Autism in the United States: a Perspective
F. Edward Yazbak, M.D., F.A.A.P.
Journal of American Physicians and Surgeons Volume 8 Number 4 WInter 2003
Once rare, autism has reached epidemic proportions in the United States. The increase cannot be attributed to changes in diagnostic criteria, which have actually become more restrictive. Already a heavy burden on educational facilities, the increasing number of patients afflicted with this serious disability will have an enormous effect on the economy as the affected children reach adulthood. Studies of all possible causes of the epidemic are urgently needed. To date, studies of a potential relationship to childhood vaccines have been limited and flawed.
The important historical observation about autism is that it was
unknown in ancient cultures, or even in medieval times, and that it just
“appeared” some 60 years ago.
Leo Kanner, while at Johns Hopkins, was first to describe autism in
1943. His article Autistic Disturbances of Affective Contact described 11
children who had an apparently rare syndrome of “extreme autistic
aloneness.”‘ Because these children’s symptoms started early, Kanner’s
Syndrome was also known as “infantile autism.” In 1944, Hans Asperger also
described a group of children with similar symptoms who were “highly
recognizable.”2 In the same year, Bruno Bettelheim theorized that children
developed autism because their “refrigerator mothers” raised them in a
non-stimulating environment, with resulting damage to their social, language
and general development. Bettelheim’s credentials were questionable,3 and
his theory has been discredited.
Bernard Rimland, Ph.D., founder of the Autism Society of America and
founding president of the Autism Research Institute (ARI), has thoroughly
analyzed the ARI database of more than 30,000 entries and reported two clear
trends: First, the incidence of autism has increased remarkably, becoming
“an explosion~~ in recent years4 and second, a distinct shift in the time of
onset of autistic symptoms has become evident. “Late onset autism (starting
in the second year) was almost unheard of in the 1950s, ’60s, and ’70s;
today such cases outnumber early onset cases five to one.”5 Parents in
increasing numbers are reporting similar stories. A child, most often a boy
who is developmentally, socially, and verbally on par for his age, suddenly
stops acquiring new words and skills in the second year of life and then
regresses, losing speech, cognitive abilities, and social dexterity.
Children in this group are said to have regressive autism. Further,
overwhelmed parents may drift apart, and siblings’ stress may be manifested
as behavior problems.’ Suggesting that a sudden and exponential increase in
autistic disorders is not real, and results only from better diagnosis,
amounts to denial.
Similarly, though some affected children have Fragile-X Syndrome or a
family history of autism, it does not seem reasonable to insist that the
present autism outbreak is solely caused by hereditary factors. Genetic
disorders have never presented as epidemics, and investing the scant
available resources solely in genetic research diverts them from the
scientific exploration of more plausible envli~nmental etiological factors.
Official U.S. Statistics
The Individuals with Disabilities Education Act (IDEA) of 1975 was
intended to ensure equal educational opportunities for children with
disabilities. State and local education districts must provide a “free
appropriate public education” based upon an “individualized education
program” geared to each student’s needs. The U.S. Department of Education is
mandated to report yearly to Congress in accordance with IDEA.
Earlier in the program, cases of autism in U.S. schools were few. As
they steadily increased, a decision was made to list autism as a separate
entity starting in 1991. As shown in Figure 1, the number of children aged 6
to 21 with autism in U.S. schools rose steadily from 5,415 in 1991-1992 to
118,602 in the latest published Department of Education report for the
2001-2002 school year.7
Table 1. States with the Largest Increases in the Number of Students with
Autism,Ages6to 21 [1 992-93 to 2002-03]
Illinois 5 5,080
Ma land 28 2,962
Mississi 0 537
Nebraska 4 481
Nevada 5 684
New Hampshire 0 491
Ohio 22 4017
Oreon 37 339
Wisconsin 18 2.739
DC 0 179
Source: U.S. Department of Education Annual Reports to Congress (IDEA)
Every state in the U.S. has recorded an increase in the number of students
with the diagnosis of autism. Nine states and the District of Columbia,
which recorded the highest increases, are shown in Table 1.
It is likely that the number of children with autism will continue to
increase as more affected children enter first grade (see Figure 2).
According to the New Jersey Department of Education, there were 14 students
of age 21 with autism in the state in the last reported school year,
2001-2002, compared to 514 students who were age 6. Other states are
reporting the same trend: an increase in the number of younger children
needing services, more programs being organized to handle the influx, and an
ever increasing number of specially trained therapists hired.
In the last 10 years, the number of children with autism between 6
and 21 years of age attending school in the U.S. rose at a much faster rate
than the number of children with disabilities in general (seeTable 2).
When IDEA passed in 1975, Congress promised to provide 40 percent of
the costs associated with the legislation. However, it has never actually
funded more than 15 percent.8 State governments have also provided a
decreasing proportion of the school budgets as state funds become scarce
nationwide. Federal and state contributions vary greatly from state to state
(see Table 3).
