American Family Physician

Home , Epidemiology of autism

Autism: A Medical Primer CHRISTOPHER D. PRATER, M.D., and ROBERT G. ZYLSTRA, ED.D., L.C.S.W. University of Tennessee College of Medicine, Chattanooga, Tennessee

Autistic disorder, a pervasive developmental disorder resulting in social, language, or sensorimotor deficits, occurs in approximately seven of 10,000 persons. Early detection and inter- O A patient infor- vention significantly improve outcome, with about one third of autistic persons achieving mation handout about autism, written some degree of independent living. Indications for developmental evaluation include no bab- by the authors of this bling, pointing, or use of other gestures by 12 months of age, no single words by 16 months article, is provided on of age, no two-word spontaneous phrases by 24 months of age, and loss of previously page 1680. learned language or social skills at any age. The differential diagnosis includes other psychiatric and pervasive developmental disorders, deafness, and profound hearing loss. Autism is frequently associated with fragile X syndrome and tuberous sclerosis, and may be caused by lead poisoning and metabolic disorders. Common comorbidities include mental retardation, seizure disorder, and psychiatric disorders such as depression and anxiety. Behavior modification programs are helpful and are usually administered by multidisciplinary teams; targeted medication is used to address behavior concerns. Many different treatment approaches can be used, some of which are unproven and have little scientific support. Parents may be encouraged to investigate national resources and local support networks. (Am Fam Physician 2002;66:1667-74,1680. Copyright© 2002 American Academy of Family Physicians.)

See page 1591 ecognition of the disorder for definitions of called autism may have its ori- Definition strength-of-evidence gin in Itard’s 1801 description levels. Autistic disorder is a pervasive developmen- of the “wild boy of Aveyron,” a tal disorder defined behaviorally as a syn- violent child with no language drome consisting of abnormal development Rskills who related to other people as if they of social skills (withdrawal, lack of interest in were objects. It was not until 1943 that Kanner peers), limitations in the use of interactive identified a complex set of characteristics language (speech as well as nonverbal com- (e.g., aberrations in social development, ver- munication), and sensorimotor deficits (in- bal and nonverbal communication, symbolic consistent responses to environmental stim- thinking) for a syndrome he labeled “autism.” uli).1,2 In this article, the more generic terms Although Kanner theorized that a single, autism and autistic refer to the broad spectrum biologically based defect was responsible for of pervasive developmental disorders that the development of autistic disorders, treat- exhibit autistic features as their primary pre- ment in the 1950s and 1960s was dominated senting behaviors. The term autistic disorder is by the psychodynamic theory of the etiology of used to describe the specific developmental autism that charged that pathologic parenting disorder that occurs at the more severely was responsible for the withdrawal of children affected end of this spectrum (Table 1).1 from their environment. Following the 1970s The development of impairments in autis- discovery of neuroreceptors, endogenous neu- tic persons is varied (Table 21) and character- rohormones, and the stereospecific binding istically uneven, resulting in good skills in sites of neuropeptides to neurons, clinicians some areas and poor skills in others. Echolalia, have discounted the psychodynamic theory of the involuntary repetition of a word or a sen- autism and repostulated Kanner’s original tence just spoken by another person, is a com- supposition that biologically based deficits are mon feature of language impairment that, responsible for the etiology of autism. when present, may cause language skills to appear better than they really are. There may See editorial on page 1610. also be deficiencies in symbolic thinking, stereotypic behaviors (e.g., repetitive nonpro-

NOVEMBER 1, 2002 / VOLUME 66, NUMBER 9 www.aafp.org/afp AMERICAN FAMILY PHYSICIAN 1667 Early detection and intervention in autistic disorders signifi- TABLE 2 cantly improve outcomes, with about one third of patients Impairments Common to Autistic Syndromes achieving some degree of independent living as adults.

Impairments in social skills Limitations in the use of interactive language TABLE 1 Sensorimotor deficiencies DSM-IV Diagnostic Criteria for Autistic Disorder Echolalia Deficiencies in symbolic thinking Stereotypic behaviors Self-injury behaviors Mental retardation The rightsholder did not Seizure disorders grant rights to reproduce this item in electronic Information from the American Psychiatric Associa- media. For the missing tion. Diagnostic and statistical manual of mental disorders. 4th ed. Washington, D.C.: American Psychi- item, see the original print atric Association, 1994:65-78. Copyright 1994. version of this publication.

ductive movements of hands and fingers, rocking, meaningless vocalizations), self-stim- ulation, self-injury behaviors, and seizures. Mental retardation is not a diagnostic crite- rion, but it is frequently present in the moder- ate to severe range.

