Vaccine Safety

The ABCs of MMR & DTP:
Is There an Association Between Vaccination and Autism?

by Eric London, M.D. Edited by Catherine Johnson, Ph.D.

One of the questions I am most frequently asked, both as a physician and as NAAR'S Vice President-Medical Affairs, is whether vaccinations can cause autism.

Many parents trace their children's first signs of autism to a moment shortly after their toddler reacted badly to a vaccine. When these families can find no other cases of autism on either side of the family tree, they reason that it must have been the vaccination that delivered the blow. But logical as this might seem, there has been little if any scientific evidence to substantiate an association between vaccination and autism.

The lack of data to support a connection between vaccine and autism makes sense given the increasing body of information concerning when the neurobiological differences associated with autism first occur. The preponderance of evidence tells us that autism happens to our children before birth, not after. Margaret Bauman's and Anthony Bailey's brain autopsies both show many brain changes that must occur before birth; Joseph Piven's MRI studies indicate a prenatal defect in brain development; embryologist Patty Rodier's work puts the date for some or many cases of autism as early as days 20 to 24 after gestation.

Furthermore, evidence indicates that autism is highly genetic. Geneticists believe that autism is a 'multifactorial, polygenic' disorder. This means that it is an inherited, genetic disorder that may, or may not, involve outside factors such as toxins, pollutants, complications of pregnancy and so on. As to how a child can inherit a disorder when no one else in his family has the disease, geneticist Irving Gottesman provides a highly read-able explanation in his book Schizophrenia Genesis: The Origins of Madness. With all 'polygenic' disorders, he writes, it is the norm for other family members to be free of the disorder, because of the odds against any particular family member inheriting all 3 or 4 or 5 of the necessary genes. Even if every living soul in an extended family carries one or two of the genes for autism, no one will actually become autistic without suffering the misfortune of acquiring all 3 (or 4 or 5) autism genes at conception.

In short: inheriting autism is like winning the lottery in reverse.

"Any family worried about the safety of vaccination...should keep in mind the profound public health implications of vaccinations... Before we had childhood immunizations, many, many thousands of children died of now—preventable disease."

Why do some children with autism seem normal until just after they have received their vaccinations?

The answer to this question could be that many children with autism develop normally for 12 to 18 months, whether or not they were vaccinated. That may be the nature of the disorder in some of its expressions; the child has a period of reasonably typical development followed by a heartbreaking plateau or outright regression after the first birthday. A third world child with autism who received no vaccinations of any kind might show exactly the same pattern.

Schizophrenia may work the same way: researchers have now found a great deal of evidence that schizophrenia happens in the womb—and yet a person who is destined to become schizophrenic can be completely normal for a good twenty years. So much evidence has accumulated concerning schizophrenia and certain critical periods in pregnancy that there is now a movement afoot to redesignate schizophrenia as a 'neurodevelopmental disorder'. Given that most of us think of development as being over and done with by the time a person reaches young adulthood, the notion that schizophrenia may soon be reclassified as a developmental disorder is counterintuitive to say the least.

The MMR (Measles, Mumps, Rubella) Vaccine Scare

Recently a paper published in the British medical journal LANCET raised the vaccine issue anew. In a preliminary report entitled "Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children", Anthony Wakefield, M.D. and his colleagues described their study of 12 children with co-occurring chronic enterocolitis (a chronic inflammation involving the small intestine and the colon) and regressive developmental disorder (in which a history of normal development is followed by a loss of acquired skills, including language, and the onset of behaviors associated with pervasive developmental disorder). Nine of the children carried a diagnosis of autism; one carried a diagnosis of integrative disorder; and two carried a diagnosis of possible
post-viral or post-vaccinal encephalitis.

Prospective developmental records showed satisfactory achievement of early milestones in all of the children. In eight of the twelve children, the onset of behavioral problems had been linked, either by the child's parents or by the child's physician, with the administration of MMR vaccine. In these cases the average interval between vaccine exposure and onset of behavioral symptoms was 6.3 days; the range was one day to fourteen days.

