicago College of Medicine; Susanne Lucae, Max Planck Institute of Psychiatry; Anna Maaser, University of Bonn; Donald J. MacIntyre, Mental Health, NHS 24, Glasgow and University of Edinburgh; Pamela B. Mahon, Johns Hopkins University School of Medicine and Brigham and Women’s Hospital; Wolfgang Maier, University of Bonn; Lina Martinsson, Karolinska University Hospital; Steve McCarroll, Broad Institute and Harvard Medical School; Peter McGuffin, King’s College London; Melvin G. McInnis, University of Michigan; James D. McKay, International Agency for Research on Cancer; Helena Medeiros, SUNY Downstate Medical Center College of Medicine; Sarah E. Medland, QIMR Berghofer Medical Research Institute; Fan Meng, University of Michigan; Lili Milani, University of Tartu; Grant W. Montgomery, University of Queensland; Derek W. Morris, National University of Ireland and Trinity College Dublin; Thomas W. Mühleisen, University of Basel and Research Centre Jülich; Niamh Mullins, King’s College London; Hoang Nguyen, Icahn School of Medicine at Mount Sinai; Caroline M. Nievergelt, University of California San Diego, La Jolla and Veterans Affairs San Diego Healthcare System; Annelie Nordin Adolfsson, Umeå University; Evaristus A. Nwulia, Howard University Hospital; Claire O'Donovan, Dalhousie University; Loes M. Olde Loohuis, University of California Los Angeles; Anil P. S. Ori, University of California Los Angeles; Lilijana Oruc, Clinical Center University of Sarajevo; Urban Ösby, Karolinska University Hospital; Roy H. Perlis, Harvard Medical School and Massachusetts General Hospital; Amy Perry, University of Worcester; Andrea Pfennig, Technische Universität Dresden; James B. Potash, Johns Hopkins University School of Medicine; Shaun M. Purcell, Icahn School of Medicine at Mount Sinai and Brigham and Women’s Hospital; Eline J. Regeer, Outpatient Clinic for Bipolar Disorder; Andreas Reif, University Hospital Frankfurt; Céline S. Reinbold, University of Basel and University Hospital Basel; John P. Rice, (Washington University in Saint Louis; Fabio Rivas, University Regional Hospital, Biomedicine Institute, Málaga; Margarita Rivera, King’s College London and University of Granada; Panos Roussos, Icahn School of Medicine at Mount Sinai; Douglas M. Ruderfer, Vanderbilt University Medical Center; Euijung Ryu, Mayo Clinic, Rochester; Cristina Sánchez-Mora, Biomedical Network Research Centre on Mental Health, Madrid, Hospital Universitari Vall d´Hebron, and Universitat Autònoma de Barcelona; Alan F. Schatzberg, Stanford University School of Medicine; William A. Scheftner, Rush University Medical Center; Nicholas J. Schork, Scripps Translational Science Institute; Cynthia Shannon Weickert, University of New South Wales and Neuroscience Research Australia; Tatyana Shehktman, University of California San Diego, La Jolla; Paul D. Shilling, University of California San Diego, La Jolla; Engilbert Sigurdsson, University of Iceland; Claire Slaney, Dalhousie University; Olav B. Smeland, University of California San Diego, La Jolla, Oslo University Hospital, and University of Oslo; Janet L. Sobell, University of Southern California, Los Angeles; Christine Søholm Hansen, the Lundbeck Foundation Initiative for Integrative Psychiatric Research and Statens Serum Institut; Anne T. Spijker, Mood Disorders, PsyQ; David St Clair, University of Aberdeen; Michael Steffens, Federal Institute for Drugs and Medical Devices; John S. Strauss, University of Toronto and Centre for Addiction and Mental Health; Fabian Streit, Heidelberg University; Jana Strohmaier, Heidelberg University; Szabolcs Szelinger, TGen; Robert C. Thompson, University of Michigan; Thorgeir E. Thorgeirsson, deCODE Genetics/Amgen; Jens Treutlein, Heidelberg University; Helmut Vedder, Psychiatrisches Zentrum Nordbaden; Weiqing Wang, Icahn School of Medicine at Mount Sinai; Stanley J. Watson, University of Michigan; Thomas W. Weickert, University of New South Wales and Neuroscience Research Australia; Stephanie H. Witt, Heidelberg University; Simon Xi, Pfizer Global Research and Development; Wei Xu, Princess Margaret Cancer Centre and University of Toronto; Allan H. Young, King’s College London; Peter Zandi, Johns Hopkins University Bloomberg School of Public Health; Peng Zhang, Johns Hopkins University School of Medicine; Sebastian Zollner, University of Michigan; Rolf Adolfsson, Umeå University; Ingrid Agartz, Karolinska Institutet, Diakonhjemmet Hospital, and University of Oslo; Martin Alda, Dalhousie University; Lena Backlund, Karolinska University Hospital; Bernhard T. Baune, University of Adelaide; Frank Bellivier, Assistance Publique-Hôpitaux de Paris, FondaMental Foundation, INSERM, and Université Paris Diderot; Wade H. Berrettini, University of Pennsylvania; Joanna M. Biernacka, Mayo Clinic; Douglas H. R. Blackwood, University of Edinburgh; Michael Boehnke, University of Michigan; Anders D. Børglum, Aarhus University and the Lundbeck Foundation Initiative for Integrative Psychiatric Research; Aiden Corvin, Trinity College Dublin; Nicholas Craddock, Cardiff University; Mark J. Daly, Broad Institute and Massachusetts General Hospital; Udo Dannlowski, University of Münster; Tõnu Esko, Broad Institute, Harvard Medical School, University of Tartu, and Children’s Hospital Boston; Bruno Etain, Assistance Publique-Hôpitaux de Paris, INSERM, Université Paris Diderot, and the Centre for Affective Disorders, Institute of Psychiatry, Psychology and Neuroscience, London; Mark Frye, Mayo Clinic; Janice M. Fullerton, Neuroscience Research Australia and University of New South Wales; Elliot S. Gershon, University of Chicago; Michael Gill, Trinity College Dublin; Fernando Goes, Johns Hopkins University School of Medicine; Maria Grigoroiu-Serbanescu, Alexandru Obregia Clinical Psychiatric Hospital; Joanna Hauser, Poznan University of Medical Sciences; David M. Hougaard, the Lundbeck Foundation Initiative for Integrative Psychiatric Research and Statens Serum Institut; Christina M. Hultman, Karolinska Institutet; Ian Jones, Cardiff University; Lisa A. Jones, University of Worcester; René S. Kahn, Icahn School of Medicine at Mount Sinai and UMC Utrecht Hersencentrum Rudolf Magnus; George Kirov, Cardiff University; Mikael Landén, Karolinska Institutet and University of Gothenburg; Marion Leboyer, Université Paris Est, Assistance Publique-Hôpitaux, and INSERM; Cathryn M. Lewis, King’s College London; Qingqin S. Li, Janssen Research and Development, LLC; Jolanta Lissowska, M. Sklodowska-Curie Cancer Center and Institute of Oncology; Nicholas G. Martin, QIMR Berghofer Medical Research Institute and the University of Queensland; Fermin Mayoral, University Regional Hospital, Biomedicine Institute, Málaga; Susan L. McElroy, Lindner Center of HOPE; Andrew M. McIntosh, University of Edinburgh; Francis J. McMahon, National Institute of Mental Health; Ingrid Melle, Oslo University Hospital and University of Oslo; Andres Metspalu, University of Tartu; Philip B. Mitchell, University of New South Wales; Gunnar Morken, Norwegian University of Science and Technology and St Olavs University Hospital; Ole Mors, the Lundbeck Foundation Initiative for Integrative Psychiatric Research and Aarhus University Hospital; Preben Bo Mortensen, Aarhus University and the Lundbeck Foundation Initiative for Integrative Psychiatric Research; Bertram Müller-Myhsok, Max Planck Institute of Psychiatry and University of Liverpool; Richard M. Myers, HudsonAlpha Institute for Biotechnology; Benjamin M. Neale, Broad Institute and Massachusetts General Hospital; Vishwajit Nimgaonkar, University of Pittsburgh; Merete Nordentoft, the Lundbeck Foundation Initiative for Integrative Psychiatric Research and University of Copenhagen; Markus M. Nöthen, University of Bonn; Michael C. O'Donovan, Cardiff University; Ketil J. Oedegaard, University of Bergen; Michael J. Owen, Cardiff University; Sara A. Paciga, Pfizer Global Research and Development; Carlos Pato, SUNY Downstate Medical Center College of Medicine; Michele T. Pato, SUNY Downstate Medical Center College of Medicine; Danielle Posthuma, Vrije Universiteit Amsterdam; Josep Antoni Ramos-Quiroga, Biomedical Network Research Centre on Mental Health, Madrid, Hospital Universitari Vall d´Hebron, Universitat Autònoma de Barcelona, and Universitat Autònoma de Barcelona; Marta Ribasés, Biomedical Network Research Centre on Mental Health, Madrid, Hospital Universitari Vall d´Hebron, and Universitat Autònoma de Barcelona; Marcella Rietschel, Heidelberg University; Guy A. Rouleau, McGill University and Montreal Neurological Institute and Hospital; Martin Schalling, Karolinska University Hospital; Peter R. Schofield, Neuroscience Research Australia and University of New South Wales; Thomas G. Schulze, University Hospital, LMU Munich, Johns Hopkins University School of Medicine, Heidelberg University, University Medical Center Göttingen, and the National Institute of Mental Health; Alessandro Serretti, University of Bologna; Jordan W. Smoller, Broad Institute and Massachusetts General Hospital; Hreinn Stefansson, deCODE Genetics/Amgen; Kari Stefansson, deCODE Genetics/Amgen and University of Iceland; Eystein Stordal, Hospital Namsos, and Norges Teknisk Naturvitenskapelige Universitet Fakultet for naturvitenskap og teknologi; Patrick F. Sullivan, Karolinska Institutet and University of North Carolina at Chapel Hill; Gustavo Turecki, McGill University; Arne E. Vaaler, Sankt Olavs Hospital Universitetssykehuset i Trondheim; Eduard Vieta, University of Barcelona; John B. Vincent, Centre for Addiction and Mental Health, Toronto; Thomas Werge, the Lundbeck Foundation Initiative for Integrative Psychiatric Research and University of Copenhagen; John I. Nurnberger, Indiana University School of Medicine; Naomi R. Wray, the University of Queensland; Arianna Di Florio, Cardiff University and University of North Carolina at Chapel Hill; Howard J. Edenberg, Indiana University School of Medicine; Sven Cichon, University of Basel, University of Bonn, University Hospital Basel, and Research Centre Jülich; Roel A. Ophoff, UMC Utrecht Hersencentrum Rudolf Magnus and University of California Los Angeles; Laura J. Scott, University of Michigan; Ole A. Andreassen, Oslo University Hospital and University of Oslo; John Kelsoe, University of California San Diego; and Pamela Sklar, Icahn School of Medicine.

