SEARCH

SEARCH BY CITATION

Keywords:

  • atypical autism;
  • autistic symptoms;
  • childhood autism;
  • high-functioning;
  • pervasive developmental disorders (PDD)

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Abstract  To assess autistic symptom differences between high-functioning atypical autism (atypical symptomatology) (HAA; IQ ≥ 70) and childhood autism (HCA), 53 HAA children (mean: 6.0 ± 0.5 years) were compared with 21 HCA children (mean: 8.2 ± 1.1 years) on the Childhood Autism Rating Scale–Tokyo version (CARS-TV). Because IQ on the Japanese version of the Stanford–Binet and CARS-TV total scores differed significantly between HAA and HCA, analysis of covariance was conducted with IQ and CARS-TV total scores controlled for. In two items of CARS-TV (relationship with people and general impressions) the HAA children were significantly less abnormal than the HCA children. Affect tended to be significantly milder in HAA than HCA. Anxiety reaction was significantly more abnormal in HAA than HCA. These findings may be useful to distinguish between HAA and HCA.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Pervasive developmental disorders (PDD) are a group of autistic conditions characterized by pervasive impairments in social interaction; verbal and non-verbal communication; and restricted and stereotyped behavior. According to International Classification of Diseases (10th revision; ICD-10),1 PDD includes eight subtypes (i.e. childhood autism, atypical autism, Rett's syndrome, other childhood disintegrative disorder, overactive disorder associated with mental retardation and stereotyped movements, Asperger's syndrome, other pervasive developmental disorders, and pervasive developmental disorders, unspecified). Until a decade ago the prevalence of PDD had been reported to be 4–5 per 10 000 children. However, the prevalence of PDD turned out to be higher than previously thought. Chakrabarti and Fombonne reported a high prevalence of PDD (62.6 per 10 000 children).2 Moreover, high-functioning (i.e. IQ = 70) PDD (HPDD) seems to be more prevalent than previously believed because the number of reports on the high prevalence of high-functioning autism ranging from 0.1%3 to 0.18%4 has increased since the middle of the 1990s.

There still exists, however, a debate on the concept of PDD, especially with that of atypical autism, which is also termed ’PDD not otherwise specified’ (PDDNOS) according to Diagnostic and Statistical Manual of Mental Disorders (4th edn; DSM-IV).5 According to ICD-10,1 atypical autism is defined as a condition in which abnormal and/or impaired development becomes manifest for the first time only after age 3 years and/or there are insufficient demonstrable abnormalities in one or two of the three areas of psychopathology required for the diagnosis of autism (i.e. reciprocal social interactions; communication; and restrictive, stereotyped, repetitive behavior). Atypical autism, however, is often described as a continuum with childhood autism, although it has similar yet less severe symptoms of childhood autism.6,7 Diagnostic criteria of atypical autism need to be established because a differential diagnosis of PDDNOS from childhood autism is frequently difficult. In addition, atypical autism seems to be more prevalent than childhood autism,2,7 high-functioning autism,8 and Asperger's syndrome.9 Chakabarti and Fombonne reported that the prevalence of PDDNOS was 36.1 per 10 000 and almost all PDDNOS children (92.4%) were high-functioning.2

For classifying milder variants of autistic type behavior, the ICD-10 provides the category of atypical autism. Buitelaar and van der Gaag showed that high-functioning childhood autism (HCA) is significantly more abnormal than high-functioning PDD-NOS in DSM-IV criteria of autistic disorder.10 In another study, Buitelaar et al. also showed that the mean total score on the Childhood Autism Rating Scale (CARS) was higher in HCA than in high-functioning atypical autism (HAA).11 Their studies, however, did not make clear which autistic symptoms were different between the two high-functioning PDD variants. In addition, it is frequently difficult for even experienced clinicians to differentiate clearly between HCA and HAA solely on the basis of clinical observation, because autistic symptoms of such high-functioning PDD children are usually much milder than those of their mentally retarded counterparts. It is important to clarify autistic symptoms that distinguish between the two important variants of high-functioning PDD, if any, to facilitate understanding of children suffering such conditions.

The purpose of the present study is to clarify autistic symptoms on the Childhood Autism Rating Scale–Tokyo version (CARS-TV), to see whether HCA and HAA can be differentiated.

METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Subjects

The subjects of the present study consisted of 74 children (mean: 6.6 ± 0.5 years; range: 2.7–22.9 years; 63 boys, 11 girls) with childhood autism or atypical autism, who were high-functioning (i.e. IQ ≥ 70). There were 53 children (mean: 6.0 ± 0.5 years; range: 2.7–22.9 years; 45 boys, eight girls) with HAA and 21 children (mean: 8.2 ± 1.1 years; range: 3.2–18.2 years; 18 boys, three girls) with HCA. Both diagnoses were based on ICD-10; all HAA children in the present study had only atypical symptomatology. The two groups did not differ significantly in age and sex ratio.