The onset of symptoms of inflammatory bowel disease (IBD)—a collective term for the various gastrointestinal conditions observed in these children—was not as distinctive or memorable. In five of the twelve cases, the timing of onset was unknown. In all but two of the remaining cases, the onset of bowel symptoms was after the onset of behavioral symptoms. In only one case did the symptoms of bowel disease immediately follow (i.e. within a couple of weeks) the MMR vaccination.

Based on their study findings and on related findings in cited research, the authors suggest that there is a real connection between inflammatory bowel disease and regressive developmental disorder and that it reflects a unique disease process. The authors state that they did not prove an association between MMR vaccine and the syndrome described, though it is clear from their discussions and their references to other research that they believe a causal link may exist.

Wakefield's report, in spite of its 'preliminary' status, has received worldwide media attention and has caused alarm among parents in the community who have young babies or who are contemplating having more children in the future. Did the MMR vaccine cause their child's autism? Should they refuse to vaccinate their new babies?

Vaccinations Save Lives

Any family worried about the safety of vaccination—and this includes families who do not have autism but who have read accounts of children suffering permanent damage as a result of vaccine—should keep in mind the profound public health implications of vaccinations.

Before we had childhood immunization, many, many thousands children died of now-preventable disease. Others suffered permanent injury and/or brain damage. The diseases against which we can now protect our children are no small thing: diphtheria, pertussis, polio, tetanus, smallpox—all of these illnesses are killers. Even measles, which most of us think of as a minor childhood ailment, has left children brain-damaged or dead. (Author Jessica Mitford published The American Way of Death in 1963 after her infant daughter died of measles.)

Table 1.
Maximum Annual vs. 1995 Cases of Vaccine-Preventable Diseases in the U.S.


Maximum Annual Reported Cases in the Pre-Vaccine Era (year)

Reported Cases in 1995**

Percentage Change in Morbidity

Congenital Rubella Syndrome

20,000* (1964-5)




206,939 (1921)



Invasive H.influenzae

20,000* (1984)




894,134 (1941)




152,209 (1968)




265,269 (1934)



Poliomyelitis (wild)

21,269 (1952)




57,686 (1969)




601 (1948)




1.64 million



*Estimated because national reporting did not exist in the prevaccine era.
**Provisional total.
Source: Advisory Committee on Immunization Practices, CDC

Table 1 outlines nine diseases and the effect vaccines have had on their incidence in the U.S. It's easy to see the result: in 1995 there were only 6,815 reported cases of all 9 of these diseases put together, compared to 1.6 million reported cases when these disorders ran unchecked through the population. The pre-vaccine figure, 1.6 million, represents an enormous degree of human suffering and loss. The moral: we vaccinate children against disease in order to keep them alive and well.

Ironically, given the alarm that has greeted Wakefield's report, we know for a fact that the MMR vaccine has already saved some children from autism. As Dr. Marie Bristol-Power of the National Institute of Child Health and Human Development has pointed out, congenital rubella syndrome, which strikes the unborn child when his mother contracts a case of German measles, is a known cause of autism. In other words, the one and only known cause of autism that has been virtually wiped out—thanks to medical research—involves those cases of autism that resulted from German measles in the pregnant mother!

The Dangers

Still, as vaccine critics point out, the tremendous success of twentieth century vaccine programs has not come without a price. Babies can suffer adverse reactions to vaccine; some of these reactions are severe.

"...Wakefield and his colleagues themselves have publicly stated that they endorse the current vaccination policy until further data are available..."

How high are the risks?

Researchers consistently find that the risk of an adverse reaction of any kind, mild or severe, is very low. Of the approximately 13 million childhood vaccinations given in 1995, for instance, only 10,594 'adverse events' were reported to the Vaccine Adverse Events Reporting System (VAERS). Ten thousand may sound like a large number, but proportionately it represents only .08% of all vaccinations given in that year. Since the phrase 'adverse events' includes reactions that were unpleasant or frightening but short—lived, the percentage of bad reactions to vaccine that permanently harmed a child's life is minute.