Conflict of Interest Disclosures: Dr Andreassen reports receiving personal fees and speaker’s honorarium from Lundbeck. No other disclosures were reported.

Funding/Support: This study was funded by the Lundbeck Foundation (grant R155-2014-1724). The Psychiatric Genomics Consortium (PGC) has received major funding from the US National Institute of Mental Health and the US National Institute of Drug Abuse (grants U01 MH109528 and U01 MH1095320).

Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Additional Contributions: We thank the Broad Institute for genotyping and the PGC Major Depression and Schizophrenia Work Groups as well as 23andMe for making summary statistics available. We thank the Bipolar Disorders Working Group of the PGC for making leave one out summary statistics available in advance of publication. We also thank Thomas Damm Als, PhD, Aarhus University, for generating the principal component variables and Jakob Grove, PhD, Aarhus University, for generating the polygenic risk scores.

^✉

Corresponding author.

PMCID: PMC6495355 PMID: 30698613

Key Points

Question

Are polygenic liabilities for major depression, bipolar disorder, and schizophrenia associated with depression in the general population?

Findings

In this case-cohort study of 34 573 individuals, each 1 SD increase in the polygenic risk for major depression was associated with a 30% increase in the hazard for a depression diagnosis in hospital-based psychiatric care. Polygenic liabilities for schizophrenia and bipolar disorder were also associated with an increased hazard of depression but to a lesser extent.

Meaning

Polygenic risk scores trained in samples of prevalent major depression cases are associated with first depression in the general population.

Abstract

Importance

Although the usefulness of polygenic risk scores as a measure of genetic liability for major depression (MD) has been established, their association with depression in the general population remains relatively unexplored.

Objective

To evaluate whether polygenic risk scores for MD, bipolar disorder (BD), and schizophrenia (SZ) are associated with depression in the general population and explore whether these polygenic liabilities are associated with heterogeneity in terms of age at onset and severity at the initial depression diagnosis.

Design, Setting, and Participants

Participants were drawn from the Danish iPSYCH2012 case-cohort study, a representative sample drawn from the population of Denmark born between May 1, 1981, and December 31, 2005. The hazard of depression was estimated using Cox regressions modified to accommodate the case-cohort design. Case-only analyses were conducted using linear and multinomial regressions. The data analysis was conducted from February 2017 to June 2018.

Exposures

Polygenic risk scores for MD, BD, and SZ trained using the most recent genome-wide association study results from the Psychiatric Genomics Consortium.

Main Outcomes and Measures

The main outcome was first depressive episode (International Statistical Classification of Diseases and Related Health Problems, Tenth Revision [ICD-10] code F32) treated in hospital-based psychiatric care. Severity at the initial diagnosis was measured using the ICD-10 code severity specifications (mild, moderate, severe without psychosis, and severe with psychosis) and treatment setting (inpatient, outpatient, and emergency).

Results

Of 34 573 participants aged 10 to 31 years at censoring, 68% of those with depression were female compared with 48.9% of participants without depression. Each SD increase in polygenic liability for MD, BD, and SZ was associated with 30% (hazard ratio [HR], 1.30; 95% CI, 1.27-1.33), 5% (HR, 1.05; 95% CI, 1.02-1.07), and 12% (HR, 1.12; 95% CI, 1.09-1.15) increases in the hazard of depression, respectively. Among cases, a higher polygenic liability for BD was associated with earlier depression onset (β = −.07; SE = .02; P = .002).

Conclusions and Relevance

Polygenic liability for MD is associated with first depression in the general population, which supports the idea that these scores tap into an underlying liability for developing the disorder. The fact that polygenic risk for BD and polygenic risk for SZ also were associated with depression is consistent with prior evidence that these disorders share some common genetic overlap. Variations in polygenic liability may contribute slightly to heterogeneity in clinical presentation, but these associations appear minimal.

This cohort study explores the association of polygenic risk scores for major depression, bipolar disorder, and schizophrenia with the risk for depression in the Danish population.

Introduction

Genes play a moderate role in the etiology of depression, with twin-based heritability estimates ranging from 30% to 40%¹ and single-nucleotide polymorphism (SNP)–based heritability estimates ranging from 9% to 29%.^2,3,4 Large empirical studies of the genetic architecture of depression indicate that it is polygenic, meaning that the contribution of genetic factors is attributable to small effects of hundreds or thousands of genetic variants spread across the genome.^3,5

To date, multiple studies have shown small but statistically significant associations between polygenic risk scores (PRSs) a weighted sum of the number of variants associated with the disorder in a different data set, and depression.^6,7,8,9 However, these studies focused on prevalent depression, which is more likely than incident depression to be recurrent or chronic.^10,11 Because the genome-wide association study (GWAS) data underpinning the PRSs are largely based on prevalent cases, this could suggest that part of the genetic architecture discovered in GWAS studies is linked to chronicity or recurrence rather than the risk of developing the disorder. Therefore, while the usefulness of PRSs as a measure of genetic liability for depression has been established, their association with depression in the general population remains, to our knowledge, relatively unexplored.

In addition to heterogeneity in chronicity and recurrence, depression is also characterized by substantial variation in characteristics, such as age at onset (AAO) and symptom severity. Research suggests that this variation may be partially due to differences in genetic liability. Family studies demonstrate that individuals with a parental history of major depression (MD) are at increased risk for onset of depression at earlier ages,^12,13,14,15 and recent results from the Psychiatric Genomics Consortium (PGC) showed that the polygenic risk for depression had a stronger association with early-onset vs late-onset depression.⁴ Research also suggests that individuals with severe MD may have a higher genetic burden than individuals with milder symptoms. A recent GWAS in Han Chinese women found an increased genetic signal among individuals with melancholia,¹⁶ and the PGC reported higher PRSs among severe vs moderate depression cases.⁴

Our primary aim in this study was to evaluate the extent to which polygenic liability is associated with risk for first depressive episode in the general population. As a secondary aim, we examined whether polygenic liability is associated with severity and AAO at first depression diagnosis. Because prior evidence suggests a possible shared genetic etiology between depression and other psychiatric disorders,^2,4,17 we also examined the extent to which PRSs for bipolar disorder (BD) and schizophrenia (SZ) are associated with the risk for developing depression in the general population. To accomplish these aims, we used data from the iPSYCH2012 sample, a unique data set that links genetic information with longitudinal phenotype data from Danish national registers.

Methods

Study Design

For a detailed description of the iPSYCH2012 sample, see Pedersen et al.¹⁸ Briefly, the iPSYCH2012 sample has a case-cohort design¹⁹ that consists of 2 parts: a random sample (ie, subcohort) of individuals drawn from a specified base population (ie, full cohort) and all additional cases from the full cohort that were not selected as part of the subcohort. Like a traditional cohort study, a case-cohort study can obtain accurate estimates of hazards and risks using a traditional survival analysis, provided the analyses are modified to address issues associated with point and variance estimations that are caused by oversampling cases.^19,20,21,22 For a more detailed description of the case-cohort design see the eMethods in the Supplement.

