Research in Developmental Disabilities 34 (2013) 2900–2908

Contents lists available at SciVerse ScienceDirect

Research in Developmental Disabilities

Autism in community pre-schoolers: Developmental profiles

Anne-Katrin Kantzer a,b,*, Elisabeth Fernell a,c, Christopher Gillberg a,Carmela Miniscalco a

a Gillberg Neuropsychiatry Centre, Sahlgrenska Academy, University of Gothenburg, Swedenb Department of Child and Adolescent Psychiatry, NU Hospital Organization, Trollhattan/Uddevalla, Swedenc Unit of Neurodevelopmental Disorders, Skaraborg Hospital, Mariestad and FoU-Center, Skovde, Sweden

A R T I C L E I N F O

Article history:

Received 4 March 2013

Received in revised form 13 June 2013

Accepted 13 June 2013

Available online 28 June 2013

Keywords:

Autism spectrum disorder

Preschool community study

Developmental disorder

Cognitive function

ESSENCE

Comorbidity

A B S T R A C T

Autism is often a complex developmental disorder. The aim of the present study was to

describe the developmental characteristics of 129 1–4-year-old children (102 boys, 27

girls) referred for clinical assessment (mean age 2.9 years) due to suspicion of autism

spectrum disorder (ASD) after community screening at Child Health Care centers. All

children were clinically assessed at the Child Neuropsychiatry Clinic (CNC) in Gothenburg

by a research team (neurodevelopmental examination, structured interviews and general

cognitive and language examinations). Of the 129 children, 100 met diagnostic criteria for

ASD (69 with autistic disorder, and 31 with atypical autism/pervasive developmental

disorder-not otherwise specified). The remaining 29 children had a variety of

developmental disorders, most often attention-deficit/hyperactivity disorder (ADHD),

language disorder, borderline intellectual functioning, and intellectual developmental

disorder (IDD) with (n = 25) or without (n = 4) autistic traits (AT). IDD was found in 36% of

the 100 children with ASD, and in 4% of the 25 children with AT. Of the children with ASD,

56% had language disorder with no or just a few words at the initial assessment at the CNC,

many of whom in combination with IDD. Hyperactivity was found in 37% of those with

ASD and in 40% of those with AT. Epilepsy was found in 6% of the total group and in 7% of

those with a diagnosis of ASD. Of the latter group 11% had a history of regression, while

none of the AT cases had a similar background. When results were compared with a non-

screened preschool ASD group of 208 children, referred for ASD intervention at a mean age

of 3.4 years, very similar developmental profiles were seen. In conclusion, early

community ASD screening appears to systematically identify those children who are in

need of intervention and follow-up.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Autism spectrum disorder (ASD), with its core symptoms in the areas of social interaction, communication and limitedinterests in activities or play, is also still referred to as pervasive developmental disorder (PDD). ASD comprises a widevariety of clinical phenotypes, presenting during the child’s first years of life. As is the case with other developmentaldisorders, ASD occurs with many other conditions (Nazeer & Ghaziuddin, 2012) and there is usually either general cognitive

* Corresponding author at: Gillberg Neuropsychiatry Centre, Sahlgrenska Academy, University of Gothenburg, Kungsgatan 12, 411 19 Gothenburg,

Sweden. Tel.: +46 31 342 5970.

E-mail address: anne-katrin.kantzer@vgregion.se (A.-K. Kantzer).

0891-4222/$ – see front matter � 2013 Elsevier Ltd. All rights reserved.

http://dx.doi.org/10.1016/j.ridd.2013.06.016

A.-K. Kantzer et al. / Research in Developmental Disabilities 34 (2013) 2900–2908 2901

dysfunction, language delay, hyper- or hypoactivity, attention problems, or deviant motor development. Thesedevelopmental problems are now frequently referred to as ESSENCE (Early Symptomatic Syndromes ElicitingNeurodevelopmental Clinical Examinations) (Gillberg, 2010). In order to understand the heterogeneity of ASD in general,and the needs of individual children in particular, it is important to increase knowledge about these children’s earlydevelopmental profiles and to follow their development and trajectory over time.

Many recent studies underline the importance of early identification and of tailoring individual interventionprograms to improve outcomes in ASD (Dawson & Gernsbacher, 2010; Eikeseth, 2009; Magiati, Charman, & Howlin,2007). The American Academy of Pediatrics (AAP) recommended in 2001 and 2006 that developmentalsurveillance should be carried out at every visit at Child Health Care (CHC) centers and that any concerns raisedduring surveillance should be addressed promptly with standardized developmental screening tests (Johnson & Myers,2007). In addition, the AAP recommends that screening tests should be administered regularly at the 9-, 18-, and 24- or30-month visits.