Of course all of us instinctively feel that even one child harmed out of many millions is still one too many; obviously the ideal for vaccine programs would be to produce no serious adverse reactions at all, ever. But since we live in the real world, where children are exposed to terrible disease, the proper calculus when weighing the risks involved in vaccination must be to compare the number of children hurt by vaccine to the number of children hurt by not receiving vaccine. 10,594 adverse reactions, most of them mild and transient, is vastly preferable to 1.6 million cases of life-threatening illness.

Both Great Britain and Japan have learned this lesson the hard way. Back in the 1970s, after 36 parent reports of severe neurological illness following pertussis vaccination, the British vaccination rate fell from 80% in 1974 to just 31% in 1978. Without the vaccine to check its progress the disease came roaring back, and the pertussis epidemic of 1977 to 1979 claimed 36 lives.

The people of Japan suffered a similar object lesson in 1975. When two deaths were reported after pertussis vaccinations, the vaccine was withdrawn. Then, in 1979, there were 41 deaths from pertussis. Again: while two deaths from vaccination is two too many, 41 deaths is many more than two. These are deadly diseases, and that is why vaccinations exist.

Measles is also a potentially dangerous disease—a fact we'll re-learn quickly if parents pursue a wholesale course of abandoning MMR vaccination. One out of every 15 individuals stricken by measles will have complications, including ear problems, bronchitis, pneumonia and seizures; one in 5000 will develop encephalitis.

The statistics tell the story. Between 1989 and 1991 there were 55,000 cases of measles in the United States. Eleven thousand people—mainly children—were hospitalized; 123 people died. The death rate from measles, 2.2 per 1000, is higher than the rate of all autism from all causes, which is 1.0 to 1.5 per 1000. So it should be obvious that allowing the virus to return is not an acceptable solution to the concerns raised by Wakefield's report—and in fact Wakefield and his colleagues themselves have publicly stated that they "endorse the current vaccination policy until further data are available." The measles vaccine, all researchers understand, saves lives.

Why do so many people fear vaccines when they are one of the greatest success stories in medical history?

Possibly because the vaccine program is a victim of its own success. Vaccinations have worked so well for so long that people have forgotten what it was like to bear children before vaccination became available. No one can tell you, parent to parent, what it's like to nurse a child through a virulent case of pertussis—or what it was like, in the 1940s, to keep your children indoors all summer long and pray that the polio virus passed them by. The only horror stories we hear today are horror stories about vaccine. Of course there are far fewer of those; that is the point to remember.

Even more important: in the uproar surrounding the Wakefield report many have lost sight of the fact that the entire point of vaccination programs is to end the need for vaccinations! If new parents faithfully vaccinate their young children against measles, measles will cease to exist. We are not far from that point now. In 1995 there were only 309 cases of measles reported in the entire United States—and the World Health Organization, the Pan American Health Organization and the Centers for Disease Control have set a global eradication goal for between 2005 and 2010. In other words, if parents continue to vaccinate at high rates, and do not become frightened off by reports like Wakefield's, within ten years measles could go the way of small-pox—and, soon, of polio.

It could disappear from the earth.

The Crippled Virus

Although the effectiveness of the vaccine program in reducing death and disability is obvious, the question still remains whether the measles vaccine in particular might cause cases of autism in some children.

A useful first step in thinking about this question is to ask how likely it is that measles itself causes autism. If you're not afraid of measles (and most parents are not) then there is no reason to be afraid of the measles vaccine.

The essential fact that seems to get lost in worries about vaccine safety is that the viruses used in vaccinations are either dead, or alive-but-attenuated. The measles vaccine uses the live-but-attenuated variety. A live-but-attenuated virus has been slightly altered; some attenuated vaccines, for instance, use only a part of the structure, not the entire virus. Once the virus has been crippled in this way, it can no longer cause disease.

But the immune system does not know this. Instead, the disabled virus tricks the body's immune system into thinking the child has caught the disease when he hasn't. After a vaccination, the immune system creates an army of antibodies so that when the child is exposed to the wild virus (researchers refer to living viruses that exist in nature as 'wild') he or she is ready to fight it off. The beauty of vaccinations is that they use the body's natural defense system to ward off death and disability.

