The EU-AIMS Longitudinal European Autism Project (LEAP) : Clinical Characterisation

Charman et al.
Molecular Autism (2017) 8:27

DOI 10.1186/s13229-017-0145-9
RESEARCH Open Access
The EU-AIMS Longitudinal European Autism

Project (LEAP): clinical characterisation
Tony Charman1*, Eva Loth2,3, Julian Tillmann1, Daisy Crawley3, Caroline Wooldridge4, David Goyard5, Jumana Ahmad2,
Bonnie Auyeung6,7, Sara Ambrosino8, Tobias Banaschewski9, Simon Baron-Cohen6, Sarah Baumeister9,
Christian Beckmann10, Sven Bölte11,12, Thomas Bourgeron13, Carsten Bours10, Michael Brammer4, Daniel Brandeis9,
Claudia Brogna14, Yvette de Bruijn10, Bhismadev Chakrabarti6,15, Ineke Cornelissen10, Flavio Dell’ Acqua2,
Guillaume Dumas13, Sarah Durston8, Christine Ecker1,16, Jessica Faulkner3, Vincent Frouin5, Pilar Garcés17, Lindsay Ham18,
Hannah Hayward3, Joerg Hipp17, Rosemary J. Holt6, Johan Isaksson11,19, Mark H. Johnson20, Emily J. H. Jones20,
Prantik Kundu21, Meng-Chuan Lai6,22, Xavier Liogier D’ardhuy17, Michael V. Lombardo6,23, David J Lythgoe4, René Mandl8,
Luke Mason20, Andreas Meyer-Lindenberg24, Carolin Moessnang24, Nico Mueller9, Laurence O’Dwyer10,
Marianne Oldehinkel10, Bob Oranje8, Gahan Pandina25, Antonio M. Persico14,26, Barbara Ruggeri27, Amber N. V. Ruigrok6,
Jessica Sabet3, Roberto Sacco14, Antonia San Jóse Cáceres3, Emily Simonoff28, Roberto Toro13, Heike Tost24,
Jack Waldman6, Steve C. R. Williams4, Marcel P. Zwiers10, Will Spooren29, Declan G. M. Murphy2,3 and Jan K. Buitelaar10
Abstract
Background: The EU-AIMS Longitudinal European Autism Project (LEAP) is to date the largest multi-centre, multi-
disciplinary observational study on biomarkers for autism spectrum disorder (ASD). The current paper describes the
clinical characteristics of the LEAP cohort and examines age, sex and IQ differences in ASD core symptoms and
common co-occurring psychiatric symptoms. A companion paper describes the overall design and experimental
protocol and outlines the strategy to identify stratification biomarkers.
Methods: From six research centres in four European countries, we recruited 437 children and adults with ASD and
300 controls between the ages of 6 and 30 years with IQs varying between 50 and 148. We conducted in-depth
clinical characterisation including a wide range of observational, interview and questionnaire measures of the ASD
phenotype, as well as co-occurring psychiatric symptoms.
Results: The cohort showed heterogeneity in ASD symptom presentation, with only minimal to moderate site differences
on core clinical and cognitive measures. On both parent-report interview and questionnaire measures, ASD
symptom severity was lower in adults compared to children and adolescents. The precise pattern of differences varied
across measures, but there was some evidence of both lower social symptoms and lower repetitive behaviour severity in
adults. Males had higher ASD symptom scores than females on clinician-rated and parent interview diagnostic measures
but not on parent-reported dimensional measures of ASD symptoms. In contrast, self-reported ASD symptom severity was
higher in adults compared to adolescents, and in adult females compared to males. Higher scores on ASD symptom
measures were moderately associated with lower IQ. Both inattentive and hyperactive/impulsive ADHD symptoms were
lower in adults than in children and adolescents, and males with ASD had higher levels of inattentive and hyperactive/
impulsive ADHD symptoms than females.
(Continued on next page)
* Correspondence: [email protected]
1
Department of Psychology, Institute of Psychiatry, Psychology and
Neuroscience, King’s College London, De Crespigny Park, Denmark Hill,
London SE5 8AF, UK
Full list of author information is available at the end of the article
© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Charman et al. Molecular Autism (2017) 8:27 Page 2 of 21
(Continued from previous page)

Conclusions: The established phenotypic heterogeneity in ASD is well captured in the LEAP cohort. Variation both in
core ASD symptom severity and in commonly co-occurring psychiatric symptoms were systematically associated with sex,
age and IQ. The pattern of ASD symptom differences with age and sex also varied by whether these were clinician ratings
or parent- or self-reported which has important implications for establishing stratification biomarkers and for
their potential use as outcome measures in clinical trials.
Keywords: Autism, Autism spectrum disorder, Phenotype, Behaviours, Heterogeneity, Sex, Age, IQ
Background and cellular architecture, brain structure and function,

Heterogeneity is a core feature of the ASD phenotype cognitive, behavioural, and clinical variation, assessing
Autism spectrum disorder (ASD) is a common neurode- individuals across development, etc. [17].
velopmental disorder, affecting ~1% of children and
adults [1–4]. The core characteristics are impairments in Variation of the ASD phenotype by sex, age and
social communication abilities, the presence of rigid, intellectual ability
repetitive and stereotyped behaviours, and atypical ASD is at least three times more prevalent in males than
sensory responses (DSM-5; [5]). However, there is wide females, and biological sex may be an important source
heterogeneity in clinical presentation, both in terms of of heterogeneity in ASD presentation. Lai and colleagues
symptom profiles and severity (hence the use of the term [18] recently summarised research on sex differences in
‘spectrum’; [6]) and levels of intellectual and functional ASD, covering potential mechanisms underlying the sex
communication ability. Commonly associated conditions differential liability to possible sex differences in brain
range from psychiatric symptoms, such as anxiety disor- structure and function. Other factors may also affect the
ders and attention-deficit/hyperactivity disorder (ADHD) recognition and presentation of ASD symptoms in males
[7] to medical conditions including epilepsy and gastro- and females, including potentially different patterns or
intestinal abnormalities [8]. Heterogeneity is present profiles of symptoms and ‘compensatory’ or ‘masking’ of
both between individuals who fulfil the diagnostic cri- symptoms in females [18]. In addition, there is evidence
teria and within individuals across development [9, 10]. from population studies that girls with similar levels of
Decomposing this heterogeneity may get us closer to symptoms to boys are less likely to be diagnosed by
more precise inferences about which subsets of individ- community services [19], unless there are more substan-
uals are best characterised by different cognitive theories tial behavioural or cognitive difficulties [20]. In terms of
of ASD [11]. Wide variability is also present at the level clinical profile and behaviour, findings have been incon-
of aetiological mechanisms. Common genetic variants of sistent. While a meta-analysis suggested lower levels of
small effect size are thought to accumulate and contrib- repetitive and restricted behaviours and interests (RRB)
ute towards enhanced risk, implicating a diverse range in females but comparable levels of social communica-
of biological pathways. Similarly, some rare genetic vari- tion difficulties in males and females [19, 21], other
ants found in a small percentage of individuals are studies have reported greater social communication diffi-
highly penetrant for ASD (i.e. copy number variants, sin- culties and lower cognitive ability and adaptive function
gle nucleotide variants) but also affect a diverse set of in females [22, 23]. Similarly, some studies have reported
biological pathways [12–14]. Thus, the genomic land- higher levels of anxiety in girls than boys with ASD and
scape of risk mechanisms is highly diverse. Environmen- more externalising symptoms in boys [24–26]—but
tal factors as well as the interplay between genetic and other studies have not [7]. Comparisons across studies
environmental risk mechanisms are also likely import- are compromised by differences between samples such
ant, though the magnitude of impact is still largely as varying rates of intellectual disability.
unknown [15]. Age is another potential source of heterogeneity in
Heterogeneity within ASD is a challenge for basic individuals with ASD. There are some reports of reduc-
science attempts to understand the pathophysiological tions in ASD symptoms over early childhood [27] but
and neurodevelopmental mechanisms that lead to the also high variability in the trajectory over childhood and
disorder and for the development of effective psycho- into early adolescence with some children showing
pharmacological or behavioural treatments [16]. Decom- stable high or low severity across development, while a
posing heterogeneity across individuals and at multiple minority significantly improve or worsen, respectively
levels of analysis requires ‘big data’ approaches that are [28–33]. Several longitudinal studies have reported a
both ‘broad’ (i.e. large numbers of people) and ‘deep’, i.e. reduction in ASD symptoms in adulthood, although
multiple levels of analysis within an individual—genetic functional outcomes for many individuals remain poor
[34–36]. A number of longitudinal studies have reported schools, mainstream schools and local communities.
lower levels of psychiatric symptoms in adolescence than Based on parent- or self-reported ethnicity, most partici-
in childhood [37, 38], and others have reported further pants were Caucasian white (73%). The remaining partici-
reductions into adulthood [39] and even throughout the pants were described as either of mixed race (6%), Asian
adult life course [40]. (2%), black (1%) or other (2%). For 16% of participants
Variation in intellectual ability is included in DSM-5 information on ethnicity was either not provided (12%) or
as a ‘clinical specifier’, indicating its importance in driv- missing (4%). Annual household income was measured on
ing heterogeneity of ASD. In many samples, lower IQ an 8-point-scale ranging from <£25,000 to >£150,000,
has been modestly but significantly associated with with the median annual household income being esti-
higher levels of ASD symptom severity [41, 42]. In con- mated at £30,000–£39,999. Highest household parental
trast to the moderate association found in the general education was coded on a 5-point scale ranging from pri-
population between low IQ and increased levels of exter- mary education to postgraduate qualifications; 61% of
nalising disorders [43, 44], some studies have reported households had at least one parent with education beyond
that in population-derived samples, this association was a high school diploma (i.e. with an undergraduate degree
only present in adolescents (and not children) with ASD from university). At each site, an independent ethics com-
[7, 38]. A meta-analysis focusing on anxiety disorders in mittee approved the study. All participants (where appro-
ASD revealed complex associations with IQ, finding that priate) and their parent/legal guardian provided written
social anxiety was more common in studies with lower informed consent.
IQ samples but that obsessive-compulsive disorder and
separation anxiety were higher in studies with higher IQ Inclusion/exclusion criteria
samples [45]. Participant inclusion criteria for the ASD sample were
an existing clinical diagnosis of ASD according to DSM-
Clinical characterisation of the EU-AIMS LEAP cohort IV [50], DSM-IV-TR [51], DSM-5 [5] or ICD-10 [52]
As described in the companion paper [46], as part of the criteria and age between 6 and 30 years. ASD diagnoses
EU-AIMS clinical research programme [47–49], we were based on a comprehensive assessment of the
established the Longitudinal European Autism Project participant’s clinical history and/or current symptom
(LEAP). Here, we report on the baseline clinical assess- profile, depending on when the participant was originally
ment of the EU-AIMS LEAP cohort. The paper will first identified at that site. In addition, we assessed ASD
describe the cohort and its clinical characteristics. Then, symptoms using the Autism Diagnostic Observation
taking advantage of the size and heterogeneity of the Schedule (ADOS; [53, 54]) and the Autism Diagnostic
cohort, we will examine whether there are sex, age and Interview-Revised (ADI-R; [55]). However, individuals
IQ differences on measures of core ASD symptoms and with a clinical ASD diagnosis who did not reach cut-offs
levels of commonly co-occurring psychiatric symptoms. on these instruments were not excluded. Clinical judge-
ment has been found to be more stable than scores on
Methods individual diagnostic instruments alone [56], reflecting
Participants the moderate-to-good but still imperfect accuracy of
In this multi-site study, participants were recruited such tools [57].
between January 2014 and March 2017 across six Exclusion criteria included significant hearing or visual
European specialist ASD centres: Institute of Psych- impairments not corrected by glasses or hearing aids, a
iatry, Psychology and Neuroscience, King’s College history of alcohol and/or substance abuse or dependence
London (IoPPN/KCL, UK), Autism Research Centre, in the past year and the presence of any MRI contraindi-
University of Cambridge (UCAM, UK), University cations (e.g. metal implants, braces, claustrophobia) or
Medical Centre Utrecht (UMCU, Netherlands), Rad- failure to give informed written consent to MRI scan-
boud University Nijmegen Medical Centre (RUNMC, ning (or to provide contact details for a primary care
Netherlands), Central Institute of Mental Health physician at centres where this is a pre-condition for
(CIMH, Germany) and the University Campus Bio- scanning). Participants were purposively sampled to
Medico (UCBM) in Rome, Italy (see Table 1 for recruit- enable in depth experimental characterisation of poten-
ment information by site). In addition, twins discordant tial biomarkers (including MRI scans). Therefore, we
for ASD were recruited at Karolinska Institutet, Swe- excluded individuals with low IQ (<50) as core measures
den—however, twins were not included in the case- (e.g. most cognitive tasks and MRI scanning without
control comparisons reported below. Participants were sedation) were deemed difficult to administer in this
recruited from a variety of sources including existing group. Participants who did not complete an IQ assess-
volunteer databases, existing research cohorts, clinical ment were excluded (controls: n = 7, ASD: n = 10). In the
referrals from local outpatient centres, special needs TD group, individuals who had a T score of 70 or higher
Table 1 Number of participants recruited by each site according to schedule and diagnostic group
Total Adults Adolescents Children Mild ID
ASD TD/ID ASD TD ASD TD ASD TD ASD ID
London (KCL) 159 89 55 38 41 19 32 14 31 18
Cambridge (UCAM) 59 34 17 14 22 10 17 10 3 0
Mannheim (CIMH) 36 38 7 5 20 25 7 8 2 0
Nijmegen (RUNMC) 117 74 24 13 31 28 32 22 30 11
Rome (UCBM) 22 19 21 19 0 0 0 0 1 0
Utrecht (UMCU) 44 46 18 20 12 12 13 14 1 0
Total 437 300 142 109 126 94 101 68 68 29
ASD autism spectrum disorder, TD typically developing, Mild ID intellectual disability
on the self-report (1 adult) or parent-report form (1 ado- Clinical measures—ASD symptomatology