In Sweden, children are followed at CHC centers during their first five years of life. The routine follow-up includesdevelopmental surveillance and screening at certain key-ages in order to see if the children meet developmental milestones.About 95–99% of the children in the Gothenburg area, Sweden, participate in the surveillance program (Arvidsson, Holmberg,Reuter, & Strombom, 2010). One of these general national screenings mainly focuses on language of 2.5-year-old children. In2008, a new screening program focusing on both language/communication, and on ASD, was introduced at the CHC centersin the Gothenburg area (Nygren, Sandberg, et al., 2012). The screening is part of a collaborative project between the CHCcenters, the Child Neuropsychiatry Clinic (CNC) and the Child Habilitation units in Gothenburg (Nygren, Cederlund, et al.,2012). The overall aim is to identify young children with suspected ASD, and to implement interventions at an early stage.The screening consists of (1) language screening with (a) direct observation of the child, performed by the nurse using toyobjects and (b) a parent questionnaire about early language milestones and, (2) an autism screening consisting of (c) a five-item joint attention observation (JA-OBS) and (d) the Modified Checklist for Autism in Toddlers (M-CHAT) (Baron-Cohen,Allen, & Gillberg, 1992; Robins & Dumont-Mathieu, 2006). The M-CHAT comprises a 23 items yes/no parent report withfollow-up interview. If ASD is suspected after the screening, children are referred to the CNC for further assessment. Childrenwho fail the language, but not the autism screening, are referred to a speech and language pathologist and for hearingassessment.

In this study, we present the developmental profiles of those children who were referred to and assessed at CNC becauseof suspected ASD following screening at the Gothenburg CHC centers between 2009 and 2011. The cohort will be followedinto school age and future studies will document the further development of the children.

2. Aims

The aim of the study was to describe the clinical profiles of a group of preschool children screening positive for ASD in acommunity surveillance program with respect to a broad panorama of developmental disorders and problems. The childrenin the cohort were then compared with a non-screened Swedish preschool cohort with ASD, with a view to establishing anydifferences in clinical profile or age at diagnosis across community-screened or non-screened ASD cases.

3. Materials and methods

3.1. Screening and subjects

The screening procedure at the age of 2.5 years at the CHC center consisted of (1) brief language screening (Miniscalco,Nygren, Hagberg, Kadesjo, & Gillberg, 2006), (2) the M-CHAT (Baron-Cohen et al., 1992, modified version Robins & Dumont-Mathieu, 2006), and (3) a joint attention observation (JA-OBS) performed by the CHC nurse, who had received targeted ‘‘ASDtraining’’ in order to increase ASD awareness (Nygren, Sandberg, et al., 2012). A nurse and/or a pediatrician evaluated theresults. All nurses at the CHC centers had been personally instructed to refer all screen positive cases for suchneuropsychiatric assessment to one and the same clinic (i.e. the CNC), for further in-depth assessments.

During 2009–2011, a total of 134 children under age four years with suspected ASD following screening at CHC centers inGothenburg were referred to the CNC for such assessments. Parents of 129/134 provided written informed consent to havetheir child participate in the assessment program at the CNC. Three children did not turn up for their booked assessment, andtwo moved abroad before assessment could take place.

The 129 children were born between 2005 and 2008 and were between 19 and 49 months of age (mean 35.3 months) atthe start of the assessment at the CNC. The boy:girl ratio was 3.8:1 (102 boys and 27 girls). More than three in five (62% – 80/129) of the study group had at least one parent coming from a country outside Sweden. A total of 28 families needed aninterpreter during the assessment. In nine families, the assessment was performed in English.

3.2. In-depth diagnostic assessments at the CNC

A team of professionals with several-many years of experience of working with children with ASD and otherneurodevelopmental disorders performed the assessment. The team consisted of at least one physician (pediatrician or child

A.-K. Kantzer et al. / Research in Developmental Disabilities 34 (2013) 2900–29082902

and adolescent psychiatrist), one child neuropsychologist, one special education teacher, and one speech and languagepathologist. After completion of all the assessments detailed under Sections 3.2.1–3.2.5 below, the team made conjointclinical diagnoses (Andersson, Gillberg, & Miniscalco, 2013). The diagnostic criteria of the DSM-IV-TR (American PsychiatricAssociation, 2000), including those for PDD/ASD, ADHD and Developmental Coordination Disorder (DCD), were used at alltimes. Autistic traits (AT) were diagnosed in cases meeting two or more obvious autism symptoms without qualifying for anASD/PDDNOS diagnosis. Clinical diagnoses were assigned in children who were socially or developmentally disabled by theirproblems, and who had C-GAS scores below 60 (see Section 3.2.6).

3.2.1. Specific ASD assessment

Two experienced professionals (special education teacher(s) or psychologist) performed an Autism DiagnosticObservation Schedule – Generic (ADOS-G) (Lord et al., 2000) for all children. In 72% of the cases (93/129), a parent wasinterviewed using the Diagnostic Interview for Social and Communication Disorders (DISCO-11, Wing, Leekam, Libby, Gould,& Larcombe, 2002). An experienced physician or psychologist performed the interview. The physician also checked a DSM-IVcriteria checklist for autistic disorder.

3.2.2. Cognitive/developmental assessment

An experienced child neuropsychologist examined the children with a view to establishing the general cognitive anddevelopmental level. The Griffiths’ test (Alin-Akerman & Nordberg, 1980) was used in 112 cases, the Wechsler Preschool andPrimary Scale of Intelligence (WPPSI-III) (Wechsler, 2004) in eight cases, and the Merrill-Palmer test (Roid & Sampers, 2005)in five cases (all three tests: mean DQ = 100, SD = 15). Three children had already been tested at the CHC center at the time ofreferral and they were therefore not retested at the CNC. One child, aged 26 months, did not cooperate at all in the testprocedure. His cognitive level was clinically rated as ‘‘general developmental delay’’ (based on observation of child andinterview with parent). Intellectual Developmental Disorder (IDD) was diagnosed at a DQ below 70. Borderline IntellectualFunctioning (BIF) was diagnosed at a DQ at 70 or higher, but below 85.

