SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Aim  The aim of this study was to investigate the stability of motor problems in a clinically referred sample of children with, or at risk of, autism spectrum disorders (ASDs), attention-deficit–hyperactivity disorder (ADHD), and/or developmental coordination disorder (DCD).

Method  Participants were 49 children (39 males, 10 females; mean age 5y 6mo, SD 10mo) with various developmental problems, a Movement Assessment Battery for Children (M-ABC) score on or below the 15th centile, and an IQ of 70 or more. Sixteen children were at risk of developing ADHD, 15 children had a diagnosis of, or were at risk of developing ASD, and 18 children had neither diagnosis. Children were reassessed 2 to 3 years later.

Results  At follow-up (mean age 7y 11mo; SD 1y), the mean M-ABC score was significantly increased, and in 22 children was no longer below the 15th centile. A general linear model to compare the difference in M-ABC scores in the three groups of children demonstrated a significant difference between groups (p=0.013), with the age at the initial assessment as a significant covariate (p=0.052). The group of children with or at risk of ASD showed less improvement in motor performance.

Interpretation  Motor problems among preschool age children are not always stable, but appear to be so in most children with ASDs.

List of Abbreviations
ASD

 Autism spectrum disorder

DCD

 Developmental coordination disorder

M-ABC

 Movement Assessment Battery for Children

A number of children exhibit poor motor performance in the first 6 years of life. In some cases the motor problems can be explained by an apparent central nervous system disorder, and these problems are unlikely to resolve. However, some children without a clear neurological disorder present with poor motor abilities before 7 years of age. Some of these children could be considered to have developmental coordination disorder (DCD), according to DSM-IV-TR1 criteria. However, it appears from the literature that the large majority of children with DCD are identified after starting school.2 In some cases, poor motor performance at preschool age is one of the first clear signs of another (comorbid) developmental disorder with more prominent behavioural features, such as autism spectrum disorders3 (ASDs), or attention-deficit–hyperactivity disorder4 (ADHD). Although motor disorders can occur in isolation, many studies have described their frequent co-occurrence with ADHD5,6 and ASD.7,8 The purpose of this study was therefore, to determine the stability of poor motor performance among younger children.

The most commonly used screening instruments for young children evaluate basic motor skills, referenced against age norms and based on early normative models of motor development.9 But is motor development really a linear process with a small range of variability? Darrah et al.10 investigated the stability of outcome on serial motor assessment and found a large variability of scores within individual infants and between infants aged 0 to 18 months. They concluded that motor development in very young children is non-linear, rather than occurring at a constant rate. They suggested that screening of infants should include multiple domains and multiple time-points before referring a child to early intervention programmes. Whether non-linearity in motor development persists at later ages (i.e. between the ages of 2–8y) to our knowledge has only rarely been documented in the literature. However, from a review of that literature, Malina11 concluded that variation in motor development within individual children, between children, and from age to age is considerable during early childhood. Variation in performance between testing periods probably reflects: (1) normal variation in growth; (2) neuromuscular maturation; (3) opportunity for practice; (4) motivation to perform in the test situation; and (5) perhaps also in the adults administering the tests and in the cooperation of young children.11 The link between motor development and the remodelling of the cerebral cortex between 6 and 8 years of age remains to be established. However, the period between the age of 6 and 8 years is critical for the development of various cognitive functions, and coincides with different developmental trajectories for different cortical regions.12

The literature on stability or change in motor performance of children with DCD has mainly concentrated on the long-term follow-up of children identified at 6 years or later.13 To date, few studies have focused on the follow-up of motor problems below the age of 6 years. Pless et al.14 performed a population-based screening of about 3000 5- and 6-year-old children by means of a short motor screening test. From this screening, 100 children were referred for an extended physiotherapy assessment. Fifty-three children participated in the study, and 37 were identified as having definite (n=19) or borderline (n=18) motor problems, corresponding to a score on the Movement Assessment Battery for Children (M-ABC) test15 at or below the 5th centile and between the 5th and 15th centiles respectively. Two years later, 20 children still had a definite motor problem, eight children had borderline motor problems, and nine children no longer showed poor motor performance. At an individual level, most children with definite motor problems remained in this category at follow-up, whereas most children with borderline motor difficulties did not.