Epidemiology In general, pervasive developmental disorders are estimated to occur at a rate of 63 per 10,000 persons.3 While the reported incidence of autistic disorder ranges from about five per 10,0004 to 20 per 10,000 persons,5 a recent meta-analysis reports the median rate for 11 surveys conducted since 1989 to be seven per 10,000 persons.6 Male-to-female ratios vary with IQ scores from 2:1 in severely handicapped persons to 4:1 in moderately handicapped persons.7 [Evidence level B, non- randomized studies] The occurrence rate in siblings is suspected to be from 3 to 7 percent, representing a 50- to 100-fold increase in risk.8

Etiology No single cause has been identified for the development of autism. Genetic origins are suggested by studies of twins and a higher

1668 AMERICAN FAMILY PHYSICIAN www.aafp.org/afp VOLUME 66, NUMBER 9 / NOVEMBER 1, 2002 Autism

incidence of recurrence among siblings.9 In routine part of the well-child examination can addition, an increased frequency of occur- enhance recognition of developmental disor- rence is noted in patients with genetic condi- ders. While the Denver screening tools19 have tions such as fragile X syndrome and tuberous historically been used in primary care settings sclerosis.10 Some reports have suggested a pos- for routine developmental surveillance, they sible association with Down syndrome.11 lack the sensitivity and specificity necessary for In addition to the implication of neuro- use as screening tools for developmental disor- transmitters, such as serotonin, in the devel- ders.20 More specific and sensitive screening opment and expression of autism,12 many surveillance tools, such as the Parents’ Evalua- other disorders may result in brain dysfunc- tion of Developmental Status (PEDS),21 are tion. Possible contributing factors in the available for assessing these conditions. development of autism include infections, Screening tools that are specific for autism errors in metabolism, immunology, lead poi- include the Checklist for Autism in Toddlers soning, and fetal alcohol syndrome.13 Concerns have been raised in recent years that immunizations, particularly measles, TABLE 3 mumps, and rubella (MMR) vaccine, may Resources for Management of Autism precipitate autism. In addition to reports from several parents who first detected autism in their children following an MMR Centers for Disease Control and Prevention National Immunization Program vaccination at 12 to 15 months of age, an Web address: www.cdc.gov/nip anecdotal study14 reported similar suspicions National Institutes of Health, National Institute of on the part of physicians who provided care Child Health and Human Development for 12 autistic patients. Subsequent studies in Web address: www.nichd.nih.gov 15,16 the United Kingdom [reference 16, Evi- Parents’ Evaluation of Developmental Status (PEDS), dence level B: epidemiologic study] and the Ellsworth & Vandermeer Press United States17 [Evidence level B: epidemio- Telephone: 615-226-4460 logic study] have failed to show an association Web address: www.pedstest.com/test/peds_ between any vaccine and the development of manual.html autism. Information about ongoing studies Checklist for Autism in Toddlers (CHAT) being conducted by the Centers for Disease Web address: www.nas.org.uk/profess/chat.html Control and Prevention and the National Pervasive Developmental Disorders Screening Institutes of Health (NIH) is available at their Test-Stage I (PDDST), Porter Psychiatric Institute Web sites (Table 3). Telephone: 415-476-7385 Association of University Centers on Disabilities, Recognition and Screening a listing of professionals by state Telephone: 301-588-8252 Indications for formal developmental eval- Web address: www.aucd.org uation include no babbling, pointing, or other The National Institute of Child Health and Human gestures by 12 months of age, no single words Development, a listing of research centers by 16 months of age, no two-word sponta- investigating treatment strategies for autism neous phrases by 24 months of age, and loss of Web address: www.nichd.nih.gov previously learned language or social skills at Autism Society of America any age.18 Parental concerns about delayed Telephone: 800-3AUTISM (800-328-8476) speech and language development, typically Web address: www.autism-society.org noticed at about 18 months of age, should Center for the Study of Autism always be taken seriously. Web address:www.autism.org Including developmental surveillance as a