3.2.3. Speech and language assessment

An experienced speech and language pathologist examined all children. Different language developmental tests wereperformed, including measures of the child’s expressive speech and language developmental level. In the present study,language and communicative quality was rated in the following way: a) no words at all; b) a few single words; c) a fewcommunicative sentences; d) talks a great deal, mostly echolalia, or e) talks a great deal, mostly in a communicativefashion.

3.2.4. Neurodevelopmental assessment

An experienced physician (pediatric specialist or child and adolescent psychiatrist) interviewed the parent(s) andexamined the child. The physician met the family one, two or three times, depending on the family history and the child’sability to cooperate. Specifically, motor function, toe walking, activity level (parents report and observation by thephysician), co-existing epilepsy (parents report, patient history from records and/or EEG), and a history of regression(parents report, information from the CHC records) were analyzed.

3.2.5. Adaptive behavior assessment

The psychologist used the Vineland Adaptive Behavior Scales (VABS) (Sparrow, Cicchetti, & Balla, 2005), including fourdomains (Communication, Daily living skills, Socialization, Motor skills and a sum of domain scores), in face-to-faceinterview with a parent (mean = 100, SD = 15).

3.2.6. Children’s Global Assessment Scale (C-GAS)

The Children’s Global Assessment Scale (C-GAS) (Schorre & Vandvik, 2004; Shaffer et al., 1983) was used for rating thechild’s general functioning in daily life. It combines communication, socialization and practical skills in a more general wayby evaluating the child’s compatibility to use them in daily life situations. Scores range from 0 to 100, 100 corresponding tothe highest function. The scale was originally developed for use with 4–20-year-old children and adolescents, and has beenadapted for younger children. The C-GAS was scored at a consensus case conference with the whole assessment team, whereall available information was shared and discussed.

3.3. Statistics

This is mainly a descriptive study. Most of the results are presented as confidence intervals, and percentages. Wheneverappropriate, group frequencies have been compared using chi square tests.

3.4. Ethics

The study was approved by the Human Ethics Committee at the Medical Faculty, University of Gothenburg, Sweden.Informed consent was obtained from at least one of the parents/responsible carers for each patient.

A.-K. Kantzer et al. / Research in Developmental Disabilities 34 (2013) 2900–2908 2903

4. Results

4.1. ASD diagnoses

Of the 129 children, 100 were diagnosed with ASD after assessment at the CNC. Of these, 69 were diagnosed with autisticdisorder (AD), 31 were diagnosed with atypical autism/pervasive developmental disorder not otherwise specified (PDDNOS).

A total of 29 children did not meet criteria for a full ASD diagnosis, but 25 of these had some marked autistic traits (AT).Nineteen children in the AT group were given at least one non-ASD DSM-IV diagnosis, viz. language disorder, IDD, BIF, ADHD,DCD or tic disorder. One of the 25 children almost met full Gillberg & Gillberg criteria for Asperger syndrome (Gillberg &Gillberg, 1989), but not those of the DSM-IV-TR because of abnormal development in the first three years of life.

In another four cases, the specialist team at the CNC did not identify autistic symptoms during the assessment, althoughASD had been suspected at the CHC. They were therefore not included in the present study. All four received non-ASD DSM-IV diagnoses, viz. language disorder, BIF, or ADHD, or a combination of these three.

The results below therefore refer to the 100 children with ASD (both AD and PDDNOS), and the 25 children with AT.

4.2. General cognitive/developmental level

Thirty-seven of the 100 children with an ASD diagnosis and 19 of the 25 (76%) of those with AT had a cognitive level in theaverage range (p < .01). Twenty-seven of the ASD subgroup had BIF, and 36 had IDD. Among those with AT, 5 (20%) had BIFand 1 (4%) had IDD.

Of the children with AD, 20% performed in the average range, compared to 74% of those with PDDNOS (p < .01); 33% and46% of the children with AD, and 13% of those with PDDNOS had BIF and IDD, respectively (p < .01).

4.2.1. Gender by subgroups

The boy:girl ratio in the ASD group was 3.3:1 (AD group 5.2:1, PDDNOS 1.6:1, AT 8.3:1).The boy:girl ratio was 2.4:1 in the group with average cognitive level, 5.8:1 in the BIF group, and 3.5:1 in the IDD group.

4.3. Speech and language

4.3.1. Speech and language development

In the AD group, 81% of the children under the age of 36 months had ‘‘no words or a few single words’’. No child in thisgroup used ‘‘mostly communicative sentences’’. Corresponding rates in the PDDNOS and AT subgroups were 50%/31% and54%/15%. Among the AD children older than 36 months, 34% had ‘‘no words or a few single words’’, and 27% used ‘‘mostlycommunicative sentences’’. Corresponding rates in the PDDNOS and AT subgroups were 26%/53% and 8%/58% (Tables 1a and1b).