In the iPSYCH2012 sample, the subcohort consists of a random sample of 30 000 individuals drawn from the full cohort of all singletons born in Denmark between May 1, 1981, and December 31, 2005, who were alive and living in Denmark on their first birthday and had known mothers (N = 1 472 762).¹⁸ The full cohort was identified using information from the Danish Civil Registration system.²³ The iPSYCH2012 study includes all individuals from the full cohort who received a diagnosis of depression in a psychiatric hospital in Denmark between 1991 and 2012 at 10 years or older. Information on psychiatric diagnoses was obtained from the Danish Central Psychiatric Research Register (DCPRR).²⁴ Diagnoses are based on the ICD-10 Classification of Mental and Behavioral Disorders: Diagnostic Criteria for Research (ICD-10).²⁵

Sample Selection

We began by selecting all members of the subcohort and all cases outside the subcohort whose first depression diagnosis in the DCPRR was F32 (depressive episode). We focused exclusively on individuals with Danish-born parents to reduce heterogeneity in genetic ancestry. The sample was further restricted to individuals who were successfully genotyped as part of the iPSYCH2012 sample, passed quality control (QC) measures, and were alive, living in Denmark, and at least age 10 years by the end of follow-up (December 31, 2012). Seventy-six individuals were removed from the subcohort because their first depression diagnosis in the register was F33 (recurrent depressive disorder) and we were thus unable to identify their first depressive episode. The final sample included 34 573 individuals. Of the 14 799 members with a depression diagnosis, 308 (2.1%) were members of the subcohort and 14 491 individuals (97.9%) were cases outside the subcohort (eFigure 1 in the Supplement). This study was approved by the Danish Data Protection Agency. By Danish law, register-based studies do not require informed consent.

Genotyping

For a more detailed description of genotyping and QC in the iPSYCH2012 cohort, see Pedersen et al.¹⁸ Briefly, DNA was collected from blood spots that were drawn as part of routine clinical testing at birth and stored at the Danish Neonatal Screening Biobank.²⁶ Blood spots were located for 80 422 of 86 189 members (93.3%) of the original iPSYCH2012 sample. Genotyping was performed at the Broad Institute of Harvard University and Massachusetts Institute of Technology (Cambridge, MA) using the Infinium PsychChip array, version 1.0 (Illumina) according to the manufacturer’s protocols.²⁷ Quality control and imputation (using 1000 genomes as the reference panel) were conducted using the Ricopili pipeline. As the iPSYCH2012 sample is population based, individuals from the same nuclear family unit were neither purposely sampled nor purposely removed.

Measures

The AAO was operationalized as the individual’s age in years at first F32 diagnosis in the DCPRR. Information on severity was obtained from ICD-10 diagnostic codes (mild, F32.0; moderate, F32.1; severe without psychotic features, F32.2; and severe with psychotic features, F32.3). We also examined differences in treatment settings (inpatient, emergency, and outpatient) as cases treated in inpatient or emergency settings are likely more severe than cases treated in an outpatient setting.

Polygenic risk scores were calculated using a standard approach²⁸ in which a linkage disequilibrium–pruned discovery data set is used to identify autosomal SNPs associated with the outcome at varying P value thresholds, and then a score is calculated for individuals in a target data set that corresponds to the weighted sum of each participant’s allelic burden at that threshold. Polygenic risk scores for MD and SZ were created using the Ricopili process with the most recently published GWAS results from the PGC (not including iPSYCH2012) as discovery data sets.^4,29 The discovery data set for BD was composed of leave one out summary statistics provided in advance of the latest GWAS publication from the Bipolar Working Group of the PGC.³⁰ Single-nucleotide polymorphisms from the discovery data sets were filtered at an INFO score of more than 0.9 and a minor allele frequency of more than .05, and the broad major histocompatibility complex region (chr6: 25-35 MB) was removed. Additionally SNPs were only included in the scores if they were reliably genotyped or imputed across all 23 waves of the iPSYCH2012 sample, at an INFO score of more than 0.6 and a minor allele frequency of more than .01. Ten PRSs were calculated for each disorder (30 total) at the following P value thresholds: P < .00000005, .000001, .0001, .001, .01, .05, .10, .20, .50, and >.99 (see eTable 1 in the Supplement for SNP numbers).³¹ Polygenic risk scores were standardized using the means and SDs from the distributions in subcohort members with Danish ancestry (eFigure in the Supplement). As the iPSYCH2012 subcohort is a simple random sample drawn from the base population, the distribution of PRSs in the subcohort approximates the distribution in the Danish-born population.

Statistical Analysis

The hazard of depression was estimated using Cox regressions with days since the participant’s tenth birthday as the timescale. Individuals entered the analysis on their tenth birthday and were censored on the date of their first F32 diagnosis in the DCPRR, death, emigration, or December 31, 2012, whichever came first. The oldest participants were age 31 years when they received their diagnosis or were censored. We used robust standard errors and Barlow weighting to account for oversampling of cases.^22,32 All models were adjusted for sex and the top 4 ancestral principal components (PCs). Additionally, all models were stratified by birth year (1981-2002) to control for secular trends in diagnostic practices and because fewer bloodspots were retrievable among individuals born during earlier years.

To examine whether polygenic liability has a stronger association with early-onset depression, we conducted separate Cox regressions in each analysis with only depression with onset in a specific age range considered as an outcome. Thus, we estimated separately the association of PRSs with hazard of depression with diagnosis at ages 10 to 15, 16 to 20, 21 to 25, and 26 to 31 years, respectively. Likewise, we conducted separate Cox regressions that considered respectively mild, moderate, severe without psychotic features, severe with psychotic features, inpatient, emergency and outpatient treatment setting at first depression as outcomes in each of the analyses. We also examined the associations between PRSs, AAO, and severity among cases only. Associations with AAO in days were estimated using linear regressions. Associations with severity measures were estimated using multinomial logistic regressions. All case-only regressions were adjusted for the first 4 PCs, sex, and birth year. Statistical significance was assessed at the Bonferroni-corrected α-level P < .017. Analyses were conducted in SAS, version 9.4 (SAS Institute; eAppendix in the Supplement).

Results

Sample Characteristics

Table 1 shows sample characteristics. Of the participants without depression, 9659 (48.9%) were female, which was expected given that the subcohort is a representative sample of the Danish population. Of the participants with MD, 10 056 (68.0%) were female, which was also expected given that the typical female to male ratio of depression in the population is 2:1.³³ The AAO ranged from 10 to 31 years, with a mean (SD) of 19.1 (4.1) years. Seventeen percent of the cases were classified as mild (n = 2573), 45% as moderate (n = 6635), 9% as severe (n = 1386), 3% as psychotic (n = 481), and 25% had no ICD-10 severity specification (n = 3724). Most (60%; n = 8880) were treated in an outpatient setting.

Table 1. Sample Characteristics.

Characteristic	Case Status, No. (%)
Characteristic	Participants Without Depression	Participants With Depression
Sex
Male	10 115 (51.2)	4743 (32.1)
Female	9659 (48.9)	10 056 (68.0)
Age at depression diagnosis, y
Noncase	19 774 (100.0)	NA
10-15	NA	2965 (20.0)
16-20	NA	6653 (45.0)
21-25	NA	3981 (26.9)
26-31	NA	1200 (8.1)
ICD-10 severity specifier
Noncase	19 774 (100.0)	NA
Mild	NA	2573 (17.4)
Moderate	NA	6635 (44.8)
Severe without psychotic symptoms	NA	1386 (9.4)
Severe with psychotic symptoms	NA	481 (3.3)
Severity unspecified	NA	3724 (25.2)
Treatment setting
Noncase	19 774 (100.0)	NA
Outpatient	NA	8880 (60.0)
Inpatient	NA	2269 (15.3)
Emergency	NA	3650 (24.7)
Total sample	19 774 (100.0)	14 799 (100.0)

Open in a new tab

Abbreviations: ICD-10, International Statistical Classification of Diseases and Related Health Problems, Tenth Revision; NA, not applicable.

Polygenic Liability and Hazard of Depression

For all 3 PRSs (PRS-MD, PRS-BD, and PRS-SZ), the strength of the association increased as the stringency of the P value threshold decreased up to P < .05 (eTable 2 and eFigures 2-7 in the Supplement). For this reason, and to maintain consistency with prior research, we present results from the P < .05 thresholds throughout this article.

Each SD increase in PRS-MD was associated with a 30% increase in the hazard of depression (95% CI, 1.27-1.33; P < .0001). Compared with an individual with average polygenic liability, an individual who is 1 SD more than the population average had a 30% increased risk of receiving a depression diagnosis before age 31 years. The corresponding values for PRS-BD and PRS-SZ were 5% (95% CI, 1.02-1.07; P < .0001) and 12% (95% CI, 1.09-1.15; P < .0001), respectively (eTable 2 in the Supplement). Figure 1 shows the associations between PRS deciles and the hazard of depression, with the bottom decile (ie, the lowest 10% of the PRS distribution) as the reference category. Compared with individuals in the bottom decile, individuals in the top decile of PRS-MD had a hazard ratio of 2.55 (95% CI, 2.28-2.85; P < .0001). The corresponding values were 1.22 (95% CI, 1.10-1.36; P < .0001) for PRS-BD and 1.49 (95% CI, 1.34-1.66; P < .0001) for PRS-SZ (eTable 3 in the Supplement).