NOVEMBER 1, 2002 / VOLUME 66, NUMBER 9 www.aafp.org/afp AMERICAN FAMILY PHYSICIAN 1669 tise in the diagnosis of autism include the Uni- Indications for formal developmental evaluation for autism versity Affiliated Program system and the include no babbling or pointing gestures by 12 months of age. National Institute of Child Health and Human Development (Table 3). Once an autistic disorder is suspected, cer- tain medical evaluations should be performed. (CHAT)22 (Table 423) and the Pervasive Devel- A family history of limited cognitive abilities or opmental Disorders Screening Test-Stage I the presence of dysmorphic features may sug- (PDDST).24 For a comprehensive review of gest the need for genetic evaluation. Wood’s available screening tools, the authors recom- light examination of the skin should be per- mend an article by Filipek and colleagues.18 formed to help identify the depigmented mac- When an autistic disorder is suspected, ules of tuberous sclerosis. Lead screening and referral should be made for further develop- metabolic testing should be considered if there mental evaluation and cognitive testing. is a history of lethargy, cyclic vomiting, early Although there is currently no cure for seizures, dysmorphic features, or mental retar- autism, early diagnosis and initiation of struc- dation. Electrophysiologic testing such as elec- tured multidisciplinary intervention can sig- troencephalography and central nervous sys- nificantly enhance functioning in later life.23 tem imaging studies are warranted to evaluate Experienced clinicians can reliably diagnose neurologic features that cannot be explained autism in children younger than three years by the diagnosis of autism alone.18 Because and, frequently, as young as two years. deafness or profound hearing loss can cause Presently no biologic markers are available to symptoms mimicking autism, a formal hear- identify patients with autistic disorders. Useful ing evaluation should be given if the diagnosis resources for identifying clinicians with exper- of autism is being considered. Autistic disorders should be distinguished from other psychiatric and pervasive developmental disorders. Table 5 1 lists a differential TABLE 4 diagnosis for several similar disorders. The Five Key Items on the CHAT Screen Clinical Course Ask the parent: The typical presenting symptoms of autistic Does your child ever pretend (for example, to make a cup of tea using a toy cup and teapot) or pretend with other things? disorder are delayed speech or challenging 7 Does your child ever use an index finger to point, to indicate interest in something? behavior before three years of age. Although Health practitioner observation: parents frequently see these signs and suspect Gain child’s attention, then point across the room at an interesting object and that something is wrong with their child by say “Oh look! There’s a (name of toy)!” Watch child’s face. Does the child 18 months of age, a diagnosis of autism is fre- look across to see what you are pointing at? quently delayed by two to three years because Gain child’s attention, then give child a toy cup and teapot and say “Can you of reluctance on the part of clinicians and make me a cup of tea?” Does the child pretend to pour out tea, drink it, etc.? families to incorrectly label a child as autis- Say to the child “Where’s the light?” or “Show me the light.” Does the child 25 point with an index finger at the light? To record “yes” on this item, the child tic. Seventy-five percent of autistic persons must have looked up at your face around the time of pointing. have some level of mental retardation.1 Devel- opmental gains in childhood and adolescence CHAT = Checklist for Autism in Toddlers. are common, but some persons have behav- Reprinted with permission from Baird G, Charman T, Cox A, Baron-Cohen S, ioral regression during adolescence. Swettenham I, Wheelwright S, et al. Current topic: screening and surveillance for Low IQ scores and failure to develop com- autism and pervasive and developmental disorders. Arch Dis Child 2001;84:471. municative language by five years of age cor- relate positively with a poor prognosis for