Of the 125 children, 78 (63%) had one or both parents speaking a native language other than Swedish. There were nosignificant differences in terms of the lowest language level as regards children of immigrant parents and those born to oneor two Swedish parents. The percentage of children with migrant parents was similar in the different diagnostic andcognitive level subgroups.

4.3.2. Regression of language skills

Language regression had occurred in 11 children, of whom 10 had AD (14% of that subgroup) and one PDDNOS (3% of thatsubgroup). Only two of the children with AD and language regression had shown normal early development of languageskills.

In total, only one child in the whole group (with AD) of 125 showed ‘‘classic autistic regression’’ with loss of both languageand social skills.

Table 1a

Language development in ASD subgroups by ages.

<30 months 31–36 months 37–42 months >42 months

No words/a few single words

AD 17/19 (90%) 18/24 (75%) 7/18 (39%) 2/8 (25%)

PDDNOS 3/4 (75%) 5/12 (42%) 3/10 (30%) 1/5 (20%)

AT 4/6 (66%) 3/7 (43%) 1/10 (10%) 0/2

Talks a great deal, mostly communicative sentences

AD 0/19 0/24 5/18 (28%) 2/8 (25%)

PDDNOS 1/4 (25%) 4/12 (33%) 4/10 (40%) 4/5 (80%)

AT 0/6 2/7 (28%) 5/10 (50%) 2/2 (100%)

Autism spectrum disorder (ASD), autistic syndrome (AD), pervasive developmental disorder-not otherwise specified (PDDNOS), autistic symptoms but

below cut off for ASD (AT).

Table 1b

Language development in cognitive subgroups by ages.

<30 months 31–36 months 37–42 months >42 months

No words/a few single words

IDD 11/11 (100%) 11/13 (84%) 5/8 (62%) 2/5 (40%)

BIF 6/7 (86%) 8/12 (66%) 4/10 (40%) 1/3 (33%)

Average 7/10 (70%) 6/17 (35%) 3/22 (14%) 0

Talks a great deal, mostly communicative sentences

IDD 0/11 0/13 0/8 1/5 (20%)

BIF 0/7 0/12 3/10 (30%) 1/3 (33%)

Average 0/10 6/17 (35%) 11/22 (50%) 6/7 (85%)

DQ � (average), DQ 70- < 85 (BIF), DQ < 70 (IDD).

A.-K. Kantzer et al. / Research in Developmental Disabilities 34 (2013) 2900–29082904

4.4. Motor function

4.4.1. Age of unsupported walk

The mean age for unsupported walking was 14.3 months (SD 3.3 months) in the ASD group. Children with AD/PDDNOSand cognitive functioning within the normal range, with BIF, and with IDD started to walk at 13.8/14.0 months, 14.6/14.8months and 14.2/16.0 months, respectively. One child with AD and IDD had not started unsupported walk at the time of theassessment at 41 months.

In the group of 25 children with AT, the mean age for unsupported walking was 13.1 months (SD 2.4 months).

4.4.2. Toe walking

Twelve of the 100 children in the ASD group had clear and persistent toe walking in all situations. Another 19 had toewalking in some but not in all situations, and a further two had shown toe walking temporarily for a short period of time. Toewalking occurred in 36% of the AD and 25% of the PDDNOS group (n.s.). In the group of children with AT, 4/25 (16%) childrenwere toe-walkers or had shown toe walking in the past.

4.5. Activity level

The activity level was ‘‘normal’’ in 61/100 children of the children with ASD, while 37 children showed hyperactivityduring observation or were reported to have hyperactive periods by the parents. Two children with ASD were hypoactive.The proportions of hyper- and hypoactivity in the AT-group were 10/1 (out of 25).

4.6. Co-existing epilepsy

Eight of the 125 children (6.4%) had co-existing epilepsy that was diagnosed before or during the assessment at the CNC.Of these children, seven had ASD (three AD and four PDDNOS). Three of these seven children had BIF, two had IDD, and twohad cognitive function in the average range. One child in the AT group with language disorder and average intellectual levelhad epilepsy.

4.7. Adaptive behavior

The complete VABS interview was performed in 93 of 100 children with ASD and in 21 of 25 children with AT. Mean totalscores and subdomain scores (Communication, Daily living Skills, Socialization and Motor skills) of the VABS differed in theASD diagnostic subgroups and according to cognitive levels (Tables 2a and 2b).

4.8. C-GAS

The mean C-GAS score in the ASD group was 40.4 (SD 9.4) compared to 55.3 (SD 10.8) in the group with AT.None of the children in the ASD group had a C-GAS score of 60 or above, compared to 32% in the AT group (p < .01)

Table 2a

Mean VABS scores in the ASD group by cognitive levels.

Vineland domain standard score Average (n = 34), mean (SD) BIF (n = 26), mean (SD) IDD (n = 33), mean (SD)

Sum of domain 82.7 (8) 75.3 (5.4) 69.2 (8.5)

Communication 79.5 (12.2) 73.6 (5.4) 65.5 (13.9)

Daily living skills 90.8 (12) 81.8 (7) 75.8 (10.2)

Socialization 78.6 (11.4) 75.8 (7.3) 68.8 (10)

Motor skills 91 (9.5) 82.4 (7.4) 65.3 (7.3)

DQ � (average), DQ 70- < 85 (BIF), DQ < 70 (IDD).