The aim of this study was to investigate if poor motor performance identified at 4 to 6 years of age in children with, or at risk of, ASD, ADHD, and/or DCD persists for at least 2 years.

Method

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Participants

The study population consisted of 106 children who were referred for various reasons (but mostly not because of a delay in motor development) to the Centre for Developmental Disabilities in Ghent, Belgium. Some of the participants were in a follow-up programme for preterm-born children. Inclusion and exclusion criteria were: (1) age between 4 and 6 years; (2) a score on the M-ABC test ≤15th centile; (3) a total IQ score at or below the 70; and (4) no medical diagnosis such as cerebral palsy (CP), epilepsy, or a genetic syndrome, as confirmed by a paediatric neurologist. Twenty-eight children could not be traced for follow-up. This was probably a random problem of parents moving to another address or changed phone numbers. Four children were excluded because a genetic disorder or CP was subsequently diagnosed, and 25 parents refused to participate, mostly because their child had already participated in multiple assessments since their initial assessment. These multiple assessments were probably due to a further problematic development. Forty-nine children participated in the study (39 males, 10 females). Nine children were born preterm, of whom eight were born at 32 to 37 weeks’ gestational age and one was born at less than 32 weeks.

Materials

The M-ABC15 consists of eight items: three measure manual dexterity, two assess ball skills, and three items measure balance. A score on or below the 5th centile is indicative of a definite motor problem. Scores between the 5th and 15th centiles suggest a degree of difficulty that is borderline.

IQ tests administered varied according to the clinical needs but included the Snijders-Oomen Non-verbal Intelligence Test – Revised16 (SON-R), the Dutch version of the Wechsler Intelligence Scale for Children – Revised17 (WISC-RN), the Dutch version of the McCarthy Developmental Scales 2½–8½18 (MOS), and the Dutch version of the Wechsler Preschool and Primary Scales of Intelligence – Revised19 (WPPSI-R NL).

Procedure

The study was approved by the ethics committee of the Arteveldehogeschool University College, Ghent, Belgium. The initial assessment was part of a clinical assessment. A physiotherapist assessed the children on the M-ABC. The children’s IQ was evaluated by a psychologist. A medical doctor was responsible for the exclusion of a medical diagnosis. Based on reason for referral, case history, information from the medical doctor, and clinical observations, the psychologist decided which behavioural assessments were carried out. A diagnosis of ASD or risk of ASD was based mostly on clinical observations, observation in school, the Autistic Diagnostic Observation Schedule – Generic20 (Ados-G), and the Autism Diagnostic Interview – Revised21 (ADI-R), and always in accordance with DSM-IV-TR1 criteria. A classification of ‘high risk of developing ADHD’ or ‘moderate risk of developing ADHD’ was based mostly on the clinical observations and the results on the Attention Scale of the Child Behaviour Checklist22 (CBCL). The rule of thumb followed for interpretation of test results was as follows: a delay or deviation of 2 standard deviations (SDs) from the norm was taken to indicate the presence of a disorder or a high risk of the disorder; a delay or deviation of 1 SD was considered to indicate risk of the disorder. Behavioural diagnoses were, in the main, made by the psychologist but always in accordance with the complete multidisciplinary team and DSM-VI-TR1 criteria.

At follow-up, after written consent from the parents, the M-ABC was administered to children, in a quiet room in their own school, by two trained testers not involved in the initial assessment. Parents were asked to complete a questionnaire regarding therapeutic interventions. The children were categorized as receiving therapy if they received at least 6 months of occupational or physiotherapy over the previous 2 years, with a minimum frequency of once a week.

Data analysis

The participants were divided in to three groups: children with a high or moderate risk of developing ADHD (ADHD group), children with a diagnosis of or at risk of developing ASD (AUTI group), and children with neither condition (no diagnosis group). χ2 analyses were used to control frequencies of sex, preterm versus term children, and whether or not children received therapy in the three groups.

The three groups were compared for initial M-ABC score, initial age, IQ, and M-ABC score at follow-up by means of a one-way analysis of variance (ANOVA). A post-hoc Tukey analysis was used.23 M-ABC results at follow-up were dichotomized into children with a M-ABC score at or below the 15th centile and those with an M-ABC score greater than the 15th centile.

