Good morning, Mr. Chairman, and Members of the Committee. I am Dr. Coleen Boyle, Chief of the Developmental Disabilities Branch of the National Center for Environmental Health (NCEH), Centers for Disease Control and Prevention (CDC). I am accompanied by Dr. Ben Schwartz, Acting Director of the Epidemiology and Surveillance Division of the National Immunization Program (NIP), CDC. Thank you for giving me an opportunity to discuss the work that we are doing at CDC in the area of preventing developmental disabilities, including autism. I have actively conducted research in this area for the past three years, directing CDC's applied research in developmental disabilities for 10 years. Developmental disabilities prevent children from achieving at the level of their peers and through no fault of their own cause a tremendous burden on families and our society in general.
Developmental disabilities are a diverse group of physical, cognitive, psychological, sensory and speech impairments that are usually identified between birth and up to age 18 years. It is estimated that about 17% of all children have a developmental disability, and 2% have a serious developmental disability such as mental retardation, cerebral palsy or autism. In most cases, we do not know the cause of the developmental disability. Several infectious diseases are known to cause developmental disabilities, including Haemophilus influenzae type B and congenital rubella syndrome, a known cause of autism. Other known causes of developmental disabilities include nutritional deficiencies such as those of iodine and iron, and environmental exposures including lead and mercury. Autism and conditions related to autism represent some of the most serious of these developmental disabilities. Autism is a life-long disability characterized by impairments in social interaction and communication and a pattern of restrictive, repetitive and stereotypic behaviors, interests, and activities. The impact and burden of autism on children and their families is tremendous. Most children with autism require long term care and services, including special education and supervised care.
Developmental disabilities are costly to society and to families. Local, state and federal education departments spent approximately $49.2 billion in the 1998-99 school year on special education programs for children with developmental disabilities. The cost of special education for a child with autism is often more than $30,000 per year to the family and the community, and the cost of residential care, which many of these children require, is $80,000 to $100,000 per year. There is no cure for autism; however, recent studies suggest that early intense behavioral interventions benefit some children with autism.
CDC's role in preventing developmental disabilities, including autism, is to determine the scope of the problem, identify preventable causes of developmental disabilities, and establish and evaluate intervention programs. We do this using the public health tools of population-based surveillance, epidemiologic investigations, and prevention programs.
PREVALENCE OF AUTISM
Much attention has been focused on whether there is a higher rate of autism than previously thought. Researchers conducted the first epidemiologic study of autism in England in 1966 and found the autism rate in the general population to be 4 to 5 per 10,000 children. Other early community studies published also yielded prevalence rates in this range. In general, these studies defined autism using a narrow set of criteria B which included marked impairment in social contact and elaborate behaviors or mannerisms. Studies published since 1985 outside the United States reported higher prevalence rates than those prior to 1985, about 12 per 10,000 children, while a few very recent studies have found rates that are even higher.
It is unclear why the more recent studies have yielded higher rates. However, we do know that more recent studies have used diagnostic criteria for autism which are considerably broader and incorporate the clinical recognition that the hallmark features for autism can occur in a wide range of severity levels and in many different manifestations. Clearly, a greater awareness and better recognition of this condition also have had an impact on the reported prevalence rates.
There have been only two population-based prevalence studies of autism in the U.S. Both studies were conducted in the 1980s and yielded prevalence rates of 3.3 and 1.2 per 10,000 children, which is lower than most European studies. The U.S. studies relied on identification of children with autism already known to service providers - which may account for the relatively low prevalence rate.
However, other U.S. data sources seem to support the idea that the prevalence of autism is higher than previously thought. A recent report from the California Health and Human Services Agency examined the number of people with autism entering the California Developmental Services system each year from 1987 to 1998 and showed an overall increase of 273%, with increases every year since 1987. Services for all other developmental disabilities increased no more than 50% during this time period. The State of California has contracted with the Medical Investigation of Neurodevelopmental Disorders (M.I.N.D.) Institute of the University of California at Davis to conduct a study to determine the factors accounting for the increased caseload of children with autism in California. Similarly, service provider data from the US Department of Education showed that the number of U.S. children with autism who received special education services increased 556% from 1991 through 1997.
