Despite numerous studies showing laboratory and anatomical differences in the brains of individuals with autism and autistic spectrum disorders (ASD), the diagnosis remains clinical, based on behavioral criteria. Key features to be assessed in evaluating a child for possible autism include impairment of the quality of social interactions; impairment of communication skills, both verbal and nonverbal; and the presence of stereotyped, repetitive patterns of behavior.
Deficits in social interactions are the hallmark of the disorder and present in several ways. These children often do not seek the attention of their caregivers, seeming to be unaware of their presence or at least disinterested in them. They often prefer to play by themselves; if involved in play with others, they tend to remain passive or engage in parallel play. Left to themselves, they often are content to engage in repetitive behaviors such as opening and closing doors or lining up objects. Many parents describe the child as being "in his own world."
Despite the primary deficit in social interaction, however, the most common presenting concern is the child's delayed language. For this reason, language screening as part of routine developmental surveillance and monitoring provides one of the earliest opportunities for identification of the child with ASD. Specific recommendations for further evaluation have been provided by the American Academy of Neurology in the publication Practice Parameters for the Diagnosis and Evaluation of Autism.1 These appear in Sidebar 1 (page 666).
Just as several instruments have been devised for developmental surveillance and monitoring, instruments have been developed to help objectify and quantify the physician's clinical impression of autistic features. Most of these instruments are dependent on parental report, which can be subject to bias but nonetheless is useful.
Autism is classified as one of the pervasive developmental disorders (PDDs) in the Diagnostic and Statistical Manual for the Mental Disorders, 4th edition (DSM-IV).2 Diagnostic criteria for autistic disorder from DSM-IV are listed in Sidebar 2. Delays or abnormal functioning must be found by age 3 in at least one of the following areas: social interaction, language as used in social communication, or symbolic or imaginative play. The child must have two characteristics from the social-interaction area, and one characteristic from each of the language and play areas. At least six characteristics are necessary to meet the criteria for diagnosis.
While the diagnostic criteria from DSM-IV are brief enough to review during an office visit, some of the items are subject to interpretation. For example, the phrases "social and emotional reciprocity" and "idiosyncratic language" should be explained by the clinician. Some examples of specific phrases for eliciting suspicious symptoms are listed in Sidebar 3 (page 668).
A number of instruments are used to diagnose autism, varying from short, simple screening tools to in-depth, comprehensive, research-oriented assessments. Some screenings are conducted by the clinician, while others are completed by parental interview. What follows is a brief description of some of these tools. (Note: When "parent" is mentioned, this implies the child's primary caregiver.)
Checklist for Autism in Toddlers (CHAT)3
This checklist can be incorporated easily into a routine well-child check up if concerns about autism should surface. The first is a series of nine questions for the parent to answer. This is followed by five items that the examiner can perform or observe from the child directly. One of the advantages of this test is that it can be used with toddlers as young as 1 8 months. One of the disadvantages of CHAT is that it is a screening device and does not give a definitive diagnosis.
Pervasive Developmental Disorder Screening Test (PDDST)4
This test has three stages. Stage 1 is designed to be used as a screening test by primary care physicians. Stage 2 is for diagnostic use by a specialist in the field of developmental disorders. Stage 3 is designed for diagnostic use by specialists in the field of autism or PDD. This is one of the newer screening tests and thus has not been as widely used as other screening tests.
Each stage consists of a series of questions that can be answered with a "yes" or "no." The stages vary in length from two to four pages. The number of positive answers is scored. If the score is above a certain number, it indicates that the child either needs further evaluation for an autistic spectrum disorder or is likely to have an autistic spectrum disorder.
The Gilliam Autism Rating Scale (GARS)5
GARS, a questionnaire answered by the parent, covers stereotyped behaviors, communication, and social interaction. The child's behavior is rated on a scale of 0 (never) to 3 (frequently) for each item, depending on the frequency of the particular behavior. The data generate an Autism Quotient, which falls into categories for the child's probability of having autism. The probability categories are unlikely, borderline, and likely. GARS is fairly easy to administer and score and requires no special training.
The Childhood Autism Rating Scale (CARS)6
The CARS is completed by conducting a structured interview and direct observations. The clinician must ask the parent how the child's behavior differs from that of typical children in 15 different areas such as body use, response to pain, and eye contact. Each item is scored on a seven-point scale, from normal to markedly abnormal. One advantage with this test is that the cumulative score is placed on a scale ranging from non-autistic to severely autistic. A disadvantage of this method is that the clinician should have some knowledge of or experience with autistic patients in order to rate subjects appropriately.