In addition, state contributions are not evenly distributed, and
cities usually receive proportionately more state dollars than suburban
communities, though the latter may have a higher concentration of affected
children.’ Local taxpayers usually end up bearing the spiraling cost of
educating the increasing number of affected children.
Special services are often provided through most of the summer, and
each child with autism can cost the school system $30,000 a year or more. If
an acceptable program is not locally available, the affected child has to be
transported to a collaborative or regional center, where the yearly tuition
may be higher. In rural areas, the cost of providing therapy and educating a
single child with autism may be considerable.”‘ Lastly, as autism is
diagnosed earlier, services must be made available starting at age 3, and
affected children must be enrolled in early intervention programs, which are
also expensive to start and maintain.
For a child to receive services in any program, a firm diagnosis of
autism must first be made by a physician. Each case is then thoroughly
reviewed by the special education department in the school district, which
may request more information or turn down the recommendation. Autism is not
a diagnosis that parents accept readily, physicians make lightly, or that
school authorities approve easily. In fact, it is probable that autism in
U.S. schools is actually under-diagnosed and that many less severe cases are
labeled behavior and communication disorders, in order to avoid the stigma
and/or the added cost.
In spite of all of the above, some “experts” still claim that the
spectacular increases in autism reported lately are simply the result of
more liberal or less stringent diagnostic criteria.
Diagnostic Criteria for Autism in the U.S.
As technical recognition of autism increased along with the number of
cases, accurate comparisons in prevalence and incidence rates became
difficult, if not impossible.
In Kanner’s conception, children with autism were affected early and
had poor or absent speech; an obsessive desire for preservation of sameness;
lack of affective contact; difficulty with interpersonal communication; and
a fascination with specific objects.11
In 1956, Kanner and Eisenberg’2 proposed that just two essential
features were required to make a diagnosis of autism: profound lack of
affective contact and repetitive, ritualistic, elaborate behavior.
In 1978, Rutter’3 proposed that a definition of autism in children
required four criteria: (1) Impaired social development out of keeping with
the child’s intellectual level; (2) impaired language development out of
keeping with the child’s intellectual level; (3) stereotyped play patterns,
abnormal preoccupations, and resistance to change; and (4) onset before the
age of 30 months.
In 1980, DSM III (Diagnostic and Statistical Manual ofMental
Disorders, Third Edition),11 was introduced, and its classification of
infantile autism requiredfive criteria: (1) Lack of responsiveness to
others; (2) language absence or abnormalities; (3) resistance to change or
attachment to objects; (4) absence of schizophrenic features; and (4) onset
before 30 months.
In 1987, the diagnostic criteria for autism were revised (DSM III-R),
and a definition of pervasive developmental disorders was introduced.” Other
countries had their own sets of criteria.
Since 1994, the required criteria for Autistic Disorder 299.00 have
been those established in the DSM iv; shown in Table 4,15
Similarly, detailed and strict criteria were outlined for Asperger’s
Syndrome (AS) and Pervasive Developmental Disorder, Not Otherwise Specified
Clearly, DSM IV criteria are more restrictive. In fact, in certain
states such as Washington, the number of cases of autism (IDEA) actually
decreased immediately after the introduction of DSM IV (Figure 3). The
subsequent steady rise in the number of affected students clearly points to
a true increase, as the criteria have remained uniform, and in all
probability the diagnosis in each case has been accepted by the same
medical, educational, and developmental teams.
The Impact of Autism
The autism explosion since 1994 and DSM IV is best documented in
California, where the Department of Developmental Services (DDS)” regularly
reports all new cases of the disorder introduced in the system (Figure 4).
As shown, there were 633 new cases of DSM IV autism in 1994. Within 5 years
(1999), the number of new cases had risen to 1,944 or 6 new cases a day, 7
days a week. There were 2,725 confirmed new cases of autism added to the
system in 2001 and 3,577 more in 2002 or ten children a day.’7 That one-year
increase of 31 percent was the highest in the 33-year history of the
department. Children with autism under age 3 and those with PDD-NOS and
Asperger’s Syndrome were not included.
Table 5. Autism, Behavior Disorders, and All Disabilities in Rhode Island
6/30/2001 6/30/2002 % Increase
Autism 407 498 23.35
Behavior Disorders 2,583 2,848 10.25
A11 Disabilities 31,793 32,78 3.13
Source: Rhode Island Department of Education
Autism has become the predominant disability for which services are
accessed in California. According to the most recent California Autism
Report released in March 2003, cases of Type I autism increased by 97
percent in the last four years compared to 16 percent for cerebral palsy and
29 percent for mental retardation.
The same is true in other states and is well documented recently in
Rhode Island, where proportionately, the one-year increase in autism was
substantially greater than the increase in behavior disorders and all
disabilities combined (see Table 5).
The only reasonable conclusion from this review is that the recent
increase in autism in the U.S. is real and significant. There is also every
reason to believe that more children will be “developing” autism in the
coming years. Educational programs will have great difficulty coping with
the flow of newly diagnosed children. In addition, when autistic children
become adults and their parents are not there, the impact on society will be
even greater and the burden on the national economy will mount into the
trillions of dollars.