A paired-samples t-test was used to compare initial and follow-up M-ABC scores. The difference between M-ABC scores (ΔM-ABC) was considered to be the main outcome measure of this study. ΔM-ABC was compared in males and females, in preterm and term children, and in children receiving therapy or not receiving therapy by means of independent-samples t-tests. A Pearson’s correlation analysis was used to analyse the relationship between ΔM-ABC and initial age, initial M-ABC score, and IQ.

The number of participants (n=49) did not allow all possible factors and covariates to be controlled in one model. Several univariate general linear models were used to compare the three groups while controlling for one single factor or covariate. The level of significance was set at p=0.05.

Results

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Initial assessment

The children were between 4 years and 6 years 11 months (mean age 5y 6mo; SD 10mo). A group of 16 children with a high or moderate risk of developing ADHD (ADHD group), a group of 15 children with a diagnosis of or at risk of developing ASD (AUTI group), and a group of 18 children not falling into either of these groups (no diagnosis group) were identified. Two of the children from the ADHD group were also considered at risk for ASD. One child from the AUTI group was also considered at risk for ADHD. The numbers of males and females and the number of preterm children (<37wks) are reported in Table I. χ2 analyses did not reveal significant differences between the three groups. Initial age, IQ, and initial M-ABC scores are reported in Table II, along with the results of the ANOVA to compare the three groups for these variables. The initial M-ABC score was significantly different between the three groups. Post-hoc Tukey analyses revealed a significantly poorer M-ABC score in the AUTI group than in the no diagnosis group (p=0.017).

Table I.   Number of males and females, number of preterm children and term children, and number of children receiving or not receiving therapy in three groups of childrena
Initial assessmentComplete group (n=49)Autism group (n=15)ADHD group (n=16)No diagnosis (n=18)
  1. aAutism group: 10 children with a diagnosis of autism spectrum disorder (ASD) and five children considered at risk of developing ASD; attention-deficit–hyperactivity disorder (ADHD) group: 10 children considered at high risk of developing ADHD and six children at possible risk of developing ADHD; no diagnosis group: children for whom no diagnosis or risk of developing ASD or ADHD was put forward.

Males/females (% males)39/10 (80)14/1 (93)12/4 (75)13/5 (72)
Preterm/term (% preterm)9/40 (18)2/13 (13)5/11 (31)2/16 (11)
Therapy/no therapy (% receiving therapy)19/21 (46)4/8 (27)9/5 3(6)6/8 (40)
Table II.   Initial age, IQ, initial M-ABC score,a follow-up M-ABC score,a and difference between M-ABC scoresa (ΔM-ABC) for the three groups of childrenb, and the results of the one-way-ANOVA to compare these groups
 Complete group (n=49), mean (SD); rangeAutism group (n=15), mean (SD); rangeADHD group (n=16), mean (SD); rangeNo diagnosis (n=18), mean (SD); rangeANOVA, F2,46; p-value
  1. aMovement Assessment Battery for Children (M-ABC), centile score. bAUTI group: 10 children with a diagnosis of autism spectrum disorder (ASD) and five children considered at risk for ASD; attention deficit–hyperactivity disorder (ADHD) group: 10 children considered at high risk of developing ADHD and six children at possible risk of developing ADHD; no diagnosis group: children for whom no diagnosis or risk of developing ASD or ADHD was put forward.

Initial assessment
 Age5y 6mo (10mo); 4y–6y 11mo5y 2mo (0.8y); 4y–6y 2mo5y 10mo (0.9y); 4y 3mo–6y 11mo5y 5mo (10.8mo); 4y–6y 11mo2.77; 0.073
 IQ85.5 (10.5); 71–11883.5 (11.1); 71–10987.0 (7.6); 75–10685.8 (12.4); 71–1180.43; 0.650
 M-ABCa5.3 (4.2); 1–153.1 (3.8); 1–155.6 (3.7); 1–127.1 (4.4); 1–154.13; 0.022
Follow-up assessment
 M-ABC-2*25.3 (29.6); 1–9610.7 (13.1); 1–5421.3 (25.6); 1–8941.2 (35.9); 1–965.37; 0.008
 ΔM-ABC*20 (28.1); –7 to 867.7 (12.6); 0–5015.7 (25.6); –7 to 8634.1 (34.1); –6 to 864.45; 0.017

Follow-up assessment

The time between initial and follow-up assessment was at least 2 years, and at most 3 years (mean 2y 4mo; SD 3mo). At follow-up the children were between 6 years and 9 years 11 months old (mean age 7y 11mo; SD 1y). Nineteen children had received physiotherapy or occupational therapy for at least 6 months in the 2 years before follow-up. Twenty-one children received no physiotherapy or occupational therapy between assessments. Information was missing for nine children (Table I). Frequencies were not significantly different in the three groups.

