Sibling Recurrence Risk and Cross-aggregation of Attention-Deficit/Hyperactivity Disorder and Autism Spectrum Disorder | Attention Deficit/Hyperactivity Disorders | JAMA Pediatrics | JAMA Network
[Skip to Content]
[Skip to Content Landing]
Table 1.  Sample Characteristics of Later-Born Siblings Based on Familial Risk Status
Sample Characteristics of Later-Born Siblings Based on Familial Risk Status
Table 2.  Hierarchical Generalized Linear Model Results for Familial Risk Status
Hierarchical Generalized Linear Model Results for Familial Risk Status
1.
Faraone  SV, Perlis  RH, Doyle  AE,  et al.  Molecular genetics of attention-deficit/hyperactivity disorder.  Biol Psychiatry. 2005;57(11):1313-1323. doi:10.1016/j.biopsych.2004.11.024PubMedGoogle ScholarCrossref
2.
Tick  B, Bolton  P, Happé  F, Rutter  M, Rijsdijk  F.  Heritability of autism spectrum disorders: a meta-analysis of twin studies.  J Child Psychol Psychiatry. 2016;57(5):585-595. doi:10.1111/jcpp.12499PubMedGoogle ScholarCrossref
3.
Willcutt  EG.  The prevalence of DSM-IV attention-deficit/hyperactivity disorder: a meta-analytic review.  Neurotherapeutics. 2012;9(3):490-499. doi:10.1007/s13311-012-0135-8PubMedGoogle ScholarCrossref
4.
Christensen  DL, Baio  J, Van Naarden Braun  K,  et al; Centers for Disease Control and Prevention (CDC).  Prevalence and characteristics of autism spectrum disorder among children aged 8 years—Autism and Developmental Disabilities Monitoring Network, 11 sites, United States, 2012.  MMWR Surveill Summ. 2016;65(3):1-23. doi:10.15585/mmwr.ss6503a1PubMedGoogle ScholarCrossref
5.
Musser  ED, Hawkey  E, Kachan-Liu  SS,  et al.  Shared familial transmission of autism spectrum and attention-deficit/hyperactivity disorders.  J Child Psychol Psychiatry. 2014;55(7):819-827. doi:10.1111/jcpp.12201PubMedGoogle ScholarCrossref
6.
Ghirardi  L, Brikell  I, Kuja-Halkola  R,  et al.  The familial co-aggregation of ASD and ADHD: a register-based cohort study.  Mol Psychiatry. 2018;23(2):257-262. doi:10.1038/mp.2017.17PubMedGoogle ScholarCrossref
7.
Rommelse  NNJ, Franke  B, Geurts  HM, Hartman  CA, Buitelaar  JK.  Shared heritability of attention-deficit/hyperactivity disorder and autism spectrum disorder.  Eur Child Adolesc Psychiatry. 2010;19(3):281-295. doi:10.1007/s00787-010-0092-xPubMedGoogle ScholarCrossref
8.
Anttila  V, Bulik-Sullivan  B, Finucane  HK,  et al; Brainstorm Consortium.  Analysis of shared heritability in common disorders of the brain.  Science. 2018;360(6395):pii:eaap8757.PubMedGoogle ScholarCrossref
9.
Stergiakouli  E, Davey Smith  G, Martin  J,  et al.  Shared genetic influences between dimensional ASD and ADHD symptoms during child and adolescent development.  Mol Autism. 2017;8(1):18. doi:10.1186/s13229-017-0131-2PubMedGoogle ScholarCrossref
10.
Martin  J, Cooper  M, Hamshere  ML,  et al.  Biological overlap of attention-deficit/hyperactivity disorder and autism spectrum disorder: evidence from copy number variants.  J Am Acad Child Adolesc Psychiatry. 2014;53(7):761-70.e26. doi:10.1016/j.jaac.2014.03.004PubMedGoogle ScholarCrossref
11.
Kim  DS, Burt  AA, Ranchalis  JE,  et al; University of Washington Center for Mendelian Genomics.  Sequencing of sporadic attention-deficit hyperactivity disorder (ADHD) identifies novel and potentially pathogenic de novo variants and excludes overlap with genes associated with autism spectrum disorder.  Am J Med Genet B Neuropsychiatr Genet. 2017;174(4):381-389. doi:10.1002/ajmg.b.32527PubMedGoogle ScholarCrossref
12.
Ray  S, Miller  M, Karalunas  S,  et al.  Structural and functional connectivity of the human brain in autism spectrum disorders and attention-deficit/hyperactivity disorder: a rich club-organization study.  Hum Brain Mapp. 2014;35(12):6032-6048. doi:10.1002/hbm.22603PubMedGoogle ScholarCrossref
13.
Karalunas  SL, Hawkey  E, Gustafsson  H,  et al.  Overlapping and distinct cognitive impairments in attention-deficit/hyperactivity and autism spectrum disorder without intellectual disability.  J Abnorm Child Psychol. 2018;46(8):1705-1716. doi:10.1007/s10802-017-0394-2PubMedGoogle ScholarCrossref
14.
Feczko  E, Balba  NM, Miranda-Dominguez  O,  et al.  Subtyping cognitive profiles in autism spectrum disorder using a functional random forest algorithm.  Neuroimage. 2018;172:674-688. doi:10.1016/j.neuroimage.2017.12.044PubMedGoogle ScholarCrossref
15.
Hoffmann  TJ, Windham  GC, Anderson  M, Croen  LA, Grether  JK, Risch  N.  Evidence of reproductive stoppage in families with autism spectrum disorder: a large, population-based cohort study.  JAMA Psychiatry. 2014;71(8):943-951. doi:10.1001/jamapsychiatry.2014.420PubMedGoogle ScholarCrossref
16.
Wood  CL, Warnell  F, Johnson  M,  et al.  Evidence for ASD recurrence rates and reproductive stoppage from large UK ASD research family databases.  Autism Res. 2015;8(1):73-81. doi:10.1002/aur.1414PubMedGoogle ScholarCrossref
