Differential diagnosis

Other pervasive developmental disorders (PDD)
Autism/autism spectrum disorders (DSM–IV)
Autism shares the same DSM criteria for abnormal social interaction and behaviour as Asperger disorder, but requires additional impairments in communication (Box 3). Delay or impairment in social interaction, communication or behaviour must arise before age 3 years. In DSM–IV, if both autism and Asperger disorder diagnoses can be made, the autism diagnosis takes precedence. Unlike Asperger disorder (which excludes individuals with delays in cognitive abilities or other skills), autism can occur at all levels of ability: the majority (70%) of cases have associated learning disabilities and almost half have an IQ below 50.

In a prevalence study of autism, Wing & Gould (1979) identified a large number of children who failed to meet the diagnostic criteria for classic autism, but had a triad of impairments involving social interaction, communication and imagination, with additional repetitive stereotyped activities. This triad of symptoms, termed the ‘autistic spectrum’, was recognised at all levels of intelligence and is included in DSM–IV as ‘pervasive developmental disorder not otherwise specified’, and in ICD–10 as ‘atypical autism’, ‘other pervasive developmental disorders’ or ‘pervasive developmental disorders, unspecified’. The estimated prevalence of autism spectrum disorders (autistic spectrum disorders) may be as high as 91 per 10 000.

At present it is not clear whether Asperger syndrome is distinct from the autistic spectrum disorders. Diagnostic uncertainty arises in patients of near normal cognitive ability (total IQ >70) with autistic spectrum disorders, who are described as having high-functioning autism (HFA). Differences are reported between Asperger syndrome and HFA, for example, verbal skills being significantly greater than non-verbal ones in Asperger syndrome – the opposite of the pattern reported in HFA. This differentiation may be simplistic, as it depends on the diagnostic system used. A sample defined using Wing or Gillberg criteria for Asperger syndrome would include children with abnormal or delayed language and, by definition, worse verbal skills than a DSM or ICD sample, which would exclude these individuals. Attempts to separate Asperger syndrome or HFA categorically based on presence or absence of language delay are artificial. Comparison studies have also failed to control adequately for IQ differences between samples in many cases. Current research data do not convincingly support the separation of Asperger syndrome and the autistic spectrum disorders as distinct disorders. Both Asperger syndrome and autistic spectrum disorders are about five times more common in boys than in girls; segregate within the same families; appear strongly genetic (American Psychiatric Association, 2000); and share similar comorbidity (Gillberg & Billstedt, 2000). If autism and Asperger syndrome differ at all, it is in the degree of impairment rather than in having discrete, specific and independent features. This is supported by outcome studies; both are associated with social difficulties persisting into adulthood, but these are less severe in Asperger syndrome. Combining Asperger syndrome and autistic spectrum disorders into an autistic spectrum is a better way forward.

In addition to the diagnostic criteria mentioned, a number of assessment instruments are available or in development for use with people with Asperger syndrome. These include the Asperger Syndrome Screening Questionnaire (ASSQ; Ehlers et al, 1999) and the Pervasive Developmental Disorder Questionnaire (PDD–Q; Baron-Cohen et al, 1996). The ASSQ is designed for completion by parents or teachers to screen for Asperger syndrome and other high-functioning autism spectrum disorders in school-age children. The PDD–Q is an 18-item questionnaire for parents, which includes questions specifically designed to identify clinical characteristics of Asperger syndrome.

These instruments were all designed for screening purposes, not to differentiate Asperger syndrome from HFA, but to identify higher-functioning individuals within the autistic spectrum. Each requires further work to fulfil psychometric requirements for sensitivity, specificity, reliability and validity. Of existing instruments used for the diagnosis of autistic spectrum disorders, the Autism Diagnostic Interview – Revised (ADI–R; Lord et al, 1994) may be useful in assessing individuals for Asperger syndrome. This instrument consists of three scales corresponding to the social, communication and behavioural impairments and is based on parent report. The Autism Diagnostic Observational Schedule – Generic (ADOS–G; Lord et al, 1999), a supplementary standardised interview and observational assessment may also be helpful. Accurate diagnosis in younger children (under 2 years of age) is difficult, despite the neurodevelopmental nature of these disorders. The Checklist for Autism in Toddlers (CHAT; Baird et al, 2000) may prove useful in identifying children at risk. Howlin (2000) provides a more extensive appraisal of available assessment instruments. Full assessment of Asperger syndrome requires a multi-disciplinary approach, as outlined in Box 4

Pervasive developmental disorder not otherwise specified (DSM–IV)
Another diagnosis of relevance is pervasive developmental disorder not otherwise specified (PDDNOS). This is characterised by “a severe and pervasive impairment in the development of reciprocal social interaction or verbal and non-verbal communication skills or when stereotyped behaviour, interests and activities are present” (American Psychiatric Association, 1994: p. 77). This category in DSM–IV needed revision as the inclusion of the word ‘or’ twice greatly diluted its meaning and grossly widened it as a category. DSM–IV–TR has corrected this error, requiring there to be an impairment in reciprocal social interaction associated with an impairment in communication skills or with stereotyped behaviour, interests or activities. In the past, particularly in the USA, PDDNOS was used as a synonym for Asperger syndrome, although for parents this title is unhelpful and confusing.

Other pervasive developmental disorders (DSM–IV)
Neither Rett disorder nor childhood disintegrative disorder are part of the autistic spectrum, and they are unlikely to represent sources of diagnostic confusion. Rett disorder is most common in girls and is characterised by apparently normal development in the first 5 months of life, with subsequent deceleration of head growth, loss of previously acquired hand skills, loss of social engagement, poorly coordinated gait and language problems (American Psychiatric Association, 2000). Similarly, childhood disintegrative disorder presents with loss of language, motor skills and bowel and bladder control following a 2-year history of normal development (American Psychiatric Association, 2000). In each case the diagnosis should be excluded by a detailed developmental history and physical examination.