lescent, 3 children) of the Social Responsiveness Scale Given the cautious conclusions of recent reviews of ASD
[58] were also excluded. symptom measures as potential endpoints for clinical
In the ASD sample, psychiatric conditions (except for trials [59–61], we used a range of different measures of
psychosis or bipolar disorder) were allowed as up to 70% ASD symptoms (a full list of all clinical measures is
of people with ASD have one or more psychiatric disor- reported in the Additional file 1: Table 3). These various
ders [7] and reflect DSM-5 that allows co-occurring ASD symptom measures have complementary strengths
psychiatric disorders alongside an ASD diagnosis [5]. and limitations, relevant to our clinical and conceptual
In future individual biomarker analyses, additional ex- understanding of measurement of ASD symptomatology
clusion criteria or sub-grouping may then be applied [57]. The parent-report ADI-R algorithm gives histor-
(e.g. ADI-R cut-offs, medication-free, etc.). ical/early developmental symptom severity; the ADOS is
Exclusion criteria of the TD/ID group were the same an observational measure of current symptom severity.
as described above for the ASD participants with the Both are diagnostic instruments. The ADOS has a stan-
exception that in the TD group parent- or (where appro- dardised ‘calibrated severity score’, that is equivalent
priate) self-report of a psychiatric disorder was also an across different modules while the ADI-R produces raw
exclusion criteria. algorithm scores in the three core ASD behavioural
domains but is more susceptible to skew. The ADI and
ADOS were not administered to the typically developing
Study schedules controls or mild ID cases without ASD. In addition,
Participants were split into four study schedules dimensional measures of ASD symptomatology were
depending on their age and cognitive ability level. derived from a variety of questionnaires (described below).
Three schedules included individuals with IQ in the Each of these questionnaires was parent rated and/or self
typical range (≥75) (children: aged 6–11 years, adoles- rated depending on age and cognitive level (see Table 2 for
cents: aged 12–17 years and adults: aged 18–30 years). a summary of parent-report and participant self-report
At two sites (KCL, RUNMC)1, adolescents and adults questionnaires). The use of both parent and self-report in
(aged 12–30 years) with ASD and mild intellectual a subsample will allow us to determine if the pattern of
disabilities (mild ID; defined by IQ between 50 and age and sex differences in ASD and associated psychiatric
742) were also recruited alongside age- and IQ- symptoms varies by respondent, which will have implica-
matched individuals without ASD (mild ID group). tions both for mapping putative biomarkers onto the ASD
Each schedule received a tailored and largely compar- phenotype and for their use as outcomes in clinical trials.
able study protocol to take into account differences in The Social Responsiveness Scale, Second Edition (SRS-2;
age and cognitive level [46]. Within each age band [58]) is a parent-reported symptom questionnaire suitable
(children, adolescents, adults), participants were re- across the whole age range (and is sex normed) that in
cruited with a similar male:female ratio (3:1) and IQ addition has a self-report companion measure suitable for
composition so that predicted cognitive/biological dif- adolescents and adults. Other questionnaire measures
ferences can be compared across sex and develop- (Autism Spectrum Quotient (AQ; [62–64]); Children’s
mental stages. Likelihood ratio tests confirmed that Social Behaviour Questionnaire (CSBQ; [65])/Adult Social
the targeted male:female ratio did not differ signifi- Behaviour Questionnaire (ASBQ; [66]) are designed as
cantly across schedules (x2(2) = 1.41, p = .494) and more dimensional/trait measures of ASD severity and
study sites (x2(5) = 2.69, p = .754), as well as between have different versions across the age span. The inclusion
ASD and TD groups within each age band (all p > .1). of multiple dimensional measures of ASD symptom
Table 2 Summary of parent-report and participant self-report questionnaires

Phenotypic measures Adults (TD) Adults (ASD) Adolescents (TD/ASD) Children (TD/ASD) Mild ID (ID/ASD)
Dimensional measures of ASD symptoms
Social Responsiveness Scale-2nd Edition (SRS-2), S S&P S&P P P
Children’s Social Behaviour Questionnaire (CSBQ) - - P P P
Adults’ Social Behaviour Questionnaire (ASBQ) S S&P - - P (>18 years)
Autism Spectrum Quotient (AQ)—adult version S S&P - - -
Autism Spectrum Quotient (AQ)—adolescent version - - P - -
Autism Spectrum Quotient (AQ)—child version - - - P P
Repetitive Behaviour Scale-Revised (RBS-R) - P P P P
Short Sensory Profile (SSP) - P P P P
Psychiatric symptoms
DSM-5 ADHD rating scale S S&P P P P
Beck Anxiety Inventory S S S P P
Beck Depression Inventory S S S P P
S self-report (completed by participant), P parent-report (completed by primary carer or parent of participant), S & P self- and parent-report administered; - not administered
severity will allow us to test which measure best over the previous 6 months. Each item is scored
relates to neurobiological or neurocognitive bio- using a ‘0’ (not true) to ‘3’ (almost always true) on a
markers and is most sensitive to change over time. Likert scale. The total raw score is transformed into
Other questionnaires measure aspects of the ASD sex-specific T scores, and here, we report both raw
phenotype not well captured by the SRS-2, including and sex-standardised scores. Parent report was used
atypical sensory responses (Short Sensory Profile (SSP; for all participants with ASD and mild ID, as well as
[67]) and repetitive, rigid and stereotyped behaviours children and adolescents with typical development.
(Repetitive Behavior Scale-Revised (RBS-R; [68]). Adults with ASD additionally completed the self-
The Autism Diagnostic Observation Schedule (ADOS; report form. Adults with typical development only
[53, 54]), a standardised social interaction observation completed the self-report form as, for feasibility
assessment, was used to assess current symptoms in reasons, in this schedule, parents were not enrolled in
ASD participants (module 2 for 2 participants, module 3 the study.
for 154 participants, module 4 for 208 participants). The Repetitive Behavior Scale-Revised (RBS-R; [68])
Calibrated Severity Scores (CSS) for Social Affect (SA), assesses restricted repetitive behaviours associated
Restricted and Repetitive Behaviours (RRB) and Overall with ASD. Parents or caregivers rate 43 behaviours
Total were computed [69, 70], which provide standardised (e.g. ‘arranges certain objects in a particular pattern
autism severity measures that account for differences in or place’; ‘need for things to be even or symmetrical’)
the modules administered. The Autism Diagnostic on a scale of 0–3, where 0 indicates the behaviour
Interview-Revised (ADI-R; [55]), a structured parent inter- does not occur and 3 indicates the behaviour does
view, was completed with parents/carers of ASD partici- occur and is a severe problem.
pants. Standard algorithm scores which combine current Sensory processing atypicalities were measured using
and historical symptom information were computed for the SSP [67]. This parent-report questionnaire comprises
Reciprocal Social Interaction (Social), Communication, 37 items, where each item is scored on a 5-point Likert-
and Restricted, Repetitive and Stereotyped Behaviours and rating scale from 1 (always occurs) to 5 (never occurs).
Interests (RRB). Current ADI-R scores were available on a The SSP is based on the sensory profile [71]. Lower
subset of the ASD sample (356/414 (86%)) but are not scores on the SSP are indicative of greater impairment.
reported in the current paper. Where ADOS and ADI-R The CSBQ [65] is a 49-item parent-report question-
scores from previous assessments were available (ADOS: naire that is specifically useful in assessing behaviour
within the past 12 months for children/past 18 months for atypicalities across the entire ASD spectrum. Adults
all other schedules; ADI-R: at any historical point since we received the ASBQ for either self or parent report, com-
report the 4 to 5 years/ever algorithm scores), these posed of 44 items [66].
assessments were not repeated. The AQ [62–64]) is a continuous self- or parent-
The Social Responsiveness Scale, Second Edition report measure that quantifies the degree to which
(SRS-2; [58]) is a quantitative measure comprising 65 children, adolescents or adults of average intelligence
items asking about characteristic autistic behaviour show behavioural characteristics associated with ASD.
The AQ consists of 50 statements asking about habits The Beck Anxiety Inventory (BAI; [80]) is a well-
and personal preferences. Each statement is rated by the validated 21-item inventory probing for common symp-
participant or parent/carer on a 4-point Likert-rating toms of anxiety. Participants rate each item along different
scale from ‘definitely agree’, ‘slightly agree’, ‘slightly levels of symptom severity experienced over the past
disagree’ to ‘definitely disagree’. While adult participants month from 0 = not at all to 3 = severely. The self-report
completed the AQ by self-report, the adolescent version version of the BAI was administered to adult participants.
is parent report but is otherwise composed of the same Children and adolescents/adults with mild ID were
items compared to the adult AQ. The AQ-Child also given the anxiety subscale of the Beck Youth Inven-
entails parent-report, yet items that were not age appro- tories (BYI-II; [79]) as parent-report, while adoles-
priate in the adolescent/adult questionnaire were revised cents completed the anxiety subscale of the BYI-II as
accordingly. self-report.
The DSM-5 rating scale of attention-deficit/hyperactivity
Intellectual ability disorder (ADHD) covers 18 items measuring the presence
Level of intellectual abilities was assessed using the of inattention and hyperactive/impulsive symptoms in the
Wechsler Abbreviated Scales of Intelligence—Second past 6 months, each evaluated on a 0–3 scale (0 = not at
Edition, WASI-II [72] or—in countries where the WASI all to 3 = very often). In children, six or more responses
is not translated (i.e. The Netherlands, Germany and scored with 2 (often) or 3 (very often) to either (or both)
Italy)—the four-subtest short forms of the German, the inattention and hyperactivity/impulsivity domains
Dutch or Italian WISC-III/IV [73, 74] for children or indicate clinical concern. Depending on age and abil-
WAIS-III/IV [75, 76] for adults. The shortened versions ity level, either parent- or self-report forms were
were used for feasibility reasons to not further prolong administered.
the testing sessions for participants. All versions
included two verbal subscales (vocabulary, similarities) Quality control procedures
and two non-verbal subscales (block design, matrix rea- Appropriate to a multi-centre, cross-national study,
soning). To standardise data across sites, IQ was pro- we established quality control procedures around
rated from two verbal subtests (vocabulary and training, data collection and data entry and checking.
similarities) and two performance subtests (matrix rea- We had cross-site training sessions for collecting clin-
soning and block design) using an algorithm developed ical data, the ADOS and ADI-R were administered
by [77] that produces an estimated IQ score that is and scored by qualified/certified personnel and the
highly correlated (r = .93) with a full-Scale IQ obtained study was regularly monitored according to good clinical
by administering the complete test. Age-appropriate practice standards. Of the total number of ADI-R assess-
national population norms were available for each partici- ments (4–5 ever/diagnostic) administered to participants
pating site, and these were used to derive standardised (N = 414), N = 162 were re-used from previous studies,
estimates of an individual’s intellectual functioning. Where while for the ADOS (N = 364), a total of N = 61 were re-
recent IQ scores from previous assessments were available used (all completed within the previous 12 months). Prior
(less than 12 months in children; less than 18 months in to data analysis, a series of quality control procedures
adolescents and adults), IQ tests were not repeated. were adopted to maximise coherence and comparability
of data. This involved initial randomised double data entry
Clinical measures—co-occurring psychiatric symptoms of 10% of cases at each site for core clinical measures
The Beck Depression Inventory—Second Edition (BDI- (e.g. ADI-R, ADOS, IQ data). If a significant level of
II; [78]) is a 21-item inventory measuring the severity incorrect/inconsistent data was identified, all data was
of characteristic attitudes and symptoms associated checked against the original paper forms. Other pro-
with depression. Each item contains four possible cedures also included impossible values/range checks
responses, which range in severity from 0 (e.g. ‘I do of all items, sub-scales and total scores for interview
not feel sad’) to 3 (e.g. ‘I am so sad or unhappy that and questionnaire measures, duplicated entry detec-
I can’t stand it’). Participants are asked to provide tion and correction, as well as data audits and checks
answers based on the way they have been feeling over of scoring algorithms. When missing data was
the past month, including the assessment day. The present, site coordinators were asked to secure the
self-report version of the BDI-II was administered to information if possible.
adult participants. Parents/caregivers completed the Across all clinical measures, we have applied a prorat-
depression subscale of the Beck Youth Inventories ing approach to deal with missing scores. Prorating
(BYI-II; [79]) for children and adolescents/adults with replaces the missing score for a given participant with
mild ID. Adolescents were given the depression her/his mean score on other items on the same sub-
subscale of the BYI-II as self-report. scale. Prorating was only applied if less than 20% of
scores on the same sub-scale were missing. For a higher measure separately) and margin plots. Log-transformed
percentage of missing scores, prorating was not applied variables were used where appropriate to meet normality
(i.e. data for these participants was recorded as missing). assumptions (RBS-R, SSP). A random effect for site was
included in all models to take into consideration the
Statistical analysis multi-level nature of the data, as well as to account for
Statistical analysis was performed with the following site heterogeneity across outcome measures. Intraclass
objectives: correlation coefficients (ICCs) reflecting the ratio of
between-site variance to total variance are reported
(1) To examine whether there are age and/or sex (see Table 4). All models included a continuous measure of
differences in the severity of ASD symptoms by IQ (full-scale IQ) as a covariate (Additional file 3: Table S2).
comparing individuals with ASD across different Linear mixed models report chi-square coefficients and
age groups (children, adolescents, adults); p value. Effect sizes were calculated following [82] by
(2) To examine whether differences in age (i.e. ADOS) dividing the difference in marginal means by the square
or sex (i.e. ADI-R, ADOS) are observed on diagnostic root of the variance at the within-participant level. This
instruments as well as on continuous measures of measure of effect size is equivalent to Cohen’s d or
ASD symptomatology (i.e. SRS-2, CSBQ/ASBQ, AQ, standardised difference [83], where an effect size of 0.2 to
RBS-R, SSP) and whether these patterns are similar or 0.3 is taken to be a small effect, 0.5 a medium effect
different across parent- and self-report measures; and greater than 0.8 a large effect. For the analyses
(3) To characterise the association between ASD reported in the (Additional file 2: Table S1) that treat age
symptoms and level of intellectual functioning; and IQ as continuous variables, we performed linear
(4) To characterise the severity of co-occurring mixed-effects models to take into account site effects yet
psychiatric symptoms (i.e. ADHD, anxiety, depression) replacing the categorical age/ability level variable with
in individuals with ASD and to examine how continuous measures of chronological age and IQ.
these relate to age, sex and IQ.
Results
Linear mixed-effects models were fit using a maximum Participant characteristics are shown in Table 3.
likelihood estimation method and were executed using
STATA software 14.0 [81]. Differences in ASD symp- Demographics
tomatology between individuals with ASD relating to In the total sample, the mean (SD) chronological age
age, sex and IQ were analysed by restricting the analysis was 16.9 (5.9) years, with similar distributions of age
to participants with ASD only since by definition ASD for individuals with ASD (M = 16.7, SD = 5.8) and TD/
participants will score more highly than controls on mild ID individuals (M = 17.2, SD = 5.9), x2(1) = 1.84,
ASD symptom measures. Each model (except for ADI-R p = .175. Of the 737 participants, 511 were men and
diagnostic scores) included fixed main effects for study 226 were woman (2.3:1 male-female ratio). While
schedules (children, adolescents, adults and mild ID) overall, the male-female ratio was significantly but
and sex (male, female), as well as their interaction. In only slightly higher across individuals with ASD
this paper, we treat age and IQ in two ways. First, both (2.6:1) relative to TD/mild ID individuals (1:9:1)
for clinical ‘face validity’ and to allow the comparison (x2(1) = 5.49, p = .019), it was not significant within
between the clinical characteristics of the LEAP cohort each age band (all p > .1). For annual household
to previously published samples—often comprised of income, there was a significant interaction between
children, adolescents or adults only, with or without diagnosis and schedule (x2(4) = 26.10, p = .0001), with
intellectual disability and not with the heterogeneity individual comparisons indicating that household
present in our cohort by design—we analyse and present income was significantly higher in TD children com-
the clinical data in the main paper according to the age/ pared to children with ASD (x2(1) = 13.61, p = .0009).
IQ-defined schedules outlined above. Second, in the For both paternal (x2(4) = 10.86, p = .028) and mater-
(Additional file 2: Table S1), we present scores on some nal education (x2(4) = 19.08, p = .0008), a significant
of the key measures continuously by age and IQ as this interaction between diagnosis and schedule was
maximises the power of the large sample and recognises found. Individual contrasts revealed that the level of
the arbitrary nature of creating age and IQ ‘groups’ by paternal and maternal education was significantly
‘binning’ the sample into pre-defined age and IQ sub- higher in TD children relative to children with ASD
groups. For the analysis by schedule, significant main (x2(1) = 5.11, p = .024 and x2(1) = 6.55, p = .042 respect-
and interaction effects were further explored using post- ively). There were no differences in ethnicity between TD/
estimation methods including contrasts (Bonferroni-cor- mild ID and ASD participants overall and within each age
rected for the number of post hoc comparisons for each band (all p > .4).
Table 3 Sample characteristics