17.
Ozonoff  S, Young  GS, Carter  A,  et al.  Recurrence risk for autism spectrum disorders: a Baby Siblings Research Consortium study.  Pediatrics. 2011;128(3):e488-e495.PubMedGoogle Scholar
18.
Chen  W, Zhou  K, Sham  P,  et al.  DSM-IV combined type ADHD shows familial association with sibling trait scores: a sampling strategy for QTL linkage.  Am J Med Genet B Neuropsychiatr Genet. 2008;147B(8):1450-1460. doi:10.1002/ajmg.b.30672PubMedGoogle ScholarCrossref
19.
Chen  Q, Brikell  I, Lichtenstein  P,  et al.  Familial aggregation of attention-deficit/hyperactivity disorder.  J Child Psychol Psychiatry. 2017;58(3):231-239. doi:10.1111/jcpp.12616PubMedGoogle ScholarCrossref
20.
Jokiranta-Olkoniemi  E, Cheslack-Postava  K, Sucksdorff  D,  et al.  Risk of psychiatric and neurodevelopmental disorders among siblings of probands with autism spectrum disorders.  JAMA Psychiatry. 2016;73(6):622-629. doi:10.1001/jamapsychiatry.2016.0495PubMedGoogle ScholarCrossref
21.
Visser  SN, Danielson  ML, Bitsko  RH,  et al.  Trends in the parent-report of health care provider-diagnosed and medicated attention-deficit/hyperactivity disorder: United States, 2003-2011.  J Am Acad Child Adolesc Psychiatry. 2014;53(1):34-46.e2. doi:10.1016/j.jaac.2013.09.001PubMedGoogle ScholarCrossref
22.
R Core Team.  R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2015.
23.
Bates  D, Mächler  M, Bolker  BM, Walker  SC.  Fitting linear mixed-effects models using lme4.  J Stat Softw. 2015;67(1):1-48. doi:10.18637/jss.v067.i01Google ScholarCrossref
24.
Biederman  J, Faraone  SV, Mick  E,  et al.  High risk for attention deficit hyperactivity disorder among children of parents with childhood onset of the disorder: a pilot study.  Am J Psychiatry. 1995;152(3):431-435. doi:10.1176/ajp.152.3.431PubMedGoogle ScholarCrossref
25.
Williams  NM, Franke  B, Mick  E,  et al.  Genome-wide analysis of copy number variants in attention deficit hyperactivity disorder: the role of rare variants and duplications at 15q13. 3.  Am J Psychiatry. 2012;169(2):195-204.PubMedGoogle Scholar
26.
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.2711PubMedGoogle ScholarCrossref
27.
American Psychiatric Association.  Diagnostic and Statistical Manual of Mental Disorders. 5th ed. Washington, DC: American Psychiatric Association; 2013.
28.
Danielson  ML, Bitsko  RH, Ghandour  RM, Holbrook  JR, Kogan  MD, Blumberg  SJ.  Prevalence of parent-reported ADHD diagnosis and associated treatment among U.S. children and adolescents, 2016.  J Clin Child Adolesc Psychol. 2018;47(2):199-212. doi:10.1080/15374416.2017.1417860PubMedGoogle ScholarCrossref
29.
Constantino  JN, Zhang  Y, Frazier  T, Abbacchi  AM, Law  P.  Sibling recurrence and the genetic epidemiology of autism.  Am J Psychiatry. 2010;167(11):1349-1356. doi:10.1176/appi.ajp.2010.09101470PubMedGoogle ScholarCrossref
30.
Grønborg  TK, Schendel  DE, Parner  ET.  Recurrence of autism spectrum disorders in full- and half-siblings and trends over time: a population-based cohort study.  JAMA Pediatr. 2013;167(10):947-953. doi:10.1001/jamapediatrics.2013.2259PubMedGoogle ScholarCrossref
31.
Risch  N, Hoffmann  TJ, Anderson  M, Croen  LA, Grether  JK, Windham  GC.  Familial recurrence of autism spectrum disorder: evaluating genetic and environmental contributions.  Am J Psychiatry. 2014;171(11):1206-1213. doi:10.1176/appi.ajp.2014.13101359PubMedGoogle ScholarCrossref
32.
Lord  C, Bishop  SL.  Recent advances in autism research as reflected in DSM-5 criteria for autism spectrum disorder.  Annu Rev Clin Psychol. 2015;11(1):53-70. doi:10.1146/annurev-clinpsy-032814-112745PubMedGoogle ScholarCrossref
33.
Polanczyk  GV, Salum  GA, Sugaya  LS, Caye  A, Rohde  LA.  Annual research review: a meta-analysis of the worldwide prevalence of mental disorders in children and adolescents.  J Child Psychol Psychiatry. 2015;56(3):345-365. doi:10.1111/jcpp.12381PubMedGoogle ScholarCrossref
34.
Martin  J, Walters  RK, Demontis  D,  et al; 23andMe Research Team; Psychiatric Genomics Consortium: ADHD Subgroup; iPSYCH–Broad ADHD Workgroup.  A genetic investigation of sex bias in the prevalence of attention-deficit/hyperactivity disorder.  Biol Psychiatry. 2018;83(12):1044-1053. doi:10.1016/j.biopsych.2017.11.026PubMedGoogle ScholarCrossref
35.
Chang  Z, Lichtenstein  P, D’Onofrio  BM,  et al.  Maternal age at childbirth and risk for ADHD in offspring: a population-based cohort study.  Int J Epidemiol. 2014;43(6):1815-1824. doi:10.1093/ije/dyu204PubMedGoogle ScholarCrossref
36.
Sandin  S, Hultman  CM, Kolevzon  A, Gross  R, MacCabe  JH, Reichenberg  A.  Advancing maternal age is associated with increasing risk for autism: a review and meta-analysis.  J Am Acad Child Adolesc Psychiatry. 2012;51(5):477-486.e1. doi:10.1016/j.jaac.2012.02.018PubMedGoogle ScholarCrossref
Original Investigation
December 10, 2018