The M-ABC centile scores at follow-up are reported in Table II. ANOVA revealed a significant effect of group on M-ABC at follow-up (F2,46=5.37, p=0.008). Tukey post-hoc analyses revealed a significant difference between the AUTI group and the no diagnosis group (p=0.007).

At follow-up, 13 children had an M-ABC score at or below the 5th centile, 14 children had a score between the 5th and 15th centiles, seven children had a score between the 15th and 25th centiles, and 15 children had a score above the 25th centile. Table III shows the frequencies of children with M-ABC scores at or below the 15th centile or above the 15th centile in the three groups.

Table III.   Frequency table of M-ABC score at or below the 15th centile versus above the 15th centile at follow-up in three groups of childrena
 AUTI groupADHD groupNo diagnosisTotal
  1. aAUTI group: 10 children with a diagnosis of autism spectrum disorder (ASD) and five children considered at risk for ASD; attention-deficit–hyperactivity disorder (ADHD) group: 10 children considered at high risk of developing ADHD and six children at possible risk of developing ADHD; no diagnosis group: children for whom no diagnosis or risk of developing ASD or ADHD was put forward. M-ABC, Movement Assessment Battery for Children.

M-ABC score ≤15th centile138627
M-ABC score >15th centile281222
Total15161849

Stability of motor performance across time

A paired-samples t-test demonstrated that the M-ABC score centile at follow-up was significantly better than the initial M-ABC score centile (p=0.012). The difference in ΔM-ABC between males and females was not significant (Table IV). The improvement in M-ABC score was lower in preterm children than in term children, but the difference was not significant (p=0.087; Table IV). Children receiving no therapy improved significantly more than children who did receive therapy (p=0.001; Table IV). The correlation coefficient (rp) for the association between ΔM-ABC and initial age, initial M-ABC score, and IQ was –0.26, –0.28, and 0.03 respectively. Only the correlation between ΔM-ABC and the initial M-ABC score was significant (p=0.048).

Table IV.   Difference between initial and follow-up M-ABC scores (ΔM-ABC) (mean and SDs) according to sex, preterm and term birth, and therapy or no therapy. The p-values of the differences between groups were analysed by independent-samples t-tests
 ΔM-ABC score, mean (SD)p-value
  1. M-ABC, Movement Assessment Battery for Children.

Males21.3 (28.4)0.52
Females14.7 (27.3) 
Preterm children10.7 (19.2)0.087
Term children22.1 (29.6) 
Children receiving therapy14.7 (20.2)0.001
Children not receiving therapy25.7 (33.8) 

ΔM-ABC in the three groups of children is reported in Table II. A one-way ANOVA to compare ΔM-ABC between the groups was significant (p=0.017). Tukey post-hoc analyses demonstrated a significant difference between the AUTI group and the no diagnosis group (p=0.017). Several general linear models were used to compare ΔM-ABC between the three groups while controlling for a single factor or covariate. Sex, preterm versus term birth, and therapy versus no therapy were not significant factors in these models. Initial M-ABC score and IQ were not significant as covariates in these models. Only initial age was a significant covariate (F1,45=3.9, p=0.052).

Discussion

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Overall, the mean M-ABC score among children aged 4 to 6 years improved significantly after 2 or 3 years. Of the 49 children with an M-ABC score below the 15th centile at preschool age, 22 no longer fell into this category 2 or 3 years later. This indicates that poor motor performance at preschool age was not always a stable condition in the current sample. Some of the children received therapy in the period between assessments. Unexpectedly, M-ABC score increased most among children who did not receive therapy. It is likely that only the children with the poorest prognosis received therapy.