While we cannot be certain of the reason for changes in prevalence rate, some percentage of the increase seems related to broadening of the definition of autism, changes in referral patterns, improved recognition and greater awareness of the condition. Other factors may involve eligibility and reporting requirements for disability services, changes in population demographics and patterns of migration. Genetic and environmental factors may also be contributing to an increase in the number of children with autism, which may result in larger numbers of individuals being identified. Ongoing population-based surveillance, using standard methods of case ascertainment, is needed to better understand trends in autism rates as well as other developmental disabilities.
CDC's EFFORTS TO DETERMINE THE PREVALENCE OF AUTISM
CDC is developing reliable estimates of the prevalence of autism. In 1998 we added autism to CDC's Metropolitan Atlanta Developmental Disabilities Surveillance Program (MADDSP). This program also monitors the prevalence of mental retardation, cerebral palsy, vision impairment, and hearing impairment among school-age children. MADDSP is the only ongoing population-based study of multiple disabilities in school-age children in the United States. Information on children with developmental disabilities, including autism, is obtained by review of records from service providers in metropolitan Atlanta and identification of children with developmental disabilities in the public school system. CDC expects to publish the results of this first population-based prevalence study later this year. CDC is also funding Marshall University to begin a surveillance program for autism in six counties in West Virginia. CDC believes population-based surveillance for developmental disabilities is a critical element in preventing this disorder.
THE BRICK TOWNSHIP INVESTIGATION
Mr. Chairman, beginning in early 1998, as you are probably aware, CDC and the Agency for Toxic Substances and Disease Registry (ATSDR) were contacted by the New Jersey Department of Health and Human Services, U.S. Senator Robert Torricelli, and U.S. Representative Christopher Smith regarding the possibility of federal assistance in addressing issues raised by the citizens of Brick Township, New Jersey. They were concerned that the number of children with autism in Brick Township could be several times higher than expected based on available U.S. prevalence rates for the disorder. CDC assistance was requested because of the complexity of investigating a behavioral disorder such as autism, and because of the fact that CDC was developing epidemiologic methods that address the unique challenges of autism. ATSDR's expertise was requested because community members felt that environmental factors related to potential hazardous waste sites might be involved. CDC investigators with expertise in population-based studies of autism were responsible for the prevalence investigation. The prevalence investigation identified children with possible autism spectrum disorders whose parents were residents of Brick Township during 1998. The autism diagnosis was verified through a clinical assessment. ATSDR has conducted a scientific literature review to determine what is known about associations between chemical contaminants and autism. ATSDR has been assessing potential exposure pathways for sources of environmental pollution in Brick Township, and also is assessing whether the potential exposure pathways could be associated with the homes where children with autism lived. Specifically, three potential sources of environmental pollution were identified: (1) the municipal drinking water supply, (2) the Metedeconk River, and (3) the Brick Township Landfill. Possible exposure pathways include recreational uses of the river where contaminants may have been discharged, as well as the use of the river as a source of drinking water supply.
The report of the findings from the investigation and environmental public health assessment are currently being reviewed. The reports will be provided to the parents whose children were a part of this investigation in Brick Township and to the general public as soon as possible. We would be pleased to brief interested Members or their staff on the results at the same time we make them available to the affected community.
CAUSES OF AUTISM
The cause of autism remains unknown for most children. Several studies support an underlying genetic mechanism for autism. Studies indicate that family members of individuals with autism also are more likely than others to be diagnosed with the disorder. An identical twin of a child with autism has a 75 to 90 percent chance of having autism as well. A fraternal twin of a child with autism has a 5 to 10 percent chance of having autism. Similar genetic susceptibility may extend to other developmental disabilities as well. Among families with autism there is a 10 to 40 percent increase in the diagnosis of other developmental and learning disabilities, such as language delays and learning disabilities. Autism tends to occur more frequently than expected among individuals with certain medical conditions such as Fragile X syndrome, untreated phenylketonuria, congenital rubella syndrome, and certain seizure disorders.
A scientific literature review has identified limited evidence that certain agents ingested by pregnant women such as lead, alcohol and the prescription drug thalidomide may cause autism in their children. Such evidence, as well as prevailing theories about autism etiology, suggest that events during development in utero, especially in the earliest stages, play a substantial role in the cause of autism. Less information is known about postnatal exposure and autism. Little research has been done in the area of environmental contaminants and more is needed. In addition, it is important to do carefully controlled studies on the potential causes of autism to determine whether children exposed to a given risk factor are more likely to develop autism than children who are not exposed.