Sex N 437 300 142 109 126 94 101 68 68 29
Males (%) 72.3 65 72.5 67 77 69.1 71.3 61.8 64.7 51.7
Females (%) 27.7 35 27.5 33 23 30.9 28.7 38.2 35.3 48.3
Age M 16.68 17.22 22.79 23.10 14.86 15.33 9.40 9.52 18.09 19.30
(in years)
SD 5.80 5.94 3.37 3.27 1.73 1.73 1.58 1.54 4.27 4.97
Range 6.08–30.60 6.24 -30.78 18.02–30.60 18.07–30.78 12.07–17.90 12.04–17.99 6.08–11.97 6.24–11.98 11.50–30.19 12.92–30.24
Full-scale M 97.61 104.57 103.99 109.15 101.59 106.58 105.29 111.46 65.84 63.39
IQ
SD 19.74 18.26 14.82 12.60 15.68 13.18 14.76 12.69 7.70 8.00
Range 40 –148
a
50–142 76–148 76–142 75–143 77–140 74–148 76–142 40 –74
a
50–74
ASD autism spectrum disorder, TD typically developing, Mild ID intellectual disability
a
There are 3 individuals with a full-scale IQ <50
Site effects Diagnostic ASD measures—sex and age effects

The random effect for site included in all the models On the ADOS, male ASD participants had significantly
was significant for all the key demographic and diagnos- higher CSS Total (x2(1) = 15.81, p = .0001, d = .46) and
tic measures except for sex and ADOS Total and Social CSS Social Affect (SA) (x2(1) = 12.71, p = .0004, d = .44)
Affect CSS (see Table 4). The ICCs shown in Table 4 than females with ASD and was approaching signifi-
indicate that while the effect of site was large for age cance for CSS Restricted and Repetitive Behaviours
(~25%), reflecting the variable recruitment targets across (RRB) (log-transformed, x2(1) = 3.15, p = .076, d = .22)
age schedules and across sites (see Table 1), for other (see Table 5 and Fig. 1). A significant interaction
measures, it was low to moderate, being less than 1% for between sex and schedule was found for CSS Total
sex ratio, less than 6% for IQ, between 9 and 15% for (x2(4) = 16.97, p = .002) and CSS SA (x2(4) = 13.32, p = .009).
ADI-R scores and less than 8% for ADOS scores. Individual comparisons indicated that only in adolescents,
Table 4 Summary of variation between sites in demographic and behavioural characteristics and level of ASD symptomatology for
individuals with ASD only
Ranges across sites Variance
Minimum Maximum Mean SD Overall mean (SD) Within sites Between sites ICCa x2 sig. value
Chronological age [years:months] 6:07–19:8 24:5–30:6 14:8–25:0 3:2–6:3 16:7 (5:8) 29.87 9.91 .249 p < .0001
Sex, % of male participants 66.1–80.6 72.3 (4.48) 0.46 <.01 <.001b n.s.c
Verbal IQ 45d–70 130–160 93–110 14–21 97 (19) 382.18 12.61 .031 p < .0001
Nonverbal IQ 45d–68 134–150 93–107 16–23 98 (21) 430.82 24.39 .054 p = .0001
Full-scale IQ 40 –73
d
128–148 96–105 12–22 98 (20) 373.45 16.35 .042 p = .001
ADI-R
Social interaction 0–4 24–29 12–19 6–7 17 (7) 42.26 4.33 .093 p < .0001
Communication 0–3 17–26 9–16 5–5 13 (6) 28.10 4.97 .150 p < .0001
RRB 0–1 8–12 3–5 2–4 4 (3) 6.06 .85 .122 p < .0001
ADOS—CSS
Total 1 10c 5–9 2–3 5 (3) 2.77 .35 <.001b n.s.
c
SA 1 10 6–7 2–3 6 (3) 6.88 .11 .016 n.s.
RRB 1 9–10 4–8 2–3 5 (3) 7.29 .60 .076 p < .0001
Sample sizee 22 159 72 54
ICC intraclass correlation coefficient, ADI-R Autism Diagnostic Interview-Revised, ADOS CSS Total, SA, RRB Autism Diagnostic Observation Schedule Calibrated Severity
Scores for Total, Social Affect and Restricted and Repetitive Behaviours, IQ intelligence quotient, n.s. not significant
a
The ratio of between-site variance to total variance
b
ICC truncated at zero
c
The highest possible score (i.e. ceiling) on the instrument
d
There are 3 individuals with a full-scale IQ <50 (All ASD)
e
Sample size variation of individuals with ASD across sites (minimum/maximum, mean and standard deviation of number of participants with ASD recruited at sites)
Table 5 Sex differences for key measures for ASD and TD/ID participants (pooled across schedules)
ASD TD/ID
Males Females Males Females
Autism symptomatology measures
ADI—Social 17.01 (6.78) 15.36 (6.89) – –
ADI—Communication 13.55 (5.86) 12.58 (5.33) – –
ADI—RRB 4.57 (2.66) 3.74 (2.52) – –
ADOS—CSS Total 5.73 (2.83) 4.60 (2.49) – –
ADOS—CSS SA 6.31 (2.66) 5.46 (2.56) – –
ADOS—CSS RRB 5.03 (2.86) 4.30 (2.69) – –
a
SRS-2 71.50 (11.70) 73.65 (12.18) 47.49 (9.97) 48.07 (9.17)
SRS-2b 62.37 (9.91) 66.48 (11.14) 48.48 (6.08) 46.55 (6.13)
a
CSBQ 46.86 (17.01) 46.94 (15.62) 7.55 (12.56) 6.30 (8.50)
ASBQa 32.78 (16.76) 32.61 (16.55) 14.67 (15.17) 22.11 (20.76)
ASBQb 30.34 (15.08) 37.37 (15.75) 8.11 (8.49) 7.53 (8.97)
AQ—child 94.26 (18.00) 92.76 (17.39) 45.21 (17.95) 29.70 (10.07)
AQ—adolescents 95.78 (17.66) 96.32 (18.02) 48.92 (20.43) 44.75 (20.67)
AQ—adults 81.03 (18.86) 88.06 (20.78) 49.46 (14.88) 43.10 (14.05)
RBS-Ra 17.16 (14.01) 15.76 (13.48) 2.58 (9.43) 2.42 (5.02)
SSPa 138.12 (27.78) 138.15 (26.83) 175.17 (17.00) 175.75 (17.46)
Psychiatric symptom measures
ADHD—inattentivenessa 4.75 (3.13) 4.05 (3.18) 1.34 (2.19) 1.23 (2.58)
a
ADHD—hyperactivity/impulsivity 2.98 (2.91) 2.47 (2.71) 0.57 (1.57) 0.54 (1.63)
Anxietya 48.52 (8.68) 49.14 (9.91) 40.27 (7.75) 38.64 (6.00)
Depressiona 51.42 (11.82) 50.44 (8.09) 41.76 (10.47) 39.77 (4.97)
ADI Autism Diagnostic Interview–Revised, ADOS CSS Total, SA, RRB Autism Diagnostic Observation Schedule Calibrated Severity Scores for Total, Social Affect and
restricted and repetitive behaviours; SRS-2 Social Responsiveness Scale–2, CSBQ, ASBQ Children’s Social Behaviour Questionnaire (parent-report, administered to
children, adolescents), Adults’ Social Behaviour Questionnaire (parent-report, administered to adults) scores cannot be pooled across age groups, RBS-R Repetitive
Behavior Scale–Revised, SSP Short Sensory Profile, AQ Autism Spectrum Quotient (children, adolescents and adult version; scores cannot be pooled across
age group
a
Parent-report
b
Self-report
males had significantly higher ADOS CSS Total than

females (x2(1) = 5.93, p = .04, d = .56).
A similar pattern of results was also observed on the
ADI-R, where male ASD participants had more severe
scores than female ASD participants on the Social (x2(1)
= 5.98, p = .015, d = .27), and Restricted and Repetitive
Behaviours (RRB) domain (x2(1) = 7.81, p = .005, d = .30)
but not Communication domain (x2(1) = 2.27, p = .131.
d = .19). No significant effect of schedule was observed
for ADI-R and ADOS scores (see Table 6).
Dimensional ASD measures—sex and age effects