The results of this study are quite different from those of Pless et al.14 In the present study the diagnosis of DCD was not included because this diagnostic term is not used clinically for children below 6 years of age. The question of whether preschool children can be diagnosed as having DCD is an important one, but one that this study cannot answer because information on motor problems in daily life was often not available. The results of Pless et al.14 suggested that motor status in children with poor motor performance at the age of 5 and 6 years is relatively stable. However, in both studies the number of participants was small and so the findings cannot easily be compared. In addition, the studies used different methods of recruiting children. Pless et al.14 investigated a group of children in whom poor motor performance was identified during population-based screening, whereas the children in our study were recruited from a group referred for diverse developmental problems. However, given that motor performance was more stable in the children with autism, we cannot attribute the difference between the studies to the difference in recruitment methods. Our study included children with a wider age range than the study of Pless et al. However, the increase in M-ABC score was lower in younger children. In the case of some children in our study, the period between assessments was longer than 2 years, which could be one possible explanation for the difference between the results of the two studies.

M-ABC scores differed among the three groups of children both initially and at follow-up. Children with a diagnosis of autism or considered at risk of developing autism had the lowest scores and the lowest increase in M-ABC score. The poor prognosis for the group with or at risk of autism is in agreement with the results reported by Wisdom et al.24 who also reported consistency of poor motor performance in children with autism.

Although several studies have reported on the developmental progression of preschool children with autism and with ADHD, to our knowledge no study has incorporated the monitoring of motor development. Recently, Ming et al.25 published a cross-sectional study on the prevalence of motor impairment in a cohort of children with autism aged between 2 and 6 years (n=83) and a cohort aged between 7 and 18 years (n=71). They reported a reduced prevalence of poor motor performance in the older group. However, as the study of Ming et al.25 was cross-sectional, no firm conclusions regarding the stability of motor performance can be drawn.

It is clear from this study that poor motor performance, assessed by means of the M-ABC, is not always stable over 2 to 3 years. Nevertheless, this current study has some important limitations. The age distributions at initial and follow-up assessment overlap. Some children were younger at follow-up than others were at the initial assessment. Thus, conclusions should be drawn with caution. For this reason, no assumption of a critical period in the course of (delayed) motor development can be made. The large number of children whose parents refused to participate in the follow-up assessment is also an important confounding factor. Most parents who refused argued that the children were tested recently and that new testing would be too demanding for the children. It is likely that many of the children who dropped out were not doing very well if they required repeated assessment. Thus, the most important conclusion from this study is that more prospective longitudinal studies are sorely needed to be able to make evidence-based decisions on the development and/or implementation of screening and/or early intervention programmes for children aged between 4 and 6 years with poor motor performance. The divergent results of this study and that of Pless et al.14 suggest the need for population-based studies as well as studies of clinically referred children. In the meantime, we suggest that these children be carefully monitored. In children in whom autism is combined with poor motor performance, we recommend that early intervention programmes focus also on motor development. Physiotherapists or occupational therapists can stimulate the learning of functional motor skills or help the child to compensate for motor skills disturbances.

What this paper adds

  • • 
    This study provides longitudinal data on the stability of motor problems in clinically referred children below the age of 6 years, while controlling for sex, initial age, IQ, preterm birth, therapy, and behavioural diagnosis.
  • • 
    It reveals that young children with ASD and a combined poor motor performance are in need for early intervention focused also on their motor skills.
  • • 
    It points to an urgent need for more prognostic studies on motor disabilities, starting at preschool age.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

The authors thank the Centre for Developmental Disabilities, Ghent, the children and their parents, and also Jennifer Desmet, Marjan Van Assche, Sarah Van Heuverzwijn, and Karen Vergucht for collecting data.