The HAA and HCA children were selected from outpatients of two facilities in Tokyo specializing in developmental disorders (i.e. the Child Guidance Center affiliated with the National Welfare Foundation for Disabled Children and the Nerima Welfare Center for the Mentally and Physically Handicapped). In these facilities, detailed history taking of children, medical/psychological evaluation of children, and diagnosis by DSM-IV/ICD-10 are made by clinical teams consisting of an experienced child psychiatrist, psychologists and other mental health professionals. Moreover, there is a system in each of the two facilities to introduce children to day care and individual remedial teaching, affiliated to the facilities, from which we also obtained important clinical information on the children.

Procedure

For measuring intellectual functioning, experienced psychologists measured the IQ of the subjects on the Japanese version of the Stanford–Binet Intelligence Scale at the time of the first visit to the facilities. Mean ages of measurement of IQ did not differ significantly between HAA (mean: 6.0 ± 3.9 years) and HCA (mean: 8.2 ± 4.9 years; t(30) = 1.89, P = 0.069).

Experienced psychologists rated children on the CARS-TV12 before examination of the children by a child psychiatrist without information about the children's ICD-10 diagnoses in the two facilities. The CARS-TV is a Japanese version12 of CARS developed by Schopler et al.13 and has a satisfactory reliability and validity. The CARS-TV consists of 15 items about autistic symptoms (i.e. relationship with people, imitation, affect, use of body, relationship to non-human objects, adaptation to environmental change, visual responsiveness, auditory responsiveness, near receptor responsiveness, anxiety reaction, verbal communication, non-verbal communication, activity level, intellectual functioning and general impressions). Each of the 15 items scores from 1 (normal) to 4 (severely abnormal) with 0.5 as the unit of measurement. The degree of autism is determined by the total of the 15 item scores ranging from 15 to 60 points. A higher score means a more severe degree of autism. Mean ages of measurement for CARS-TV did not differ significantly between HAA (mean: 6.3 ± 3.7 years) and HCA children (mean: 7.7 ± 4.1 years; t(72) = 1.44, P = 0.155).

Clinical data used in the present study were gathered to plan remedial teaching of patients and guidance of parents during routine clinical evaluations; the administrators of the two facilities gave us permission to use these data in the present study.

spss 11.0J for Windows (SPSS Japan, Tokyo, Japan) was used for statistic analysis. In comparison between HAA and HCA, we used t-test for continuous variable. When necessary, we used analysis of covariance (ancova). A significant level was set at P < 0.05 (two-tailed).

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Of the 15 CARS-TV items, the HAA children differed significantly from the HCA children in three items. Relationship with people was significantly less abnormal in HAA (mean: 1.67 ± 0.26) than HCA (mean: 1.93 ± 0.29; t(72) = 3.76, P = 0.000). Affect was significantly less abnormal in HAA (mean: 1.78 ± 0.30) than HCA (mean: 2.02 ± 0.25; t(72) = 3.52, P = 0.001). General impression was significantly less abnormal in HAA (mean: 1.64 ± 0.27) than HCA (mean: 1.90 ± 0.25; t(72) = 3.87, P = 0.000).

The HAA and HCA children did not have a significant difference in the other 12 CARS-TV items (imitation: HAA, 1.7 ± 0.4; HCA, 1.8 ± 0.4; use of body: HAA, 1.7 ± 0.3; HCA, 1.8 ± 0.3; relationship to non-human objects: HAA, 1.7 ± 0.4; HCA: 1.8 ± 0.4; adaptation to environmental change: HAA, 1.9 ± 0.4; HCA, 2.0 ± 0.4; visual responsiveness: HAA, 1.6 ± 0.3; HCA, 1.8 ± 0.4; auditory responsiveness: HAA, 1.7 ± 0.4; HCA, 1.8 ± 0.4; near receptor responsiveness: HAA, 1.7 ± 0.6; HCA, 1.7 ± 0.5; anxiety reaction: HAA, 1.6 ± 0.3; HCA, 1.6 ± 0.4; verbal communication: HAA, 1.8 ± 0.3; HCA, 1.9 ± 0.3; non-verbal communication: HAA, 2.0 ± 0.3; HCA, 2.1 ± 0.4; activity level: HAA, 1.6 ± 0.3; HCA, 1.7 ± 0.3; and intellectual functioning: HAA, 1.9 ± 1.0; HCA, 2.2 ± 1.6).

The mean IQ was significantly lower in the HAA children (mean: 81.5 ± 8.8) than the HCA children (mean: 86.7 ± 11.9); t(72) = 2.10, P = 0.042). The total CARS-TV score was significantly lower in the HAA children (mean: 26.03 ± 3.23) than the HCA children (mean: 28.26 ± 3.64; t(72) = 2.57, P = 0.012).