Possible Etiologies Control of epidemics is the responsibility of
public health authorities. The Centers for Disease Control and Prevention
(CDC), the central agency in charge of the nation’s health, has previously
played a vital role in a variety of national and international crises and
programs ranging from chemical explosions in Texas City, Texas, in 1947, to
family planning (1967), famine relief in Nigeria (1968), birth defects
monitoring (1970), occupational safety (1973), ship sanitation (1975), and
the health complications from the Mount St. Helens volcanic eruption in
l980.’~ Since the 1980s, the agency has become increasingly involved in
promotion and regulation of vaccines. However, the CDC has done little to
control the autism epidemic or discover its causes. So far it has funded
only three studies, and all three were epidemiologic. In the first two, both
conducted in the U.S., serious increases in autism were reported.
In one, the investigators concluded that there were “high rates of
autistic disorder and ASD in Brick Township, New Jersey, relative to rates
from previously published studies. The rates from the majority of recent
studies are several fold lower than the rate in Brick Township.””
In the second, examining autistic syndromes in the area near Atlanta,
Ga., the authors commented, “The overall rate (of autism) is 10 times higher
than rates from three other U.S. studies … in the l980s and early
1990s.~~2 The primary purpose of the third study, from Denmark,2′ appears to
be exoneration of the measles-mumps-rubella (MMR) Several methodological
problems marred the research and, despite the fact that there was a higher
prevalence of autism among the children who had received the MMR vaccine,
the authors asserted that there was no MMR-autism connection. Moreover, the
study is not relevant to the situation in the U.S. Vaccines in Denmark have
not contained thimerosal since 1992, and Danish children only received six
doses of vaccine in the first year of life. In the U.S., children received
12 or more doses of vaccine before their first birthday, and many of those
vaccines contained thiomerosal, including one administered in the nursery.
So far, though the CDC does not know what causes autism and its
neurological, endocrine, gastrointestinal, and immune symptoms, it appears
determined, without a single clinical study of its own, to deny the
potential role of MMIR vaccination and mercury preservatives. The
possibility that mercury may affect the immune system~~ of certain
genetically predisposed children and trigger autism upon their exposure to
MMR29’2 has never been conclusively ruled out.
The decision to remove thimerosal from pediatric vaccines in 1999 was
wise. It is difficult to believe that, to assure sterility, a known poison
had been added for more than 60 years to products intended for infants.
Safer “preservatives” should have been used.
Serious independent research is urgently needed. It cannot be expected
from people with financial ties to the vaccine industry and the vaccine
Research should not be restricted to looking at epidemiologic data on
computer spreadsheets. It must include parents’ interviews and a careful
examination and evaluation of the affected children. Endoscopies and
colonoscopies should be carefully performed, and biopsies of the gut wall
should be tested for evidence of measles.
It was revealed at the December 10, 2002, bearing of the House
Government Reform Committee that Arthur Krigsman, M.D., had performed some
250 colonoscopies on children with autism and had identified pathological
and histological findings similar to those described by Andrew Wakefield in
England. Unfortunately, his hospital did not allow him to have the specimens
further tested, or specifically tested for evidence of measles.
A committee member, Representative Dave Weldon, M.D., of Florida’s
District 15, requested from Dr. Stephen Foote of NIH, who was present, to
find a way to have these specimens tested promptly. It is not known whether
this was done.
A complete cerebrospinal spinal fluid examination including serology
is needed. Serum specimens should be carefully obtained and tested for
antibodies to myelin basic protein (MBP) and neuron-axon filament proteins
(NAFP). Antibody levels of measles virus (MV) and human herpesvirus-6
(Ill-IV-6) should be determined; evaluation of serotonin, serotonm receptor
antibody, interleukin- 12 (IL-12), and interferon-gamma (LFN-g) levels would
also be helpful.
Testing for urinary indolyl-acryloylglycine (lAG) and polypepfldes is
of value both diagnostically and therapeutically, as it identifies those
children who would benefit from diet restrictions. Checking for heavy metals
is always indicated. Other appropriate testing should be individualized.
There has been a true and significant increase in autism in the U.S.
To date, the CDC and other governmental health authorities have not given
enough attention to this serious epidemic and its present and future impact.
They must face their responsibility now.
Emerging evidence suggests some relationship between MMR and
thimerosal-containing vaccines and regressive autism. Additional independent
and unbiased clinical studies must be conducted in order to determine all
Information about the autism epidemic and its potential causes should
be widely disseminated.
F. Edward Yazbak, MD, FAAP, founded TL Autism Research, 70 Viewcrest
Dr., Falmouth, MA 02540, firstname.lastname@example.org.
Competing interests: Dr. Yazbak is the grandfather of a boy with
regressive autism, typical ‘autistic” enterocolltis, and evidence of measles
genomic RNA in the gut wall.