Figure 1. — A-C, The hazard of depression for each decile of polygenic liability compared with the bottom decile (ie, the bottom 10% of the polygenic liability in the Danish population). The bars represent 95% confidence intervals.

AAO

As shown in Figure 2 and Table 2, the hazard of depression per SD increase in PRS-MD was slightly higher for first diagnosis between age 16 to 20 years (1.31; 95% CI, 1.27-1.35) and age 21 to 25 years (1.32; 95% CI, 1.27-1.38) compared with age 10 to 15 years (1.27; 95% CI, 1.22-1.33) or age 26 to 31 years (1.24; 95% CI, 1.16-1.32). The hazard ratio of PRS-BD was highest for diagnosis between ages 10 to 15 years (1.08; 95% CI, 1.04-1.12) and decreased linearly with age. The hazard ratio between PRS-SZ and was highest for diagnoses between age 10 to 15 years (1.17; 95% CI, 1.12-1.22) and lowest for diagnoses between age 21 to 25 years (1.08; 95% CI, 1.04-1.13). Case-only analyses showed small associations between higher PRSs and earlier AAO across all scores, but only the association for PRS-BD survived correction for multiple testing (Table 3).

Table 2. Associations Between PRS-MD, PRS-BD, and PRS-SZ and Hazard of Depression Within Different Age at Onset and Severity Subgroups.

Characteristic	MD		BD		SZ
Characteristic	HR (95% CI)	P Value	HR (95% CI)	P Value	HR (95% CI)	P Value
Age at depression diagnosis, y
10-15	1.27 (1.22-1.33)	<.0001	1.08 (1.04-1.12)	.0002	1.17 (1.12-1.22)	<.0001
16-20	1.31 (1.27-1.35)	<.0001	1.04 (1.01-1.07)	.008	1.12 (1.09-1.16)	<.0001
21-25	1.32 (1.27-1.38)	<.0001	1.04 (1.00-1.08)	.05	1.08 (1.04-1.13)	0.0001
26-31	1.24 (1.16-1.32)	<.0001	1.02 (0.95-1.10)	.55	1.12 (1.04-1.19)	0.0018
ICD-10 severity specifier
Mild	1.31 (1.26-1.37)	<.0001	1.06 (1.02-1.11)	.005	1.09 (1.04-1.14)	0.0002
Moderate	1.31 (1.27-1.35)	<.0001	1.03 (1.00-1.06)	.04	1.13 (1.10-1.17)	<.0001
Severe	1.24 (1.18-1.32)	<.0001	1.08 (1.02-1.14)	.009	1.11 (1.05-1.18)	0.0004
Psychotic	1.33 (1.23-1.45)	<.0001	1.13 (1.03-1.24)	.01	1.20 (1.09-1.32)	<.0001
Unspecified	1.30 (1.25-1.35)	<.0001	1.04 (1.01-1.08)	.02	1.11 (1.07-1.15)	<.0001
Treatment setting
Emergency	1.34 (1.29-1.39)	<.0001	1.07 (1.03-1.11)	.001	1.14 (1.10-1.18)	<.0001
Inpatient	1.33 (1.27-1.39)	<.0001	1.06 (1.01-1.11)	.01	1.13 (1.08-1.18)	<.0001
Outpatient	1.28 (1.24-1.32)	<.0001	1.04 (1.01-1.07)	.008	1.11 (1.08-1.14)	<.0001

Open in a new tab

Abbreviations: BD, bipolar disorder; HR, hazard ratio; ICD-10, International Statistical Classification of Diseases and Related Health Problems, Tenth Revision; MD, major depression; PRS, polygenic risk score; SZ, schizophrenia.

Table 3. Case-Only Analyses of the Associations of PRS-MD, PRS-BD, and PRS-SZ With Age at Onset and Severity at First Depression Diagnosis.

Characteristic	MD			BD			SZ
Characteristic	β (SE)	OR (95% CI)	P Value	β (SE)	OR (95% CI)	P Value	β (SE)	OR (95% CI)	P Value
Age at depression diagnosis	−0.05 (0.02)	NA	.04	−0.07 (0.02)	NA	.002	−0.05 (0.02)	NA	.04
ICD-10 severity specifier
Mild	Reference Category
Moderate	−0.00 (0.02)	1.00 (0.95-1.05)	.99	−0.03 (0.02)	0.97 (0.93-1.02)	.24	0.04 (0.02)	1.04 (0.99-1.09)	.08
Severe	−0.05 (0.03)	0.95 (0.89-1.01)	.13	0.02 (0.03)	1.02 (0.95-1.08)	.62	0.02 (0.03)	1.02 (0.95-1.09)	.58
Psychotic	0.02 (0.05)	1.02 (0.92-1.12)	.71	0.06 (0.05)	1.06 (0.97-1.17)	.21	0.09 (0.05)	1.10 (1.00-1.21)	.06
Unspecified	−0.01 (0.03)	0.99 (0.94-1.04)	.77	−0.02 (0.03)	0.98 (0.93-1.03)	.48	0.02 (0.03)	1.02 (0.97-1.07)	.51
Treatment setting
Outpatient	Reference Category
Emergency	0.05 (0.02)	1.05 (1.01-1.09)	.02	0.03 (0.02)	1.03 (0.99-1.07)	.18	0.03 (0.02)	1.03 (0.99-1.07)	.13
Inpatient	0.04 (0.02)	1.04 (0.99-1.09)	.13	0.02 (0.02)	1.02 (0.98-1.07)	.37	0.02 (0.02)	1.02 (0.97-1.07)	.46

Open in a new tab

Abbreviations: BD, bipolar disorder; ICD-10, International Statistical Classification of Diseases and Related Health Problems, Tenth Revision; MD, major depression; NA, not applicable; OR, odds ratio; PRS, polygenic risk score; SZ, schizophrenia.

Severity

The hazard ratios for PRS-MD, PRS-BD, and PRS-SZ were highest for psychotic depression (Figure 2, Table 2). The differences by severity were most pronounced for PRS-SZ (1.20; 95% CI, 1.09-1.32) and least pronounced for PRS-MD (1.33; 95% CI, 1.23-1.45). In the case-only analyses, none of the associations were statistically significant; however, the association between higher PRS-SZ and an increased odds of psychotic depression was suggestive (odds ratio, 1.10; 95% CI, 1.00-1.21; P = .06) (Table 3).

The hazard ratios for PRS-MD, PRS-BD, and PRS-SZ were higher for treatment in inpatient and emergency settings; however, these differences were small (Figure 2). Polygenic risk scores for MD were marginally associated with an increased odds of emergency treatment in the case-only analyses (odds ratio, 1.05; 95% CI, 1.01-1.09; P = .02) (Table 3).

Discussion

In this study, we found that PRSs trained using aggregated results from selected samples of prevalent, often recurrent depression cases contributed to the risk for first depression in the Danish general population. For each SD increase in polygenic liability, the hazard of depression increased by 30%. Compared with individuals in the bottom 10% of the polygenic liability distribution, the hazard of depression was 2.55 times higher among individuals in the top 10%. These results suggest that estimates of genetic liability ascertained using prevalent samples are tapping in to an underlying genetic predisposition for developing depression, not just a predisposition to maintain the disorder once it has been established. Polygenic liability for BD and SZ were associated with depression to a lesser extent than PRS-MD, which supports the well-documented finding that these disorders share some common genetic etiology.

The association of polygenic liability with AAO in this study was much smaller than prior findings from family studies would suggest. This could reflect that PRSs and family background are not entirely overlapping measures of genetic risk. Previous research suggests that some of the effect of family history of schizophrenia is mediated by polygenic risk.³⁴ If the same holds true for depression, the larger effects identified in family studies may be attributable to the portion of the family history effect not captured by a PRS, or possibly to an increased vigilance for psychiatric disorders among multiplex families. The oldest members of the iPSYCH2012 cohort were only age 31 years at the end of follow-up, which is around the median age of onset for depression.³⁵ As a result, the entire sample could be considered early-onset. It may be that the more pronounced differences exist in the effects of PRS-MD with the risk for depression at different points across the lifespan, but that these differences are less apparent when comparing groups of younger individuals.

We found little association between polygenic liability and greater severity at the initial depression diagnosis, which is inconsistent with recent findings from the PGC.⁴ In general, past studies with positive findings in this area have focused on severity measures that are associated with the illness course, such as the number of depressive episodes and the chronicity of depressive symptoms.^36,37,38 It could be that polygenic liability has less of an effect on characteristics of the first depressive episode than it does on the characteristics of course.

