Pediatric Annals

Epidemiology and Etiology of Autistic Spectrum Disorders Difficult to Determine

Daniel L Coury, MD; Patricia L Nash, MD

Abstract

Physicians tend to remember the first time they saw a patient with a diagnosis of autism. The unusual social behaviors, severely affected communication skills, and other commonly seen abnormalities in behavior are striking, and often leave an indelible image.

Controversy surrounds reports of the incidence and prevalence of autism and autistic spectrum disorders (ASD). For those professionals with mental images of the more severe end of the spectrum, new criteria and categories for diagnosis can be difficult to unravel. Even more confusing is the concern regarding an apparent increase in the prevalence of the condition, often attributed to a variety of possible causes. A review of the available evidence on these topics is essential not only for managing individuals with autism and ASD but also for addressing the questions and concerns of their families.

INCIDENCE AND PREVALENCE OF AUTISTIC DISORDERS

Since Kanner's early descriptions of autism in the 1940s, nearly 40 epidemiological studies have focused on autism and ASD. Most were conducted from the mid-1960s to the late 1980s and indicated autism was a rare condition with a prevalence of approximately two to four per 10,000 children.

However, a change in the apparent prevalence surfaced in the late 1980s, when a Japanese report indicated a rate of 15.5 per 10,000 children.1 Additional studies conducted in the late 1990s found rates ranging from 7.2 to 60 per 10,000 children in European countries and Japan.2"4 Another study, in Brick Township, New Jersey, found a rate of 40 cases of autistic disorder per 10,000 children.5 A more recent study, from the Centers for Disease Control and Prevention, found a rate of 34 per 10,000 children in the Atlanta metropolitan area.6

What could possibly be responsible for this marked increase in identified cases of autism? Several possible explanations exist, including changes in diagnostic criteria, variations in case finding and identification between different epidemiologic studies, greater public and professional awareness, wider availability of treatment services, and increases in the actual incidence of the disorders.

Changes in Diagnostic Criteria

Kanner's criteria were used in many studies until the mid-1980s. The third edition of the widely accepted psychiatric manual Diagnostic and Statistical Manual of Mental Disorders, published in 1980, introduced the phrase "pervasive developmental disorders" (PDD). This group of disorders were further classified as "infantile autism," the most similar disorder to Kanner's original description, and "childhood onset pervasive developmental disorder," classified as onset after an age of 30 months. A revision to the manual was published in 1987; the current edition, DSM-IV, has been in use since 1994.

DSM-IV categorizes the pervasive developmental disorders as autistic disorder, Rett's disorder, childhood disintegrative disorder, Asperger's disorder, and pervasive developmental disorder - not otherwise specified (PDD-NOS). Asperger's disorder was a new addition to the PDD family in 1994, although it had been described and identified in Europe for many years. At present, PDD-NOS is the most widely used diagnosis in the autistic spectrum, and, along with Asperger's, signals an increase in the identification of milder forms of the pervasive developmental disorders. (Note: In Europe, the International Classification of Diseases (ICD) terminology for these disorders has been popular, with the ICD-10 version used in several studies.)

It is important to consider these variations in diagnostic criteria and terms when reviewing published studies on prevalence, because changes in prevalence over time are difficult to interpret when the diagnostic criteria seems to be changing. In addition, the increased use of the newest categories has led to identification of individuals who would not have been diagnosed before 1994.

Study Methods

The method of case finding used in various epidemiologic studies can affect prevalence reports.…

Physicians tend to remember the first time they saw a patient with a diagnosis of autism. The unusual social behaviors, severely affected communication skills, and other commonly seen abnormalities in behavior are striking, and often leave an indelible image.

Controversy surrounds reports of the incidence and prevalence of autism and autistic spectrum disorders (ASD). For those professionals with mental images of the more severe end of the spectrum, new criteria and categories for diagnosis can be difficult to unravel. Even more confusing is the concern regarding an apparent increase in the prevalence of the condition, often attributed to a variety of possible causes. A review of the available evidence on these topics is essential not only for managing individuals with autism and ASD but also for addressing the questions and concerns of their families.

