Family History of Schizophrenia and Bipolar Disorder As Risk Factors for Autism
Total Page:16
File Type:pdf, Size:1020Kb
ORIGINAL ARTICLE Family History of Schizophrenia and Bipolar Disorder as Risk Factors for Autism Patrick F. Sullivan, MD, FRANZCP; Cecilia Magnusson, MD, PhD; Abraham Reichenberg, PhD; Marcus Boman, BS; Christina Dalman, MD, PhD; Michael Davidson, MD; Eyal Fruchter, MD; Christina M. Hultman, PhD; Michael Lundberg, MPH; Niklas La˚ngstro¨m, MD, PhD; Mark Weiser, MD; Anna C. Svensson, PhD; Paul Lichtenstein, PhD Context: The clinical and etiologic relation between au- Results: The presence of schizophrenia in parents was as- tism spectrum disorders (ASDs) and schizophrenia is un- sociated with an increased risk for ASD in a Swedish na- clear. The degree to which these disorders share a basis tional cohort (odds ratio [OR], 2.9; 95% CI, 2.5-3.4) and in etiology has important implications for clinicians, re- a Stockholm County cohort (OR, 2.9; 95% CI, 2.0-4.1). searchers, and those affected by the disorders. Similarly, schizophrenia in a sibling was associated with an increased risk for ASD in a Swedish national cohort Objective: To determine whether a family history of (OR, 2.6; 95% CI, 2.0-3.2) and an Israeli conscription co- schizophrenia and/or bipolar disorder is a risk factor hort (OR, 12.1; 95% CI, 4.5-32.0). Bipolar disorder showed for ASD. a similar pattern of associations but of lesser magnitude. Design, Setting, and Participants: We conducted Conclusions: Findings from these 3 registers along with a case-control evaluation of histories of schizophrenia consistent findings from a similar study in Denmark sug- or bipolar disorder in first-degree relatives of probands gest that ASD, schizophrenia, and bipolar disorder share in 3 samples—population registers in Sweden, Stock- common etiologic factors. holm County (in Sweden), and Israel. Probands met criteria for ASD, and affection status of parents and Arch Gen Psychiatry. 2012;69(11):1099-1103. siblings for schizophrenia and bipolar disorder were Published online July 2, 2012. established. doi:10.1001/archgenpsychiatry.2012.730 UTISM SPECTRUM DISOR- are important overlaps between ASD and ders (ASDs) and schizo- schizophrenia. Some family history stud- phrenia are currently con- ies8,9 (although not all)10 found that the rela- sidered as distinctive and tives of probands with ASD were more likely Author Affiliations: infrequently overlap- to have a history of schizophrenia. How- Author Affil 1,2 Department of Genetics, Aping. Historically, ASD was often re- ever, these studies tended to be small and Department University of North Carolina at garded as childhood schizophrenia be- relatively underpowered. Childhood- University o Chapel Hill (Dr Sullivan); cause the impaired social interactions and onset schizophrenia is a rare subtype, and Chapel Hill Department of Medical bizarre behavior found in ASD were remi- a sizable fraction have premorbid ASD.11 Department Epidemiology and Biostatistics niscent of symptoms of schizophrenia.3 In- More directly, genomewide copy number Epidemiolog (Drs Sullivan, Hultman, (Drs Sullivan La˚ngstro¨m, and Lichtenstein, deed, the psychiatrist who coined the term variation studies have identified rare mu- La˚ngstro¨m, a and Mr Boman), and schizophrenia counted autism (an active tations that are strong risk factors for and Mr Bom Department of Public Health turning away from the external world) as both ASD and schizophrenia (eg, 15q13.3, Department Sciences (Drs Magnusson, an important distinguishing feature of 16p11.2, 22q11.21, and in neurexin 1).12-15 Sciences (Dr Dalman, and Svensson, and schizophrenia.4,5 Around 1980, the noso- These points of overlap are quite uncom- Dalman, and Mr Lundberg), Karolinska logic status of ASD and schizophrenia was mon, and they apply to only a fraction of Mr Lundberg Institutet, Stockholm, Sweden; Institutet, St revised so that these disorders were sepa- clinical samples. Institute of Psychiatry, King’s 6 Institute of P College London, London, rated. The separation was strongly influ- The degree to which ASD and schizo- College Lond England (Dr Reichenberg); enced by developmental trajectory and de- phrenia share etiologic factors has impor- England (Dr Department of Psychiatry, Sheba lineated infantile autism present from very tant implications for clinicians, research- Department Medical Center and Tel-Aviv early in life7 from schizophrenia where psy- ers, and those affected by these diseases. Medical Cen University, Tel-Aviv chotic symptoms developed after an ex- Therefore, we investigated whether a fam- University, T (Dr Davidson and Weiser), and (Dr Davidso Department of Mental Health, tended period of normal or near-normal ily history of schizophrenia and/or bipo- Department Israeli Defense Force Medical development. lar disorder