Because HAA and HCA children differed significantly in IQ and total CARS-TV score, we employed ancova to compare differences on the CARS-TV items, with IQ and total CARS-TV controlled for. As shown in Table 1, HAA children were significantly less abnormal in two CARS-TV items (i.e. relationship with people and general impressions) than HCA children, and HAA children tended to be less abnormal in affect than HCA children. In contrast, HAA children were significantly more abnormal in anxiety reaction than HCA children. The two groups did not differ significantly in the other 11 CARS-TV items.

Table 1.  Comparison of CARS-TV scores, controlling for IQ and CARS-TV total score, between HAA and HCA at IQ ≥ 70 by ancova
 HAA (n = 53) (Mean ± SD)HCA (n = 21) (Mean ± SD)F (1, 72)P
  1. CARS-TV, Childhood Autism Rating Scale–Tokyo version; HAA, high-functioning atypical autism; HCA, high-functioning childhood autism.

  2. Mean ± SD was estimated by ancova, with the mean total CARS-TV scores in HAA and HCA being 26.03 and 28.26, respectively. Raw scores are given in Results.

CARS-TV items
 Relationship with people1.71 ± 0.031.83 ± 0.055.200.026
 Imitation1.71 ± 0.051.72 ± 0.080.010.922
 Affect1.82 ± 0.031.93 ± 0.054.000.050
 Use of body1.80 ± 0.041.72 ± 0.060.900.345
 Relationship to non-human objects1.78 ± 0.051.65 ± 0.081.570.214
 Adaptation to environmental change1.91 ± 0.051.90 ± 0.080.000.976
 Visual responsiveness1.67 ± 0.031.67 ± 0.050.000.967
 Auditory visual responsiveness1.72 ± 0.041.70 ± 0.070.070.794
 Near receptor responsiveness1.75 ± 0.061.57 ± 0.092.410.125
 Anxiety reaction1.63 ± 0.041.47 ± 0.064.960.029
 Verbal communication1.88 ± 0.031.85 ± 0.050.340.562
 Non-verbal communication2.10 ± 0.042.02 ± 0.061.290.260
 Activity level1.65 ± 0.031.62 ± 0.050.300.586
 Intellectual functioning1.88 ± 0.102.20 ± 0.162.700.105
 General impressions1.68 ± 0.031.82 ± 0.056.360.014

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

To our knowledge, the present study is the first to compare autistic symptoms in detail between HAA and HCA. On the CARS-TV with IQ and total CARS-TV score controlled for, HAA children was significantly less abnormal in two items (i.e. relationship with people and general impressions) and tended to be significantly less abnormal in one CARS-TV item (affect) concerning disturbances in social relationships, than HCA children, while HAA children were significantly more abnormal in anxiety reaction than HCA children.

It is reasonable that HAA children were less abnormal in social relationship, which is a core autistic symptom in PDD, than HCA children, because all of the HAA patients in the present study were diagnosed as having atypical autism with atypicality in symptomatology (i.e. having a smaller number of autistic symptoms to satisfy the diagnostic criteria for childhood autism). Even if they were high-functioning, compared with HAA patients, the HCA patients still had a more severe core autistic symptom (impairment in social relationship), in accordance with previous studies that showed more severe impairment in social interaction in CA than AA.14,15

The finding that anxiety reaction was significantly higher in HAA than HCA seems to be consistent with the findings of Buitelaar and van der Gaag that high-functioning children with PDDNOS exhibited markedly disturbed/ambivalent attachments than those with childhood autism.10 The difference in anxiety reaction between HAA and HCA in the present study may be accounted for by the relationship between anxiety reaction and sociality in both PDD subtypes. The majority of anxiety reactions found in the present PDD children was separation anxiety from mothers. Because the HAA children were less impaired in interpersonal relationships and emotions (i.e. having a greater interest in and affection for their mothers), the HAA children could express stronger separation anxiety than the HCA children. The disparity between impairment in social interaction and anxiety reaction may be useful for differential diagnosis between HAA and HCA.

Tachimori et al. noted that the CARS-TV cut-off for discriminating autistic disorder from PDDNOS is 30.0/30.5.16 The percentage of children under this cut-off was 67% in the HCA group and 89% in the HAA group. This means that the majority of HAA or HCA children in the present study were classified into the non-autistic category according to either the original CARS13 or CARS-TV,12 and such children usually do not look autistic clinically. Even in such mild autistic conditions, it seems to be possible to distinguish between HAA and HCA by observing the disparity of anxiety reaction from interpersonal relations and affect in childhood. Such disparity may also influence the long-term outcome in HCA and HAA children, especially in terms of social adaptation, because HCA children were impaired more severely than HAA children in social interaction. For clarification of this, a long-term follow-up study is needed.