INCIDENCE AND PREVALENCE OF AUTISTIC DISORDERS

Since Kanner's early descriptions of autism in the 1940s, nearly 40 epidemiological studies have focused on autism and ASD. Most were conducted from the mid-1960s to the late 1980s and indicated autism was a rare condition with a prevalence of approximately two to four per 10,000 children.

However, a change in the apparent prevalence surfaced in the late 1980s, when a Japanese report indicated a rate of 15.5 per 10,000 children.1 Additional studies conducted in the late 1990s found rates ranging from 7.2 to 60 per 10,000 children in European countries and Japan.2"4 Another study, in Brick Township, New Jersey, found a rate of 40 cases of autistic disorder per 10,000 children.5 A more recent study, from the Centers for Disease Control and Prevention, found a rate of 34 per 10,000 children in the Atlanta metropolitan area.6

What could possibly be responsible for this marked increase in identified cases of autism? Several possible explanations exist, including changes in diagnostic criteria, variations in case finding and identification between different epidemiologic studies, greater public and professional awareness, wider availability of treatment services, and increases in the actual incidence of the disorders.

Changes in Diagnostic Criteria

Kanner's criteria were used in many studies until the mid-1980s. The third edition of the widely accepted psychiatric manual Diagnostic and Statistical Manual of Mental Disorders, published in 1980, introduced the phrase "pervasive developmental disorders" (PDD). This group of disorders were further classified as "infantile autism," the most similar disorder to Kanner's original description, and "childhood onset pervasive developmental disorder," classified as onset after an age of 30 months. A revision to the manual was published in 1987; the current edition, DSM-IV, has been in use since 1994.

DSM-IV categorizes the pervasive developmental disorders as autistic disorder, Rett's disorder, childhood disintegrative disorder, Asperger's disorder, and pervasive developmental disorder - not otherwise specified (PDD-NOS). Asperger's disorder was a new addition to the PDD family in 1994, although it had been described and identified in Europe for many years. At present, PDD-NOS is the most widely used diagnosis in the autistic spectrum, and, along with Asperger's, signals an increase in the identification of milder forms of the pervasive developmental disorders. (Note: In Europe, the International Classification of Diseases (ICD) terminology for these disorders has been popular, with the ICD-10 version used in several studies.)

It is important to consider these variations in diagnostic criteria and terms when reviewing published studies on prevalence, because changes in prevalence over time are difficult to interpret when the diagnostic criteria seems to be changing. In addition, the increased use of the newest categories has led to identification of individuals who would not have been diagnosed before 1994.

Study Methods

The method of case finding used in various epidemiologic studies can affect prevalence reports. Studies that rely on a single source to identify cases provide less reliable data than those examining multiple sources that allow for more assurance of meeting diagnostic criteria and not missing cases. For example, the Atlanta study reviewed records from multiple medical and educational sources, and then had case status determined by expert review.6 Some early studies used records of children known to state agencies as receiving services but who had not necessarily received a diagnosis by a qualified professional.

Other recent studies may have overlooked other factors, such as changes in diagnostic practices and the effect of availability of services.7 A California study primarily used referral statistics and showed an increase in the number of children receiving public services for a diagnosis of autism, suggesting an increasing prevalence of ASD. However, a re-analysis of the dataset from 1987 to 1998 showed that diagnostic substitution had most likely occurred.8

Public Awareness and Availability of Services

While the prevalence of autism showed an increase from 5.8 to 14.9 per 10,000 children from 1987 to 1998, the prevalence for mental retardation decreased from 28.8 to 19.5 per 10,000 children. The changes in each prevalence rate essentially cancelled out the other. Why would this happen? Perhaps it had to do with increased availability of services.

In 1991, the federal Individuals with Disabilities Education Act (IDEA) included autism as a covered condition. The availability of effective early-intervention services, especially behavioral interventions, for those children meeting referral criteria may very well have played a role in the increasing use of the diagnosis of autism.