BENEFITS OF VACCINES
One of the most effective, proven ways parents can protect their children from certain infectious diseases is to ensure they receive all of their recommended vaccines on schedule. The threats posed by vaccine-preventable diseases are known and real. The viruses and bacteria that cause vaccine-preventable diseases still circulate in the U.S. and around the world. High rates of measles vaccination have protected U.S. children and communities from sustained outbreaks. In 1998 and 1999, 16 measles outbreaks and 13 rubella outbreaks in the U.S. were reported to CDC. Measles is not a benign childhood illness. It can have serious complications, including pneumonia, ear infections, brain damage, seizures and even death. A measles epidemic in the United States led to more than 55,000 cases of measles, more than 11,000 hospitalizations, and more than 120 deaths in the three years from 1989 to 1991. Measles still accounts for approximately 750,000 deaths each year globally, and congenital rubella syndrome results in severe disability or death for over 200,000 infants per year. Common manifestations of congenital rubella syndrome include deafness, blindness, heart defects, and mental retardation. Continued high U.S. vaccination rates are crucial to prevent the spread of these diseases among U.S. children. Vaccinations protect individuals and communities from diseases spread by person-to-person transmission. An individual's vaccination not only protects that individual from diseases spread by person-to-person transmission, but also adds to the protection of the rest of the community. A decision to not vaccinate places individuals and communities at risk.
CDC's MISSION TO ENSURE SAFE VACCINES
The public should expect that we will do all we can to deliver safe vaccines. While vaccines are among the safest pharmacologic interventions for disease prevention available, no drug or vaccine is 100 percent without risk. Extensive pre-market studies of vaccine safety are conducted and carefully reviewed by FDA before a vaccine is licensed. Post-marketing evaluation of vaccine safety is also conducted. CDC and FDA have developed an infrastructure to continue monitoring vaccine safety. This includes the Vaccine Adverse Event Reporting System (VAERS), a nationwide, passive surveillance system that provides "signals" of potential vaccine adverse events; and the Vaccine Safety Datalink, an active surveillance system that links vaccination and health data from four large Health Maintenance Organizations and which can be analyzed to assess whether an association exists between vaccination and an adverse event. In addition, targeted epidemiological studies using case-control or cohort methodologies are also conducted to investigate specific concerns.
These methods have been used to further enhance the safety of the vaccination program. Examples of changes that have been made to improve the safety of the vaccination program include a shift from whole cell pertussis vaccines to a less reactogenic acellular vaccine; a change from oral to injected poliovirus vaccine to prevent vaccine-associated paralytic polio; and the withdrawal of recommendations for rotavirus vaccine after investigations showed that vaccination increased the risk of intussusception, a form of bowel obstruction among infants. These examples demonstrate the commitment of CDC and partner agencies to assuring vaccine safety and the flexibility of the program to respond to safety issues that have been scientifically documented.
SCIENTIFIC EVIDENCE REGARDING VACCINES AND AUTISM
Currently available scientific evidence does not support a link between vaccination and autism or any other behavior disorder. This statement is based on the following synopsis of published data:
- An initial observation linking autism and MMR vaccine was reported by Dr. Andrew Wakefield and colleagues, who had first attempted to link measles disease and vaccination to inflammatory bowel diseases such as Crohn's Disease. Dr. Wakefield suggested that measles/mumps/rubella (MMR) vaccination led to intestinal abnormalities, resulting in impaired intestinal function and developmental regression within 24 hours to a few weeks of vaccination. This hypothesis, which suggested that children developed autism shortly after receipt of MMR vaccine, was based on a case-series (a collection of patients with limited comparison or control groups) reporting data from 12 children.
- The Working Party on MMR vaccine of the British Committee on Safety of Medicines conducted a systematic review of reports of autism, gastrointestinal disease and similar disorders after receipt of MMR or measles/rubella (MR) vaccine. In 1999, the Working Party concluded that the information available "...did not support the suggested causal association or give cause for concern about the safety of MMR or MR vaccines."
- A study published in 1999 in The Lancet by Dr. Brent Taylor and colleagues provides population-based evidence that overcomes a number of limitations that the Working Party and the Wakefield group experienced. The authors identified all 498 known cases of autism spectrum disorders (ASD) in children living in certain districts of London who were born in 1979 or later and correlated the cases to an independent vaccination registry. The results of this study were:
- The first diagnosis of autism or initial signs of behavioral regression were not more likely to occur within time periods following MMR vaccination than during other time periods.