References

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  • 1
    American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders: DSM-IV-TR, 4th edn. Washington: American Psychiatric Association, 2000.
  • 2
    Geuze RH, Jongmans MJ, Schoemaker MM, Smits-Engelsman BC. Clinical and research diagnostic criteria for developmental coordination disorder: a review and discussion. Hum Mov Sci 2001; 20: 747.
  • 3
    Esposito G, Venuti P, Maestro S, Muratori F. An exploration of symmetry in early autism spectrum disorders: analysis of lying. Brain Dev 2009; 31: 1318.
  • 4
    Kroes M, Kessels AG, Kalff AC, et al. Quality of movement as predictor of ADHD: results from a prospective population study in 5- and 6-year-old children. Dev Med Child Neurol 2002; 44: 75360.
  • 5
    Kadesjo B, Gillberg C. The comorbidity of ADHD in the general population of Swedish school-age children. J Child Psychol Psychiatry 2001; 42: 48792.
  • 6
    Flapper BC, Houwen S, Schoemaker MM. Fine motor skills and effects of methylphenidate in children with attention-deficit–hyperactivity disorder and developmental coordination disorder. Dev Med Child Neurol 2006; 48: 1659.
  • 7
    Rinehart NJ, Tonge BJ, Iansek R, et al. Gait function in newly diagnosed children with autism: cerebellar and basal ganglia related motor disorder. Dev Med Child Neurol 2006; 48: 81924.
  • 8
    Vanvuchelen M, Roeyers H, De Weerdt W. Nature of motor imitation problems in school-aged boys with autism: a motor or a cognitive problem? Autism 2007; 11: 22540.
  • 9
    Wilson PH. Practitioner review: approaches to assessment and treatment of children with DCD: an evaluative review. J Child Psychol Psychiatry 2005; 46: 80623.
  • 10
    Darrah J, Hodge M, Magill-Evans J, Kembhavi G. Stability of serial assessments of motor and communication abilities in typically developing infants – implications for screening. Early Hum Dev 2003; 72: 97110.
  • 11
    Malina RM. Motor development during infancy and early childhood: overview and suggested directions for research. Int J Sport Health Sci 2004; 2: 5066.
  • 12
    Shaw P, Kabini NP, Lerch JP, et al. Neurodevelopmental trajectories of the human cerebral cortex. J Neurosci 2008; 28: 358694.
  • 13
    Cousins M, Smyth MM. Progression and development in developmental coordination disorder. In: SugdenDA, ChambersM, editors. Children with Developmental Coordination Disorder. London: Whurr Publishing, 2005: 11943.
  • 14
    Pless M, Carlsson M, Sundelin C, Persson K. Preschool children with developmental coordination disorder: a short-term follow-up of motor status at seven to eight years of age. Acta Paediatr 2002; 91: 5218.
  • 15
    Henderson SE, Sugden DA. Movement Assessment Battery for Children: Manual. London: Psychological Corporation, 1992.
  • 16
    Tellegen PJ, Winkel M, Wijnberg-Williams BJ, Laros JA. Snijders-Oomen, niet verbale intelligentietest: SON-R 2½–7: Handleiding en verantwoording (SON-R 2½–7). Lissse: Swets Test Publishers, 1998.
  • 17
    Wechsler DA. WISC-NR: Wechsler Intelligence Scale for Children – Revised: Nederlandstalige versie. Lisse: Swets & Zeitlinger, 1986.
  • 18
    Van der Meulen BF, Smrkovsky M. MOS 2½–8½ Mc Carthy Ontwikkelingsschalen: Handleiding. Lisse: Swets & Zeitlinger, 1985.
  • 19
    Vander Steene G, Bos A. Wechsler Preschool and Primary Scale of Intelligence. Vlaams-Nederlandse Aanpassing, 2nd edn. Testinstructie. Lisse: Swets Test Publishers, 1997.
  • 20
    Lord C, Rutter M, Dilavore PC, Risis S. Autism Diagnostic Observation Schedule. Los Angeles: Western Psychological Services, 2002.
  • 21
    Lord C, Rutter M, Le Couteur A. Autism Diagnostic Interview – Revised. A revised version of a diagnostic interview for caregivers of individuals with possible pervasive developmental disorders. J Autism Dev Disord 1994; 30: 20523.
  • 22
    Achenbach TM. Manual for the Child Behaviour Checklist 6–18. Burlington: University of Vermont, 1991.
  • 23
    Neter J, Kutner MH, Nachtsheim CJ, Wasserman W. Applied Linear Statistical Models, 4th edn. Boston: McGraw Hill, 1996.
  • 24
    Wisdom SN, Dyck MJ, Piek JP, Hay D, Hallmayer J. Can autism, language and coordination disorders be differentiated based on ability profiles? Eur Child Adolesc Psychiatry 2007; 16: 17886.
  • 25
    Ming X, Brimacombe M, Wagner GC. Prevalence of motor impairment in autism spectrum disorders. Brain Dev 2007; 29: 56570.