Increasing awareness of autistic spectrum disorders by parents and professionals is also at work. Parents often approach pediatricians with developmental and behavioral concerns. With increased public awareness of autism comes an increase in questions regarding the possibility that a child might have the condition or a variant. Increased professional awareness results in increased case finding and diagnosis. Patients who previously might have been identified as merely "odd," or as having attention-deficit/hyperactivity disorder (ADHD) with severe social skill deficiency or mental retardation with unusual behaviors, might be correctly identified as having an ASD.

Better identification of autism among persons with other disorders, such as mental retardation, is likely part of the rising prevalence. All of these new identifications may contribute to an apparent increase. This increase is actually a more accurate representation of what has always been there, as has been seen with other conditions.

A recent study suggests that the increase in autism incidence is occurring in the context of an increase in all developmental disorders, suggesting an increase in awareness of all developmental disorders during the past 2 decades.9

Increase in Prevalence

The apparent increase in prevalence of ASDs may also be due to a true increase in the disorder. This is the most popular theory among the public and is related to the multiple proposed etiologies of autism and ASDs. These theories often attribute increased prevalence to the increased use of a variety of medications or medical treatments, or exposure to other environmental agents. The theory that has received most publicity over the past few years has been the proposed association with administration of vaccines, specifically the measles, mumps, rubella (MMR) vaccine.

A 1998 case series reported by Wakefield et al.10 suggested a relationship between gastrointestinal problems and PDD. Twelve children with a history of gastrointestinal problems and developmental regression were evaluated and reported to have a variety of abnormal biopsy findings. History provided by the families indicated normal development followed by the onset of developmental regression, abdominal pain, and diarrhea. In eight of the subjects, parents reported the onset of these symptoms began following the administration of the MMR vaccine. Onset was reported in one child following measles infection. All but one had intestinal biopsies that were described as abnormal.

The predominant endoscopic findings were lymphoid nodular hyperplasia; the most common histological findings were acute and chronic non-specific colitis and reactive lymphoid hyperplasia. Based largely on the parents' reports of onset of both behavioral and gastrointestinal symptoms following administration of the MMR vaccine, the authors reported a possible relationship among PDDs, gastrointestinal abnormalities, and the vaccine. In particular, the authors suggested MMR vaccine may damage the bowel, resulting in altered absorption of nutrients, vitamins, and minerals. The altered absorption eventually has an effect on neurodevelopment that leads to manifestations of autism. A later study by the same group reported a relationship between the measles virus and autism, describing the presence of measles virus genome detected in intestinal biopsies.11

The Wakefield study suffers from several flaws, many of them similar to the problems seen in trying to ascertain the changes in prevalence of the autistic disorders. The report is a case series of children referred for gastroenterologie consultation, indicating selection bias. Many but not all individuals with autism have gastrointestinal (GI) symptoms, suggesting that the hypothesis does not apply to all cases. In addition, many of the parents in the report indicated onset of behavioral symptoms prior to GI symptoms, apparently refuting the proposed mechanism of this disorder. Finally, the association with administration of MMR vaccine was based on parental recall and not on objective data. In all cases, the reports of behavioral symptoms began at 2 years of age, the age range when signs of autism classically occur, known since long before the availability of MMR vaccine. Temporal association is not proof of causation; it is only evidence of temporal association. Further evaluation is needed to determine causation and the validity of the hypothesis.

Subsequent studies have failed to show an increased risk of GI symptoms or inflammatory bowel disease in children following administration of the MMR vaccine, refuting the theory of bowel damage.1213 However, given a supposition that a small portion of the population may be susceptible, perhaps due to some genetic predisposition, this small population with GI symptoms might be missed in a large study examining overall gastrointestinal symptomatology, but the theory could still be valid. This would require examining the relationship of the MMR vaccine to the diagnosis of autism.

Several studies have reviewed the possibility of a link between the introduction of the MMR vaccine and the incidence of autism. All have failed to show a relationship. In Britain, Taylor et al.14 were unable to find an increase in incidence of autism in those individuals who received the MMR vaccine. Other studies have also failed to find evidence to support an increase in the incidence of autism related to use of the MMR vaccine. A study conducted in California reviewed the percentage of children receiving the vaccine between 1980 and 1994 and compared this to the incidence of autism during that period.15 While the authors noted a clear increase in the prevalence of autism (as discussed earlier in this article), the proportion of children receiving the MMR vaccine did not change.