Sibling Recurrence Risk and Cross-aggregation of Attention-Deficit/Hyperactivity Disorder and Autism Spectrum Disorder

Author Affiliations
  • 1MIND Institute, Department of Psychiatry & Behavioral Sciences and University of California, Davis, Sacramento
  • 2Department of Psychology, Florida International University, Miami
  • 3Marshfield Clinic Research Institute, Marshfield, Wisconsin
  • 4Department of Pediatrics, School of Medicine and Public Health, University of Wisconsin, Marshfield
  • 5Department of Psychiatry, Oregon Health & Science University, Portland
  • 6Department of Behavioral Neuroscience, Oregon Health & Science University, Portland
JAMA Pediatr. 2019;173(2):147-152. doi:10.1001/jamapediatrics.2018.4076
Key Points

Question  What are the rates of within-diagnosis sibling recurrence and sibling cross-aggregation of attention-deficit/hyperactivity disorder and autism spectrum disorder among later-born siblings of diagnosed children?

Findings  In this population-based study of 15 175 US children, compared with later-born siblings of nondiagnosed children, later-born siblings of those with autism spectrum disorder were more likely to be diagnosed with autism spectrum disorder or attention-deficit/hyperactivity disorder. In addition, compared with later-born siblings of nondiagnosed children, later-born siblings of children with attention-deficit/hyperactivity disorder were more likely to be diagnosed with attention-deficit/hyperactivity disorder or autism spectrum disorder.

Meaning  Later-born siblings of children with autism spectrum disorder appear to be at elevated risk of attention-deficit/hyperactivity disorder and vice versa with implications for etiologic overlap and clinical monitoring.

Abstract

Importance  Attention-deficit/hyperactivity disorder (ADHD) and autism spectrum disorder (ASD) are believed to partially share genetic factors and biological influences. As the number of children with these diagnoses rises, so does the number of younger siblings at presumed risk for ADHD and ASD; reliable recurrence risk estimates within and across diagnoses may aid screening and early detection efforts and enhance understanding of potential shared causes.

Objective  To examine within-diagnosis sibling recurrence risk and sibling cross-aggregation of ADHD and ASD among later-born siblings of children with either disorder.

Design, Setting, and Participants  Using data extracted from medical records of 2 large health care systems in the United States, estimates of recurrence risk and cross-aggregation in later-born siblings of children with ADHD or ASD were compared with later-born siblings of children without these diagnoses. One data set included children seen between January 1, 1995, and December 31, 2013; the other included children born between January 1, 1998, and May 17, 2010. Participants included 15 175 later-born siblings of children with ADHD, ASD, and no known diagnosis. The study was conducted from October 2, 2017, to August 14, 2018.

Main Outcomes and Measures  Diagnoses of ASD or ADHD in the later-born sibling, ascertained from medical records, were the primary outcomes of interest; moderators included sex, gestational age, and maternal age.

Results  A total of 15 175 later-born siblings were classified by familial risk status based on the older child’s diagnostic status: ADHD risk (n = 730; male [51.92%]), ASD risk (n = 158; male [48.10%]), and no known risk (n = 14 287; male [50.73%]). Compared with later-born siblings of children without ADHD or ASD, later-born siblings of children with ASD were more likely to be diagnosed with ASD (odds ratio [OR], 30.38; 95% CI, 17.73-52.06) or ADHD in the absence of ASD (OR, 3.70; 95% CI, 1.67-8.21). Compared with later-born siblings of children without a diagnosis, later-born siblings of children with ADHD were more likely to be diagnosed with ADHD (OR, 13.05; 95% CI, 9.86-17.27) or ASD in the absence of ADHD (OR, 4.35; 95% CI, 2.43-7.79).

Conclusions and Relevance  Later-born siblings of children with ASD or ADHD appear to be at elevated risk for the same disorder, but also of being diagnosed with the other disorder. These findings provide further support for shared familial mechanisms underlying ASD and ADHD, which may be useful for genetic and prospective developmental studies. Later-born siblings of children with ADHD or ASD should be monitored for both conditions.

Introduction

Attention-deficit/hyperactivity disorder (ADHD) and autism spectrum disorder (ASD) are 2 common and highly heritable1,2 neurodevelopmental disorders.3,4 Evidence of their shared heritability is mixed, with familial and twin studies suggesting an association,5-7 but molecular studies calling this into question.8 These disorders appear to share some genetic factors9 and biological pathways,10 although divergences in their neurobiological, cognitive, and genetic profiles are apparent.11-13 Although some findings may be accounted for by the co-occurrence of ADHD and ASD13 and others by heterogeneity within the 2 conditions,14 it remains unclear how related these disorders are etiologically. It is important to further characterize the degree of shared etiologic risk across these impairing, prevalent, and complex conditions.