- Despite an increase in the number of diagnosed ASD cases since 1979, no jump occurred after the introduction of the MMR vaccine in 1988. Such a jump would have been expected if MMR was causing a substantial increase in autism causes, but this was not the case.
- Children who were vaccinated before 18 months of age were diagnosed with autism at ages similar to children who were vaccinated after 18 months of age, indicating that the vaccination did not result in earlier expression of ASD characteristics. If MMR were causing many autism cases, it would have been expected that children vaccinated at a younger age would develop autism at a younger age than children vaccinated at older ages, but this was not the case.
- At age two, the MMR vaccination coverage rates among ASD cases were nearly identical to vaccination coverage rates of children in the same age group in the whole region, providing evidence of a lack of overall association between ASD and the MMR vaccination. If MMR was a major cause of autism, then it would have been expected that cases of autism would be more likely to be vaccinated than the general population, but this was not the case.
Another study, conducted by Dr. Christopher Gillberg and Dr. Harald Heijbel, also showed no evidence of association between the MMR vaccine and autism. The study compared autism prevalence rates in populations of children from two communities in Sweden. The results indicated no difference in autism prevalence between children born after the introduction of the MMR vaccine in Sweden and those born before the vaccine was used. The study was published in the journal Autism in 1998.
At this time, the weight of scientific evidence does not support an association between MMR and autism.
WHAT RESEARCH IS CDC CONDUCTING ON VACCINATION AND AUTISM?
Many parents remain concerned that their child's autistic behaviors seemed to occur or to worsen shortly after vaccination. Despite the fact that the available evidence, on balance, does not support a causal link between vaccines and autism, given the level of concern, it is critical to investigate this issue more fully using the best scientific methods available. The challenge of vaccine safety research is to use scientifically sound methodologies to investigate uncommon events and to distinguish events that would have occurred anyway even without vaccination from those that are truly causally related.
One important ongoing CDC study is based on CDC's population-based autism surveillance system in metropolitan Atlanta B one of few such surveillance systems in the world. The system began to identify cases of autism in 1998 among children between 3 and 10 years old in the metropolitan Atlanta area. Children with autism are identified through special education records of all area public schools and other sources. Capitalizing on this unique resource, CDC has initiated an epidemiologic study with the primary objective of evaluating the association between MMR vaccination and autism. The study will compare the vaccination histories of over 500 children with autism and over 1500 control children matched on school, age, gender, and the children's schools. Data collection is nearing completion and the study results are expected later this year.
In addition, CDC is working with the National Institutes of Health to develop a study that will evaluate whether vaccination is linked with developmental regression. These studies, and a CDC study of inflammatory bowel disease and MMR vaccination using the Vaccine Safety Datalink, demonstrate CDC's ongoing commitment to assuring the safety of the vaccination program.
Given the balance of available scientific data and the known serious and ongoing risk of vaccine preventable diseases, the best public health policy is to continue vaccination schedules unchanged while we continue to monitor vaccine safety. This approach, we believe, is most effective in protecting the overall health of children and reflects the strategy of other countries such as the United Kingdom, which also has considered these questions.
CONCLUSION
Mr. Chairman, in the past 40 years, public health has made significant advances in preventing developmental disabilities. Identification and prevention of congenital syphilis and congenital rubella syndrome have spared the lives and prevented disabilities for thousands of children. Newborn screening programs have prevented life-long mental retardation and prolonged the lives of tens of thousands of children with metabolic disorders, such as hypothyroidism, PKU, and sickle cell disease. However, even though this is good news, we still don=t know the causes of most developmental disabilities, including autism. CDC agrees with the Committee that autism is a serious developmental disability that has a significant and profound impact on the lives of not only the child with the disorder, but also the child's family. We must develop a better understanding of autism and its causes. We need to track this disorder, identify preventable causes, and institute effective intervention programs. CDC is using available resources to develop surveillance systems to collect information so that we can better understand what is happening with the prevalence rate over time. We must also conduct epidemiological studies to begin to uncover the causes of this serious developmental disability. It is my hope that this work, combined with scientific investigations conducted by other federal agencies such as NIH and within the academic community, will lead to identifying what causes autism, preventing this disorder, and enabling these children to live full and productive lives.
Thank you, Mr. Chairman and members of the Committee, for the opportunity to testify before you today about CDC's efforts to better understand and prevent this serious developmental disability. I am happy to answer any questions you might have.