Table 2b

Mean VABS scores in the AT group by cognitive levels.

Vineland domain standard score Average (n = 16), mean (SD) BIF (n = 4), mean IDD (n = 1), mean

Sum of domain 85.9 (10.9) 78 74

Communication 86.3 (14) 72.3 59

Daily living skills 91.6 (10.6) 86.5 77

Socialization 84.5 (10) 77.5 63

Motor skills 90.1 (11) 85.5 96

DQ � (average), DQ 70- < 85 (BIF), DQ < 70 (IDD).

A.-K. Kantzer et al. / Research in Developmental Disabilities 34 (2013) 2900–2908 2905

(Table 3). Conversely, none of the AT-group had a C-GAS score under 30, whereas this level of functioning applied in 10% ofthe ASD-group.

5. Discussion

All the 129 children neuropsychiatrically assessed in the present study had screened positive for ASD at CHC centers. Nochild was excluded because of the parent(s) speaking a native language other than Swedish. The cohort can be regarded asalmost population-representative of children in the age cohort raising suspicion of suffering from ASD, and those diagnosedwith ASD as representative of all young children with ASD in the community. The total prevalence of ASD for all two-year oldchildren in the city of Gothenburg was 0.8 in 2010, and had significally increased since before the screening method wasintroduced (Nygren, Cederlund, et al., 2012). However, it is possible that a small number of the most multiple handicappedindividuals with autism and very severe comorbidities (including difficult-to-treat epilepsy, major neuromuscular disordersand profound IDD) could have been excluded, since it is known that such cases are often catered for in habilitation servicesand individuals do not always attend the regular surveillance programs at CHC centers once their neurological/neurodevelopmental diagnosis has been made.

Before the introduction of the screening method, all nurses and physicians at the CHC centers were educated at repeatedoccasions to increase awareness of autistic symptoms. Some of the study children had very clear symptoms leading to anearlier screening and referral (the youngest child in the study was 19 months of age at the time of investigation). Somechildren were older than 2.5 years of age when the new screening instrument was introduced, but the CHC team decided toscreen them when noticing ‘‘autism-suspect’’ symptoms (the oldest child in the study was 49 months of age at the time ofinvestigation at CNC). Between referral from CHC and start of investigation at CNC, there was a maximal delay of threemonths due to general Swedish policy of health care guarantee, but in very few cases the parents wanted to postpone theassessment due to vacation abroad or sickness. Children who received an ASD diagnosis were about one month younger thanchildren who did not meet the DSM-IV criteria, and children with AD about two months younger than children in thePDDNOS group. This (small) difference accords with previously published reports that children with very clear autistic coresymptoms and/or language delay may elicit parental concerns at an earlier age (Chung et al., 2011; Horovitz, Matson, & Sipes,2011).

All 129 children had been assessed in accordance with the same comprehensive protocol at the CNC. In the 100 childrenwho received an ASD diagnosis, 69 children met criteria for autistic disorder and 31 for PDDNOS. Asperger’s syndrome wasnot diagnosed in any of these young children, but it is expected that some of them will meet the clinical phenotype ofAsperger syndrome during follow-up assessments. Twenty-five children had clear autism symptoms (AT) but below cut-offfor ASD according to DSM-IV TR (American Psychiatric Association, 2000). The specialist team, however, could not documentany autism symptoms at all in 4 of the 129 patients. All 129 children are targeted for comprehensive follow-up assessmentstwo years after the comprehensive assessment reported here. Future studies will investigate their further development intoschool age.

Table 3

C-GAS scores in the diagnostic subgroups.

Diagnosis Cognitive level C-GAS

�70 60–69 50–59 40–49 30–39 20–29 <20

ASD Average (n = 37) 0 0 11 20 6 0 0

BIF (n = 27) 0 0 2 15 10 0 0

IDD (n = 36) 0 0 3 10 13 8 2

ASD total group (n = 100) 0 0 16 45 29 8 2

AT Average (n = 19) 3 3 9 4 0 0 0

BIF (n = 5) 0 2 1 1 1 0 0

IDD (n = 1) 0 0 0 1 0 0 0

Total AT group (n = 25) 3 5 10 6 1 0 0

Autism spectrum disorder (ASD), autistic symptoms but below cut off for ASD (AT).

DQ � (average), DQ 70- < 85 (BIF), DQ < 70 (IDD).

A.-K. Kantzer et al. / Research in Developmental Disabilities 34 (2013) 2900–29082906

There were significantly more children with BIF and IDD in the ASD group than in the group with AT. There are indicationsof both stability and change of IQ scores in young children with autism (Dietz, Swinkels, Buitelaar, van Daalen, & vanEngeland, 2007). All young children need a follow-up if they have been assessed with regard to cognitive level at preschoolage. The children in our cohort will be retested when they have their follow-up assessment two years after the first one.

Most clinical studies of children with ASD report a boy:girl ratio of 3–4:1. In this study, we found a boy:girl ratio of 3.8:1.Girls tend to be diagnosed at later ages than boys, especially when they have no general cognitive impairment (Giarelli et al.,2010), which usually results in a very high boy:girl ratio in this young age group. The boy:girl ratio in our ‘‘average IQ’’ ASDsubgroup was 2.4:1. This result suggests that girls with ASD and average cognitive function abilities may benefit from ageneral screening, which probably facilitates their identification.