Some limitations of the present study need to be addressed. The findings of the present cross-sectional study need to be verified by a longitudinal study because autistic symptoms change over time. Our findings also need verification with a more stringent criterion for high-functioning. When we compared HAA and HCA groups by employing a stricter criterion of IQ ≥ 85, we found that the HCA group tended to be more severely impaired than the HAA group in affect. However, we could not replicate the other significant differences due to the small sample size of high-functioning PDD according to the stricter criterion.

In the present study IQ was significantly lower in the HAA children than the HCA children, in discordance with findings by Chakrabarti and Fombonne and Mayes et al. that IQ was higher in children with PDDNOS or atypical autism than children with autism.2,7 However, Buitelaar et al. reported that children with high-functioning autism have a higher IQ than children with high-functioning PDDNOS,11 which is consistent with our results. Therefore in regard to that discordance, our sample does not appear to be unusual in terms of IQ difference between HAA and HCA.

In conclusion, HAA children were significantly less abnormal in social relationships than HCA children, while HAA children were significantly more abnormal in anxiety reaction than HCA children. This disparity may be useful for distinguishing between the two conditions in clinical settings, although verification with a longitudinal study is needed.

ACKNOWLEDGMENTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

The present study was supported in part by a Grant-in-Aid for Scientific Research (C) (No. 09670977) from the Ministry of Education, Science, Sports and Culture, Japan. The authors are very thankful to staff in the Child Guidance Center affiliated with the National Welfare Foundation for Disabled Children and the Nerima Welfare Center for the Mentally and Physically Handicapped for their assistance in data collection.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  • 1
    World Health Organization. The ICD-10 Classification of Mental and Behavioural Disorders: Diagnostic Criteria for Research. WHO, Geneva 1993.
  • 2
    Chakrabarti S, Fombonne E. Pervasive developmental disorders in preschool children. JAMA 2001; 285: 30933099.
  • 3
    Honda H, Shimizu Y, Misumi K et al. Cumulative incidence and prevalence of childhood autism in children in Japan. Br. J. Psychiatry 1996; 169: 228235.
  • 4
    Baird G, Charman T, Baron-Cohen S et al. A screening instrument for autism at 18 months of age: a 6-year follow-up study. J. Am. Acad. Child Adolesc. Psychiatry 2000; 39: 694702.
  • 5
    American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 4th edn. Amercian Psychiatric Association, Washington, DC, 1994.
  • 6
    Dahl EK, Cohen DJ, Provence S. Atypical development: DSM-III and the pervasive developmental disorders of early onset. J. Am. Acad. Child Adolesc. Psychiatry 1986; 26: 170180.
  • 7
    Mayes L, Volkmar F, Hooks M et al. Differentiating pervasive developmental disorders not otherwise specified from autism and language disorders, L. J. Autism Dev. Disord. 1993; 23: 7990.
  • 8
    Fombonne E. The epidemiology of autism: a review. Psychol. Med. 1999; 29: 769786.
  • 9
    Ehlers HD, Gillberg CL. The epidemiology of Asperger's syndrome. A total population study. J. Child Psychol. Psychiatry 1993; 34: 13271350.
  • 10
    Buitelaar JK, Van Der Gaag RJ. Diagnostic rules for children with PDD-NOS and multiple complex developmental disorder. J. Child Psychol. Psychiatry 1998; 39: 911919.
  • 11
    Buitelaar JK, Van Der Wee M, Swaab-Barneveld H et al. Verbal memory and performance IQ predict theory of mind and emotion recognition ability in children with autistic spectrum disorders and in psychiatric control children. J. Child Psychol. Psychiatry 1999; 40: 869881.
  • 12
    Kurita H, Miyake Y, Katsuno K. Reliability and validity of the childhood autism rating scale–Tokyo version (CARS-TV). J. Autism Dev. Disord. 1989; 19: 389396.
  • 13
    Schopler E, Reichler RJ, De Vellis RF et al. Toward objective classification on childhood autism: childhood autism rating scale (CARS). J. Autism Dev. Disord. 1980; 10: 91103.
  • 14
    Hobson R. The emotional origins of social understanding. Philos. Psychol. 1993; 6: 227245.
  • 15
    Luteijn E, Luteijn F, Jackson S et al. The children's Social Behavior Questionnaire for milder variants of PDD problems: evaluation of the psychometric characteristics. J. Autism Dev. Disord. 2000; 4: 317330.
  • 16
    Tachimori H, Osada H, Kurita H. Childhood autism rating scale–Tokyo version for screening pervasive developmental disorders. Psychiatry Clin. Neurosci. 2003; 57: 113118.