It has been suggested previously that stratifying on the phenotype may be a viable method to increase statistical power for identifying genetic variants that are associated with depression.³⁹ This method has been used with some success by the CONVERGE consortium, which identified a locus that was significantly associated with depression in Han Chinese women by selecting for highly severe, recurrent female cases.¹⁶ Polygenic risk was also found to be differentially associated with subtypes in autism,³¹ BD, and SZ.³⁰ However, for depression, greater success in gene discovery was achieved by increasing the sample size at the expense of a carefully defined phenotype.^4,40 In this vein, the results of this study indicate that the usefulness of further stratification on the phenotype for gene discovery in depression might be more limited than we may have wished.

We found some evidence of genetic heterogeneity among depression cases in terms of polygenic liabilities for BD and SZ: there was a suggestive (although not significant) association between PRS-SZ and depression with psychotic symptoms, which makes intuitive sense. The PRS for BD was significantly associated with earlier age at MD onset in the case-only analyses, and the case-cohort analyses in separate age groups suggest that PRS-BD and PRS-SZ may be particularly elevated among individuals who receive a diagnosis with MD between the ages of 10 to 15 years. These results are consistent with past studies^41,42 and could suggest that a person’s degree of genetic liability for BD or SZ may place them at an increased risk for different clinical manifestations of depression. However, psychiatric diagnoses are often unstable over time, and both early AAO and greater severity/psychotic symptoms are robust risk factors for converting to BD or SZ.^43,44,45 It is therefore possible that these results reflect that many individuals with BD and SZ receive a depression diagnosis during the early stages of their illness.

Strengths and Limitations

The iPSYCH2012 sample has many strengths, including a large sample size, population-based sampling, and a uniform case definition. The fact that cases were identified through clinical records rather than selected specifically for research increases the relevance for clinical practice. In addition, the ancestral homogeneity of the Danish population reduces the likelihood of confounding by population stratification.

However, it should be noted that although cases in the iPSYCH2012 sample are representative of individuals who received treatment for depression in psychiatric hospitals, they do not include people with depression who are untreated or only treated by general practitioners.^46,47 To put this into perspective, currently unpublished results show that most (85%) individuals who are medically treated for depression in Denmark are treated first by their primary care doctors, although this proportion was lower in younger age groups (unpublished data, 2018). The cases in the iPSYCH2012 sample therefore represent the severe end of the depression distribution in Denmark, which is both a strength and a limitation. Severe cases are likely enriched for genetic determinants¹⁶; however, the results may not generalize to milder forms of depression, and they could be biased toward the null because of misclassification. Additionally, there may be too little variation in severity to assess the associations between polygenic liability and severity in this sample. The analyses may also be subject to selection bias if people with specialty-treated depression are more likely than people with untreated or primary-care treated depression to experience recurrent episodes.

Because we focused exclusively on first depression, we did not account for subsequent diagnostic conversions to BD or SZ. Individuals who convert may have different genetic profiles; however, there was no way to account for this without conditioning on the future, which can introduce bias.⁴⁸ Further research is needed to investigate the associations between polygenic liability and the characteristics of course and outcome, including progression to other psychiatric disorders. Finally, the discovery data sets used to create the polygenic risk scores for MD, BD, and SZ had different sample sizes, which affects their statistical power.⁴⁹

Conclusions

We found that polygenic liability is associated with depression in the general Danish population. Polygenic liabilities for BD and SZ were also associated with depression, which supports the idea that there is a shared genetic predisposition across these disorders. Heterogeneity in AAO might be partially attributable to underlying genetic differences among depression cases, but these associations appear minimal.

Supplement.

eMethods. Case-cohort study design

eTable 1. Number of SNPs in the polygenic risk scores

eTable 2. Main effects of PRS-MD, PRS-BD and PRS-SZ on hazard of depression across all p(t) thresholds

e Table 3. Associations between PRS-MD, PRS-BD and PRS-SZ deciles and hazard of depression

eFigure 1. Sample selection

eFigure 2. Associations between PRS-MD and hazard of depression across all p(t) thresholds: panel graph

eFigure 3. Associations between PRS-BD and hazard of depression across all p(t) thresholds: panel graph

eFigure 4. Associations between PRS-SZ and hazard of depression across all p(t) thresholds: panel graph

eFigure 5. Associations between PRS-MD and hazard of depression across all p(t) thresholds: overlay graph

eFigure 6. Associations between PRS-BD and hazard of depression across all p(t) thresholds: overlay graph

eFigure 7. Associations between PRS-SZ and hazard of depression across all p(t) thresholds: overlay graph