"Catching" Autism

Are there likely to be any children on earth, even a very tiny group, who have "caught" autism from a measles vaccine, given the fact that the measles vaccine is a crippled version of the wild virus?

First off, it is possible, as vaccine critics point out, to acquire autism. We know that a perfectly healthy baby in the womb can become autistic if his mother catches German measles; we know that perfectly healthy babies whose mothers took thalidomide during pregnancy became autistic; we know that a perfectly healthy young child can emerge from a case of encephalitis with brain damage and autistic-like symptoms, though these children are far more likely to appear simply retarded.

It is even possible, Oliver Sacks writes, though extremely rare, for a healthy adult to become autistic-like after a case of encephalitis. Some of his Awakenings patients, Sacks observes, had 'elements' of autism. And we know that measles can cause encephalitis. It stands to reason, then, that there exists a small subset of people with autism today who may have become autistic after having caught the measles and developed encephalitis as a complication. The question is: can the measles vaccine cause autism?

"...Viruses used in vaccinations are either dead, or alive-but—attenuated... Once the virus has been can no longer cause disease... But the immune system does not know this. Instead, the disabled virus tricks the body's immune system into thinking the child has caught the disease when he hasn't... The immune system creates an army of antibodies so that when the child is exposed...he or she is ready to fight it off..."

The Measles Vaccine and Autism?

We do have answers to this question. The National Childhood Vaccination Injury Act of 1986 mandated that the Institute of Medicine (IOM), a private, independent nonprofit organization that advises the federal government on health policy, review evidence regarding adverse reactions of all vaccines administered to children. Its findings on the measles vaccine are these:

Between 5 and 15 percent of children vaccinated against measles develop a fever 5 to 12 days afterwards from which they fully recover. Transient rashes have been reported in approximately 5% of children vaccinated. For parents in our community this is the important finding: central nervous system conditions—meaning conditions such as encephalitis that can cause brain damage—occurred in fewer than one per one million doses administered. In reality, a figure this tiny is impossible to interpret; the one—per-one—million number could be pure coincidence since we know that children who have not received a measles vaccination actually have a higher rate of encephalitis! 'Encephalitis of unknown origin,' meaning a case for which no one can find a cause, can and does happen to small children, with or without vaccine. One day the child is healthy, the next he or she is fighting for his life.

"...Wakefield's report has provoked a storm of criticism... Some critics even question whether it should have been published at all...given the prediction among public health experts that its publication and the resulting publicity would cause a decline in vaccination rates... Health officials...have already noticed such a decline."

What about an even smaller group of children, small enough to escape detection by the IOM—smaller even than one-in-one-million—who might have the specific syndrome suggested by Wakefield and his colleagues?

Criticism of Wakefield's Report

The publication of Wakefield's report has provoked a storm of criticism, much of it published in the Lancet itself, and the validity of the Wakefield findings has been seriously questioned on a number of grounds. Some critics even question whether it should have been published at all, given its preliminary status and given the prediction among public health experts that its publication and the resulting publicity would cause a decline in vaccination rates. Not surprisingly, health officials in Wales (UK) have already noted such a decline.

Some critics believe that there may have been 'ascertainment bias' in the selection of study subjects. Wakefield et al state that the study group was comprised of a consecutive series of children with both chronic enterocolitis and regressive developmental disorder who had been referred to the department of pediatric gastroenterology at the Royal Free Hospital and School of Medicine in London. There is no statistical meaning in the fact that all subjects had both conditions—they were selected to participate for this very reason. But, critics suggest, there might also be no statistical meaning, due to ascertainment bias, in the fact that such a high percentage of the subjects also had experienced an observable response to MMR vaccine.