Parent-report and self-report data were analysed separ-
ately. For parent-reported SRS-2 raw scores, no signifi-
cant sex differences were observed within the ASD
group (x2(1) = 0.01, p = .939). There were however
significant differences in SRS-2 raw scores across the
Fig. 1 Boxplot of ADOS CSS Total scores by sex and for each schedule
various schedules (x2(3) = 16.82, p = .0008). Follow-up
(ASD participants only)
contrasts (Bonferroni-corrected p values) indicated that
Table 6 ADI-R and ADOS scores by schedule for individuals with ASD only
ADI-R—Social 16.54 (6.85) 15.31 (6.87) 17.27 (6.55) 15.38 (6.76) 19.45 (6.57)
n = 411 n = 132 n = 123 n = 94 n = 62
ADI-R—Communication 13.13 (5.72) 12.19 (5.76) 13.63 (5.63) 13.29 (5.75) 13.87 (5.62)
n = 414 n = 132 n = 123 n = 96 n = 63
ADI-R—RRB 4.33 (2.65) 4.23 (2.62) 4.28 (2.71) 4.68 (2.79) 4.14 (2.36)
n = 414 n = 132 n = 123 n = 96 n = 63
ADOS—CSS Total 5.39 (2.78) 4.84 (2.80) 5.78 (2.77) 4.98 (2.65) 6.39 (2.65)
n = 362 n = 110 n = 102 n = 91 n = 59
ADOS—CSS SA 6.06 (2.65) 5.49 (2.70) 6.44 (2.55) 5.55 (2.58) 7.25 (2.38)
n = 362 n = 110 n = 102 n = 91 n = 59
ADOS—CSS RRB 4.81 (2.83) 4.80 (2.76) 4.84 (2.61) 4.84 (3.07) 4.75 (3.00)
n = 362 n = 110 n = 102 n = 91 n = 59
ASD (autism spectrum disorder), Mild ID (intellectual disability), ADI-R Autism Diagnostic Interview–Revised, ADOS Autism Diagnostic Observation Schedule
adults had significantly lower SRS-2 raw scores com- higher SRS-2 raw scores (x2(1) = 6.81, p = .009, d = .49)
pared to children (x2(1) = 13.93, p = .0006, d = .62) and and T scores (x2(1) = 7.02, p = .008, d = .50) than males
adolescents (x2(1) = 10.34, p = .0039, d = .52) (see Fig. 2) overall. A significant interaction between schedule and
but not compared to adolescents/adults with ASD and sex was also observed for SRS-2 raw scores (x2(1) = 9.60,
mild ID (x2(1) = 4.82, p = .084, d = .49). For parent- p = .008) and SRS-2 T scores (x2(1) = 9.89, p = .007).
reported SRS-2 T scores (age- and sex-adjusted), while Follow-up tests revealed that adult ASD females
there were no significant sex differences within the ASD reported significantly higher SRS-2 raw scores (x2(1) =
group (x2(1) = 2.58, p = .108), SRS-2 T scores differed 8.38, p = .008, d = .60) and T scores (x2(1) = 8.63, p = .007,
significantly across the various schedules (x2(3) = 65.70, d = .60) than adult ASD males, but there were no sex
p < .0001). Follow-up contrasts indicated that adults differences in adolescents.
had significantly lower SRS-2 T scores compared to In contrast to parent-reported SRS-2 T scores, adults
children (x2(1) = 51.16, p < .0001, d = 1.19) and adoles- had significantly higher self-reported SRS-2 T scores
cents (x2(1) = 46.52, p < .0001, d = 1.10), as well as (x2(1) = 6.57, p = .010, d = .36) and SRS-2 raw scores
compared to adolescents/adults with ASD and mild (x2(1) = 6.55, p = .011, d = .36) than adolescents. On both
ID (x2(1) = 12.43, p = .001, d = .80) (see Fig. 3). The the parent-report versions of the CSBQ and ASBQ,
interaction between sex and schedule was not signifi- which were analysed separately due to differences in
cant (x2(3) = 6.43, p = .169). item and sub-scale structure, no main effect of sex or
Adolescents and adults also completed the SRS-2 as schedule and no significant sex by schedule interaction
self-report. On this measure, females had significantly were observed. In contrast, for adults with ASD
Fig. 2 SRS-2 raw scores (parent-report) by chronological age Fig. 3 SRS-2 Total scores (parent-report) by chronological age
(ASD participants only) (ASD participants only)
completing the ASBQ as self-report, females reported ASBQ Total scores (parent-report) and IQ (r = −.38; n
significantly higher scores than males (x 2(1) = 7.57, = 94; p = .0002) and between RBS-R Total scores and IQ
p = .006, d = .48). (r = −190; n = 340; p = .0003) were observed. Scores on the
Data on the AQ was analysed separately for children, SRS-2 (T scores and raw scores for self-report), AQ (child,
adolescents and adults because different versions of the adolescent and adult version), SSP and CSBQ (parent-re-
measure were used. On the Adult-AQ (self-report), sex port)/ASBQ (self-report) were not significantly associated
differences were approaching significance with females with level of intellectual functioning.
having higher scores than males (x2(1) = 3.40, p = .065,
d = .39). Some group effects were found on the AQ- Psychiatric symptom measures (analysed within the ASD
Adolescent, where adolescents with ASD and ID had participants only)
significantly higher AQ scores than adolescents with Due to limited availability of self-report data (TD: n = 14;
ASD without ID (x2(1) = 7.69, p = .006, d = .93). ASD: n = 18), only parent-reported levels of ADHD symp-
Prior to analysis, total scores of the RBS-R were log toms were analysed. A large proportion of children with
transformed to meet normality assumptions. There ASD (here defined as chronological age <17 years accord-
was no significant effect of sex (x2(1) = .32, p = .569) ing to the ADHD symptom checklist) scored in the clin-
but a significant main effect of schedule (x2(3) = ical range on the inattentiveness (51%) and hyperactivity/
27.13, p < .0001), with adults having significantly lower impulsivity ADHD domains (28%). In contrast, the num-
RBS-R scores relative to children (x2(1) = 26.20, p < .0001, ber of adolescents and adults with ASD that met clinical
d = .91) and adolescents (x2(1) = 11.98, p = .001, d = .57). cut-off on these measures was somewhat lower (inatten-
There was no significant interaction effect between sex tiveness 41%; hyperactivity/impulsivity 13%). Among
and schedule. participants with ASD, males scored significantly higher
On the SSP (using log-transformed total scores), no than females on the inattentiveness domain (x2(1) = 4.73,
main effect of sex or schedule and no significant sex by p = .030, d = .22) and hyperactivity/impulsivity domain
schedule interaction were observed. (x2(1) = 3.99, p = .046, d = .22). There was also a significant
effect of schedule on both the inattentiveness domain
Intellectual functioning (x2(3) = 26.30, p < .0001) and hyperactivity/impulsivity
The mixed-effects analysis revealed a significant inter- domain (x2(3) = 71.73, p < .0001), with adults with
action between schedule and diagnosis for full-scale IQ ASD having significantly lower symptom levels across
scores (x2(4) = 25.13, p = .0001, see Table 3), with signifi- these domains compared to children (inattentiveness:
cantly higher IQ scores in TD individuals compared to x2(1) = 20.72, p < .0001, d = .78; hyperactivity/impulsiv-
participants with ASD in the adult (x2(1) = 8.60, p = .01, ity: x2(1) = 69.35, p < .0001, d = 1.32) and adolescents
d = .39), adolescent (x2(1) = 7.79, p = .02, d = .38) and (inattentiveness: x2(1) = 14.94, p = .0003, d = .54; hyper-
children age groups (x2(1) = 8.23, p = .017, d = .37). No activity/impulsivity: x2(1) = 11.80, p = .002, d = .50).
significant differences in intellectual functioning were However, while no differences were observed between
found between individuals with/without ASD and children and adolescents in inattentive symptom
mild ID. levels (x2(1) = 0.60, p = .438), children with ASD had
Examining the association between measures of ASD significantly higher levels of hyperactivity/impulsivity
symptomatology and IQ (full-scale IQ) in individuals symptoms compared to adolescents with ASD (x2(1)
with ASD only, there were significant, albeit weak = 24.98, p < .0001, d = .87). There was no significant
negative correlations between ADOS Total CSS and IQ interaction effect between sex and schedule.
(r = −.23; n = 358; p < .0001), as well as between ADOS Among participants with ASD completing the BAI or
Social Affect CSS and IQ (r = −.23; n = 358; p < .0001), BYI-II as self-report, 24% of adults (26 of 108; i.e. raw
with higher IQs being associated with lower symptom anxiety scores 21+) and 18% of adolescents (12 of 66;
levels. There was no significant association between sex-and age-adjusted T score 60+) scored in the moder-
ADOS RRB CSS and IQ. Scores on the ADI-R Social ate/severe clinical range. In children (TD: n = 51; ASD:
domain (r = −.22; n = 404; p < .0001) and ADI-R Commu- n = 83) and adolescents/adults with mild ID (mild ID: n =
nication domain (r = −.12; n = 407; p = .04), but not ADI-R 10; ASD: n = 29), symptoms of anxiety were assessed by
RRB domain (r = .01; n = 407; p = .782), were also signifi- the BYI-II through parent-report. In addition, some ado-
cantly associated with IQ. On dimensional measures of lescents without ID (TD: n = 4; ASD: n = 17) received the
ASD symptom severity significant negative correla- BYI-II as parent-report. The proportion of individuals
tions between SRS-2 Total T scores (parent-report) with ASD considered to present with a moderate/severe
and IQ (r = −.23; n = 350; p < .0001) see (Additional severity level in anxiety symptoms (same clinical cut-offs
file 3: Table S2), between SRS-2 raw scores (parent- apply as above) was 12% for children (10 of 83), 7% for
report) and IQ (r = −.26; n = 350; p < .0001), between adolescents (2 of 29) and 27% for adolescents/adults with
mild ID (4 of 15). No significant effects of sex or schedule of anxiety, no significant correlation was found be-
were found across all anxiety scales. tween self-report measures and IQ in adolescents (r =
For depressive symptoms as measured by the BDI-II −.10; n = 66; p = .421), as well as between parent-report
or BYI-II as self-report, it was found that among par- measures and IQ in children, adolescents and adoles-
ticipants with ASD, 22% of adults (24 of 107; raw de- cents/adults with mild ID (r = −.05; n = 125; p = .555).
pression scores of 21+) and 27% of adolescents (18 of There was however a significant, albeit weak negative
67; i.e. T score 60+) scored in the moderate to severe correlation between anxiety symptoms (self-report) and
clinical range. In adults with ASD, females reported IQ in adults with ASD (r = −.23; n = 108; p = .017). No
significantly higher depressive symptoms than males significant association between depressive symptoms
(x2(1) = 11.66, p = .0006, d = .72) but not in adoles- (parent- or self-report) and IQ was observed across all
cents (x2(1) = .44, p = .507). The depression subscale of schedules (all p > 0.1).
the BYI-II was administered to children (TD: n = 53;
ASD: n = 86), adolescents/adults with mild ID (mild
ID: n = 10; ASD: n = 29) and adolescents without ID Associations between ASD measures
(TD: n = 4; ASD: n = 17) and completed by their par- Figure 4 shows the associations between the different
ents. Sixteen percent of children (14 of 86), 29% of questionnaire ASD symptom measures separately for the
adolescents (5 of 17) and 28% of adolescents/adults ASD and TD/ID participants. Within the ASD group, as
with mild ID (8 of 29) had scores in the moderate/severe expected, the parent-report global ASD symptom
clinical range (i.e. sex- and age-adjusted T score of 60+). measures (SRS, CSBQ,/ASBQ, AQ) were highly inter-
correlated (all r values >.60, p < .0001). The RBS-R meas-
Association between psychiatric symptoms and uring repetitive behaviour symptoms (r from .56 to .73,
intellectual functioning all p < .0001) and the SSP measuring sensory symptoms
Among participants with ASD, the association between (higher scores on the SSP indicate lower symptomatol-
psychiatric symptoms (depression, anxiety, inattention ogy; r from −.44 to −.70, all p < .0001) were also strongly
and hyperactivity/impulsivity) and intellectual function- inter-correlated with the global symptom measures.
ing (full-scale IQ) was also assessed. There were Parent-report of ASD symptoms (SRS, CSBQ/ASBQ)
significant but weak negative correlations between parent- was moderately to strongly associated with parent-report
reported symptoms of inattention and IQ (r = −.20; n = of both ADHD inattention and hyperactivity/impulsivity
345; p < .0001), as well as between hyperactivity/impul- symptoms (all r > .38, p < .0001) but the parent-report
sivity and IQ (r = −.17; n = 345; p = .001). On measures AQ less so (see Fig. 4).
Fig. 4 Heatmap of correlations between ASD and psychiatric symptom measures (ASD left diagonal; TD/ID participants right diagonal)
Discussion repetitive (RBS-R) and sensory (SSP) symptoms. Further-