Recurrence risk among siblings is a common way to measure shared genetic contributions. Recurrence risk estimates in families, not to be confused with recurrence within an individual, are also helpful clinically (eg, screening, genetic counseling). However, numerous confounders can obscure recurrence risk estimates for familial disorders. The most important confounder, and the one addressed herein, is that these estimates are often based on the total number of siblings within a family rather than being limited to later-born siblings and can therefore be influenced by reproductive stoppage. That is, if families decide to have fewer children after a child develops ADHD or ASD, risk is underestimated. This limitation can be addressed by examining sibling recurrence in families who have had at least 1 child after the diagnosis of ADHD or ASD in an older child.15,16 Previous studies focused only on later-born siblings of children with ASD have documented higher—and potentially more accurate—rates of recurrence of ASD16,17 than those relying on other methods. To our knowledge, none have used the later-born sibling method to examine recurrence risk of ADHD, although studies examining concordance among all siblings of children with ADHD, irrespective of birth order, have documented elevated rates of ADHD.18,19

The crucial issue of shared etiologic factors, however, concerns cross-aggregation of ADHD and ASD among later-born siblings. This, to our knowledge, has not been studied, even though ADHD rates generally are elevated in siblings of children with ASD,6,20 and higher rates of ASD have been reported in first-degree relatives of individuals with ADHD.5 As implied above, prior reports documenting familial cross-aggregation—a potential proxy for shared genetic influences—could also be biased by stoppage. This study sought to close this gap in the literature.

Using population-based medical records, our goals were to examine the degree of familial aggregation in ASD and ADHD by (1) replicating prior studies of recurrence risk of ASD among later-born siblings of children with ASD, (2) providing what we believe to be the first estimate of recurrence risk of ADHD among later-born siblings of children with ADHD, and (3) providing initial estimates of familial cross-aggregation of recurrence of both disorders among later-born siblings.

Methods
Data Extraction and Participants

Deidentified data were extracted from medical records of the Marshfield Clinic, a large regional health care system in the upper Midwest, and Kaiser Pacific Northwest, an integrated, group-model, nonprofit health plan in Oregon and Washington. Patient bases of both health care systems are demographically representative of the service areas in which they are located (approximately 95% and 80% white, respectively). The institutional review boards of both health care systems approved this study and no written informed consent was needed. Data extraction and participant identification processes are depicted in the eFigure in the Supplement.

The full data set extracted from Marshfield Clinic included all children aged 0 to 18 years seen between January 1, 1995, and December 31, 2013 (n = 10 522). The full data set from Kaiser Pacific Northwest included all children born to parents with the health plan between January 1, 1998, and May 17, 2010 (n = 46 032). The present study was conducted from October 2, 2017, to August 14, 2018. Both systems included a maternal identifier variable for the identification of siblings. Each data set was originally extracted to address a unique, separate set of hypotheses from those addressed herein, resulting in the following overlapping variables examined in this study: child ADHD diagnosis, child ASD diagnosis, child biological sex, maternal age at delivery, and gestation duration. To reduce false-positives, child diagnosis was indicated when International Classification of Diseases, Ninth Revision–Clinical Modification, codes 314.00 or 314.01 for ADHD or 299.00, 299.80, or 299.81 for ASD were recorded at least twice in the medical record. Both practices completed universal ASD screenings for children aged 18 to 24 months; those screening positive underwent a multidisciplinary evaluation. If a child was identified as having both ASD and ADHD, the ASD diagnosis was considered primary for analyses; at the time these children were evaluated, Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, did not include dual diagnosis of ASD and ADHD, although some children had both documented. Children with neither ADHD nor ASD were considered nondiagnosed comparison youths.

In selecting the final sample, all children younger than 5 years at the time of data extraction were excluded. We reasoned that this criterion was a sensible compromise to avoid most false-negatives, given the median age of diagnosis of ASD (4.2 years4) and ADHD (6.2 years21). Families with 2 or more siblings, all aged 5 years or older, were included in the final data set, given the necessity of identifying later-born (ie, younger) siblings’ familial risk status by way of an earlier-born (ie, older) sibling’s diagnosis. If there were no other siblings within the family, twin pairs were treated as singletons and excluded since neither could be considered later-born; otherwise, 1 twin was randomly selected for inclusion.

Proband and Later-Born Sibling Identification

An earlier-born (older) sibling proband from each family was identified to determine each later-born sibling’s familial risk status. We use the term familial risk throughout to conform with convention while acknowledging that it does not take into account nonsibling familial risk. Probands were considered to be either the first (ie, oldest) diagnosed child in affected families or, among families with no children with either diagnosis, the earliest-born child. More specifically, in 2-child families, the proband was always the earlier-born child. In families with 3 or more children, the ADHD or ASD proband was considered to be the oldest child with the diagnosis. That is, if a middle child was the oldest child identified with ASD or ADHD, that child was treated as the proband and earlier-born children were excluded, because the perceived liability for later-born siblings changed when the middle child was diagnosed.

Later-born siblings were classified into 1 of 3 groups based on their earlier-born sibling’s diagnosis: ADHD risk (ie, later-born siblings of a proband with ADHD), ASD risk (ie, later-born siblings of a proband with ASD), and no known risk (ie, later-born siblings of a proband with no known diagnosis and no other earlier-born siblings with a diagnosis of ADHD or ASD).

After applying these criteria, the final sample consisted of 26 871 individual children (n = 875 with ADHD; n = 223 with ASD) born to 11 696 different mothers. Of these 11 696 mothers, 9004 (77.0%) contributed data for 2 children, 2100 (18.0%) contributed data for 3 children, 425 (3.6%) contributed data for 4 children, and 167 (1.4%) contributed data for 5 or more children. Of the 26 871 children, 559 (2.1%) were identified as ADHD probands, 126 (0.5%) as ASD probands, and 11 011 (41.0%) as nondiagnosed probands. The remaining 15 175 later-born siblings were classified by familial risk status based on the proband’s diagnostic status: 730 (4.8%) with ADHD risk, 158 (1.0%) with ASD risk, and 14 287 (94.1%) with no known risk. Table 1 reports sample characteristics.

Statistical Analysis

We first conducted 3 separate hierarchical generalized linear models for binomial variables to examine the diagnostic outcomes of ADHD, ASD, or no diagnosis in later-born siblings. Models focused on estimating the probability of diagnostic outcomes as a function of familial risk status (ie, no known risk, ASD risk, ADHD risk). In the logistic model examining ADHD outcomes, the comparison group included children with no diagnosis as well as those with ASD. Conversely, for the model examining ASD outcomes, the comparison group included children with no diagnosis as well as those with ADHD. The third logistic model involved a comparison between children with either ASD or ADHD (ie, diagnosed cases) vs children with no diagnosis. Random effects for site and family (to account for younger siblings from the same family)—functionally similar to covariance—were included.