in first-degree relatives was a Israeli Defen Corp, Ramat Gan Several lines of evidence suggest that this risk factor for ASD. Bipolar disorder was Corp, Ramat (Dr Fruchter), Israel. distinction is not absolute, and that there included given its etiologic and clinical (Dr Fruchter ARCH GEN PSYCHIATRY/ VOL 69 (NO. 11), NOV 2012 WWW.ARCHGENPSYCHIATRY.COM 1099 ©2012 American Medical Association. All rights reserved. overlap with schizophrenia.16,17 We conducted parallel Study 2 (Stockholm inpatient and outpatient) was also con- analyses in 3 samples to evaluate the consistency and gen- ducted using Swedish registries.25 The base for study 2 was all eralizability of the findings. youth aged 0 to 17 years registered in Stockholm County from 1984 through 2007 (N=589 114). Cases were drawn from health services registers from Stockholm County and they had DSM-IV METHODS diagnosis of pervasive development disorder in the Stockholm County Council Child and Adolescent Mental Health Service The overall goal of this study was to evaluate the impact of the Register26; treatment at the Autism Centre for Young Chil- exposure of family history of schizophrenia and bipolar disor- dren, Asperger Centre, or Autism Centre within Stockholm der (psychotic disorders) on the outcome of ASD. This rela- County Council Handicap and Habilitation services; or NPR tion was evaluated in 3 complementary samples. Study 1 has diagnosis of ASD (ICD-9 code 299 or ICD-10 code F84). Be- Swedish national coverage of inpatient and outpatient admis- cause mental retardation is a source of heterogeneity in ASD,27 sions. Study 2 is based on an inclusive set of primarily outpa- cases were also stratified by the presence or absence of clini- tient ASD treatment facilities but limited to the largest popu- cally significant mental retardation (ie, treatment at clinical cen- lation center in Sweden (Stockholm County). Study 3 uses ters specifically for ASD comorbid with mental retardation, a standardized psychiatric assessment of conscripts in Israel. Au- diagnosis of mental retardation in any register, or ASD sub- tism spectrum disorder outcomes were assessed using register type other than Asperger syndrome). We randomly selected 10 data, and exposures were measured using register data on par- control relative pairs for each case. Control participants also ents (studies 1 and 2) or siblings (studies 1 and 3). These stud- resided in Stockholm County, had no evidence of ASD (ie, no ies were approved by research ethics committees at Karolin- treatment or NPR diagnosis), and were matched to cases by year ska Institutet and Sheba Medical Center. of birth and sex as well as by the year of birth of the father and Study 1 (Sweden National Patient Register [NPR]) was mother. The main exposures were schizophrenia (ICD-8 codes conducted using Swedish national registers. An analysis of 295.0-295.9, ICD-9 codes 295A-295X, and ICD-10 code F20) 27% of these cases was published previously.8 The primary or bipolar disorder (ICD-8 codes 296.0-296.9, ICD-9 codes 296A- key for register linkage was the unique personal identification 296W, and ICD-10 code F31) in the parents of cases and con- number assigned to each Swedish citizen at birth or upon ar- trol participants. These exposures were assessed using the NPR rival in Sweden for immigrants.18 The NPR19,20 contains dis- supplemented by psychiatric outpatient treatment registers avail- charge diagnoses for all inpatient (since 1973) and outpatient able in Stockholm County. The statistical analysis paralleled (since 2001) psychiatric treatment in Sweden including ad- that used in study 1. missions for assessment or treatment to any psychiatric or Study 3 (Israeli conscripts) included all Jewish persons born general medical hospital (including forensic psychiatric hos- in Israel who are required to participate in an assessment at age pitals and the few private providers of inpatient health care). 17 years prior to compulsory military service. Participation rates Biological relationships were established using the Multi- are 98% for males and 75% for females (women adhering to Generation Register,21 which identifies the relatives of an in- Orthodox Judaism are exempted).28 The preinduction assess- dex person through linkage of a child to his or her parents and ment determines intellectual, medical, and psychiatric eligi- includes all individuals born in Sweden since 1932, those reg- bility for compulsory military service. The draft board registry istered as living in Sweden after 1960, and immigrants who data were used to obtain the diagnostic outcomes for ASD pro- became Swedish citizens before age 18 years. Vital status was bands and their siblings at age 17 years. Psychiatric diagnostic defined using the national Cause of Death Register. Cases procedures are described elsewhere, and