One of the core symptoms in severe ASD is a delayed or deviant language development. Language and speech problemsbefore the age of three years appear to be a strong risk factor for neuropsychiatric diagnosis at the age of seven years(Miniscalco et al., 2006). The language development was therefore in focus both in the screening at the CHC centers and inthe further assessment at the CNC. Children with ASD performed at lower speech and language levels during the assessmentat CNC than children with AT, and those with AD performed at lower levels than those with PDDNOS. However, children withPDDNOS and AT had similar levels. We expect some children to move from the AT group to the PDDNOS group and vice versaduring follow-up. Our results also suggest that the level of speech and language performance is related to autism spectrumproblem and cognitive level, and that delayed language development is a risk marker important to screen for in youngchildren.

As more than 60% of the children in our study had at least one parent coming from a country outside of Sweden, thecondition of bi- or multilingual upbringing had to be considered when evaluating language development. The correspondingrate of children coming from countries outside Scandinavia registered at all CHC centers in the Gothenburg area in 2010 was40% (Arvidsson et al., 2010). The same phenomenon of more children with ASD having migrant parents was observed in apopulation study of 208 registered cases with ASD performed in Stockholm a few years ago (Fernell et al., 2010). Recentstudies suggest that bilingual exposure is not correlated to language delay in children with ASD (Hambly & Fombonne, 2012;Petersen, Marinova-Todd, & Mirenda, 2012). In accordance with these studies, we found no significant differences as regardslanguage levels in our cohort between the children whose two parents were Swedish native speakers and those who had atleast one parent speaking another mother tongue.

Regression was an uncommon phenomenon in this young population-representative group of individuals with ASD. Thissuggests that clinical studies reporting a high rate of regression in autism are possibly not representative of the situation inthe general population. Only one child in the whole sample of 100 ASD cases showed ‘‘classic autistic regression’’ with loss ofboth language and social skills (Barger, Campbell, & McDonough, 2013).

Age at unsupported walking did not differ significantly between the AD, PDDNOS and the AT group. There was a tendencythat those with BIF and IDD had a later start of unsupported walking. Toe-walking, from persistent to transient, was presentin one third of the children with ASD.

Almost 40% of the children had either hyperactivity (37%) or hypoactivity (2%) and the rates were similar in the ASD andthe AT group. ADHD is an important and common comorbidity in ASD and needs specific considerations in intervention andtreatment (Handen, Taylor, & Tumuluru, 2011).

Of the 125 children with ASD or AT, 8 (6.4%) had co-existing epilepsy. Seven of these had ASD and one had AT. Cognitive/developmental levels, IDD, BIF and average, were equally distributed in these children. The rate of epilepsy in this youngcohort of children can be expected to rise considerably during childhood and adolescence, as the lifetime prevalence rate ofepilepsy in ASD is much higher than found in the young children reported here (Kohane et al., 2012).

It has been shown that children with ASD show characteristic VABS profiles with lower results on Communication/Socialization results, while they score higher on Daily Living and Motor Skills domains (Perry, Flanagan, Dunn Geier, &Freeman, 2009). The ASD children in our study also showed this profile when they had average intelligence or BIF. Schatz andHamdan-Allen have shown that both children with autism and children with IDD in general have low results in theSocialization Domain. With increasing IQ, children with autism show smaller increases of social functioning than childrenwith mental retardation/IDD (Schatz & Hamdan-Allen, 1995). Children with ASD and an IQ below 70 are more likely to haveimpairment of motor skills (Green et al., 2009). In this group, a more general impairment of development may occur. TheASD/IDD children in our sample scored as low on Motor Skills domains as on Communication and Socialization, but hadrelatively higher results on Daily Living Skills. The AT group with average intellectual functioning showed a similar VABS-profile as the ASD children, with lower results on Communication/Socialization Skills and higher results on Daily Living/Motor Skills. The difference between these levels was however not as big as in the ASD group. The group of children with ATand BIF was very small and can therefore not be compared to the ASD/BIF group. The only child with IDD in the group with AThad an unusually high VABS due to a Motor Skills standard score of 96. This child had quite normal motor development withan age of unsupported walk at 13 months.

Regarding their daily life functioning reflected by the C-GAS level, children in the ASD group scored lower than children inthe group with AT. Children in the AD group scored lower than the PDDNOS group. The C-GAS level increased with increasingcognitive level in the different groups. Thus, the C-GAS gives a realistic image of the child’s functioning in daily life situations.