The Autism Diagnostic Interview - Revised (ADI-R)7-10
ADI-R is a semi-structured interview conducted with the child's parent, focusing on reciprocal social interactions, communication and language, and repetitive, restricted, stereotyped behaviors. The interviewer must be able to elicit a description of certain behaviors. Interviewers should be trained specifically to ensure reliability. This, plus the fact that the interview can easily take 90 minutes, does not make it very useful in the primary care setting. It is a diagnostic tool frequently used in autism specialty clinics.
The Autism Diagnostic Observation Schedule - Generic (ADOS-G)11,12
This semi-structured assessment involves an interview and directed activities of communication, social interaction, and imaginative play. There are four modules, each designed for subjects at different ages and developmental levels. This instrument also requires training and validation procedures.
The ADI-R and ADOS-G are considered the "gold standard" diagnostic instruments and are therefore used in research protocols. They are typically administered by non-physician professionals, such as psychologists.
While the diagnosis of autism is based on behavioral criteria, it still requires a medical evaluation for possible conditions that might mimic symptoms of autism. In addition, comorbid conditions may require assessment. The child's developmental and behavioral history should be taken in detail. Emphasis should be placed on ages of acquiring skills and any evidence of loss of skills, as well as history of difficulty with sleep, eating, tantrums, or general irritability. These may indicate longstanding behavioral difficulties not previously appreciated by family members. Any history of neurologic insult should be noted, whether via trauma, infection, or other means.
A family history of neurologic disorders is also important. An increased incidence of autism has been noted in persons with Fragile X syndrome, tuberous sclerosis, and a history of autism in other family members. A history of mental retardation is also important, as some diagnoses of mental retardation with behavioral problems in past decades may have been misdiagnosed cases of autism.
The physical examination in a case of suspected autism should look for not only signs of disorders previously described but also other congenital anomalies or disorders associated with mental retardation. Dysmorphic facies, abnormalities of ears and eyes, and other craniofacial anomalies should be noted. Presence of two or more major anomalies warrants referral to a geneticist for further evaluation. The head circumference in persons with autism tends to be larger than in typically developing children.13 The skin should be examined with an ultraviolet lamp for the presence of ash-leaf spots, which are depigmented macules that are among the neurocutaneous signs of tuberous sclerosis.
Further evaluation focuses on those disorders that might mimic the symptoms of autism. Complete audiologic and speech pathology evaluations are needed. These are helpful not only in the diagnostic phase but also in the treatment planning phase. An experienced speech pathologist uses a variety of techniques to determine the child's level of function, including naturalistic observation and interviews of other adults acquainted with the child. Children with autism have more limited nonverbal communication skills than others, and thus present somewhat differently from hearing impaired children who demonstrate other communication strategies. Nonetheless, children with autism can have hearing impairment, and hearing must be evaluated as part of their overall assessment.
Epilepsy is known to occur more commonly in persons with autism than in the general population. However, electroencephalogram evaluation should be conducted only on those children with a strong history of possible seizure episodes. Although several studies have indicated abnormalities of brain morphology, imaging of the brain is not helpful diagnostically. The abnormal patterns seen to date consist of changes in mean sizes of certain areas of the brain. These mean differences from control populations are statistically significant but have such overlap with meaurements in control populations that individual findings are meaningless. Neuroimaging is similarly not recommended for simple macrocephaly without other neurologic findings.14
Metabolic disorders such as phenylketonuria are known to have symptoms seen in autism. These are part of routine newbom screening in many states but can at times be overlooked or missed in the neonatal period. When present, these disorders usually manifest with other medical problems, such as failure to thrive, seizures, cyclic vomiting, or other metabolic crises. The percentage of children with an identifiable metabolic disorder is less than 5%.15 As a result, the current recommendation is for selective metabolic testing only in the presence of suggestive clinical and physical findings.16
Genetic factors appear to account for the majority of cases of autism.17 Despite advances in genomics, an autism-specific gene has not been identified. Several candidate genes have been identified, but no single gene has been able to explain the constellation of symptoms seen in autism.
At this time, chromosome studies to rule out Fragile X syndrome are recommended, but no more specific genetic studies are advised. Even with studies for Fragile X, chromosomal testing usually is not helpful. Early reports that noted a high association between Fragile X and autism have not been corroborated,18 and few children with autism are found to have Fragile X syndrome.19 It is possible that more specific genetic testing will be recommended in the next several years, reflecting increased knowledge of the genetic factors involved in the disorder.