Clinical characteristics of the EU-AIMS LEAP cohort more, we have also acquired questionnaire measures of
The EU-AIMS LEAP cohort is a large, well-characterised the most commonly occurring psychiatric symptoms
sample of individuals with ASD and controls ranging from found in individuals with ASD [7, 40]—ADHD, anxiety
young children to adults with a fairly wide range of IQ. and depression. In terms of the biomarker discovery
The main groups of adult, adolescent and child partici- aims of the EU-AIMS LEAP project overall [46–49], this
pants with ASD and controls have IQs in the typical range comprehensive clinical characterisation of such a large
with means close to the population average. The group of sample will enable us to test for associations between
purposively sampled participants with and without ASD putative biomarkers while including potential moderat-
with mild ID (IQ range 50 to 74) is relatively small (n = 68 ing or stratification factors including sex, age, IQ and
ASD; n = 29 non-ASD). Although the LEAP sample has co-occurring psychiatric symptoms.
an elevated IQ compared to the total population of indi-
viduals with ASD, of whom around 50% have an intellec- Sex differences in ASD symptoms
tual disability [8, 9], it is rare for experimental studies of We examined sex differences in ASD severity that have
biomarkers to include any participants with an IQ below been reported in some but not all previous studies [18].
75. Participants were purposively sampled to enable in Across the whole sample, males with ASD had more
depth experimental characterisation of potential bio- severe symptom scores than females on some domains
markers (including MRI scans), and therefore we set a of the ADOS and the ADI-R, including both social com-
lower IQ limit of 50; however, we enrolled 3 participants munication and repetitive behaviours. Some previous
with lower IQ but who were capable of completing all our studies have found higher levels of repetitive behaviours
minimal assessments. It is a notable limitation of the rep- but not higher social communication symptoms in
resentativeness of the current sample that in common males vs. females [19, 21], but others have reported
with many studies, we excluded ASD participants with higher levels of social communication symptoms in
severe intellectual disability and this remains a challenge females [22, 23]. In contrast, we found no sex differences
to scientific enquiry, in particular perhaps in the domain on the parent-report questionnaire measures of ASD
of cognitive neuroscience [84]. Related to this point, we symptoms (SRS-2, CSBQ/ASBQ, RBS-R and SSP). Diag-
note that the ADOS CSS scores were somewhat lower nostic measures like the ADOS and ADI-R differ from the
overall in the current LEAP sample (Table 7) compared to parent-report ASD symptom questionnaires in several
other large cohorts such as the Simons Simplex Collection ways, including that the ADOS is an observer-rated
[85] which predominantly consists of clinically ascertained measure of current ASD symptoms and the ADI-R
samples and included participants with lower IQ than in algorithm domain scores assess historical symptom
the present volunteer research sample where IQ was severity (4 to-5 years and ever). The parent-report and
restricted to IQ ≥50 due to the experimental protocol. self-report questionnaires by design are intended to meas-
Reflecting recruitment from multiple research sites in ure symptoms or traits in a more continuous or dimen-
four countries from existing research cohorts and from sional fashion compared to these diagnostic tools.
different clinic and volunteer sources, there were signifi- However, it remains unclear as to why males had higher
cant site effects on the core characterisation measures ASD symptom severity scores on the diagnostic measures
identified in the mixed-effects models. However, ICCs but not the questionnaire measures. One possible explan-
were mostly below 10% (the exception was age which ation is a bias or expectation of researchers administering
reflects that some sites only sampled across some of the the ADOS and ADI-R, perhaps due to expectations about
schedule groups). This reflects that there was consider- sex differences—for example awareness of female com-
able heterogeneity of cognitive ability levels and scores pensatory behaviours and strengths—in ASD symptom
on core diagnostic measures within each site but system- profiles. Another possibility is that parent-reported ques-
atic differences between sites on these measures ranged tionnaire measures are influenced by parents’ gender
from minimal to moderate only. The quality control stereotypes. Alternatively, diagnostic measures that tap
procedures we implemented give us confidence in the variation in clinical level symptoms and ‘trait’ measures of
coherence and comparability of data collected across individual differences across populations of the ASD
six sites. phenotype are of a different kind, although recent twin
In addition to the well-established diagnostic measures studies suggest that they share a common genetic archi-
ADI-R and ADOS, we have further characterised ASD tecture [86]. A final point to note is that, with the notable
symptomatology using a range of dimensional parent- exception of the SRS-2, none of the other measures have
report (and, in adolescents and adults, self-report) sex-specific norms which should be a future goal for
measures of global ASD symptom severity (SRS-2, further psychometric development of ASD symptom
CSBQ/ASBQ, AQ) as well as specific measures of measures (Table 8) [18, 87].
Table 7 Summary of ASD dimensional measures by schedule and group
Charman et al. Molecular Autism (2017) 8:27

Autism symptomatology measures
SRS-2 72.12a (11.75) 47.96a (9.64) 64.39a (10.85) 47.59b (5.89) 74.31a (10.71) 45.71a (6.38) 75.04a (11.25) 44.86a (5.28) 76.41a (9.93) 64.55a (10.29)
n = 357 n = 158 n = 99 n = 90 n = 105 n = 75 n = 90 n = 57 n = 63 n = 20
CSBQ/ASBQ 46.88a (16.59) 7.10a (9.94) 32.05b (14.88) 7.85b (8.65) 44.95a (16.87) 5.88a (8.71) 45.93a (16.37) 5.21a (7.91) 55.55a (13.74) 24.55a (10.61)
n = 225 n = 139 n = 102 n = 82 n = 105 n = 75 n = 87 n = 53 n = 33 n = 11
AQ – – 82.53 be (19.49) 47.26be (14.92) 94.66ad (17.92) 42.63ad (16.49) 92.74ac (17.18) 39.46ac (17.15) 99.36ad (17.33) 81.30ad (12.02)
n = 105 n = 82 n = 99 n = 70 n = 86 n = 54 n = 28 n = 10
RBS-R 16.75a (13.85) 2.52a (8.04) 10.73a (10.13) – 16.53a (13.95) 1.01a (2.09) 19.21a (12.94) 1.07a (2.17) 22.55a (16.82) 11.86a (18.62)
n = 346 n = 157 n = 91 n = 106 n = 73 n = 87 n = 56 n = 62 n = 22
SSP 138.13a (27.46) 175.39a (17.12) 155.62a (23.88) – 137.35a (26.85) 180.52a(11.00) 130.31a (24.77) 175.68a (12.75) 132.17a (28.37) 155.11a (28.43)
n = 293 n = 141 n = 60 n = 96 n = 62 n = 85 n = 56 n = 52 n = 19
ns are lower for some measures due to missing data
SRS-2 Social Responsiveness Scale–2, CSBQ/ASBQ Children’s Social Behaviour Questionnaire (parent-report, administered to children and adolescents), Adults’ Social Behaviour Questionnaire (self-report, administered to
adults), RBS-R Repetitive Behavior Scale–Revised, SSP Short Sensory Profile, AQ Autism Spectrum Quotient (children, adolescents or adult version)
a
Parent-report
b
Self-report
c
AQ-Child
d
AQ-Adolescent
e
AQ-Adult
Page 14 of 21
Charman et al. Molecular Autism (2017) 8:27
Table 8 Summary of psychiatric symptom measures by schedule and group