eAppendix. SAS code

Click here for additional data file.^{(342.9KB, pdf)}

References

1.Sullivan PF, Neale MC, Kendler KS. Genetic epidemiology of major depression: review and meta-analysis. Am J Psychiatry. 2000;157(10):1552-1562. doi: 10.1176/appi.ajp.157.10.1552 [DOI] [PubMed] [Google Scholar]
2.Lee SH, Ripke S, Neale BM, et al. ; Cross-Disorder Group of the Psychiatric Genomics Consortium; International Inflammatory Bowel Disease Genetics Consortium (IIBDGC) . Genetic relationship between five psychiatric disorders estimated from genome-wide SNPs. Nat Genet. 2013;45(9):984-994. doi: 10.1038/ng.2711 [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Peterson RE, Cai N, Bigdeli TB, et al. . The genetic architecture of major depressive disorder in Han Chinese women. JAMA Psychiatry. 2017;74(2):162-168. doi: 10.1001/jamapsychiatry.2016.3578 [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Wray NR, Ripke S, Mattheisen M, et al. ; eQTLGen; 23andMe; Major Depressive Disorder Working Group of the Psychiatric Genomics Consortium . Genome-wide association analyses identify 44 risk variants and refine the genetic architecture of major depression. Nat Genet. 2018;50(5):668-681. doi: 10.1038/s41588-018-0090-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Sullivan PF, Daly MJ, O’Donovan M. Genetic architectures of psychiatric disorders: the emerging picture and its implications. Nat Rev Genet. 2012;13(8):537-551. doi: 10.1038/nrg3240 [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Peyrot WJ, Milaneschi Y, Abdellaoui A, et al. . Effect of polygenic risk scores on depression in childhood trauma. Br J Psychiatry. 2014;205(2):113-119. doi: 10.1192/bjp.bp.113.143081 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Chang SC, Glymour MM, Walter S, et al. . Genome-wide polygenic scoring for a 14-year long-term average depression phenotype. Brain Behav. 2014;4(2):298-311. doi: 10.1002/brb3.205 [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Musliner KL, Seifuddin F, Judy JA, Pirooznia M, Goes FS, Zandi PP. Polygenic risk, stressful life events and depressive symptoms in older adults: a polygenic score analysis. Psychol Med. 2015;45(8):1709-1720. doi: 10.1017/S0033291714002839 [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Ripke S, Wray NR, Lewis CM, et al. ; Major Depressive Disorder Working Group of the Psychiatric GWAS Consortium . A mega-analysis of genome-wide association studies for major depressive disorder. Mol Psychiatry. 2013;18(4):497-511. doi: 10.1038/mp.2012.21 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Cohen P, Cohen J. The clinician’s illusion. Arch Gen Psychiatry. 1984;41(12):1178-1182. doi: 10.1001/archpsyc.1984.01790230064010 [DOI] [PubMed] [Google Scholar]
11.Waltoft BL, Pedersen CB, Nyegaard M, Hobolth A. The importance of distinguishing between the odds ratio and the incidence rate ratio in GWAS. BMC Med Genet. 2015;16:71. doi: 10.1186/s12881-015-0210-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Weissman MM, Wickramaratne P, Gameroff MJ, et al. . Offspring of depressed parents: 30 years later. Am J Psychiatry. 2016;173(10):1024-1032. doi: 10.1176/appi.ajp.2016.15101327 [DOI] [PubMed] [Google Scholar]
13.Weissman MM, Wickramaratne P, Merikangas KR, et al. . Onset of major depression in early adulthood. increased familial loading and specificity. Arch Gen Psychiatry. 1984;41(12):1136-1143. doi: 10.1001/archpsyc.1984.01790230022003 [DOI] [PubMed] [Google Scholar]
14.Lieb R, Isensee B, Höfler M, Wittchen HU. Parental depression and depression in offspring: evidence for familial characteristics and subtypes? J Psychiatr Res. 2002;36(4):237-246. doi: 10.1016/S0022-3956(02)00015-8 [DOI] [PubMed] [Google Scholar]
15.Musliner KL, Trabjerg BB, Waltoft BL, et al. . Parental history of psychiatric diagnoses and unipolar depression: a Danish National Register-based cohort study. Psychol Med. 2015;45(13):2781-2791. doi: 10.1017/S0033291715000744 [DOI] [PMC free article] [PubMed] [Google Scholar]
16.CONVERGE consortium Sparse whole-genome sequencing identifies two loci for major depressive disorder. Nature. 2015;523(7562):588-591. doi: 10.1038/nature14659 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Pettersson E, Larsson H, Lichtenstein P. Common psychiatric disorders share the same genetic origin: a multivariate sibling study of the Swedish population. Mol Psychiatry. 2016;21(5):717-721. doi: 10.1038/mp.2015.116 [DOI] [PubMed] [Google Scholar]
18.Pedersen CB, Bybjerg-Grauholm J, Pedersen MG, et al. . The iPSYCH2012 case-cohort sample: new directions for unravelling genetic and environmental architectures of severe mental disorders. Mol Psychiatry. 2018;23(1):6-14. doi: 10.1038/mp.2017.196 [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Prentice RL. A case-cohort design for epidemiologic cohort studies and disease prevention trials. Biometrika. 1986;73(1):1-11. doi: 10.1093/biomet/73.1.1 [DOI] [Google Scholar]
20.Petersen L, Sørensen TI, Andersen PK. Comparison of case-cohort estimators based on data on premature death of adult adoptees. Stat Med. 2003;22(24):3795-3803. doi: 10.1002/sim.1672 [DOI] [PubMed] [Google Scholar]
21.Self SG, Prentice RL. Asymptotic distribution theory and efficiency results for case-cohort studies. Ann Stat. 1988;16(1):64-81. doi: 10.1214/aos/1176350691 [DOI] [Google Scholar]
22.Barlow WE. Robust variance estimation for the case-cohort design. Biometrics. 1994;50(4):1064-1072. doi: 10.2307/2533444 [DOI] [PubMed] [Google Scholar]
23.Pedersen CB. The Danish Civil Registration System. Scand J Public Health. 2011;39(7)(suppl):22-25. doi: 10.1177/1403494810387965 [DOI] [PubMed] [Google Scholar]
24.Mors O, Perto GP, Mortensen PB. The Danish Psychiatric Central Research Register. Scand J Public Health. 2011;39(7)(suppl):54-57. doi: 10.1177/1403494810395825 [DOI] [PubMed] [Google Scholar]
25.World Health Organization The ICD-10 Classification of Mental and Behavioral Disorders: Diagnostic Criteria for Research. Geneva, Switzerland: World Health Organization; 1993. [Google Scholar]
26.Nørgaard-Pedersen B, Hougaard DM. Storage policies and use of the Danish Newborn Screening Biobank. J Inherit Metab Dis. 2007;30(4):530-536. doi: 10.1007/s10545-007-0631-x [DOI] [PubMed] [Google Scholar]
27.Gunderson KL, Steemers FJ, Ren H, et al. . Whole-genome genotyping. Methods Enzymol. 2006;410:359-376. doi: 10.1016/S0076-6879(06)10017-8 [DOI] [PubMed] [Google Scholar]
28.Purcell SM, Wray NR, Stone JL, et al. ; International Schizophrenia Consortium . Common polygenic variation contributes to risk of schizophrenia and bipolar disorder. Nature. 2009;460(7256):748-752. [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Schizophrenia Working Group of the Psychiatric Genomics Consortium Biological insights from 108 schizophrenia-associated genetic loci. Nature. 2014;511(7510):421-427. doi: 10.1038/nature13595 [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Bipolar D, Schizophrenia Working Group of the Psychiatric Genomics Consortium . Genomic dissection of bipolar disorder and schizophrenia, including 28 subphenotypes. Cell. 2018;173(7):1705-1715.e1716. [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Grove J, Ripke S, Als TD, et al. . Common risk variants identified in autism spectrum disorder [published online November 17, 2017]. bioRxiv. doi: 10.1101/224774 [DOI] [Google Scholar]
32.Barlow WE, Ichikawa L, Rosner D, Izumi S. Analysis of case-cohort designs. J Clin Epidemiol. 1999;52(12):1165-1172. doi: 10.1016/S0895-4356(99)00102-X [DOI] [PubMed] [Google Scholar]
33.Pedersen CB, Mors O, Bertelsen A, et al. . A comprehensive nationwide study of the incidence rate and lifetime risk for treated mental disorders. JAMA Psychiatry. 2014;71(5):573-581. doi: 10.1001/jamapsychiatry.2014.16 [DOI] [PubMed] [Google Scholar]
34.Agerbo E, Sullivan PF, Vilhjálmsson BJ, et al. . Polygenic risk score, parental socioeconomic status, family history of psychiatric disorders, and the risk for schizophrenia: a Danish population-based study and meta-analysis. JAMA Psychiatry. 2015;72(7):635-641. doi: 10.1001/jamapsychiatry.2015.0346 [DOI] [PubMed] [Google Scholar]
35.Kessler RC, Berglund P, Demler O, Jin R, Merikangas KR, Walters EE. Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry. 2005;62(6):593-602. doi: 10.1001/archpsyc.62.6.593 [DOI] [PubMed] [Google Scholar]
36.Levine ME, Crimmins EM, Prescott CA, Phillips D, Arpawong TE, Lee J. A polygenic risk score associated with measures of depressive symptoms among older adults. Biodemography Soc Biol. 2014;60(2):199-211. doi: 10.1080/19485565.2014.952705 [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Mistry S, Harrison JR, Smith DJ, Escott-Price V, Zammit S. The use of polygenic risk scores to identify phenotypes associated with genetic risk of bipolar disorder and depression: a systematic review. J Affect Disord. 2018;234:148-155. doi: 10.1016/j.jad.2018.02.005 [DOI] [PubMed] [Google Scholar]
38.Ferentinos P, Rivera M, Ising M, et al. . Investigating the genetic variation underlying episodicity in major depressive disorder: suggestive evidence for a bipolar contribution. J Affect Disord. 2014;155:81-89. doi: 10.1016/j.jad.2013.10.027 [DOI] [PubMed] [Google Scholar]
39.Levinson DF, Mostafavi S, Milaneschi Y, et al. . Genetic studies of major depressive disorder: why are there no genome-wide association study findings and what can we do about it? Biol Psychiatry. 2014;76(7):510-512. doi: 10.1016/j.biopsych.2014.07.029 [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Hyde CL, Nagle MW, Tian C, et al. . Identification of 15 genetic loci associated with risk of major depression in individuals of European descent. Nat Genet. 2016;48(9):1031-1036. doi: 10.1038/ng.3623 [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Verduijn J, Milaneschi Y, Peyrot WJ, et al. . Using clinical characteristics to identify which patients with major depressive disorder have a higher genetic load for three psychiatric disorders. Biol Psychiatry. 2017;81(4):316-324. doi: 10.1016/j.biopsych.2016.05.024 [DOI] [PubMed] [Google Scholar]
42.Power RA, Tansey KE, Buttenschøn HN, et al. ; CONVERGE Consortium, Cardiogram Consortium, GERAD1 Consortium . Genome-wide association for major depression through age at onset stratification: major depressive disorder working group of the psychiatric genomics consortium. Biol Psychiatry. 2017;81(4):325-335. doi: 10.1016/j.biopsych.2016.05.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Musliner KL, Munk-Olsen T, Mors O, Østergaard SD. Progression from unipolar depression to schizophrenia. Acta Psychiatr Scand. 2017;135(1):42-50. doi: 10.1111/acps.12663 [DOI] [PubMed] [Google Scholar]
44.Musliner KL, Østergaard SD. Patterns and predictors of conversion to bipolar disorder in 91 587 individuals diagnosed with unipolar depression. Acta Psychiatr Scand. 2018;137(5):422-432. doi: 10.1111/acps.12869 [DOI] [PubMed] [Google Scholar]
45.Ratheesh A, Davey C, Hetrick S, et al. . A systematic review and meta-analysis of prospective transition from major depression to bipolar disorder. Acta Psychiatr Scand. 2017;135(4):273-284. doi: 10.1111/acps.12686 [DOI] [PubMed] [Google Scholar]
46.Kuramoto-Crawford SJ, Han B, Jacobus-Kantor L, Mojtabai R. Receipt of depression treatment from general medical providers and specialty mental health providers. Psychiatr Serv. 2016;67(7):758-765. doi: 10.1176/appi.ps.201500113 [DOI] [PubMed] [Google Scholar]
47.Thornicroft G, Chatterji S, Evans-Lacko S, et al. . Undertreatment of people with major depressive disorder in 21 countries. Br J Psychiatry. 2017;210(2):119-124. doi: 10.1192/bjp.bp.116.188078 [DOI] [PMC free article] [PubMed] [Google Scholar]
48.Andersen PK, Keiding N. Interpretability and importance of functionals in competing risks and multistate models. Stat Med. 2012;31(11-12):1074-1088. doi: 10.1002/sim.4385 [DOI] [PubMed] [Google Scholar]
49.Dudbridge F. Power and predictive accuracy of polygenic risk scores. PLoS Genet. 2013;9(3):e1003348. doi: 10.1371/journal.pgen.1003348 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement.