A similar study was conducted in Britain. The incidence of autism between 1988 and 1993 was reviewed and compared with MMR immunization rates.16 As was seen in the California study, the rates of identification of autism increased but the MMR immunization rate remained the same. A study of autism prevalence in Sweden over the time period in which the MMR vaccine was introduced failed to show any increase in the rate of autism.17 The data regarding these and other studies seeking a relationship between MMR and autism have been reviewed by the Medical Research Council in the United Kingdom18, and by a panel of experts convened by the American Academy of Pediatrics and the Centers for Disease Control and Prevention (CDC) in the United States.19 These expert panels have failed to find a causal relationship between MMR vaccine and autism.

THIMEROSAL LEVELS

If the MMR component is not responsible for autism, could it be some other component of vaccines such as thimerosal? Thimerosal is an organic compound of ethyl mercury that has been used in multiple dose vials of vaccines as a preservative to inhibit bacterial and fungal contamination. It has been included in several vaccines since the 1930s, and continues to be present in some medicines and products such as throat and nose sprays and some contact lens solutions. Bernard et al. have promoted the hypothesis that autism is the result of mercury toxicity resulting from the thimerosal in vaccines.20 They have suggested that the increase in prevalence in autism, and the temporal onset of the disorder, reflects increasing use of vaccines in the first 6 months of life.

The CDC and the Institute of Medicine Immunization Safety Review Committee have reviewed available data and found no evidence supporting a causal relationship between thimerosal and autism.21 There was no increase in autism rates relative to the amount of thimerosal a child may have received in the first 6 months of life. A weak but statistically significant association was found between cumulative exposure to thimerosal and speech delay and ADHD, but a later independent panel of experts felt that study favored a connection where none may have existed.

Subsequent studies have not shown an association between thimerosal and ASD. Nonetheless, efforts to reduce mercury exposure are considered beneficial to all. For this reason, efforts to remove thimerosal and other mercurybased preservatives from vaccines are under way. Since 2001, all routinely recommended vaccines manufactured for administration to infants in the United States are either thimerosal-free or contain extremely small amounts of thimerosal. Many vaccines have never contained thimerosal, including inactivated polio (IPV) and acellular pertussis (DTaP).22

SEARCHING FOR A CAUSATIVE FACTOR

If MMR and thimerosal aren't responsible, what the causative factor for autism? The simple answer is that it is not known. Genetic factors appear to be the most important component. Researchers have long recognized traits of persons with autism are often seen in the parents. Twin studies indicate genetic factors are of major importance in more than 90% of cases.23

Data from numerous studies have implicated several genes, rather than a single gene, for ASD. The complex interplay of multiple genes and the environment has not been elucidated, but research in this area is likely to be fruitful during the next decade.

SUMMARY

The epidemiology of the autistic spectrum disorders is changing. A clear increase in prevalence has been noted during the past 2 decades. What is less clear is the cause for this increase. Multiple factors appear to be responsible. The preponderance of evidence suggests most of the rise in incidence and prevalence is related to changes in diagnostic criteria and greater awareness on the part of both professionals and parents. Proposed theories of causation, which also seek to explain the increase in prevalence, have not been substantiated. Further research is needed to better determine the incidence and prevalence of these disorders and their etiologic factors.

REFERENCES

1. Maisuishi T, Shiotsuki Y, Yoshimura K, et al. High prevalence of infantile autism in Kurume City. Japanese J Child Neurol. 1987;2:268-271.

2. Webb E, Lobo S, Hervas A, et al. The changing prevalence of autistic disorder in a Welsh health district. Dev Med Child Neurol. 1997;39(3): 150-153.

3. Kadesjo B, Gillberg C, Hagberg B. Brief report: autism and Asperger syndrome in seven-yearold children: a total population study. J Autism Dev Disorder. 1999;29(4):327-332.