Although the high rate of MMR vaccine reaction in these children was presented as a finding, it may have been due to the fact that the pool of clinic patients from which study subjects were chosen might have contained a disproportionately large number of patients with prior MMR vaccine response. It had been known in the community that Wakefield's team had a special interest in studying the relation of MMR vaccine and inflammatory bowel disease. Thus, patients with inflammatory bowel disease (and possibly other conditions as well) who were also known to have had a vaccine response may have been referred specially to that clinic. Such biased case-ascertainment would exaggerate the significance of associations found in the study.

Had Wakefield's team been known to be interested in studying an association between IBD and blue eyes, for example, disproportionately large numbers of patients with IBD and blue eyes might have been referred to the clinic. If, then, a consecutive series of IBD patients was selected for study, the research findings might indicate an exaggerated association between IBD and blue eyes.

Critics point out that the most common endoscopic and/or pathological finding in this study group—ileal lymphoid-nodular hyperplasia—is also observed with significant frequency in the general pediatric population, both in connection with a variety of non-specific bowel complaints and without any corresponding symptomology. They suggest that, in the absence of evidence of a causal relationship between IBD and neurodevelopmental disorder, the observed co-occurrence of these conditions in this study population may be coincidental.

Critics also note the absence of any evidence of a causal relationship between MMR vaccination and IBD and/or pervasive developmental disorder (such as, for example, a finding of measles virus or measles vaccine fragments in intestinal tissue) as well as the absence of suitable study controls. These factors, along with possible ascertainment bias, create the likelihood that the observed associations are merely coincidental.

While the Wakefield report does describe a process by which MMR vaccination might cause IBD, which in turn might cause neuropsychiatric dysfunction, evidence presented in the report seems to contradict this theory: the observed onset of symptoms of IBD seems to occur significantly after, not before, the onset of behavioral symptoms.

The Wakefield report states that all twelve study subjects have regressive developmental disorder and that eight of them were observed to have begun to exhibit behavioral problems shortly after an MMR vaccination. The validity of using retrospective reports, mainly of parents, to establish a strong temporal association between vaccination and onset of neurodevelopmental disability is questioned by some critics.

While it may be a bit demoralizing, it is well known that our memories are not accurate enough to support medical hypotheses one way or the other. Researchers have conducted studies in which they cross-checked parents' accounts of their children's medical histories with the actual pediatric and/or hospital records—and often the parents have remembered things differently from the way they actually occurred. This is true of all parents, on all subjects, not just
parents of children with autism. Memory is a reconstruction, not a photographic record.

Memory also works by establishing landmarks. Any parent who experiences the trauma of regressive autism will naturally, due to the very nature of memory, seek to associate that devastating event with another significant event, such as his child's bad reaction to a vaccination. This is the way our minds work.

Evidence of No Association Between MMR Vaccine, Inflammatory Bowel Disease, or Autism

In the wake of the controversy surrounding the publication of the Wakefield report, researcher Eric Fombonne reviewed two large datasets. In the first, a review of the records of 174 French children with autism born between 1976 and 1985 showed no incidence of IBD. The second, a review of records of 201 British children with autism spectrum disorder born in 1987 or later (and thus assumed to have been exposed to MMR vaccine), also showed no incidence of IBD. In the May 2, 1998 issue LANCET, Heikki Peltola, M.D. and colleagues described a 14-year prospective study between 1982 and 1996 in which the health of three million vaccinated children was followed. In a prospective study researchers do not have to rely upon inaccurate memories or potentially incomplete medical records; instead they track subjects from Day One. Of 3 million vaccines given, Peltola et al found only 31 cases of gastrointestinal symptoms, all of which were short-lived. Not one of these 31 children developed autism. Over a decade's effort to detect all severe adverse events associated with MMR vaccine could find no data supporting the hypothesis that the MMR can cause either pervasive developmental disorder or inflammatory bowel disease.

"...I am often asked ... whether I have vaccinated my own children. The answer is yes ... and if we were to have another child we would vaccinate him or her as well..."