Next, in a second set of analyses, we added to the models described above and, as a sensitivity analysis, examined, individually, later-born sibling sex, proband sex, gestational age, and maternal age as additional explanatory variables above and beyond familial risk status and whether these additional variables moderated the association between familial risk status and later-born sibling diagnosis. Main and interaction effects between familial risk status and putative moderators, as well as the interaction between proband sex and later-born sibling sex, were evaluated by comparing fit statistics between successive nested models with and without the effect of interest (where the difference of log-likelihood ratios was tested using χ2 analysis). All tests were 2-tailed with α = .05. Tests were unpaired given that there were no repeated measures. Analyses were conducted in R, version, 3.2.0,22 using the glmer function from the lme4 package.23

Results
Outcome Diagnoses Among Risk Groups

Across all children in the sample (ie, later-born siblings and probands), the incidence of ASD was 0.8% and of ADHD was 3.3%. Results revealed an association with familial risk status for each model (P < .001). Absolute percentages, odds ratios (ORs), and 95% CIs are presented in Table 2.

With respect to a diagnostic outcome of ASD compared with non-ASD, the odds of an ASD diagnosis were approximately 30 times higher (OR, 30.38; 95% CI, 17.73-52.06) for the ASD-risk siblings than the no-known-risk siblings and almost 7 times higher (OR, 6.99; 95% CI, 3.42-14.27) than the ADHD-risk siblings. Although the specificity of risk to ASD confirms some separation of etiologic factors, the odds of an ASD diagnosis were more than 4 times higher (OR, 4.35; 95% CI, 2.43-7.79) for the ADHD-risk siblings than the no-known-risk siblings, suggesting partially shared underpinnings of ADHD and ASD.

For an outcome diagnosis of ADHD, the odds were 13 times higher (OR, 13.05; 95% CI, 9.86-17.27) for the ADHD-risk siblings compared with the no-known-risk siblings and, compared with the ASD-risk siblings, the odds of an ADHD diagnosis were approximately 3.5 times higher (OR, 3.53; 95% CI, 1.57-7.94), again supporting partial specificity of risk. Yet, the odds of an ADHD diagnosis were more than 3.5 times higher (OR, 3.70; 95% CI, 1.67-8.21) for the ASD-risk siblings compared with the no-known-risk siblings, again suggesting partially overlapping familial risk.

The third model, examining the prevalence of either ASD or ADHD among risk groups (ASD-risk vs no known risk: OR, 12.96; 95% CI, 8.17-20.55; ADHD-risk vs no known risk: OR, 10.90; 95% CI, 8.44-14.08), confirmed the greater likelihood of either outcome for both elevated risk groups, but revealed no difference in the odds of having either diagnosis between ASD-risk and ADHD-risk siblings (OR, 0.84; 95% CI, 0.52-1.36), suggesting similar degrees of liability in these elevated risk groups.

Thus, our hypotheses were supported: later-born siblings of children with ASD had higher odds of being diagnosed with ASD or ADHD compared with no-known-risk siblings, and later-born siblings of children with ADHD had higher odds of being diagnosed with ADHD or ASD compared with no-known-risk siblings. Liability for either diagnosis did not differ between risk groups. A small number of later-born siblings had both ADHD and ASD (n = 19). Results did not change when removing comorbid cases from analyses; they were therefore retained.

Moderators of Outcome Diagnosis

As a sensitivity analysis, we next considered whether later-born sibling sex, proband sex, later-born sibling gestational age, and/or maternal age predicted diagnostic outcomes beyond familial risk status, and whether these variables moderated the predictive effect of familial risk status.

For the model examining ASD outcomes, only the main effect of later-born sibling sex was significant; the odds of an ASD diagnosis were 5.65 times higher (95% CI, 3.24-9.85) for boys than girls. No other main or interaction effects were significantly related to ASD outcomes.

For the model examining ADHD outcomes, the effect of later-born sibling sex was again seen, with the odds of an ADHD diagnosis 2.60 times higher (95% CI, 2.05-3.31) for boys than girls. There was a significant main effect for proband sex, with odds of an ADHD diagnosis 1.34 times higher (95% CI, 1.07-1.68) for later-born siblings of female probands than male probands. The interaction between proband sex and later-born sibling sex was marginally significant such that male siblings of female probands were at greatest risk for ADHD (male vs female siblings of female probands: OR, 3.37; 95% CI, 1.48-7.67). Maternal age had an additive effect and interacted with familial risk status; as maternal age increased from 20 to 50 years, the probability of ADHD diagnosis decreased for the no-known-risk (from 2.8% to 0.25%) and ADHD-risk groups (from 18.5% to 13.3%) but increased for the ASD-risk group (from 4.0% to 10.4%). There was no association with gestational age.

Discussion

We examined sibling recurrence risk for ASD and ADHD in a population sample. Analyses were restricted to families with at least 1 child born after the proband with a diagnosis, addressing a key methodologic problem with prior investigations (ie, stoppage).

As expected, we found elevated rates of ASD recurrence in later-born siblings of children with ASD, replicating prior work; 12.03% of the later-born siblings of children with ASD were diagnosed with ASD vs 0.45% of the later-born siblings of nondiagnosed probands. This estimate among the later-born siblings of children with ASD is somewhat lower than estimates from prospective infant sibling designs,17 possibly owing to their use of direct assessment and standard diagnostic procedures to which we did not have access, or to sampling biases (eg, medical record data vs volunteer enrollment). Nonetheless, our findings confirm elevated odds of ASD among later-born siblings of children with ASD.