In 2010, developmental data from a non-screened cohort of 208 preschool children with clinically diagnosed ASD inStockholm were reported (Fernell et al., 2010). These children came to the autism specialist center in Stockholm when theyhad already been given their ASD diagnoses. Of these 208 children, 191 were found to meet full criteria for ASD when

A.-K. Kantzer et al. / Research in Developmental Disabilities 34 (2013) 2900–2908 2907

assessed at the specialist center, while 17 had AT below cut-off for ASD. That cohort, even though believed to berepresentative of the majority of clinically diagnosed cases of ASD, was not population-based. The Stockholm children were,on average, 5.5 months older at the time of referral than our cohort. The boy: girl ratio in the Stockholm ASD group washigher (5.5:1) than in the corresponding Gothenburg group (3.3:1). Fernell suggested that the high ratio in the Stockholmstudy in some part could be accounted for the exclusion from in-depth study of 25 children with more general impairmentwho had a boy-girl ratio of only 1.7:1. The rate of children with at least one parent with immigrant status was lower in theStockholm cohort (42%). The study also excluded 15 children because of their parents not speaking Swedish or English.Interestingly, in spite of these differences in respect of some demographic factors the children in the Stockholm cohortshowed similar profiles as the children we present in this study. Developmental profiles with regard to distribution ofgeneral cognitive levels, expressive language, adaptive functions, motor functions, activity regulation problems and epilepsywere very similar in the two cohorts. The rate of regression was lower in the screened cohort, which in part may be due to thelower age at assessment in the Gothenburg cohort, some children in that group potentially developing regression later.

6. Conclusions and clinical implications

This study confirms the very heterogeneous clinical picture seen in children presenting with symptoms in the autismspectrum. It was striking that developmental profiles in different domains were very similar across this group of childrenwho had been community-screened for ASD at CHC centers and the group of non-screened children, diagnosed in preschool ageon clinical grounds. Thus, early community ASD screening appears to correctly identify those children with ASD who are inneed of intervention and follow-up. It is possible that almost half a year might be ‘‘gained’’ as regards early ASDidentification. Broad assessments are necessary to provide the foundation for early appropriate interventions.

The cohorts referred to here will be followed prospectively into school age. Further studies will focus on the developmentof the children in the different clinical subgroups and the diagnostic stability.

Acknowledgments

We are grateful to the children, parents and the colleagues at the CNC, for their help in making this study possible. Thestudy has been supported by a Swedish Council Grant for Christopher Gillberg (grant no. K2013-62X-11251-19-5). We aregrateful for the support of Dr. Bjorn Kadesjo who initiated the clinical changes that made this study possible and who securedpart-funding for the research from the AnnMari and Per Ahlqvist foundation.

References

Alin-Akerman, B., & Nordberg, L. (1980). Griffiths’ Developmental Scales I and II. Stockholm: Psykologiforlaget AB.American Academy of Pediatrics. (2001). Developmental surveillance and screening of infants and young children. Pediatrics, 108, 192–195.American Academy of Pediatrics. (2006). Identifying infants and young children with developmental disorders in the medical home: An algorithm for

developmental surveillance and screening. Pediatrics, 118, 405–420.American Psychiatric Association. (2000). Diagnostic and Statistical Manual of Mental Disorders, (4th edition 1994, text revision), DSM IV-TR. Washington, DC:

American Psychiatric Association.Andersson, G. W., Gillberg, C., & Miniscalco, C. (2013). Pre-school children with suspected autism spectrum disorders: Do girls and boys have the same profiles?

Research in Developmental Disabilities, 34, 413–422.Arvidsson, T., Holmberg, L., Reuter, A., & Strombom, A. (2010). Barnhalsovardsrapport verksamhetsaret 2010.Barger, B. D., Campbell, J. M., & McDonough, J. D. (2013). Prevalence and onset of regression within autism spectrum disorders: A meta-analytic review. Journal of

Autism and Developmental Disorders, 43, 817–828.Baron-Cohen, S., Allen, J., & Gillberg, C. (1992). Can autism be detected at 18 months? The needle, the Haystack, and the CHAT. British Journal of Psychiatry, 161,

839–843.Chung, C. Y., Liu, W. Y., Chang, C. J., Chen, C. L., Tang, S. F., & Wong, A. M. (2011). The relationship between parental concerns and final diagnosis in children with

developmental delay. Journal of Child Neurology, 26, 413–419.Dawson, M., & Gernsbacher, M. A. (2010). Effectiveness of intensive autism programmes. Lancet, 375, 722–723.Dietz, C., Swinkels, S. H., Buitelaar, J. K., van Daalen, E., & van Engeland, H. (2007). Stability and change of IQ scores in preschool children diagnosed with autistic

spectrum disorder. European Child & Adolescent Psychiatry, 16, 405–410.Eikeseth, S. (2009). Outcome of comprehensive psycho-educational interventions for young children with autism. Research in Developmental Disabilities, 30, 158–

178.Fernell, E., Hedvall, A. , Norrelgen, F., Eriksson, M., Hoglund-Carlsson, L., Barnevik-Olsson, M., et al. (2010). Developmental profiles in preschool children with

autism spectrum disorders referred for intervention. Research in Developmental Disabilities, 31, 790–799.Giarelli, E., Wiggins, L. D., Rice, C. E., Levy, S. E., Kirby, R. S., Pinto Martin, J., et al. (2010). Sex differences in the evaluation and diagnosis of autism spectrum

disorders among children. Disability and Health Journal, 3, 107–116.Gillberg, C. (2010). The ESSENCE in child psychiatry: Early symptomatic syndromes eliciting neurodevelopmental clinical examinations. Research in Develop-

mental Disabilities, 31, 1543–1551.Gillberg, I. C., & Gillberg, C. (1989). Asperger Syndrome – some epidemiological considerations: A research note. Journal of Child Psychology and Psychiatry, and

Allied Disciplines, 30, 631–638.Green, D., Charman, T., Pickles, A., Chandler, S., Loucas, T., Simonoff, E., et al. (2009). Impairment in movement skills of children with autistic disorders.