1670 AMERICAN FAMILY PHYSICIAN www.aafp.org/afp VOLUME 66, NUMBER 9 / NOVEMBER 1, 2002 Autism

response to treatment.1 About one third of autistic persons can achieve some degree The goals of treatment for patients with autism are to of independent living,1 although fewer than improve language and social skills, decrease problem behav- 5 percent go on to become self-sufficient iors, support parents, and foster independence. adults.13 Development of stereotypic behavior, self-injury behavior, and selective attention toward distracting stimuli (e.g., a ticking clock) markedly interfere with structured critical. Public Law 99-457 and the Individuals learning and working environments.13 with Disabilities Education Act29 mandate Many autistic persons develop seizures in referral to the special services departments of their first year of life in the form of infantile local preschool or school systems. spasms, a particularly severe form of seizure Because no treatment protocol meets the that is difficult to treat. There is also a signifi- needs of every autistic child, it is helpful to get cant incidence of first occurrence of seizures suggestions from a variety of sources. Organi- during adolescence,26 and as many as 35 per- zations available to help families and educa- cent may develop seizures by adulthood.18 tors are listed in Table 3. Comorbid anxiety is common,27 as are depression and obsessional behavior.28 NEUROPHARMACOLOGIC TREATMENT Primary care physicians are commonly Management of Autism asked to address the stereotypic or disruptive and Comorbid Conditions behaviors of autistic patients. While numer- The general goals of treatment for autistic ous medications have been used to treat autis- patients are to improve language and social tic symptoms, no single medication has been skills, decrease problem behaviors, support shown to be universally effective. Historically, parents and families in their adjustment to psychotropic medications have been reserved and education of autistic children, and foster for use in situations where all attempts at independence. Because autistic children who behavior management have failed, and the begin treatment at a young age have signifi- patients are considered to be harmful to cantly better outcomes,24 early intervention is themselves or others. While use of behavior modification programs is often the primary method of manag- TABLE 5 ing challenging behaviors in autistic children, Differential Diagnosis of Autism supportive medication use has been found to help reduce behavior problems. Obtaining a Other pervasive developmental disorders correct diagnosis is important before initiating Rett’s syndrome any pharmacologic intervention. For example, Childhood disintegrative disorder attention deficit with or without hyperactivity Asperger’s disorder can coexist with autism and may possibly be Disorders of infancy, childhood, and adolescence managed with the use of methylphenidate Selective mutism (Ritalin)30 [Evidence level C: consensus opin- Stereotypic movement disorder 31 Schizophrenia with childhood onset ion] or clonidine (Catapres). [Evidence level C: consensus opinion] It can be difficult to dis- Information from the American Psychiatric Associa- tinguish between the behaviors associated with tion. Diagnostic and statistical manual of mental dis- autism; attention-deficit/hyperactivity disor- orders. 4th ed. Washington, D.C.: American Psychi- der; and mania, and an appropriate treatment atric Association, 1994:65-78. Copyright 1994. for one disorder may be ineffective or exacer- bate the symptoms of another.