Also as expected, later-born siblings of children with ADHD were at heightened risk for developing ADHD; 12.47% of the later-born siblings of children with ADHD were diagnosed with ADHD vs 1.53% of the later-born siblings of nondiagnosed probands. To our knowledge, no prior work has examined rates of ADHD among later-born siblings of children with ADHD. However, our finding mirrors prior research documenting higher rates of ADHD among family members of diagnosed individuals5,18,24 while addressing the issue of reproductive stoppage. These findings suggest the potential utility of prospective family risk designs to ADHD, while also providing more useful estimates of recurrence risk for genetic counseling.

The most novel finding concerns later-born sibling cross-aggregation. In addition to being at greater risk for ASD, later-born siblings of children with ASD were also more likely than no-known-risk siblings to be diagnosed with ADHD. Similarly, later-born siblings of children with ADHD were at heightened risk for ASD compared with children at no known risk. The results provide 2 types of validation. The fact that ASD was more common than ADHD in the siblings of ASD probands and ADHD was more common than ASD in the siblings of ADHD probands supports the distinction between these disorders. Conversely, the excess cross-aggregation in siblings indicates shared etiologic (familial) factors. This finding is consistent with numerous studies using genetically informed methods that have suggested shared genetic influences on both ASD and ADHD traits and diagnoses,5,6,9 and several that have highlighted shared copy number variants,10,25 although other studies have not identified genetic overlap.8,26

Results from our models collapsing across outcome diagnosis also suggest similar degrees of neurodevelopmental liability between these elevated risk groups, with implications for nosology and in support of the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5), decision to place ADHD in the neurodevelopmental category with ASD.27

We found a relatively low incidence of both ASD (0.8%) and ADHD (3.3%) in our full sample compared with population identification rates.4,28 This level may be due to our conservative definitions of ADHD and ASD: cases were counted as positive only if identified at 2 or more points. We thus hope to have avoided false-positives, but potentially at the expense of false-negatives. Although the recurrence risk estimate for ASD in the present study is lower than estimates from prospective studies on high-risk infants, it is similar to estimates from studies using comparable methods (ie, medical records).29-31 Recent changes in diagnostic criteria for ASD were also not captured since data were extracted before the publication of DSM-5; changes in diagnostic definitions may have influenced prevalence rates.32 Likewise, the identification rate seen here for ADHD is similar to the most sophisticated estimates from meta-analyses of ADHD’s true prevalence,33 suggesting our case selection method was reasonable.

With respect to moderators, our findings are generally consistent with previous work. Boys were at increased risk for both ADHD and ASD, which is well documented.4,28 Later-born siblings of girls with ADHD were marginally more likely to receive an ADHD diagnosis than later-born siblings of male probands, echoing recent population-based work.34 Maternal age interacted with risk status as expected for ADHD outcomes in the ADHD-risk and no-known-risk groups,35 although not for ASD outcomes.36 Overall, the variables examined generally did not moderate recurrence risk effects.

Limitations

This study has limitations. Although this was a population sample, it was not epidemiologically ascertained, nor was it necessarily ascertained before parental concerns about the later-born siblings. This limitation is mitigated somewhat by our decision to use samples of all children born to parents within the health plan or seen in a general clinic setting vs psychiatry or mental health clinics. We also did not have information regarding half- vs full-sibling status, potentially leading to underestimation of familial effects, nor did we have full information on parental psychopathology, comorbidities, birth complications, or other potential confounders, all of which are worthy of consideration. Whether reproductive stoppage or other related factors (eg, interpregnancy interval) are differentially influenced by ASD vs ADHD is beyond the scope of the present study but is an important future direction for research. We could not account for comorbidity between ASD and ADHD because of small cell sizes and issues surrounding prior diagnostic criteria that prohibited comorbid diagnoses. This lack of information is unlikely to account for the ADHD-to-ASD effect or the ADHD-to-ADHD effect, but we are unable to assess how it might alter the ASD-to-ADHD transmission observed. Finally, data were drawn from general medical records, so clinicians did not necessarily use standardized diagnostic procedures. However, the relatively low identification rates achieved by our requiring 2 or more instances of a given diagnosis, and the fact that these correspond relatively well to true prevalence estimates, suggest that this approach did not result in much overidentification.

Conclusions

Overall, the results of this study enhance etiologic and clinical understanding. Etiologically, our findings of sibling cross-aggregation are consistent with partially shared genetic mechanisms underlying ASD and ADHD. Thus, results may be useful for future genetic studies and prospective, genetically informed research. Clinically, later-born siblings of children with ASD and ADHD appear to be at elevated risk within and across diagnostic categories and thus should be monitored for both disorders. This finding replicates prior work in ASD17 and provides, to our knowledge, the first estimate of recurrence risk for ADHD among later-born siblings. Practitioners may wish to share such information with families given the potential relevance of monitoring social communication, attention, and behavior regulation skills in later-born siblings of children with ASD or ADHD.

Back to top
Article Information

Accepted for Publication: September 12, 2018.

Corresponding Author: Meghan Miller, PhD, MIND Institute, University of California Davis Health System, 2825 50th St, Sacramento, CA 95817 (mrhmiller@ucdavis.edu).

Published Online: December 10, 2018. doi:10.1001/jamapediatrics.2018.4076

Author Contributions: Dr Miller 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: Miller, Musser, Young, Steiner, Nigg.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Miller, Musser, Young, Nigg.

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

Statistical analysis: Miller, Young.

Obtained funding: Nigg.

Administrative, technical, or material support: Miller, Young.

Supervision: Miller, Nigg.

Conflict of Interest Disclosures: None reported.

Funding/Support: This research was supported in part by grants from the National Institute of Mental Health R00 MH106642 (Dr Miller), R03 MH110812 (Dr Musser), and R37 MH059105 (Dr Nigg).