Developmental Medicine & Child Neurology, 51, 311–316.Hambly, C., & Fombonne, E. (2012). The impact of bilingual environments on language development in children with autism spectrum disorders. Journal of Autism

and Developmental Disorders, 42, 1342–1352.Handen, B. L., Taylor, J., & Tumuluru, R. (2011). Psychopharmacological treatment of ADHD symptoms in children with autism spectrum disorder. International

Journal of Adolescent Medicine and Health, 23, 167–173.Horovitz, M., Matson, J. L., & Sipes, M. (2011). The relationship between parents’ first concerns and symptoms of autism spectrum disorders. Developmental

Neurorehabilitation, 14, 371–377.

A.-K. Kantzer et al. / Research in Developmental Disabilities 34 (2013) 2900–29082908

Johnson, C. P., & Myers, S. M. (2007). Identification and evaluation of children with autism spectrum disorders. Pediatrics, 120, 1183–1215.Kohane, I. S., McMurry, A., Weber, G., MacFadden, D., Rappaport, L., Kunkel, L., et al. (2012). The co-morbidity burden of children and young adults with autism

spectrum disorders. PLoS One, 7, e33224.Lord, C., Risi, S., Lambrecht, L., Cook, E. H., Jr., Leventhal, B. L., DiLavore, P. C., et al. (2000). The autism diagnostic observation schedule-generic: A standard measure

of social and communication deficits associated with the spectrum of autism. Journal of Autism and Developmental Disorders, 30, 205–223.Magiati, I., Charman, T., & Howlin, P. (2007). A two-year prospective follow-up study of community-based early intensive behavioral intervention and specialist

nursery provision for children with autism spectrum disorders. Journal of Child Psychology and Psychiatry, 48, 803–812.Miniscalco, C., Nygren, G., Hagberg, B., Kadesjo, B., & Gillberg, C. (2006). Neuropsychiatric and neurodevelopmental outcome of children at age 6 and 7 who

screened positive for language problems at 30 months. Developmental Medicine & Child Neurology, 48, 361–366.Nazeer, A., & Ghaziuddin, M. (2012). Autism spectrum disorders: Clinical features and diagnosis. Pediatric Clinics of North America, 59, 19–25.Nygren, G., Cederlund, M., Sandberg, E., Gillstedt, F., Arvidsson, T., & Gillberg, C. (2012). The prevalence of autism spectrum disorders in toddlers: A population

study of 2-year old Swedish children. Journal of Autism and Developmental Disorders, 42, 1491–1497.Nygren, G., Sandberg, E., Gillstedt, F., Ekeroth, G., Arvidsson, T., & Gillberg, C. (2012). A new screening programme for autism in a general population of Swedish

toddlers. Research in Developmental Disabilities, 33, 1200–1210.Perry, A., Flanagan, H. E., Dunn Geier, J., & Freeman, N. L. (2009). Brief report: The Vineland Adaptive Behavior Scales in young children with autism spectrum

disorders at different cognitive levels. Journal of Autism and Developmental Disorders, 39, 1066–1978.Petersen, J. M., Marinova-Todd, S. H., & Mirenda, P. (2012). Brief report: An exploratory study of lexical skills in bilingual children with autism spectrum disorder.

Journal of Autism and Developmental Disorders, 42, 1499–1503.Robins, D. L., & Dumont-Mathieu, T. M. (2006). Early screening for autism spectrum disorders: Update on the modified checklist for autism in toddlers and other

measures. Journal of Developmental and Behavioral Pediatrics, 27, 111–119.Roid, G. H., & Sampers, J. L. (2005). Merrill-Palmer Revised Scales of Development. Swedish version. Hofgrefe Psykologiforlaget.Schatz, J., & Hamdan-Allen, G. (1995). Effects of age and IQ on adaptive behavior domains for children with autism. Journal of Autism and Developmental Disorders,

25, 51–60.Schorre, B. E., & Vandvik, I. H. (2004). Global assessment of psychosocial functioning in child and adolescent psychiatry. A review of three unidimensional scales

(CGAS, GAF, GAPD). European Journal of Child & Adolescent Psychiatry, 13, 273–286.Shaffer, D., Gould, M. S., Brasic, J., Ambrosini, P., Fisher, P., Bird, H., et al. (1983). A Children’s Global Assessment Scale (CGAS). Archive of General Psychiatry, 40,

1228–1231.Sparrow, S. S., Cichetti, D. V., & Balla, D. A. (2005). Vineland Adaptive Behavior Scales (2nd ed.). Circle Pines, MN: American Guidance Service.Wechsler, D. (2004). Wechsler Preschool and Primary Scale of Intelligence, revised, Swedish version. Stockholm: Psykologiforlaget.Wing, L., Leekam, S. R., Libby, S. J., Gould, J., & Larcombe, M. (2002). The diagnostic interview for social and communication disorders: Background, inter-rater

reliability and clinical use. Journal of Child Psychology and Psychiatry, 43, 307–325.