NOVEMBER 1, 2002 / VOLUME 66, NUMBER 9 www.aafp.org/afp AMERICAN FAMILY PHYSICIAN 1671 When behavior management programs or is augmented communication, a method the use of supportive medications are unsuc- whereby nonverbal persons are assisted in cessful in correcting potentially dangerous communication by means of a letter board or behavior, the use of sedating medications may a computer keyboard. When a facilitator helps be necessary for brief periods while less invasive the person choose letters, words, etc., the interventions are attempted. Sedative-hypnotics process is referred to as facilitated communi- and neuroleptics such as buspirone (BuSpar), in cation. While augmented communication a dosage of 5 to 20 mg two to three times a day32 devices have markedly improved communica- [Evidence level C: consensus opinion], or tion potential in some patients, numerous risperidone (Risperdal), in a dosage of 0.5 to 2 controlled studies have failed to show that mg twice a day33 are commonly used for this facilitated communication is reliably useful.13 purpose. Benzodiazepines should be used with Another treatment currently being advo- caution because they can cause disinhibition, cated is auditory integration training (AIT), resulting in more excitable behavior.32 whereby persons with autism typically spend Objective data collection and outcome mon- 10 hours during a two-week period listening itoring are important because of the variable to music that has been computer-modified to nature of individual responses to medication. remove sensitive frequencies and reduce pre- Information should be collected by persons dictable patterns. AIT is said to improve audi- who have regular contact with the patient— tory processing capabilities by correcting dis- family members, support personnel in day care tortions in hearing and by conditioning and residential programs, case managers, and patients to focus their attention more appro- physicians. Given the multiple developmental, priately. Unfortunately, this technique also has behavior, and medical problems associated little supporting scientific documentation.35 with autism, coordination of services by a mul- Another popular behavior-based interven- tidisciplinary team is highly recommended.34 tion is the Lovaas program,36 sometimes [Evidence level C: consensus opinion] referred to as DTT because of its use of positive reinforcement through a series of inten- ALTERNATIVE THERAPIES sive discrete trial training sessions. While ini- A number of methods for teaching com- tial reports suggested a 47 percent recovery munication and socialization skills have been rate from autism when preschoolers were developed over the years. One recent example treated,36 subsequent studies have been unable to document long-term gains.37 Studies using similar behavior interventions, however, have The Authors been able to document short-term improve- ments.37 Verification of long-term success CHRISTOPHER D. PRATER, M.D., is assistant clinical professor in the Department of Family Medicine at the University of Tennessee College of Medicine, Chattanooga becomes important in view of the cost of such Unit. He also serves as medical director for the Orange Grove Center, a treatment facil- intensive treatment programs, which typically ity in Chattanooga for developmentally delayed children and adults. Dr. Prater received require extensive one-on-one training with his medical degree from the University of Tennessee College of Medicine, Memphis, and completed a family practice residency at the University of Tennessee in Knoxville. autistic children for 40 hours a week for a minimum of two years—a cost of approxi- ROBERT G. ZYLSTRA, ED.D., L.C.S.W., is assistant professor and director of behavioral 38 science in the Department of Family Medicine at the University of Tennessee College mately $40,000 a year. Controversies about of Medicine, Chattanooga Unit. Dr. Zylstra earned a master’s degree in social work at fiscal responsibility are ongoing because some the University of Michigan, Ann Arbor, and a doctorate in education at the University parents feel local school systems should make of Memphis. this level of care available for their children.38 Address correspondence to Robert G. Zylstra, Ed.D., L.C.S.W., Department of Family Other interventional strategies include deep Medicine, Chattanooga Unit, University of Tennessee College of Medicine, 1100 E. Third St., Chattanooga, TN 37403 (e-mail: [email protected]). Reprints are not pressure therapy, nutritional supplements, available from the authors. and specialty diets.39-42 [reference 40, Evidence