Occupational Therapy Assessment
Occupational therapy assessment is often useful when functional skill deficiencies are present. Evaluation by an occupational therapist can assist in the management plan for developing skills needed for daily routines. While sensory integration therapy is often recommended, no good controlled studies suggest its routine use for persons with autism.
Perhaps the most helpful component of the evaluation process is psychological testing to determine cognitive abilities and assist in the formulation of a comprehensive management plan. This should include appropriate tests of intelligence, often requiring use of non-verbal measures rather than more standard tests such as the Stanford-Binet. Measures of adaptive functioning are also needed, using standardized instruments such as the Vlneland Adaptive Behavior Scales. Psychological evaluation is an important part of developing an intensive behavioral intervention using applied behavior analysis, an effective treatment strategy.
THE OTHER PDDs
In the course of assessing a child or adolescent for autism, a diagnosis of one of the other PDDs may be made. The four other disorders in the category of PDDs are Asperger's disorder, Rett's disorder, childhood disintegrative disorder, and pervasive developmental disorder - not otherwise specified (PDD-NOS). There is not as much information in the literature on these disorders. They are, however, defined by DSM-IV, and diagnostic instruments exist for some of them.
Asperger's disorder was first described by Hans Asperger in 1944, around the same time Leo Kanner described autism. Asperger recognized children who exhibited difficulty forming friendships, difficulty relating to others, and developed intense interests in very specific topics. DSM-IV diagnostic criteria for Asperger's disorder are similar to those of autistic disorder in terms of the impairments in social interaction and the repetitive stereotyped patterns of behavior. The major difference is that Asperger's disorder does not include delay in language development. Initial use of single words and short phrases occurs on time.
However, it is clear that individuals with Asperger's disorder do have difficulty with pragmatic language skills. For example, they may not understand expressions or sayings, such as, "It is raining cats and dogs." Speech is taken very literally. Also, they may not understand sarcasm. Nonverbal forms of communication such as body language and facial expressions may be lost on them. This has obvious effects on their social interaction skills. Many of these people have significant difficulty making friends and relating to others.
While an individual with an autistic disorder may engage in repetitive motor activities, individuals with Asperger's disorder have more complicated patterns of interest. For example, they may try to read everything they can about dinosaurs, or check out the same book on trains from the library repeatedly. They may try to talk with others about their areas of interest even if the other party is clearly uninterested. Individuals with Asperger's disorder can be very intelligent and highachieving in their fields of study.
A few diagnostic instruments may be used to evaluate a child or adolescent for Asperger's disorder. One is the Gilliam Asperger's Disorder Scale (GADS), which is a standardized, norm-referenced instrument used to assess persons with Asperger's disorder and other behavioral disorders.20 Similar to the Gilliam Autism Rating Scale, the GADS is a questionnaire answered by the caregiver that rates the child's behavior on a scale of O to 3, depending on the frequency of a particular behavior. Areas rated include social interaction, restricted patterns of behavior, cognitive patterns, and pragmatic skills. The scores are used to generate an Asperger's Disorder Quotient, which then translates to low, borderline, or high probability for having Asperger's disorder.
Another method for screening for this disorder is the Australian Scale for Asperger's Syndrome.21 It describes twenty-four characteristics seen in Asperger's disorder. The parent rates the frequency that the characteristic is seen on a scale of 0, meaning rarely, to 6, meaning frequently. It is completed by either parents or teachers, so it can be conducted before the office visit for review at the next encounter with the family.
Pervasive Developmental Disorder - Not Otherwise Specified (PDD-NOS)
This category is used to describe those individuals who do not meet the full criteria for a diagnosis of a pervasive developmental disorder, but whose behavioral symptoms nonetheless are severe and display pervasive impairments in social interaction or in verbal or nonverbal communication. This category is used for individuals who have atypical presentations such as late age of onset or subthreshold symptomatology. PDD-NOS and Asperger's disorder are the two most commonly identified diagnoses among the PDDs.
Rett! s Disorder
Rett's disorder is characterized by apparently normal development and head circumference during the first 5 months of life, with onset of deceleration of head growth and loss of language and social skills beginning between 5 and 30 months. There are also clear difficulties with fine and gross motor skills, most notably with the loss of purposeful hand skills and development of hand stereotypies, such as hand-wringing motions. Severe psychomotor retardation can occur, with many such patients losing ambulatory skills.