Finally, critics of the Wakefield paper have pointed out that it is merely a collection of case reports, not an epidemiologic study attempting to determine whether this phenomenon is real. A collection of reports can neither prove, nor disprove, a hypothesis. To prove a connection between MMR vaccine, IBD and neurodevelopmental disability we need sound epidemiologic research, a research priority NAAR has been actively promoting. (See "Parent Groups Propose New Autism Research Legislation" on page 4.)

The Autoimmune Question

Because children with autism have been thought to have a number of immune system anomalies many parents wonder whether autism may in fact be an autoimmune disease. In this scenario, an otherwise normal child's 'overactive' immune system mistakenly attacks his brain, leaving him with a full-blown case of autism.

This hypothesis is not far-fetched; researchers hypothesize that children can develop obsessive-compulsive disorder in this manner. Although so far no one has found evidence that autism can be caused by an autoimmune disorder, researchers have not ruled it out.

Could the MMR vaccine so overheat an already disordered immune system that it attacks the newly-vaccinated child's brain, leaving him or her with a lifelong disability?

So far, given the vast amount of data on the safety and dangers of vaccine, the answer is no. Is an autoimmune-vaccine theory of autism plausible? Yes. Do the facts to date support the theory? No.

Where Do We Go From Here?

I am also asked, both as a doctor and as Vice President-Medical Affairs for NAAR, whether I have vaccinated my own children. The answer is yes; both of our children received the full schedule of vaccinations, and if we were to have another child we would vaccinate him or her as well.

Nevertheless, at NAAR we are committed to investigating all possible causes of autism. As I noted above, the only way to resolve the issue of a connection between vaccination and autism is through rigorous epidemiological study of large populations of children. To that end, we have urged Congress to appropriate funds to undertake an epidemiologic study of the prevalence and causes of autism. This study could certainly include data on the vaccination histories of children, as well as the vital issue of children's exposure to environmental toxins, medical disease, and a host of other issues. As a parent of a child with autism, I feel the same passion to push the science forward that drives us all. Let's direct that passion into productive channels and get the results that will finally give us the answers—and the help—we seek.



Bailey, A. et al (1998). A clinicopathological study of autism. Brain, 121: 889-905.

Bauman, Margaret and Kemper, Thomas. (1994). Neuroanatomic observations of the brain in autism. The Neurobiology of Autism, 119-145.

Correspondence to the Lancet (1998): v. 351, no.9106; v. 351, no. 9112; v.352, bo. 9121; v. 352, no.9123.

Fombonne, Eric (1998). Inflammatory bowel disease and autism. Lancet, v. 351, no.9107.

Peltola, H. et. al. (1997). No measles in Finland. Lancet, v. 350, no 9088.

Piven, J. et. al. (1990). Magnetic resonance imaging: evidence for a defect of cerebral cortical development in autism. American Journal of Psychiatry, 147 (6): 734-739.

Tuttle et. al. (1996). Update: Vaccine Side Effects, Adverse Reactions, Contraindications, and Precautions. Recommendations of the Advisory Committee on Immunization Practices.

Wakefield et. al. (1998). Ileal-Lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children. Lancet, 351: 637-41.

Wentz, KR and Marcuse, EK. (1991). Diptheria-Pertussis vaccine and serious neurologic illness.... Pediatrics, 87: 287-297.

Eric London, M.D., is co-founder and Vice President-Medical Affairs of NAAR. He is a psychiatrist in private practice with a special interest in developmental disorders and an Adjunct Assistant Professor in Psychiatry at the University of Medicine and Dentistry of NJ. Dr. London is the father of a ten year old son with autism.

Catherine Johnson, Ph.D. is co-author, with John Ratey, M.D., of Shadow Syndromes. She is author of two other books and has published in numerous magazines. She has taught at UCLA and at UC-Irvine. Dr. Johnson is a member of NAAR's Board of Trustees and the mother of two children with autism.


Important disclaimer: The information on is for educational purposes only and should not be considered to be medical advice. It is not meant to replace the advice of the physician who cares for your child. All medical advice and information should be considered to be incomplete without a physical exam, which is not possible without a visit to your doctor.

Immunizations stop
disease from spreading.

Check with your family
doctor to see if you could benefit.