Psychiatric symptom measures
ADHD—inattentiveness 4.55a (3.15) 1.29a (2.34) 3.18a (3.19) 0.88b (1.67) 4.82a (3.19) 0.88a (1.79) 5.20a (2.98) 0.67a (1.60) 5.21a (2.67) 5.17a (2.85)
n = 350 n = 153 n = 94 n = 84 n = 106 n = 75 n = 88 n = 54 n = 62 n = 18
ADHD—hyperactivity/impulsivity 2.83a (2.86) 0.56a (1.59) 1.22a (1.72) 0.55b (1.32) 2.70a (2.78) 0.19a (0.82) 4.41a (2.93) 0.35a (1.15) 3.27a (2.99) 2.94a (2.94)
n = 350 n = 153 n = 94 n = 84 n = 106 n = 75 n = 88 n = 54 n = 62 n = 18
Anxiety – – 15.05bc (12.76) 4.41bc (5.19) 49.92bd (10.63) 44.42bd (7.55) 47.81ad (8.96) 38.33ad (5.18) 49.17ad (9.45) 45.40ad (12.45)
n = 104 n = 85 n = 65 n = 65 n = 83 n = 51 n = 29 n = 10
Depression – – 14.11bc (12.41) 3.88bc (4.93) 51.48bd (10.55) 45.56bd (8.05) 49.98ad (10.22) 39.29ad (4.71) 53.17ad (10.07) 50.20ad (18.01)
n = 103 n = 85 n = 66 n = 66 n = 86 n = 56 n = 29 n = 10
ns are lower for some measures due to missing data
ADHD—inattentiveness (ADHD rating scale—inattentiveness subscale), ADHD—hyperactivity/impulsivity (ADHD rating scale—hyperactivity/impulsivity subscale), Anxiety Beck Anxiety Inventory, Depression (Beck
Depression Inventory–Second Edition)
a
Parent-report
b
Self-report
c
Raw scores
d
Standardised scores
Page 15 of 21
Age and IQ differences in ASD symptoms females and both inattentive and hyperactive/impulsive
On the diagnostic measures (ADOS and ADI-R), there symptoms were lower in adults than in adolescents, as
were no age differences in symptom severity. However, has been found in non-ASD samples [88]. Female adults
on the SRS-2 (a parent-report global measure of ASD with ASD reported higher levels of depressive, but not
symptoms), adults with ASD had lower symptom sever- anxiety, symptoms than males. This finding is potentially
ity than adolescents and children and the ASD group important to emphasise so that clinicians do not over-
with mild ID. A similar pattern was found on the look possible symptoms of depression in adult females
parent-report measure of restricted, repetitive and ste- with ASD. The proportion of individuals with elevated
reotyped behaviour, the RBS-R, with adults with ASD anxiety scores is lower in the current sample than in
scoring lower than all other groups. The findings were many previous studies, but note that we were using
corroborated when age was analysed in a continuous questionnaire screening measures of psychiatric symp-
fashion rather than according to the age and ability toms and not diagnostic instruments where 30 to 40% of
schedule presented here (see Additional file 2: Table S1). individuals with ASD have met criteria for an anxiety
This is consistent with a number of other studies show- disorder [7, 89]. Parent-report and self-report of co-
ing reduced ASD symptoms in adulthood, including occurring psychiatric symptoms were weakly negatively
samples followed longitudinally since childhood [34–36]. correlated with IQ, consistent with some previous stud-
With only one time-point of data, we cannot yet deter- ies [38, 45]. Most parent-report measures of ASD symp-
mine if the age differences in symptom severity are due to toms were moderately to strongly associated with
cross-sectional differences in sampling or true in nature parent-report of both ADHD inattention and hyperactiv-
but the accelerated longitudinal design of the LEAP study ity/impulsivity symptoms [90] (and similarly for self-
will allow us to investigate this in the future. reported ASD symptoms and self-reported associated
Social communication symptoms as measured by the psychiatric symptoms) but the AQ somewhat less so
ADOS Social Affect CSS and ADI-R Social and Commu- (see Fig. 4). Parent-report of ASD symptoms was only
nication domain scores were moderately negatively asso- moderately associated with self-reported anxiety and
ciated with IQ—with higher scores in those with lower depression, as has been previously reported in ASD [91]
IQ—but this was not the case for the ADOS RRB CSS and non-ASD samples [92]. We note that the validity of
or the ADI-R RRB domain. On the continuous measures assessments of psychiatric symptoms in samples of individ-
of ASD symptomatology, the SRS-2 and RBS-R were uals with ASD is unknown, perhaps especially with respect
also correlated negatively with IQ but the AQ and SSP to anxiety symptoms, although the measures we chose are
were not. Note, however, that even when these associa- widely used, including in previous studies in ASD.
tions were significant in this large and well-powered
sample, the variance in common between IQ and symp- Self-report measures of the ASD phenotype
tom measures (r-squared) was only ~5%. This is in line In contrast to the higher symptom scores in males com-
with previous studies where low IQ has been modestly pared to females on the diagnostic measures the ADOS
but significantly associated with higher levels of ASD and ADI-R (but not on parent-report questionnaire
symptom severity [41, 42]. This may, in part, reflect the measures of symptom severity), in a sub-sample of
fact that many diagnostic and dimensional measures of adults and adolescents with ASD able to self-report on
ASD symptomatology include a mixture of developmen- the SRS-2, ASBQ and AQ female adults reported higher
tal abilities or skills and frank atypical behaviours, in levels of symptoms than males. A similar pattern has
particular for children and adolescents. Alternatively, been reported in previous studies [93, 94] and may be
individuals with ASD with higher cognitive ability due to higher self-reflective ability in adult females than
might develop compensatory or alternative strategies males with ASD, identity-driven ‘biases’ or truly height-
to develop social communication skills resulting in ened ASD traits. The different pattern of findings for
slightly reduced symptom presentation. When looking self- vs. parent-report of ASD symptoms might also indi-
at associations between putative ASD biomarkers and cate an effect described as ‘masking’ or ‘camouflage’ in
measures of the core ASD phenotype and co-occurring (adult and adolescent) females with ASD whereby symp-
psychiatric symptoms, it will be important to consider the toms appear ameliorated to observers (in this case par-
effect of IQ as associations dependent or independent of ents) due to compensatory social engagement skills [18].
intellectual ability might indicate different neurobiological We also found contrasting patterns of self- vs. parent-
mechanisms. report of ASD symptoms with respect to age, with
parent report SRS-2 scores showing lower symptoms in
Co-occurring psychiatric symptoms adults than adolescents but self-report finding the
Among individuals with ASD, males had higher levels of reverse. One important contribution the current study
inattentive and hyperactive/impulsive symptoms than makes is the inclusion of a range of ascertainment
methods of ASD symptoms including clinician observa- The diagnostic measures have particular characteristics
tion and both parent- and self-report. These are import- that might make them useful at different levels/stages in
ant considerations both for identifying biomarkers the biomarker validation process. For example, the ADI-R
associated with the ASD phenotype and potentially for diagnostic algorithm domain scores are based on past his-
use as outcome measures in future clinical trials. The tory and in particular the early developmental period (age 4
issues raised are complex and go beyond the sample to 5 years) when it has been proposed that ASD presenta-
description contained in the current paper but in- tion is most prototypical [97]. On the other hand, the
clude what it might mean if biomarkers relate to one ADOS is a researcher/clinician-rated observational measure
type of measure but not another and what measures and is therefore less likely to suffer from the same potential
(e.g. clinician-report vs. parent-report vs. self-report) ‘halo effect’ when a parent is rating (for example, on two
should be used as outcome measures in clinical trials questionnaires) different behavioural characteristics (e.g.
and who gets to make these choices [61]. ASD and ADHD), thus reducing systematic rater bias.
We have also found modest but robust associations
Relevance of in-depth clinical characterisation for biomarker between severity of ASD symptoms and participant
analysis characteristics such as age, sex and IQ as well as with
Within the framework of the NIH Research Domain levels of co-occurring psychiatric symptoms. These con-
Criteria (RDoC; [95]) initiative, core neurobiological siderations will be important for considering the sensi-
or genetic systems vulnerabilities might map better tivity and specificity of any associations found between
onto neurodevelopmental or neurocognitive systems the ASD phenotype and potential biomarkers. The asso-
than the disorder-specific behavioural domains. This ciations between potential stratification biomarkers and
guided the ‘deep phenotyping’ approach we have ASD symptoms can be tested in models that include
taken in the EU-AIMS LEAP study to characterise the these factors where they are associated with the ASD
cohort not only comprehensively in terms of their ASD phenotypic scores themselves. The LEAP cohort has
and co-occurring disorders behavioural phenotype but purposively been ‘deep phenotyped’ at a number of levels
also at the level of structural and functional brain so that biomarker detection analysis in this large sample
development, neurocognitive function and biochemical can take account of these factors.
and genomic assays [46], consistent with other ‘big
data’ approaches in psychiatry [17]. Conclusions
Choices as to which ASD symptom measures should The in-depth clinical characterisation of the EU-AIMS
be used for biomarker validation need to be informed by LEAP cohort will allow us to test how a wide range of
a number of considerations. These include statistically potential biological and neurocognitive biomarkers [46–49]
guided principles regarding distributions (in both cases are associated with both diagnostic and more dimensional
and controls). Across the range of ASD phenotypic mea- measures of the core ASD phenotype. We will be able to
sures acquired in the LEAP sample, some are highly test whether these associations are influenced by the pres-
skewed even in the ASD sample (e.g. SSP), while other ence of commonly co-occurring psychiatric symptoms, as
measures are dimensional and more akin to ‘trait’ mea- well as whether they differ across males and females or ac-
sures and have considerable variation in both the ASD cording to age or intellectual ability. In addition, the pattern
and control samples (e.g. SRS-2, CSBQ/ASBQ, AQ). of associations we have found in the LEAP cohort differs
Although skewed data can be statistically transformed across the clinician observational and parent- vs. self-report
back towards normality, non-parametric, ordinal or questionnaire measures and both conceptual and methodo-
categorical approaches can also be adopted but this logical considerations should guide how these issues are
needs to be mapped back onto the clinical phenomena addressed in stratification biomarker analysis. The inclusion
that any phenotypic measure is assaying. Another con- of multiple dimensional measures of ASD symptom
sideration will be the extent to which potential bio- severity will allow us to test which measure relates best to
markers are examined in terms of their association with neurobiological or neurocognitive biomarkers and is most
‘domains’ or ‘sub-domains’ of the ASD phenotype, for sensitive to change over time. This would have important
example within the repetitive behaviours domain there is implications for choosing appropriate outcome measures in
some evidence at the genetic level that different genes future clinical trials. We anticipate that as the EU-AIMS
might associated with ‘lower’ vs. ‘higher’ levels of repeti- LEAP cohort is followed into the future, it will become a
tive behaviour [96]. Finally, we have reported both raw key resource of autism discovery science.
and age and sex-normed T scores on an instrument such
as the SRS-2 in this clinical paper but for biomarker Endnotes
1
analysis raw un-adjusted scores allows a more neutral At four additional sites (UCAM, CIMH, UCBM and
mapping onto the phenotypic behaviour. UMCU) following assessment, a minority of participants
with ASD were allocated to the Mild ID group due to and JKB contributed to writing the manuscript. All authors read and approved
measured IQ falling in the 50–74 range (see Table 1). the final manuscript.
2
There are three individuals with a Full-scale IQ <50
in the sample (all ASD). Competing interests
Jan Buitelaar has been in the past 3 years a consultant to/member of
advisory board of/and/or speaker for Janssen Cilag BV, Eli Lilly, Lundbeck,
Additional files Shire, Roche, Novartis, Medice and Servier. He is not an employee of any of
these companies and not a stock shareholder of any of these companies. He
has no other financial or material support, including expert testimony, patents,
Additional file 1: Table S3. (DOCX 22 kb)
and royalties. Sven Bölte receives royalties for the German and Swedish
Additional file 2: Table S1. Predicted effect of age and IQ on ADOS KONTAKT manuals and adaptations of the ADI-R, ADOS, and SRS from Hogrefe
Calibrated Severity Scores. (DOCX kb) Publishers. Bölte has in the last 3 years acted as an author, consultant or lecturer
Additional file 3: Table S2. Predicted effect of age and IQ on parent- for Shire, Medice, Roche, Eli Lilly, Prima Psychiatry, GLGroup, System Analytic,
and self-report ASD measures. (DOCX kb) Kompetento, Expo Medica, Prophase and receives royalties for text books and
diagnostic tolls from Huber/Hogrefe, Kohlhammer and UTB. Lindsay Ham,
Xavier Liogier D’Ardhuy, Joerg Hipp, Pilar Garcés and Will Spooren are employees
Abbreviations at F. Hoffmann-La Roche Ltd. Gahan Pandina is an employee at Janssen. Andreas
ADHD: Attention-deficit/hyperactivity disorder; ADI-R: Autism Diagnostic Meyer-Lindenburg has received consultant fees and travel expenses from Alexza
Interview-Revised; ADOS: Autism Diagnostic Observation Schedule; Pharmaceuticals, AstraZeneca, Bristol-Myers Squibb, Defined Health, Decision
AQ: Autism Spectrum Quotient; ASBQ: Adult Social Behaviour Resources, Desitin Arzneimittel, Elsevier, F. Hoffmann-La Roche, Gerson Lehrman
Questionnaire; ASD: Autism Spectrum Disorder; BAI: Beck Anxiety Inventory; Group, Grupo Ferrer, Les Laboratoires Servier, Lilly Deutschland, Lundbeck
BDI-II: Beck Depression Inventory—Second Edition; BYI-II: Beck Youth Foundation, Outcome Sciences, Outcome Europe, PriceSpective, and Roche
Inventories; CIMH: Central Institute of Mental Health; CSBQ: Children’s Pharma and has received speaker’s fees from Abbott, AstraZeneca, BASF,
Social Behaviour Questionnaire; CSS: Calibrated Severity Scores; Bristol-Myers Squibb, GlaxoSmithKline, Janssen-Cilag, Lundbeck, Pfizer Pharma,
DSM-5: Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition; and Servier Deutschland. Tobias Banaschewski has served in an advisory or
DSM-IV: Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition; consultancy role for Actelion, Hexal Pharma, Lilly, Medice, Novartis, Oxford
DSM-IV-TR: Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition outcomes, Otsuka, PCM scientific, Shire and Viforpharma. He received conference
text-revision; EU-AIMS: European Autism Interventions—A Multicentre Study for support or speaker’s fee by Medice, Novartis and Shire. He is/has been involved
Developing New Medications; ICC: Intraclass correlation coefficient; in clinical trials conducted by Shire and Viforpharma. He received royalties from
ICD-10: International Statistical Classification of Diseases and Related Health Hogrefe, Kohlhammer, CIP Medien, and Oxford University Press. The present work
Problems, 10th Revision; ID: Intellectual disability; IoPPN: Institute of Psychiatry, is unrelated to the above grants and relationships. The other authors declare that
Psychology and Neuroscience; IQ: Intelligent quotient; KCL: King’s College they have no competing interests.
London; LEAP: Longitudinal European Autism Project; MRI: Magnetic resonance
imaging; P: Parent-report; RBS-R: Repetitive Behavior Scale-Revised;
RDoC: Research Domain Criteria; RRB: Repetitive and Restricted Behaviours; Consent for publication
RUNMC: Radboud University Nijmegen Medical Centre; S: Self-report; SA: Social Consent for publication was obtained from all participants prior to the study.
Affect; SRS-2: Social Responsiveness Scale, Second Edition; SSP: Short Sensory
Profile; TD: Typically developing; UCAM: Autism Research Centre, University of
Ethics approval and consent to participate
Cambridge; UCBM: University Campus Bio-Medico; UMCU: University Medical
All participants (where appropriate) and their parent/legal guardian provided
Centre Utrecht; WAIS-III/IV: Wechsler Adult Intelligence Scale—Third Edition/
written informed consent. Ethical approval for this study was obtained
Fourth Edition; WASI-II: Wechsler Abbreviated Scales of Intelligence—Second
through ethics committees at each site (Table a):
Edition; WISC-III/IV: Wechsler Intelligence Scale for Children—Third Edition/
Fourth Edition
Acknowledgements
We thank all participants and their families for their efforts to participate in Site Ethics committee ID/reference no.
the study. KCL London Queen Square Health Research 13/LO/1156
Authority Research Ethics Committee
Funding UCAM
This work was supported by EU-AIMS (European Autism Interventions), RUNMC Radboud Universitair Medisch Centrum 2013/455
which receives support from the Innovative Medicines Initiative Joint
Undertaking under grant agreement no. 115300, the resources of which UMCU Instituut Waarborging Kwaliteit en Veiligheid
are composed of financial contributions from the European Union’s Sev- Commissie Mensgebonden Onderzoek
enth Framework Programme (grant FP7/2007-2013), from the European Regio Arnhem-Nijmegen (Radboud
Federation of Pharmaceutical Industries and Associations companies’ in- University Medical Centre Institute Ensuring
kind contributions, and from Autism Speaks. Quality and Safety
Committee on Research Involving Human
Availability of data and materials Subjects
The datasets generated and/or analysed during the current study are not Arnhem-Nijmegen)
publicly available due to an embargo period but are available from the
CIMH UMM Universitatsmedizin Mannheim, 2014-540N-MA
corresponding author on reasonable request.
Medizinishe Ethik Commission II
(UMM University Medical Mannheim,
Authors’ contributions Medical Ethics Commission II)
TC, EL, JT, DG, BA, ToB, SBC, CB, SB, TB, MB, DB, BC, FDA, GD, SD, CE, VF, PG,
LH, JH, RJH, MHJ, EJHJ, PK, MCL, XLD, MVL, DJL, RM, LM, AML, CM, LOD, MO, UCBM Universita Campus Bio-Medica De Roma 18/14 PAR
BO, GP, AMP, RS, ES, RT, HT, JW, SCRW, MPZ, WS, DGM and JKB designed the Comitato Etico (University Campus ComET CBM
study and developed acquisition and/or analysis protocols. CW, DC, JA, SA, Bio-Medical Ethics Committee De Roma)
SB, CB, CB, YdB, IC, JF, HH, JI, NM, BR, ANVR, JS and ASJC collected the data.
KI Centrala Etikprovningsnamnden (Central 32-2010
TC wrote the first and final draft of the manuscript. TC, EL, JT, EJHJ, JB, ToB,
Ethical Review Board)
SBC, SB, SB, TB, DG, LMH, MHJ, MCL, BO, GP, AMP, ASJC, ES, HT, WS, DGM
Publisher’s Note 2. Brugha TS, McManus S, Bankart J, Scott F, Purdon S, Smith J, Bebbington P,
Springer Nature remains neutral with regard to jurisdictional claims in Jenkins R, Meltzer H. Epidemiology of autism spectrum disorders in adults
published maps and institutional affiliations. in the community in England. Arch Gen Psychiatry. 2011;68:459–65.
3. Christensen DL. Prevalence and characteristics of autism spectrum disorder
Author details among children aged 8 years—autism and developmental disabilities
1
Department of Psychology, Institute of Psychiatry, Psychology and monitoring network, 11 sites, United States, 2012. MMWR Surveill Summ.
Neuroscience, King’s College London, De Crespigny Park, Denmark Hill, 2016;65:1–23.
London SE5 8AF, UK. 2Sackler Institute for Translational Neurodevelopment, 4. Gokoolparsadh A, Sutton G, A. C, Green N, Pardy C, Voineagu I. Searching
Institute of Psychiatry, Psychology and Neuroscience, King’s College London, for convergent pathways in autism spectrum disorders: insights from
De Crespigny Park, Denmark Hill, London SE5 8AF, UK. 3Department of human brain transcriptome studies. Cell Mol Life Sci. 2016;73:4517–30.
Forensic and Neurodevelopmental Sciences, Institute of Psychiatry, 5. American Psychiatric Association. Diagnostic and statistical manual of
Psychology and Neuroscience, King’s College London, De Crespigny Park, mental disorders. 5th ed. Arlington: Author; 2013.
Denmark Hill, London SE5 8AF, UK. 4Department of Neuroimaging, Institute 6. Lai M-C, Lombardo MV, Chakrabarti B, Baron-Cohen S. Subgrouping the
of Psychiatry, Psychology and Neuroscience, King’s College London, De autism “spectrum”: reflections on DSM-5. PLoS Biol. 2013;11:e1001544.
Crespigny Park, Denmark Hill, London SE5 8AF, UK. 5Neurospin Centre CEA, 7. Simonoff E, Pickles A, Charman T, Chandler S, Loucas T, Baird G. Psychiatric
Saclay 91191, Gif sur Yvette, France. 6Autism Research Centre, Department of disorders in children with autism spectrum disorders: prevalence,
Psychiatry, University of Cambridge, Douglas House, 18b Trumpington Road, comorbidity, and associated factors in a population-derived sample. J Am
Cambridge CB2 8AH, UK. 7The School of Philosophy, Psychology, and Acad Child Adolesc Psychiatry. 2008;47:921–9.
Language Sciences, University of Edinburgh, Dugald Stewart Building, 3 8. Lai M-C, Lombardo MV, Baron-Cohen S. Autism. Lancet. 2014;383:896–910.
Charles Street, Edinburgh EH8 9AD, UK. 8Department of Psychiatry, Brain 9. Charman T. Variability in neuro-developmental disorders evidence from
Center Rudolf Magnus, University Medical Center Utrecht, Universiteitsweg autism spectrum disorders. In: Van Herwegen J, Riby DM, editors.
100, 3584 CG, Utrecht, The Netherlands. 9Child and Adolescent Psychiatry, Neurodevelopmental Disorders : Research Challenges and Solutions. Hove:
Central Institute of Mental Health, University of Heidelberg, Medical Faculty Psychology Press; 2015.
Mannheim, J5, 68159 Mannheim, Germany. 10Donders Institute for Brain, 10. Lord C, Bishop S, Anderson D. Developmental trajectories as autism
Cognition and Behaviour, Radboud University Nijmegen Medical Centre, phenotypes. Am J Med Genet C: Semin Med Genet. 2015;169:198–208.
Kapittelweg 29, 6525 EN Nijmegen, The Netherlands. 11Center for 11. LLombardo MV, Lai M-C, Auyeung B, Holt RJ, Allison C, Smith P, Chakrabarti B,
Neurodevelopmental Disorders at Karolinska Institutet (KIND), Stockholm, Ruigrok AN, Suckling J, Bullmore ET, MRC AIMS Consortium, Ecker C, Craig MC,
Sweden. 12Child and Adolescent Psychiatry, Center of Psychiatry Research, Murphy DGM, Happé F, Baron-Cohen S. Unsupervised data-driven stratification
Stockholm County Council, Stockholm, Sweden. 13Institut Pasteur, Human of mentalizing heterogeneity in autism. Scientific Reports. 2016;6:35333.
Genetics and Cognitive Functions Unit, 25 Rue du Docteur Roux, Paris Cedex 12. Gaugler T, Klei L, Sanders SJ, Bodea CA, Goldberg AP, Lee AB, Mahajan M,
15, France. 14University Campus Bio-Medico, Via Álvaro del Portillo, 21, Rome, Manaa D, Pawitan Y, Reichert J. Most genetic risk for autism resides with
Italy. 15School of Psychology and Clinical Language Sciences, University of common variation. Nat Genet. 2014;46:881–5.
Reading, Whiteknights, Reading RG6 6AL, UK. 16Department of Child and 13. Geschwind DH, State MW. Gene hunting in autism spectrum disorder: on
Adolescent Psychiatry, Psychosomatics and Psychotherapy, University the path to precision medicine. Lancet Neurol. 2015;14:1109–20.
Hospital Frankfurt am Main, Goethe University, Deutschordenstrasse 50, 14. Bourgeron T. From the genetic architecture to synaptic plasticity in autism
60528 Frankfurt, Germany. 17Roche Pharma Research and Early Development, spectrum disorder. Nat Rev Neurosci. 2015;16:551–63.
Neuroscience, Ophthalmology and Rare Diseases, Roche Innovation Center 15. Mandy W, Lai MC. Annual research review: the role of the environment in
Basel, Grenzacherstrasse 124, B.001 N.667, CH-4070 Basel, Switzerland. the developmental psychopathology of autism spectrum condition. J Child
18
Regulatory Affairs, Pharmaceutical Development, F. Hoffmann-La Roche Psychol Psychiatry. 2016;57:271–92.
Pharmaceuticals, Grenzacherstrasse 124, CH-4070 Basel, Switzerland. 16. Jeste SS, Geschwind DH. Disentangling the heterogeneity of autism
19
Department of Neuroscience, Uppsala University, Uppsala, Sweden. spectrum disorder through genetic findings. Nat Rev Neurol. 2014;10:74–81.
20
Centre for Brain and Cognitive Development, Birkbeck, University of
17. Zhao Y, Castellanos FX. Annual research review: discovery science
London, 32 Torrington Square, London WC1E 7JL, UK. 21Department of
strategies in studies of the pathophysiology of child and adolescent
Radiology, Icahn School of Medicine at Mount Sinai, NY, USA. 22Child and
psychiatric disorders‐promises and limitations. J Child Psychol
Youth Mental Health Collaborative, Centre for Addiction and Mental Health
Psychiatry. 2016;57:421–39.
and The Hospital for Sick Children, Department of Psychiatry, University of
18. Lai M-C, Lombardo MV, Auyeung B, Chakrabarti B, Baron-Cohen S. Sex/
Toronto, 80 Workman Way, Toronto, ON M6J 1H4, Canada. 23Center for
gender differences and autism: setting the scene for future research. J Am
Applied Neuroscience, Department of Psychology, University of Cyprus, PO
Acad Child Adolesc Psychiatry. 2015;54:11–24.
Box 205371678 Nicosia, Cyprus. 24Department of Psychiatry and
19. Van Wijngaarden-Cremers PJ, van Eeten E, Groen WB, Van Deurzen PA,
Psychotherapy, Central Institute of Mental Health, Medical Faculty Mannheim,
Oosterling IJ, Van der Gaag RJ. Gender and age differences in the core triad
University of Heidelberg, 68159 Mannheim, Germany. 25Janssen Research &
of impairments in autism spectrum disorders: a systematic review and
Development, 1125 Trenton Harbourton Road, Titusville, NJ 08560, USA.
meta-analysis. J Autism Dev Disord. 2014;44:627–35.
26
Child and Adolescent Neuropsychiatry Unit, “Gaetano Martino” University
20. Dworzynski K, Ronald A, Bolton P, Happé F. How different are girls and boys
Hospital, University of Messina, via Consolare Valeria 1, I-98125 Messina, Italy.
27 above and below the diagnostic threshold for autism spectrum disorders?
Social, Genetic and Developmental Psychiatry Centre, Institute of
J Am Acad Child Adolesc Psychiatry. 2012;51:788–97.
Psychiatry, Psychology and Neuroscience, King’s College London, Denmark
21. Wilson CE, Murphy CM, McAlonan G, Robertson DM, Spain D, Hayward H,
Hill, London, UK. 28Department of Child and Adolescent Psychiatry, Institute
Woodhouse E, Deeley PQ, Gillan N, Ohlsen JC. Does sex influence the diagnostic
of Psychology, Psychiatry and Neuroscience, King’s College London, De
evaluation of autism spectrum disorder in adults? Autism. 2016;20:808–19.
Crespigny Park, Denmark Hill, London SE5 8LF, UK. 29Roche Pharmaceutical
22. Frazier TW, Georgiades S, Bishop SL, Hardan AY. Behavioral and cognitive
Research and Early Development, NORD Discovery and Translational Area,
characteristics of females and males with autism in the Simons Simplex
Roche Innovation Center Basel, Grenzacherstrasse 124, CH-4070 Basel,
Switzerland. Collection. J Am Acad Child Adolesc Psychiatry. 2014;53:329–40. e323.
23. Szatmari P, Liu XQ, Goldberg J, Zwaigenbaum L, Paterson AD,
Received: 14 November 2016 Accepted: 19 May 2017 Woodbury‐Smith M, Georgiades S, Duku E, Thompson A. Sex differences
in repetitive stereotyped behaviors in autism: implications for genetic
liability. Am J Med Genet B Neuropsychiatr Genet. 2012;159:5–12.
24. Hiller RM, Young RL, Weber N. Sex differences in autism spectrum disorder
References based on DSM-5 criteria: evidence from clinician and teacher reporting.
1. Baird G, Simonoff E, Pickles A, Chandler S, Loucas T, Meldrum D, Charman T. J Abnorm Child Psychol. 2014;42:1381–93.
Prevalence of disorders of the autism spectrum in a population cohort of 25. Mandy W, Chilvers R, Chowdhury U, Salter G, Seigal A, Skuse D. Sex
children in South Thames: the Special Needs and Autism Project (SNAP). differences in autism spectrum disorder: evidence from a large sample of
Lancet. 2006;368:210–5. children and adolescents. J Autism Dev Disord. 2012;42:1304–13.
26. May T, Cornish K, Rinehart N. Does gender matter? A one year follow-up of 49. Loth E, Spooren W, Ham LM, Isaac MB, Auriche-Benichou C,
autistic, attention and anxiety symptoms in high-functioning children with Banaschewski T, Baron-Cohen S, Broich K, Bölte S, Bourgeron T.
autism spectrum disorder. J Autism Dev Disord. 2014;44:1077–86. Identification and validation of biomarkers for autism spectrum
27. Charman T, Taylor E, Drew A, Cockerill H, Brown J-A, Baird G. Outcome at disorders. Nat Rev Drug Discov. 2016;15:70–3.
7 years of children diagnosed with autism at age 2: predictive validity of 50. American Psychiatric Association. Diagnostic and statistical manual of
assessments conducted at 2 and 3 years of age and pattern of symptom mental disorders. 4th ed. Washington, DC: Author; 1994.
change over time. J Child Psychol Psychiatry. 2005;46:500–13. 51. Association AP. Diagnostic and statistical manual of mental disorders. 4th
28. Fountain C, Winter AS, Bearman PS. Six developmental trajectories ed. Washington, DC: American Psychiatric Association; 2000.
characterize children with autism. Pediatrics. 2012;129:e1112–20. 52. Organization WH. The ICD-10 Classification of mental and behavioural
29. Gotham K, Pickles A, Lord C. Trajectories of autism severity in children using disorders: diagnostic criteria for research. Geneva: WHO; 1993. 2013.
standardized ADOS scores. Pediatrics. 2012;130:e1278–84. 53. Lord C, Risi S, Lambrecht L, Cook EH, Leventhal BL, DiLavore PC, Pickles A,
30. Lombardo MV, Pierce K, Eyler LT, Barnes CC, Ahrens-Barbeau C, Solso S, Rutter M. The Autism Diagnostic Observation Schedule—Generic: a
Campbell K, Courchesne E. Different functional neural substrates for good standard measure of social and communication deficits associated with the
and poor language outcome in autism. Neuron. 2015;86:567–77. spectrum of autism. J Autism Dev Disord. 2000;30:205–23.
31. Szatmari P, Georgiades S, Duku E, Bennett TA, Bryson S, Fombonne E, 54. Lord C, Rutter M, DiLavore PC, Risi S, Gotham K, Bishop S. Autism Diagnostic
Mirenda P, Roberts W, Smith IM, Vaillancourt T. Developmental trajectories Observation Schedule, Second Edition (ADOS-2) manual (part I): modules
of symptom severity and adaptive functioning in an inception cohort of 1–4. Torrance: Western Psychological Services; 2012.
preschool children with autism spectrum disorder. JAMA Psychiatry. 2015; 55. Rutter M, Le Couteur A, Lord C. Autism Diagnostic Interview-Revised. Los
72:276–83. Angeles: Western Psychological Services; 2003.
32. Bal VH, Kim S-H, Cheong D, Lord C. Daily living skills in individuals with autism 56. Lord C, Risi S, DiLavore PS, Shulman C, Thurm A, Pickles A. Autism from 2 to
spectrum disorder from 2 to 21 years of age. Autism. 2015;19:774–84. 9 years of age. Arch Gen Psychiatry. 2006;63:694–701.
33. Szatmari P, Bryson S, Duku E, Vaccarella L, Zwaigenbaum L, Bennett T, Boyle 57. Charman T, Gotham K. Measurement issues: screening and diagnostic
MH. Similar developmental trajectories in autism and Asperger syndrome: instruments for autism spectrum disorders—lessons from research and
from early childhood to adolescence. J Child Psychol Psychiatry. 2009;50: practise. Child Adolesc Mental Health. 2013;18:52–63.
1459–67. 58. Constantino JN, Gruber CP. Social Responsiveness Scale. 2nd ed. Los
34. Billstedt E, Carina Gillberg I, Gillberg C. Autism in adults: symptom patterns Angeles: Western Psychological Services; 2012.
and early childhood predictors. Use of the DISCO in a community sample 59. Anagnostou E, Jones N, Huerta M, Halladay AK, Wang P, Scahill L, Horrigan
followed from childhood. J Child Psychol Psychiatry. 2007;48:1102–10. JP, Kasari C, Lord C, Choi D. Measuring social communication behaviors as a
35. Howlin P, Moss P, Savage S, Rutter M. Social outcomes in mid-to later treatment endpoint in individuals with autism spectrum disorder. Autism.
adulthood among individuals diagnosed with autism and average 2015;19:622–36.
nonverbal IQ as children. J Am Acad Child Adolesc Psychiatry. 2013;52: 60. McConachie H, Parr JR, Glod M, Hanratty J, Livingstone N, Oono IP, Robalino S,
572–81. e571. Baird G, Beresford B, Charman T, Garland D, Green J, Gringras P, Jones G, Law J,
36. Shattuck PT, Seltzer MM, Greenberg JS, Orsmond GI, Bolt D, Kring S, Lounds Le Couteur AS, Macdonald G, McColl EM, Morris C, Rodgers J, Simonoff E,
J, Lord C. Change in autism symptoms and maladaptive behaviors in Terwee CB, Williams K. Systematic review of tools to measure outcomes for
adolescents and adults with an autism spectrum disorder. J Autism Dev young children with autism spectrum disorder. Health Technol Assess.
Disord. 2007;37:1735–47. 2015;41:1–506.
37. Verheij C, Louwerse A, van der Ende J, Eussen M, Van Gool A, Verheij F, 61. Scahill L, Hallett V, Aman MG, McDougle CJ, Arnold LE, McCracken JT,
Verhulst F, Greaves-Lord K. The stability of comorbid psychiatric disorders: Tierney E, Deng Y, Dziura J, Vitiello B. Brief report: Social disability in autism
a 7 year follow up of children with pervasive developmental disorder-not spectrum disorder: Results from research units on pediatric
otherwise specified. J Autism Dev Disord. 2015;45:3939–48. psychopharmacology (RUPP) autism network trials. J Autism Dev Disord.
38. Simonoff E, Jones CR, Baird G, Pickles A, Happé F, Charman T. The 2013;43:739–46.
persistence and stability of psychiatric problems in adolescents with autism 62. Baron-Cohen S, Wheelwright S, Skinner R, Martin J, Clubley E. The
spectrum disorders. J Child Psychol Psychiatry. 2013;54:186–94. Autism-Spectrum Quotient (AQ): evidence from Asperger syndrome/
39. Taylor JL, Seltzer MM. Changes in the autism behavioral phenotype during high-functioning autism, males and females, scientists and
the transition to adulthood. J Autism Dev Disord. 2010;40:1431–46. mathematicians. J Autism Dev Disord. 2001;31:5–17.
40. Lever AG, Geurts HM. Psychiatric co-occurring symptoms and disorders in 63. Baron-Cohen S, Hoekstra RA, Knickmeyer R, Wheelwright S. The
young, middle-aged, and older adults with autism spectrum disorder. J autism-spectrum quotient (AQ)—adolescent version. J Autism Dev
Autism Dev Disord. 2016;46:1916–30. Disord. 2006;36:343–50.
41. Bishop SL, Richler J, Lord C. association between restricted and repetitive 64. Auyeung B, Baron-Cohen S, Wheelwright S, Allison C. The autism spectrum
behaviors and nonverbal IQ in children with autism spectrum disorders. quotient: children’s version (AQ-Child). J Autism Dev Disord. 2008;38:1230–40.
Child Neuropsychol. 2006;12:247–67. 65. Hartman CA, Luteijn E, Serra M, Minderaa R. Refinement of the Children’s
42. Mayes SD, Calhoun SL, Murray MJ, Zahid J. Variables associated with anxiety Social Behavior Questionnaire (CSBQ): an instrument that describes the
and depression in children with autism. J Dev Phys Disabil. 2011;23:325–37. diverse problems seen in milder forms of PDD. J Autism Dev Disord. 2006;
43. Ford T, Goodman R, Meltzer H. The relative importance of child, family, 36:325–42.
school and neighbourhood correlates of childhood psychiatric disorder. Soc 66. Horwitz E, Schoevers R, Ketelaars C, Kan C, van Lammeren A, Meesters Y, Spek
Psychiatry Psychiatr Epidemiol. 2004;39:487–96. A, Wouters S, Teunisse J, Cuppen L. Clinical assessment of ASD in adults using
44. Hinshaw SP. Externalizing behavior problems and academic self-and other-report: psychometric properties and validity of the Adult Social
underachievement in childhood and adolescence: causal relationships and Behavior Questionnaire (ASBQ). Res Autism Spectr Disord. 2016;24:17–28.
underlying mechanisms. Psychol Bull. 1992;111:127. 67. Tomchek SD, Dunn W. Sensory processing in children with and without
45. van Steensel FJA, Bögels SM, Perrin S. Anxiety disorders in children and autism: a comparative study using the short sensory profile. Am J Occup
adolescents with autistic spectrum disorders: a meta-analysis. Clin Child Fam Ther. 2007;61:190–200.
Psychol Rev. 2011;14:302. 68. Bodfish JW, Symons FJ, Parker DE, Lewis MH. Varieties of repetitive behavior
46. Loth E, Charman T, Mason L, Jones EJH, Wooldridge C, Amad J, Auyeung B, in autism: comparisons to mental retardation. J Autism Dev Disord. 2000;30:
Brogna C, Ambrosino S, Banaschewski T, et al.. The EU-AIMS Longitudinal 237–43.
European Autism Project (LEAP): design and methodologies to identify and 69. Gotham K, Pickles A, Lord C. Standardizing ADOS scores for a measure of
validate stratification biomarkers for autism spectrum disorders. Molecular severity in autism spectrum disorders. J Autism Dev Disord. 2009;39:693–705.
Autism (submitted). 70. Hus V, Gotham K, Lord C. Standardizing ADOS domain scores: separating
47. Murphy D, Spooren W. EU-AIMS: a boost to autism research. Nat Rev Drug severity of social affect and restricted and repetitive behaviors. J Autism Dev
Discov. 2012;11:815. Disord. 2014;44:2400–12.
48. Loth E, Spooren W, Murphy DG. New treatment targets for autism spectrum 71. Dunn W. The sensory profile manual. San Antonio: Psychological
disorders: EU-AIMS. Lancet Psychiatry. 2014;1:413–5. Corporation; 1999.
72. Wechsler D. Wechsler Abbreviated Scale of Intelligence–Second Edition