Role of the Funder/Sponsor: The National Institute of Mental Health 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: Katrina Goddard, PhD (Kaiser Permanente Center for Health Research), contributed data to this project. No compensation was received.

References
1.
Faraone  SV, Perlis  RH, Doyle  AE,  et al.  Molecular genetics of attention-deficit/hyperactivity disorder.  Biol Psychiatry. 2005;57(11):1313-1323. doi:10.1016/j.biopsych.2004.11.024PubMedGoogle ScholarCrossref
2.
Tick  B, Bolton  P, Happé  F, Rutter  M, Rijsdijk  F.  Heritability of autism spectrum disorders: a meta-analysis of twin studies.  J Child Psychol Psychiatry. 2016;57(5):585-595. doi:10.1111/jcpp.12499PubMedGoogle ScholarCrossref
3.
Willcutt  EG.  The prevalence of DSM-IV attention-deficit/hyperactivity disorder: a meta-analytic review.  Neurotherapeutics. 2012;9(3):490-499. doi:10.1007/s13311-012-0135-8PubMedGoogle ScholarCrossref
4.
Christensen  DL, Baio  J, Van Naarden Braun  K,  et al; Centers for Disease Control and Prevention (CDC).  Prevalence and characteristics of autism spectrum disorder among children aged 8 years—Autism and Developmental Disabilities Monitoring Network, 11 sites, United States, 2012.  MMWR Surveill Summ. 2016;65(3):1-23. doi:10.15585/mmwr.ss6503a1PubMedGoogle ScholarCrossref
5.
Musser  ED, Hawkey  E, Kachan-Liu  SS,  et al.  Shared familial transmission of autism spectrum and attention-deficit/hyperactivity disorders.  J Child Psychol Psychiatry. 2014;55(7):819-827. doi:10.1111/jcpp.12201PubMedGoogle ScholarCrossref
6.
Ghirardi  L, Brikell  I, Kuja-Halkola  R,  et al.  The familial co-aggregation of ASD and ADHD: a register-based cohort study.  Mol Psychiatry. 2018;23(2):257-262. doi:10.1038/mp.2017.17PubMedGoogle ScholarCrossref
7.
Rommelse  NNJ, Franke  B, Geurts  HM, Hartman  CA, Buitelaar  JK.  Shared heritability of attention-deficit/hyperactivity disorder and autism spectrum disorder.  Eur Child Adolesc Psychiatry. 2010;19(3):281-295. doi:10.1007/s00787-010-0092-xPubMedGoogle ScholarCrossref
8.
Anttila  V, Bulik-Sullivan  B, Finucane  HK,  et al; Brainstorm Consortium.  Analysis of shared heritability in common disorders of the brain.  Science. 2018;360(6395):pii:eaap8757.PubMedGoogle ScholarCrossref
9.
Stergiakouli  E, Davey Smith  G, Martin  J,  et al.  Shared genetic influences between dimensional ASD and ADHD symptoms during child and adolescent development.  Mol Autism. 2017;8(1):18. doi:10.1186/s13229-017-0131-2PubMedGoogle ScholarCrossref
10.
Martin  J, Cooper  M, Hamshere  ML,  et al.  Biological overlap of attention-deficit/hyperactivity disorder and autism spectrum disorder: evidence from copy number variants.  J Am Acad Child Adolesc Psychiatry. 2014;53(7):761-70.e26. doi:10.1016/j.jaac.2014.03.004PubMedGoogle ScholarCrossref
11.
Kim  DS, Burt  AA, Ranchalis  JE,  et al; University of Washington Center for Mendelian Genomics.  Sequencing of sporadic attention-deficit hyperactivity disorder (ADHD) identifies novel and potentially pathogenic de novo variants and excludes overlap with genes associated with autism spectrum disorder.  Am J Med Genet B Neuropsychiatr Genet. 2017;174(4):381-389. doi:10.1002/ajmg.b.32527PubMedGoogle ScholarCrossref
12.
Ray  S, Miller  M, Karalunas  S,  et al.  Structural and functional connectivity of the human brain in autism spectrum disorders and attention-deficit/hyperactivity disorder: a rich club-organization study.  Hum Brain Mapp. 2014;35(12):6032-6048. doi:10.1002/hbm.22603PubMedGoogle ScholarCrossref
13.
Karalunas  SL, Hawkey  E, Gustafsson  H,  et al.  Overlapping and distinct cognitive impairments in attention-deficit/hyperactivity and autism spectrum disorder without intellectual disability.  J Abnorm Child Psychol. 2018;46(8):1705-1716. doi:10.1007/s10802-017-0394-2PubMedGoogle ScholarCrossref
14.
Feczko  E, Balba  NM, Miranda-Dominguez  O,  et al.  Subtyping cognitive profiles in autism spectrum disorder using a functional random forest algorithm.  Neuroimage. 2018;172:674-688. doi:10.1016/j.neuroimage.2017.12.044PubMedGoogle ScholarCrossref
15.
Hoffmann  TJ, Windham  GC, Anderson  M, Croen  LA, Grether  JK, Risch  N.  Evidence of reproductive stoppage in families with autism spectrum disorder: a large, population-based cohort study.  JAMA Psychiatry. 2014;71(8):943-951. doi:10.1001/jamapsychiatry.2014.420PubMedGoogle ScholarCrossref
16.
Wood  CL, Warnell  F, Johnson  M,  et al.  Evidence for ASD recurrence rates and reproductive stoppage from large UK ASD research family databases.  Autism Res. 2015;8(1):73-81. doi:10.1002/aur.1414PubMedGoogle ScholarCrossref
17.
Ozonoff  S, Young  GS, Carter  A,  et al.  Recurrence risk for autism spectrum disorders: a Baby Siblings Research Consortium study.  Pediatrics. 2011;128(3):e488-e495.PubMedGoogle Scholar