eMethods. Case-cohort study design

eTable 1. Number of SNPs in the polygenic risk scores

eTable 2. Main effects of PRS-MD, PRS-BD and PRS-SZ on hazard of depression across all p(t) thresholds

e Table 3. Associations between PRS-MD, PRS-BD and PRS-SZ deciles and hazard of depression

eFigure 1. Sample selection

eFigure 2. Associations between PRS-MD and hazard of depression across all p(t) thresholds: panel graph

eFigure 3. Associations between PRS-BD and hazard of depression across all p(t) thresholds: panel graph

eFigure 4. Associations between PRS-SZ and hazard of depression across all p(t) thresholds: panel graph

eFigure 5. Associations between PRS-MD and hazard of depression across all p(t) thresholds: overlay graph

eFigure 6. Associations between PRS-BD and hazard of depression across all p(t) thresholds: overlay graph

eFigure 7. Associations between PRS-SZ and hazard of depression across all p(t) thresholds: overlay graph

eAppendix. SAS code

Click here for additional data file.^{(342.9KB, pdf)}

[yoi180103r1] 1.Sullivan PF, Neale MC, Kendler KS. Genetic epidemiology of major depression: review and meta-analysis. Am J Psychiatry. 2000;157(10):1552-1562. doi: 10.1176/appi.ajp.157.10.1552 [DOI] [PubMed] [Google Scholar]

[yoi180103r2] 2.Lee SH, Ripke S, Neale BM, et al. ; Cross-Disorder Group of the Psychiatric Genomics Consortium; International Inflammatory Bowel Disease Genetics Consortium (IIBDGC) . Genetic relationship between five psychiatric disorders estimated from genome-wide SNPs. Nat Genet. 2013;45(9):984-994. doi: 10.1038/ng.2711 [DOI] [PMC free article] [PubMed] [Google Scholar]

[yoi180103r3] 3.Peterson RE, Cai N, Bigdeli TB, et al. . The genetic architecture of major depressive disorder in Han Chinese women. JAMA Psychiatry. 2017;74(2):162-168. doi: 10.1001/jamapsychiatry.2016.3578 [DOI] [PMC free article] [PubMed] [Google Scholar]

[yoi180103r4] 4.Wray NR, Ripke S, Mattheisen M, et al. ; eQTLGen; 23andMe; Major Depressive Disorder Working Group of the Psychiatric Genomics Consortium . Genome-wide association analyses identify 44 risk variants and refine the genetic architecture of major depression. Nat Genet. 2018;50(5):668-681. doi: 10.1038/s41588-018-0090-3 [DOI] [PMC free article] [PubMed] [Google Scholar]

[yoi180103r5] 5.Sullivan PF, Daly MJ, O’Donovan M. Genetic architectures of psychiatric disorders: the emerging picture and its implications. Nat Rev Genet. 2012;13(8):537-551. doi: 10.1038/nrg3240 [DOI] [PMC free article] [PubMed] [Google Scholar]

[yoi180103r6] 6.Peyrot WJ, Milaneschi Y, Abdellaoui A, et al. . Effect of polygenic risk scores on depression in childhood trauma. Br J Psychiatry. 2014;205(2):113-119. doi: 10.1192/bjp.bp.113.143081 [DOI] [PMC free article] [PubMed] [Google Scholar]

[yoi180103r7] 7.Chang SC, Glymour MM, Walter S, et al. . Genome-wide polygenic scoring for a 14-year long-term average depression phenotype. Brain Behav. 2014;4(2):298-311. doi: 10.1002/brb3.205 [DOI] [PMC free article] [PubMed] [Google Scholar]

[yoi180103r8] 8.Musliner KL, Seifuddin F, Judy JA, Pirooznia M, Goes FS, Zandi PP. Polygenic risk, stressful life events and depressive symptoms in older adults: a polygenic score analysis. Psychol Med. 2015;45(8):1709-1720. doi: 10.1017/S0033291714002839 [DOI] [PMC free article] [PubMed] [Google Scholar]

[yoi180103r9] 9.Ripke S, Wray NR, Lewis CM, et al. ; Major Depressive Disorder Working Group of the Psychiatric GWAS Consortium . A mega-analysis of genome-wide association studies for major depressive disorder. Mol Psychiatry. 2013;18(4):497-511. doi: 10.1038/mp.2012.21 [DOI] [PMC free article] [PubMed] [Google Scholar]

[yoi180103r10] 10.Cohen P, Cohen J. The clinician’s illusion. Arch Gen Psychiatry. 1984;41(12):1178-1182. doi: 10.1001/archpsyc.1984.01790230064010 [DOI] [PubMed] [Google Scholar]

[yoi180103r11] 11.Waltoft BL, Pedersen CB, Nyegaard M, Hobolth A. The importance of distinguishing between the odds ratio and the incidence rate ratio in GWAS. BMC Med Genet. 2015;16:71. doi: 10.1186/s12881-015-0210-1 [DOI] [PMC free article] [PubMed] [Google Scholar]

[yoi180103r12] 12.Weissman MM, Wickramaratne P, Gameroff MJ, et al. . Offspring of depressed parents: 30 years later. Am J Psychiatry. 2016;173(10):1024-1032. doi: 10.1176/appi.ajp.2016.15101327 [DOI] [PubMed] [Google Scholar]

[yoi180103r13] 13.Weissman MM, Wickramaratne P, Merikangas KR, et al. . Onset of major depression in early adulthood. increased familial loading and specificity. Arch Gen Psychiatry. 1984;41(12):1136-1143. doi: 10.1001/archpsyc.1984.01790230022003 [DOI] [PubMed] [Google Scholar]

[yoi180103r14] 14.Lieb R, Isensee B, Höfler M, Wittchen HU. Parental depression and depression in offspring: evidence for familial characteristics and subtypes? J Psychiatr Res. 2002;36(4):237-246. doi: 10.1016/S0022-3956(02)00015-8 [DOI] [PubMed] [Google Scholar]

[yoi180103r15] 15.Musliner KL, Trabjerg BB, Waltoft BL, et al. . Parental history of psychiatric diagnoses and unipolar depression: a Danish National Register-based cohort study. Psychol Med. 2015;45(13):2781-2791. doi: 10.1017/S0033291715000744 [DOI] [PMC free article] [PubMed] [Google Scholar]

[yoi180103r16] 16.CONVERGE consortium Sparse whole-genome sequencing identifies two loci for major depressive disorder. Nature. 2015;523(7562):588-591. doi: 10.1038/nature14659 [DOI] [PMC free article] [PubMed] [Google Scholar]

[yoi180103r17] 17.Pettersson E, Larsson H, Lichtenstein P. Common psychiatric disorders share the same genetic origin: a multivariate sibling study of the Swedish population. Mol Psychiatry. 2016;21(5):717-721. doi: 10.1038/mp.2015.116 [DOI] [PubMed] [Google Scholar]

[yoi180103r18] 18.Pedersen CB, Bybjerg-Grauholm J, Pedersen MG, et al. . The iPSYCH2012 case-cohort sample: new directions for unravelling genetic and environmental architectures of severe mental disorders. Mol Psychiatry. 2018;23(1):6-14. doi: 10.1038/mp.2017.196 [DOI] [PMC free article] [PubMed] [Google Scholar]

[yoi180103r19] 19.Prentice RL. A case-cohort design for epidemiologic cohort studies and disease prevention trials. Biometrika. 1986;73(1):1-11. doi: 10.1093/biomet/73.1.1 [DOI] [Google Scholar]

[yoi180103r20] 20.Petersen L, Sørensen TI, Andersen PK. Comparison of case-cohort estimators based on data on premature death of adult adoptees. Stat Med. 2003;22(24):3795-3803. doi: 10.1002/sim.1672 [DOI] [PubMed] [Google Scholar]

[yoi180103r21] 21.Self SG, Prentice RL. Asymptotic distribution theory and efficiency results for case-cohort studies. Ann Stat. 1988;16(1):64-81. doi: 10.1214/aos/1176350691 [DOI] [Google Scholar]

[yoi180103r22] 22.Barlow WE. Robust variance estimation for the case-cohort design. Biometrics. 1994;50(4):1064-1072. doi: 10.2307/2533444 [DOI] [PubMed] [Google Scholar]

[yoi180103r23] 23.Pedersen CB. The Danish Civil Registration System. Scand J Public Health. 2011;39(7)(suppl):22-25. doi: 10.1177/1403494810387965 [DOI] [PubMed] [Google Scholar]

[yoi180103r24] 24.Mors O, Perto GP, Mortensen PB. The Danish Psychiatric Central Research Register. Scand J Public Health. 2011;39(7)(suppl):54-57. doi: 10.1177/1403494810395825 [DOI] [PubMed] [Google Scholar]

[yoi180103r25] 25.World Health Organization The ICD-10 Classification of Mental and Behavioral Disorders: Diagnostic Criteria for Research. Geneva, Switzerland: World Health Organization; 1993. [Google Scholar]

[yoi180103r26] 26.Nørgaard-Pedersen B, Hougaard DM. Storage policies and use of the Danish Newborn Screening Biobank. J Inherit Metab Dis. 2007;30(4):530-536. doi: 10.1007/s10545-007-0631-x [DOI] [PubMed] [Google Scholar]

[yoi180103r27] 27.Gunderson KL, Steemers FJ, Ren H, et al. . Whole-genome genotyping. Methods Enzymol. 2006;410:359-376. doi: 10.1016/S0076-6879(06)10017-8 [DOI] [PubMed] [Google Scholar]