1672 AMERICAN FAMILY PHYSICIAN www.aafp.org/afp VOLUME 66, NUMBER 9 / NOVEMBER 1, 2002 Autism

REFERENCES TABLE 6 1. American Psychiatric Association. Diagnostic and Behavior Modification statistical manual of mental disorders, 4th ed. in the Management of Autism Washington, DC: American Psychiatric Association, 1994:65-78. 2. World Health Organization. ICD-10: international Structuring the environment statistical classification of diseases and related Providing consistent responses to behaviors health problems. Geneva: World Health Organiza- tion, 1992. Positive reinforcement—rewarding a desired behavior 3. Chakrabarti S, Fombonne E. Pervasive develop- Negative reinforcement—not rewarding an mental disorders in preschool children. JAMA undesirable behavior 2001;285:3093-9. Punishment—application of an adverse stimulus to 4. Fombonne E, Du Mazaubrun C, Cans C, Grandjean deter an unwanted response H. Autism and associated medical disorders in a French epidemiological survey. J Am Acad Child Shaping—reinforcing closer and closer Adolesc Psychiatry 1997;36:1561-9. approximations to the desired behavior 5. Bryson SE. Epidemiology of autism. J Autism Dev Disord 1996;26:165-7. Information from Farber JM. Autism and other com- 6. Fombonne E. The epidemiology of autism: a re- munication disorders. In: Capute AJ, Accardo PJ, eds. view. Psychol Med 1999;29:769-86. 7. Bryson SE. Epidemiology of autism: Overview and Developmental disabilities in infancy and childhood. issues outstanding. In: Cohen DJ, Volkmar FR, eds. 2d ed. Baltimore, Md.: Brookes, 1996:347-64. Handbook of autism and pervasive developmental disorders. 2d ed. New York: Wiley, 1997:41-6. 8. Rutter M, Bailey A, Simonoff E, Pickles A. Genetic influences and autism. In: Cohen DJ, Volkmar FR, level A: randomized controlled trial (RCT); eds. Handbook of autism and pervasive developmental disorders. 2d ed. New York: Wiley, 1997: reference 42, Evidence level C: consensus 370-87. opinion] Anecdotal reports of success with 9. Szatmari P, Jones MB, Zwaigenbaum L, MacLean alternative or complementary interventions JE. Genetics of autism: overview and new direc- tions. J Autism Dev Disord 1998;28:351-68. are common, but efficacy in most cases 10. Baker P, Piven J, Sato Y. Autism and tuberous scle- remains clinically unproven. Generally speak- rosis complex: prevalence and clinical features. ing, most successful programs have several J Autism Dev Disord 1998;28:279-85. 11. Howlin P, Wing L, Gould J. The recognition of common components: (1) recognition of the autism in children with Down syndrome—implica- importance of early identification and inter- tions for intervention and some speculations about vention; (2) use of behavior-oriented strate- pathology. Dev Med Child Neurol 1995;37:406-14. 13 12. Anderson GM, Hoshino Y. Neurochemical studies gies (Table 6 ); (3) development of social of autism. In: Cohen DJ, Volkmar FR, eds. Hand- communication; and (4) active involvement book of autism and pervasive developmental disor- of parents and families. ders. 2d ed. New York: Wiley, 1997:325-43. 13. Farber JM. Autism and other communication disor- Recently, there has been discussion about a ders. In: Capute AJ, Accardo PJ, eds. Developmen- possible role of the gastric hormone secretin as tal disabilities in infancy and childhood. 2d ed. Bal- a pharmacologic intervention in the treatment timore, Md.: Brookes, 1996:347-64. 14. Wakefield AJ, Murch SH, Anthony A, Linnell J, Cas- of autism. This information was based on one son DM, Malik M, et al. Ileal-lymphoid-nodular study of three autistic persons.43 Unfortu- hyperplasia, non-specific colitis, and pervasive nately, a subsequent study involving 56 autistic developmental disorder in children. Lancet 1998; 44 351:637-41. persons failed to support the initial findings. 15. Miller D, Wadsworth J, Diamond J, Ross E. Measles [Evidence level B: lower quality RCT] At pre- vaccination and neurological events. Lancet 1997; sent, most investigators do not see a role for 349:730-1. 16. Taylor B, Miller E, Farrington CP, Petropoulos MC, secretin in the treatment of autism, an opinion Favot-Mayaud I, Li J, et al. Autism and measles, supported by ongoing research at the NIH. mumps, and rubella vaccine: no epidemiological evidence for a causal association. Lancet 1999;353:2026-9. The authors indicate that they do not have any con- 17. DeStefano F, Chen RT. Autism and measles, flicts of interest. Sources of funding: none reported. mumps, and rubella vaccine: no epidemiological

NOVEMBER 1, 2002 / VOLUME 66, NUMBER 9 www.aafp.org/afp AMERICAN FAMILY PHYSICIAN 1673 Autism