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 (6):694702.

5. Bertrand J, Mars A, Boyle C, et al. Prevalence of autism in a United States population: the Brick Township, New Jersey, investigation. Pediatrics. 2001;108(5):1155-1161.

6. Yeargin-Allsopp M, Rice C, Karapurkar T, et al. Prevalence of autism in a US metropolitan area. JAMA. 2003:289(1):49-55.

7. Department of Developmental Services. Changes in the population of persons with autism and pervasive developmental disorders in California's Developmental Services System: 1987 through 1998. Report to die Legislature March 1, 1999:1-9. 1999. Available at: htto://www.dds.ca.gov/autism/autism_main.c fm. Accessed August 27, 2003.

8. Croen LA, Gredier JK, Hoogstrate J, Selvin S. The changing prevalence of autism in California. J Autism Dev Disord. 2002;32(3):207-215.

9. Barbaresi WJ, Katusic SK, Colligan RC, et al. Explaining the apparent increase in die incidence of autism in Olmstead County, Minnesota, from 1976 to 1997. J Dev Behav Pediatr. 2002;23:(5)398-399.

10. Wakefield AJ, Murch SH, Anthony A, et al. Deal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children. Lancet. 1998;351(9103):637-641.

11. Uhlmann V, Martin CM, Shiels O, et al. Potential viral pathogenic mechanism for new variant inflammatory bowel disease. Mol Pathol. 2002;55(2):1-6.

12. Davis RL. Kramarz P, Böhlke K, et al. Measles-mumps-rubella and other measlescontaining vaccines do not increase the risk for inflammatory bowel disease: a case-control study from the Vaccine Safety Datalink project. Arch Pediatr Adolesc Med. 2001;155(3):354-359.

13. Morris DL, Montgomery SM, Thompson NP, et al. Measles vaccination and inflammatory bowel disease: a national British Cohort Study. Am J Gastroenterol. 2000;95(12):3507-3512.

14. Taylor B, Miller E, Lingam R, et al. Measles, mumps, and rubella vaccination and bowel problems or developmental regression in children with autism: population study. BMJ. 2002;324(7334):393-396.

15. Dales L, Hammer SJ, Smith NJ. Time trends in autism and in MMR immunization coverage in California. JAMA. 200 1;285(9): 11831185.

16. Kaye JA, del Mar Melero-Montes M, Jick H. Mumps, measles, and rubella vaccine and me incidence of autism recorded by general practitioners: a time trend analysis. BMJ. 2001;322(7284):460-463.

17. Gillberg C, Steffenburg S, Schaumann H. Is autism more common now than ten years ago? Br J Psychiatry. 1991;158:403-409.

18. Medical Research Council. Review of Autism Research: Epidemiology and Causes. London, England: Medical Research Council; 2001. Available at: http://www.mrc.ac.uk/prn/ pdfautism-reportpdf. Accessed August 20, 2003.

19. Halsey NA, Hyman SL; Omference Writing Panel. Measles-mumps-rubella vaccine and autistic spectrum disorder: report from die New Challenges in Childhood Immunizations Conference convened in Oak Brook, Illinois, June 12-13, 2000. Pediatrics. 2001;107(5):e84.

20. Bernard S, Enayati A, Redwood L, Roger H, Binstock T. Autism: A novel form of mercury poisoning. Med Hypothesis. 2001;56(4):462471.

21. Stratton K, Gable A, McCormick M, eds. Immunization Safety Review Committee, Board on Healm Promotion and Disease Prevention. Immunization Safety Review: Thimerosal-Containing Vaccines and Neurodevelopmental Disorders. Washington, DC: The National Academies Press; 2001.

22. Nelson KB, Bauman ML. Thimerosal and autism? Pediatrics. 2003;111:(3)674-678.

23. Rutter M. Genetic studies of autism: from the 1970s into the millennium. J Abnorm Child Psychol. 2000;28(1):3-14.

10.3928/0090-4481-20031001-11

Sign up to receive

Journal E-contents