(WASI-II). San Antonio: NCS Pearson; 2011.
73. Wechsler D. Wechsler Intelligence Scale for Children–Third Edition. San
Antonio: Psychological Corporation; 1991.
74. Wechsler D. Wechsler Intelligence Scale for Children–Fourth Edition. San
Antonio: Psychological Corporation; 2003.
75. Wechsler D. Wechsler Adult Intelligence Scale– Third Edition. San Antonio:
The Psychological Corporation; 1997.
76. Wechsler D. Wechsler Adult Intelligence Scale–Fourth Edition. San Antonio:
Pearson; 2008.
77. Sattler JM. Assessment of children (revised and updated. 3rd ed. San Diego:
Jerome M. Sattler; 1992.
78. Beck AT, Steer RA, Brown GK. Beck Depression Inventory-II. San Antonio: The
Psychological Corporation; 1996.
79. Beck JS, Beck AT, Jolly JB, Steer RA. Beck Youth Inventories. 2nd ed. San
Antonio: Harcourt Assessment; 2005.
80. Beck AT, Steer RA. Beck Anxiety Inventory Manual. San Antonio: The
Psychological Corporation Harcourt Brace & Company; 1993.
81. StataCorp. Stata statistical software: release 14. College Station:
StataCorp LP; 2015.
82. Tymms P. Effect sizes in multilevel models. In: Schagen I, Elliot K, editors. But
what does it mean? The use of effect sizes in educational research. Slough:
National Foundation for Educational Research; 2004.
83. Cohen J. A power primer. Psychol Bull. 1992;112:155.
84. Jack A, Pelphrey KA. Annual Research Review: Understudied populations
within the autism spectrum–current trends and future directions in
neuroimaging research. J Child Psychol Psychiatry. J Child Psychol
Psychiatry. 2017;58:411–35.
85. Bal VH, Lord C. Replication of standardized ADOS domain scores in the
Simons simplex collection. Autism Res. 2015;8:583–92.
86. Robinson EB, St Pourcain B, Anttila V, Kosmicki JA, Bulik-Sullivan B, Grove J,
Maller J, Samocha KE, Sanders SJ, Ripke S. Genetic risk for autism spectrum
disorders and neuropsychiatric variation in the general population. Nat
Genet. 2016;48:552–5.
87. Constantino JN, Charman T. Diagnosis of autism spectrum disorder:
reconciling the syndrome, its diverse origins, and variation in expression.
Lancet Neurol. 2016;15:279–91.
88. Costello EJ, Copeland W, Angold A. Trends in psychopathology across the
adolescent years: what changes when children become adolescents, and
when adolescents become adults? J Child Psychol Psychiatry. 2011;52:1015–25.
89. Leyfer OT, Folstein SE, Bacalman S, Davis NO, Dinh E, Morgan J, Tager-
Flusberg H, Lainhart JE. Comorbid psychiatric disorders in children with
autism: interview development and rates of disorders. J Autism Dev Disord.
2006;36:849–61.
90. Hus V, Bishop S, Gotham K, Huerta M, Lord C. Factors influencing scores on
the social responsiveness scale. J Child Psychol Psychiatry. 2013;54:216–24.
91. Mazefsky CA, Kao J, Oswald DP. Preliminary evidence suggesting caution in
the use of psychiatric self-report measures with adolescents with high-
functioning autism spectrum disorders. Res Autism Spectr Disord. 2011;5:
164–74.
92. Achenbach TM, McConaughy SH, Howell CT. Child/adolescent behavioral
and emotional problems: implications of cross-informant correlations for
situational specificity. Psychol Bull. 1987;101:213–32.
93. Baron-Cohen S, Cassidy S, Auyeung B, Allison C, Achoukhi M, Robertson S,
Pohl A, Lai M-C. Attenuation of typical sex differences in 800 adults with
autism vs. 3,900 controls. PLoS One. 2014;9:e102251.
94. Lai M-C, Lombardo MV, Pasco G, Ruigrok AN, Wheelwright SJ, Sadek SA,
Chakrabarti B, Baron-Cohen S, Consortium MA. A behavioral comparison of Submit your next manuscript to BioMed Central
male and female adults with high functioning autism spectrum conditions.
PLoS One. 2011;6:e20835. and we will help you at every step:
95. Cuthbert BN, Insel TR. Toward the future of psychiatric diagnosis: the seven
• We accept pre-submission inquiries
pillars of RDoC. BMC Med. 2013;11:1.
96. Staal WG, de Krom M, de Jonge MV. Brief report: the dopamine-3-receptor • Our selector tool helps you to find the most relevant journal
gene (DRD3) is associated with specific repetitive behavior in autism • We provide round the clock customer support
spectrum disorder (ASD). J Autism Dev Disord. 2012;42:885–8.
• Convenient online submission
97. Lord C, Rutter M, Le Couteur A. Autism Diagnostic Interview-Revised: a
revised version of a diagnostic interview for caregivers of individuals with • Thorough peer review
possible pervasive developmental disorders. J Autism Dev Disord. 1994;24: • Inclusion in PubMed and all major indexing services
659–85.
• Maximum visibility for your research
Submit your manuscript at