evidence for a causal association. J Pediatr 2000; and developmental disabilities: the international 136:125-6. consensus handbook. Columbus, Ohio: Ohio State 18. Filipek PA, Accardo PJ, Baranek GT, Cook EH Jr, University Nisonger Center, 1998:259-69. Dawson G, Gordon B, et al. The screening and 32. Werry JS. Anxiolytics and sedatives. In: Reiss S, diagnosis of autistic spectrum disorders. J Autism Aman MG, eds. Psychotropic medications and Dev Disord 1999;29:439-84. developmental disabilities: the international con- 19. Frankenburg WK, Dodds J, Archer P, Shapiro H, sensus handbook. Columbus, Ohio: Ohio State Bresnick B. The Denver II: a major revision and University Nisonger Center, 1998:201-14. restandardization of the Denver Developmental 33. Demb HB, Espiritu C. The use of risperidone with Screening Test. Pediatrics 1992;89:91-7. children and adolescents with developmental dis- 20. Sciarillo WG, Brown MM, Robinson NM, Bennett abilities: effects on some common challenging FC, Sells CJ. Effectiveness of the Denver Develop- behaviors. Ment Health Aspects Dev Disabil 1999; mental Screening Test with biologically vulnerable 2:73-82. infants. J Dev Behav Pediatr 1986;7:77-83. 34. Volkmar F, Cook EH Jr, Pomeroy J, Realmuto G, 21. Glascoe FP. Collaborating with parents: using par- Tanguay P. Practice parameters for the assessment ents’ evaluation of developmental status to detect and treatment of children, adolescents, and adults and address developmental and behavioral prob- with autism and other pervasive developmental lems. Nashville: Ellsworth & Vandermeer, 1998. disorders. J Am Acad Child Adolesc Psychiatry 22. Baron-Cohen S, Allen J, Gillberg C. Can autism be 1999;38(12 suppl):32S-54S. detected at 18 months? The needle, the haystack, 35. Rimland B, Edelson SM. Auditory integration train- and the CHAT. Br J Psychiatry 1992;161:839-43. ing in autism: a pilot study. Autism Research Insti- 23. Baird G, Charman T, Cox A, Baron-Cohen S, Swet- tute 1992;112:1-7. tenham J, Wheelwright S, et al. Current topic: 36. Lovaas OI, Smith T. A comprehensive behavioral screening and surveillance for autism and pervasive theory of autistic children: paradigm for research developmental disorders. Arch Dis Child 2001; and treatment. J Behav Ther Exp Psychiatry 1989; 84:468-75. 20:17-29. 24. Siegel B. Early screening and diagnosis in autism 37. Sheinkopf SJ, Siegel B. Home-based behavioral spectrum disorders: the pervasive developmental treatment of young children with autism. J Autism disorders screening test (PDDST). NIH State of the Dev Disord 1998;28:15-23. Science in Autism: Screening and Diagnosis Work- 38. Mauk JE, Reber M, Batshaw ML. Autism and other ing Conference, Bethesda, Md., June 15-17, 1998. pervasive developmental disorders. In: Batshaw 25. Filipek PA, Accardo PJ, Ashwal S, Baranek GT, Cook ML, ed. Children with disabilities. 4th ed. Balti- EH Jr, Dawson G, et al. Practice parameter: screen- more, Md.: Brookes, 1997:425-47. ing and diagnosis of autism: report of the Quality 39. Grandin T. Thinking in pictures: and other reports Standards Subcommittee of the American Acad- from my life with autism. New York: Vintage emy of Neurology and the Child Neurology Society. Books, 1996:62-81. Neurology 2000;55:468-79. 40. Rimland B, Callaway E, Dreyfus P. The effect of

26. Gillberg C, Steffenburg S. Outcome and prognos- high doses of vitamin B6 on autistic children: a tic factors in infantile autism and similar conditions: double-blind crossover study. Am J Psychiatry 1978; a population-based study of 46 cases followed 135: 472-5.

through puberty. J Autism Dev Disord 1987;17: 41. Rimland B. Megavitamin B6 and magnesium in the 273-87. treatment of autistic children and adults. In: 27. Muris P, Steerneman P, Merckelbach H, Holdrinet I, Schopler E, Mesibov GB, eds. Neurobiological issues Meesters C. Comorbid anxiety symptoms in chil- in autism. New York: Plenum, 1987:389-405. dren with pervasive developmental disorders. J Anx- 42. Singh NN, Ellis CR, Mulick JA, Poling A. Vitamin, iety Disord 1998;12:387-93. mineral and dietary treatments. In: Reiss S, Aman 28. Kobayashi R, Murata T. Behavioral characteristics of MG, eds. Psychotropic medications and develop- 187 young adults with autism. Psychiatry Clin Neu- mental disabilities: the international consensus rosci 1998;52:383-90. handbook. Columbus, Ohio: Ohio State University 29. Individuals with Disabilities Education Act–IDEA Nisonger Center, 1997:311-20. ’97. Public Law 1997;105-17. 43. Horvath K, Stefanatos G, Sokolski KN, Wachtel R, 30. Arnold LE, Gadow K, Pearson D, Varley C. Stimu- Nabors L, Tildon JT. Improved social and language lants. In: Reiss S, Aman MG, eds. Psychotropic skills after secretin administration in patients with medications and developmental disabilities: the autistic spectrum disorders. J Assoc Acad Minor international consensus handbook. Columbus, Phys 1998;9:9-15. Ohio: Ohio State University Nisonger Center, 1998: 44. Chez MG, Buchanan CP, Bagan BT, Hammer MS, 229-43. McCarthy KS, Ovrutskaya I, et al. Secretin and 31. Hagerman R, Bregman J, Tirosh E. Clonidine. In: autism: a two-part clinical investigation. J Autism Reiss S, Aman MG, eds. Psychotropic medications Dev Disord 2000;30:87-94.

1674 AMERICAN FAMILY PHYSICIAN www.aafp.org/afp VOLUME 66, NUMBER 9 / NOVEMBER 1, 2002