18.
Chen  W, Zhou  K, Sham  P,  et al.  DSM-IV combined type ADHD shows familial association with sibling trait scores: a sampling strategy for QTL linkage.  Am J Med Genet B Neuropsychiatr Genet. 2008;147B(8):1450-1460. doi:10.1002/ajmg.b.30672PubMedGoogle ScholarCrossref
19.
Chen  Q, Brikell  I, Lichtenstein  P,  et al.  Familial aggregation of attention-deficit/hyperactivity disorder.  J Child Psychol Psychiatry. 2017;58(3):231-239. doi:10.1111/jcpp.12616PubMedGoogle ScholarCrossref
20.
Jokiranta-Olkoniemi  E, Cheslack-Postava  K, Sucksdorff  D,  et al.  Risk of psychiatric and neurodevelopmental disorders among siblings of probands with autism spectrum disorders.  JAMA Psychiatry. 2016;73(6):622-629. doi:10.1001/jamapsychiatry.2016.0495PubMedGoogle ScholarCrossref
21.
Visser  SN, Danielson  ML, Bitsko  RH,  et al.  Trends in the parent-report of health care provider-diagnosed and medicated attention-deficit/hyperactivity disorder: United States, 2003-2011.  J Am Acad Child Adolesc Psychiatry. 2014;53(1):34-46.e2. doi:10.1016/j.jaac.2013.09.001PubMedGoogle ScholarCrossref
22.
R Core Team.  R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2015.
23.
Bates  D, Mächler  M, Bolker  BM, Walker  SC.  Fitting linear mixed-effects models using lme4.  J Stat Softw. 2015;67(1):1-48. doi:10.18637/jss.v067.i01Google ScholarCrossref
24.
Biederman  J, Faraone  SV, Mick  E,  et al.  High risk for attention deficit hyperactivity disorder among children of parents with childhood onset of the disorder: a pilot study.  Am J Psychiatry. 1995;152(3):431-435. doi:10.1176/ajp.152.3.431PubMedGoogle ScholarCrossref
25.
Williams  NM, Franke  B, Mick  E,  et al.  Genome-wide analysis of copy number variants in attention deficit hyperactivity disorder: the role of rare variants and duplications at 15q13. 3.  Am J Psychiatry. 2012;169(2):195-204.PubMedGoogle Scholar
26.
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.2711PubMedGoogle ScholarCrossref
27.
American Psychiatric Association.  Diagnostic and Statistical Manual of Mental Disorders. 5th ed. Washington, DC: American Psychiatric Association; 2013.
28.
Danielson  ML, Bitsko  RH, Ghandour  RM, Holbrook  JR, Kogan  MD, Blumberg  SJ.  Prevalence of parent-reported ADHD diagnosis and associated treatment among U.S. children and adolescents, 2016.  J Clin Child Adolesc Psychol. 2018;47(2):199-212. doi:10.1080/15374416.2017.1417860PubMedGoogle ScholarCrossref
29.
Constantino  JN, Zhang  Y, Frazier  T, Abbacchi  AM, Law  P.  Sibling recurrence and the genetic epidemiology of autism.  Am J Psychiatry. 2010;167(11):1349-1356. doi:10.1176/appi.ajp.2010.09101470PubMedGoogle ScholarCrossref
30.
Grønborg  TK, Schendel  DE, Parner  ET.  Recurrence of autism spectrum disorders in full- and half-siblings and trends over time: a population-based cohort study.  JAMA Pediatr. 2013;167(10):947-953. doi:10.1001/jamapediatrics.2013.2259PubMedGoogle ScholarCrossref
31.
Risch  N, Hoffmann  TJ, Anderson  M, Croen  LA, Grether  JK, Windham  GC.  Familial recurrence of autism spectrum disorder: evaluating genetic and environmental contributions.  Am J Psychiatry. 2014;171(11):1206-1213. doi:10.1176/appi.ajp.2014.13101359PubMedGoogle ScholarCrossref
32.
Lord  C, Bishop  SL.  Recent advances in autism research as reflected in DSM-5 criteria for autism spectrum disorder.  Annu Rev Clin Psychol. 2015;11(1):53-70. doi:10.1146/annurev-clinpsy-032814-112745PubMedGoogle ScholarCrossref
33.
Polanczyk  GV, Salum  GA, Sugaya  LS, Caye  A, Rohde  LA.  Annual research review: a meta-analysis of the worldwide prevalence of mental disorders in children and adolescents.  J Child Psychol Psychiatry. 2015;56(3):345-365. doi:10.1111/jcpp.12381PubMedGoogle ScholarCrossref
34.
Martin  J, Walters  RK, Demontis  D,  et al; 23andMe Research Team; Psychiatric Genomics Consortium: ADHD Subgroup; iPSYCH–Broad ADHD Workgroup.  A genetic investigation of sex bias in the prevalence of attention-deficit/hyperactivity disorder.  Biol Psychiatry. 2018;83(12):1044-1053. doi:10.1016/j.biopsych.2017.11.026PubMedGoogle ScholarCrossref
35.
Chang  Z, Lichtenstein  P, D’Onofrio  BM,  et al.  Maternal age at childbirth and risk for ADHD in offspring: a population-based cohort study.  Int J Epidemiol. 2014;43(6):1815-1824. doi:10.1093/ije/dyu204PubMedGoogle ScholarCrossref
36.
Sandin  S, Hultman  CM, Kolevzon  A, Gross  R, MacCabe  JH, Reichenberg  A.  Advancing maternal age is associated with increasing risk for autism: a review and meta-analysis.  J Am Acad Child Adolesc Psychiatry. 2012;51(5):477-486.e1. doi:10.1016/j.jaac.2012.02.018PubMedGoogle ScholarCrossref
×