[yoi180103r28] 28.Purcell SM, Wray NR, Stone JL, et al. ; International Schizophrenia Consortium . Common polygenic variation contributes to risk of schizophrenia and bipolar disorder. Nature. 2009;460(7256):748-752. [DOI] [PMC free article] [PubMed] [Google Scholar]

[yoi180103r29] 29.Schizophrenia Working Group of the Psychiatric Genomics Consortium Biological insights from 108 schizophrenia-associated genetic loci. Nature. 2014;511(7510):421-427. doi: 10.1038/nature13595 [DOI] [PMC free article] [PubMed] [Google Scholar]

[yoi180103r30] 30.Bipolar D, Schizophrenia Working Group of the Psychiatric Genomics Consortium . Genomic dissection of bipolar disorder and schizophrenia, including 28 subphenotypes. Cell. 2018;173(7):1705-1715.e1716. [DOI] [PMC free article] [PubMed] [Google Scholar]

[yoi180103r31] 31.Grove J, Ripke S, Als TD, et al. . Common risk variants identified in autism spectrum disorder [published online November 17, 2017]. bioRxiv. doi: 10.1101/224774 [DOI] [Google Scholar]

[yoi180103r32] 32.Barlow WE, Ichikawa L, Rosner D, Izumi S. Analysis of case-cohort designs. J Clin Epidemiol. 1999;52(12):1165-1172. doi: 10.1016/S0895-4356(99)00102-X [DOI] [PubMed] [Google Scholar]

[yoi180103r33] 33.Pedersen CB, Mors O, Bertelsen A, et al. . A comprehensive nationwide study of the incidence rate and lifetime risk for treated mental disorders. JAMA Psychiatry. 2014;71(5):573-581. doi: 10.1001/jamapsychiatry.2014.16 [DOI] [PubMed] [Google Scholar]

[yoi180103r34] 34.Agerbo E, Sullivan PF, Vilhjálmsson BJ, et al. . Polygenic risk score, parental socioeconomic status, family history of psychiatric disorders, and the risk for schizophrenia: a Danish population-based study and meta-analysis. JAMA Psychiatry. 2015;72(7):635-641. doi: 10.1001/jamapsychiatry.2015.0346 [DOI] [PubMed] [Google Scholar]

[yoi180103r35] 35.Kessler RC, Berglund P, Demler O, Jin R, Merikangas KR, Walters EE. Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry. 2005;62(6):593-602. doi: 10.1001/archpsyc.62.6.593 [DOI] [PubMed] [Google Scholar]

[yoi180103r36] 36.Levine ME, Crimmins EM, Prescott CA, Phillips D, Arpawong TE, Lee J. A polygenic risk score associated with measures of depressive symptoms among older adults. Biodemography Soc Biol. 2014;60(2):199-211. doi: 10.1080/19485565.2014.952705 [DOI] [PMC free article] [PubMed] [Google Scholar]

[yoi180103r37] 37.Mistry S, Harrison JR, Smith DJ, Escott-Price V, Zammit S. The use of polygenic risk scores to identify phenotypes associated with genetic risk of bipolar disorder and depression: a systematic review. J Affect Disord. 2018;234:148-155. doi: 10.1016/j.jad.2018.02.005 [DOI] [PubMed] [Google Scholar]

[yoi180103r38] 38.Ferentinos P, Rivera M, Ising M, et al. . Investigating the genetic variation underlying episodicity in major depressive disorder: suggestive evidence for a bipolar contribution. J Affect Disord. 2014;155:81-89. doi: 10.1016/j.jad.2013.10.027 [DOI] [PubMed] [Google Scholar]

[yoi180103r39] 39.Levinson DF, Mostafavi S, Milaneschi Y, et al. . Genetic studies of major depressive disorder: why are there no genome-wide association study findings and what can we do about it? Biol Psychiatry. 2014;76(7):510-512. doi: 10.1016/j.biopsych.2014.07.029 [DOI] [PMC free article] [PubMed] [Google Scholar]

[yoi180103r40] 40.Hyde CL, Nagle MW, Tian C, et al. . Identification of 15 genetic loci associated with risk of major depression in individuals of European descent. Nat Genet. 2016;48(9):1031-1036. doi: 10.1038/ng.3623 [DOI] [PMC free article] [PubMed] [Google Scholar]

[yoi180103r41] 41.Verduijn J, Milaneschi Y, Peyrot WJ, et al. . Using clinical characteristics to identify which patients with major depressive disorder have a higher genetic load for three psychiatric disorders. Biol Psychiatry. 2017;81(4):316-324. doi: 10.1016/j.biopsych.2016.05.024 [DOI] [PubMed] [Google Scholar]

[yoi180103r42] 42.Power RA, Tansey KE, Buttenschøn HN, et al. ; CONVERGE Consortium, Cardiogram Consortium, GERAD1 Consortium . Genome-wide association for major depression through age at onset stratification: major depressive disorder working group of the psychiatric genomics consortium. Biol Psychiatry. 2017;81(4):325-335. doi: 10.1016/j.biopsych.2016.05.010 [DOI] [PMC free article] [PubMed] [Google Scholar]

[yoi180103r43] 43.Musliner KL, Munk-Olsen T, Mors O, Østergaard SD. Progression from unipolar depression to schizophrenia. Acta Psychiatr Scand. 2017;135(1):42-50. doi: 10.1111/acps.12663 [DOI] [PubMed] [Google Scholar]

[yoi180103r44] 44.Musliner KL, Østergaard SD. Patterns and predictors of conversion to bipolar disorder in 91 587 individuals diagnosed with unipolar depression. Acta Psychiatr Scand. 2018;137(5):422-432. doi: 10.1111/acps.12869 [DOI] [PubMed] [Google Scholar]

[yoi180103r45] 45.Ratheesh A, Davey C, Hetrick S, et al. . A systematic review and meta-analysis of prospective transition from major depression to bipolar disorder. Acta Psychiatr Scand. 2017;135(4):273-284. doi: 10.1111/acps.12686 [DOI] [PubMed] [Google Scholar]

[yoi180103r46] 46.Kuramoto-Crawford SJ, Han B, Jacobus-Kantor L, Mojtabai R. Receipt of depression treatment from general medical providers and specialty mental health providers. Psychiatr Serv. 2016;67(7):758-765. doi: 10.1176/appi.ps.201500113 [DOI] [PubMed] [Google Scholar]

[yoi180103r47] 47.Thornicroft G, Chatterji S, Evans-Lacko S, et al. . Undertreatment of people with major depressive disorder in 21 countries. Br J Psychiatry. 2017;210(2):119-124. doi: 10.1192/bjp.bp.116.188078 [DOI] [PMC free article] [PubMed] [Google Scholar]

[yoi180103r48] 48.Andersen PK, Keiding N. Interpretability and importance of functionals in competing risks and multistate models. Stat Med. 2012;31(11-12):1074-1088. doi: 10.1002/sim.4385 [DOI] [PubMed] [Google Scholar]

[yoi180103r49] 49.Dudbridge F. Power and predictive accuracy of polygenic risk scores. PLoS Genet. 2013;9(3):e1003348. doi: 10.1371/journal.pgen.1003348 [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Association of Polygenic Liabilities for Major Depression, Bipolar Disorder, and Schizophrenia With Risk for Depression in the Danish Population

Katherine L Musliner, PhD

Preben B Mortensen, DrMedSc

John J McGrath, PhD

Nis P Suppli, MD, PhD

David M Hougaard, DrMedSc

Jonas Bybjerg-Grauholm, MSE

Marie Bækvad-Hansen, PhD

Ole Andreassen, MD

Carsten B Pedersen, PhD

Marianne G Pedersen, MSc

Ole Mors, MD

Merete Nordentoft, PhD

Anders D Børglum, MD, PhD

Thomas Werge, PhD

Esben Agerbo, PhD

Key Points

Question

Findings

Meaning

Abstract

Importance

Objective

Design, Setting, and Participants

Exposures

Main Outcomes and Measures

Results

Conclusions and Relevance

Introduction

Methods

Study Design

Sample Selection

Genotyping

Measures

Statistical Analysis

Results

Sample Characteristics

Table 1. Sample Characteristics.

Polygenic Liability and Hazard of Depression

Figure 1. Polygenic Liability for Major Depressive Disorder (MD), Bipolar Disorder (BD), and Schizophrenia (SZ) and Hazard of Depression in the iPSYCH2012 Cohort.

AAO

Figure 2. Polygenic Liability for Major Depressive Disorder (MD), Bipolar Disorder (BD), and Schizophrenia (SZ) and Hazard of Depression by Age and Severity.

Table 2. Associations Between PRS-MD, PRS-BD, and PRS-SZ and Hazard of Depression Within Different Age at Onset and Severity Subgroups.

Table 3. Case-Only Analyses of the Associations of PRS-MD, PRS-BD, and PRS-SZ With Age at Onset and Severity at First Depression Diagnosis.

Severity

Discussion

Strengths and Limitations

Conclusions

References

Associated Data

Supplementary Materials

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases