Published: Dec. 8, 2009
Updated: May 17, 2010
During office visits when its time for an immunization shot, parents frequently ask me whether vaccines cause autism.
It is a highly charged discussion, involving great emotion for many parents.
Jeffrey P. Baker, MD, PhD, director of Duke's History of Medicine Program and an associate clinical professor of pediatrics, discusses the evidence.
--Dennis Clements, MD, PhD
The annals of medicine are full of stories about scientists who stubbornly cling to a “great idea” despite evidence to the contrary. The history of autism provides a classic example.
From the 1940s through at least the 1960s, autism was widely viewed as a psychiatric condition, typically attributed to highly educated mothers lacking the capacity to provide warmth or affection for their infants. The so-called “refrigerator mother” theory turned out to be based on little reality beyond the imaginations of its originators. It survived as long as it did because it promised (falsely, as it turned out) the possibility of cure through psychotherapy.
Today, another hypothesis has captured the imagination of many parents in the autism community. It is the conviction that vaccines are responsible for the dramatic rise in the disorder’s visibility over the past 20 years. Despite the failure of 10 years of scientific study to provide support, this belief remains powerful among parents’ groups and the Internet. Why?
Although the cause of autism remains unknown, the vast majority of researchers believe that genetics play a central role. Siblings of children with autism have a 2 to 7 percent chance of the disorder, at least 50 times the rate of the general population. The concordance rate is higher for fraternal (5-10 percent) and highest of all for identical twins (60-90 percent).
Other studies have found a higher family risk of problems in communication, social relations, and anxiety, suggesting that a broader form of the disorder may be inherited that presents as classical autism only in its most severe form. Collectively, this research underlines that genetics likely represents the most critical factor leading to autism. The question is whether an environmental trigger may play a secondary role in genetically-predisposed children.
If autism is largely genetic, why has it seemingly become “epidemic” in recent years? There is no doubt that the disorder is diagnosed far more commonly (by a factor of at least 10) today than was the case 20 years ago. It is equally clear that its definition has been expanded tremendously, encompassing both higher functioning children (such as those with Asperger’s syndrome) and others who one would have been diagnosed with mental retardation.
At the same time, there has been a great push for physicians and schools to identify autism at earlier ages and with milder presentations. Many experts in the field believe that the rise of autism reflects the success of early intervention and school-based programs to heighten awareness of the disorder. Still, the possible role of an environmental exposure cannot be eliminated. And of the hundreds of agents to which pregnant women or infants are exposed, none are quite as visible as vaccines.
Activists in the last 10 years have promoted two particular theories linking immunizations and autism. One concerns the MMR vaccine against measles, mumps, and rubella. It became controversial in Great Britain following the publication in 1998 of a case report by Dr. Andrew Wakefield describing several children who developed signs of diarrhea and autistic regression following this vaccine. MMR immunization rates fell in the U.K., and outbreaks of measles followed.
The other hypothesis, originating in the U.S., concerns the preservative thimerosal, which was removed from infant vaccines between 1999 and 2001 as part of broader public health efforts to reduce infant exposure to environmental mercury.
These two theories are not easily reconciled. Parents blaming the MMR typically described infants who were normal prior to this particular vaccine; thimerosal/autism narratives told of infants who developed autism after any vaccine combination.
British parents often noted the correlation between the rise of autism and the use of the MMR, introduced in 1987 and the focus of national immunization drives in the early 1990s. The fact that MMR had been used widely in the U.S. since 1971, long before talk of an autism epidemic, was generally ignored.
Both hypotheses have been subjected an extraordinary amount of study in large populations. Comprehensive reviews by expert panels, most notably the U.S. Institute of Medicine, have concluded that the evidence simply does not support either vaccine/autism hypothesis.
In Britain, 10 of the 12 co-authors of Dr. Wakefield’s 1998 report have disavowed its conclusion regarding MMR. Wakefield himself is under investigation for serious professional misconduct for not having revealed his relationship to anti-MMR litigation groups when submitting his article to The Lancet. In the U.S., data published in 2008 from the state of California showed that the elimination of thimerosal from all routine infant vaccines in 2001 had no effect slowing down the rise of autism, despite many predictions to the contrary.
Among mainstream health researchers, the MMR and thimerosal autism hypotheses are in tatters. Yet like the hydra of ancient mythology that grew two heads whenever one was severed, the belief in a vaccine/autism connection continues to survive by taking new forms. Some activists are focusing on other vaccine additives, such as the aluminum salts used to boost the immune response.
Others are arguing that giving too many vaccines somehow overwhelms the child’s immune system. This was the question at the heart of the Federal Vaccine Court’s decision to award damages to Hannah Poling, a girl with a mitochondrial disease (a very uncommon disorder disrupting her ability to process nutrients) who regressed developmentally and developed signs of autism after receiving several vaccines at age 19 months.
Cases such as Hannah’s are tragic, but raise more questions than they answer. There are in fact rare children with silent metabolic disorders who may develop normally until suddenly regressing after the stress of a childhood infection. Whether vaccines are a risk has not been proven. Certainly, the infections that vaccines prevent do constitute a danger for these children. Even if we could identify at-risk children, it is far from clear that holding or splitting vaccines would do them a service.
Before accepting the “multiple vaccine” hypothesis, it is worth remembering that more vaccines does not mean more stress on the immune system. The 14 vaccines given to young children expose them to a total of about 150 immunological units, or antigens. The MMR, for all the ballyhoo, contains only 24. In contrast, the old smallpox vaccine included 200 proteins, and the whole cell pertussis vaccine used before the 1990s contained 3,000.
In a nutshell, while more vaccines are being given to infants, these vaccines are far more targeted and purified than was the case twenty years ago. This is why giving vaccines separately makes so little sense to the scientific community. Splitting vaccines certainly makes the schedule even more complicated, and will likely lead to lower immunization rates. When these rates fall below a certain threshold in a community, outbreaks become possible. This has already happened with respect to whooping cough and measles in various locations in the United States.
It will always be possible to think of new mechanisms linking vaccines and autism as others are disproven. But after 10 years of extensive research on vaccines, it is time to entertain other ideas regarding environmental exposures. Vaccine opponents consistently disparage the positive benefits of vaccines, which the vast majority of physicians and public health leaders regard as one of our most powerful tools to protect the health of our children. Deferring or declining vaccines has consequences for our neighbors’ children as well as our own. It is important to learn about the diseases they prevent prior to questioning their benefits.
-- Jeffrey P. Baker, MD, PhD, is director of Duke's History of Medicine Program and an associate clinical professor of pediatrics.
-- Dennis Clements, MD, PhD, is the chief of primary care pediatrics at Duke Children's Hospital.