Neuropathologic findings indicate a failure of neuronal maturation, with too small neurons and too few dendritic arbors, and no evidence of a progressive neurodegenerative process. Recent research has indicated mutations in the gene MECP2 occur in more than 80% of females with Rett's disorder. Previously only diagnosed in females, new research has identified a few rare male Rett syndrome cases.
Childhood Disintegrative Disorder
This disorder is characterized by marked regression in multiple areas of functioning following a period of at least 2 years of apparently normal development. These individuals present with behavioral features generally seen in autistic disorder, but the critical difference is the delayed onset Evaluation of these children is the same as described earlier, but with more emphasis on neuroimaging and consideration of disorders, such as metachromatic leukodystrophy because of the delayed onset of symptoms. There is no loss of motor skills as is seen in Rett's disorder, and the severity of symptoms is similar to that for autistic disorder.
1 . Rlipek PA, Accanto PJ, Baranek GT, et al. The screening and diagnosis of autistic spectrum disorders. J Autism Dev Disord. 1999;29<6):439-484.
2. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Washington, DC: APA; 1994.
3. Baron-Cohen S, Allen J, Gillberg C. Can autism be detected at 18 months? The needle, the haystack, and the CHAT. Br J Psychiatry. 1992;161:839-843.
4. Siegel B. Early screening and diagnosis in autism spectrum disorders: the Pervasive Developmental Disorder Screening Test (PDDST). Paper presented at the National Institutes of Health State of the Science in Autism Screening and Diagnosis Working Conference. Bethesda. Md; June 15-17, 1998.
5. Gilliam JE. Gilliam Autism Rating Scale (GARS). Austin, Tex: Pro-Ed; 1995.
6. Schopler E, Reichler R, Rochen-Renner B. The Childhood Autism Rating Scale (CARS). Los Angeles, Calif: Western Psychological Services; 1988.
7. Le Couteur A, Rutter M, Lord C, et al. Autism diagnoistic interview: a standardized investigator-based instrument. J Autism Dev Disord. 1989;19(3):363-387.
8. Lord C, Storoschuk S, Rutter M, Pickles A. Using the ADI-R to diagnose autism in preschool children. Infant Ment Health J. 1993;14:234-252.
9. Lord C, Pickles A, McLennan J, et al. Diagnosing autism: analyses of data from the autism diagnosis interview. J Autism Dev Disord. 1997;27(5):501-517.
10. Lord C, Rutter M, Le Couteur A. Autism Diagnostic Interview - Revised: a revised version of the diagnostic interview for caregivers and individuals with possible pervasive developmental disorders. J Autism Dev Disord. 1994;24(5):659-685.
1 1. DiLavore PC, Lord C, Rutter M. The pre-linguistic autism diagnostic observation schedule. J Autism Dev Disord. 1995;25(4):355-379.
12. Lord C, Risi S, Lambrecht L, et al. The Autism Diagnostic Observation Schedule - Generic: a standard measure of social and communication deficits associated with the spectrum of autism. J Autism Dev Disord. 20O0;3O(3):205-223.
13. Davidovich M, Patterson B, Gartside P. Head circumference measurements in children with autism. J Child Neurol 1996; 1 1(5):389-393.
14. Filipek PA. Neuroimaging in the developmental disorders: the state of the science. J Child Psychol Psychiatr. 1 999;40( 1 ): 1 1 3- 1 28.
15. Dykens EM, Volkmar FR. Medical condition associated with autism. Ih: Cohen DJ, Volkmar FR, eds. Handbook of Autism and Pervasive Developmental Disorders. 2nd ed. New York, NY: Wiley; 1997:388-110.
16. Curry CJ, Stevenson RE, Aughton D, et al. Evaluation of mental retardation: recommendations of a Consensus Conference: American College of Medical Genetics. Am J Med Genet. 1997;72(4):468-477.
17. Rutter M. Genetic studies of autism: from the 1970s into the millennium. J Abnorm Child Psychol. 2000;28(1):3-14.
18. Bailey A, Bolton P, Butler L, et al. Prevalence of the fragile X anomaly amongst autistic twins and singletons. / Child Psychol Psychiatry. I993;34(5):673-688.
19. Feinstein C, Reiss AL. Autism: the point of view from fragile X studies. J Autism Dev Disord. 1998;28<5):393^05.
20. Gilliam JE. Gilliam Asperger's Disorder Scale. Austin. Tex: Pro-Ed; 2001 .
21. Garnen MS, Attwood AJ. The Australian scale for Asperger's syndrome. In: Attwood T, ed. Asperger's Syndrome: A Guide for Parents and Professionals. London, England: Kingsley; 1998:17-19.