www.biomedcentral.com/submit

The EU-AIMS Longitudinal European Autism Project (LEAP) : Clinical Characterisation

Uploaded by

Copyright:

Available Formats

The EU-AIMS Longitudinal European Autism Project (LEAP) : Clinical Characterisation

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The EU-AIMS Longitudinal European Autism Project (LEAP) : Clinical Characterisation

Uploaded by

Copyright:

Available Formats

Charman et al.

Molecular Autism (2017) 8:27

RESEARCH Open Access

The EU-AIMS Longitudinal European Autism

(Continued from previous page)

Background and cellular architecture, brain structure and function,

on the self-report (1 adult) or parent-report form (1 ado- Clinical measures—ASD symptomatology

Table 2 Summary of parent-report and participant self-report questionnaires

Table 3 Sample characteristics

Site effects Diagnostic ASD measures—sex and age effects

males had significantly higher ADOS CSS Total than

Dimensional ASD measures—sex and age effects

Discussion repetitive (RBS-R) and sensory (SSP) symptoms. Further-

ASD TD/ID ASD TD ASD TD ASD TD ASD ID

Table 8 Summary of psychiatric symptom measures by schedule and group

72. Wechsler D. Wechsler Abbreviated Scale of Intelligence–Second Edition

Submit your manuscript at

You might also like