Fear Not: Vaccines Do Not Cause Autism

June 2005
by Steven Novella, MD

One of the most memorable scenes in The Big Lebowski , the classic 1998 Coen Brothers movie starring Jeff Bridges as “the Dude,” is when Walter Sobchak, played with overweight aplomb by John Goodman, shows up at the house of the young boy he believes stole the Dude’s car. Little Larry is, it turns out, the son of one Arthur Digby Sellers, a serial writer from the early days of television who happens to have written one of Sobchak’s favorite shows. But Sobchak can’t exactly address Arthur, not to his face anyway, because Arthur resides at the far end of the living room in an iron lung.

To most people alive today, the iron lung is not even a distant memory, it’s a relic of history, from the days when polio swept across the country in periodic waves, leading parents to yank their children from poolsides and seashores and sequester them inside, leaving desolate beaches in their wake. It was the disease that crippled a president, though most Americans had no idea how much trouble it was for Franklin D. Roosevelt just to walk from one side of the room to another. Most terrifying of all, it was a disease that made its victims wait in terror for the final verdict: The virus would come and then leave, and only a week or two later would paralysis set in – if, in fact, it was going to set in. Some survived polio with no lasting effects; others ended up in an iron lung.

Thanks to the good doctors Salk and Sabin, polio is practically gone in America, not a menace since the early 1950s. Smallpox, too, has vanished, meaning that nobody is scarred anymore in the manner of Sadie Burke, the smallpox victim from Robert Penn Warren’s great 1946 novel All the King’s Men . Were it not for vaccinations, thousands, even millions, of people would still be crippled and scarred by these diseases every year; thanks to vaccines, very few are.

Vaccines are one of the most successful programs in modern health-care, reducing and in some cases even eliminating serious infectious diseases. Public support for the vaccination program remains strong, especially in the United States, where vaccination rates are currently at an all-time high of 93 percent. Yet despite their long history of safety and effectiveness, vaccines have always had their critics – some parents and a tiny fringe of doctors question whether vaccinating children is worth what they perceive to be the risks.

Some fears of vaccines have a basis in truth. The old polio vaccine could, on extremely rare occasion, cause polio (although the current polio vaccine does not carry this risk). Millions have been spared this scourge, but that’s small consolation to the parents of a sick child. The swine flu vaccine caused an outbreak of Guillaine Barre syndrome (an inflammatory illness causing nerve damage) Today, the flu vaccine can make people achy, making one sick to prevent sickness.

But the media and the internet continue spreading fears about vaccines that are both unreasonable and unproven. In the past, false fears about vaccine safety have resulted in plummeting compliance rates, in Nigeria and Great Britain, followed, predictably, by disease outbreaks. The United States, by contrast, has proven very resilient to such fears. The reasons for this are unclear, but some believe it may be due to the fact that vaccinations here are compulsory for school attendance.

In recent years, concern has focused on a possible link between childhood vaccines and the dreaded neurological disorder autism. The stakes are high on both sides of this issue. If vaccines are unsafe, then millions of children are being placed at risk. The resulting injuries could have a catastrophic cost, not only in the lives they harm but in healthcare and special services costs to an already overburdened society. Victims and their families would require and deserve compensation. Further, since the United States exports vaccines to much of the world, there are potential implications for our foreign relations.

On the other side, unwarranted fear about vaccine safety could reduce public confidence and threaten the vaccine program. A decrease in vaccine compliance would result in an increase in serious and potentially fatal infectious illnesses. Baseless class action suits against vaccine manufacturers could unfairly bankrupt companies or keep pharmaceutical companies out of the risky vaccine business, leading to potential shortages and a greater increase in preventable infections.

We can’t afford to be wrong on this issue. We cannot simply err on the side of caution, because caution resides equally on both sides. We need to know if our vaccines are safe and we need to have confidence in the institutions that monitor and regulate the vaccine program. Fortunately, we have a tool that can reliably answer such important questions: It is called science.

1: Do vaccines cause autism or other developmental disorders?

Autistic children typically develop normally for the first year or two of life, and then start to lose cognitive ground. They often retreat into their own world, stop talking or making eye contact, and do not interact with others. The disorder is devastating to parents; having an autistic child has a profound effect on the entire family. Although the exact cause or causes of autism are not clearly known, there is solid consensus among researchers that the evidence for a genetic cause is very strong and growing. The evidence for this consensus is too casually dismissed by vaccine-autism proponents, who simply state that today scientists are overly enamored of genetics.

Historically, the incidence of autism has been reported to be around 3-4 cases per 10,000 people. Starting in the early 1990’s, however, the number of autism diagnoses began to climb steadily, to about 13 per 10,000. When you consider not only autism but also all related disorders (collectively called pervasive developmental disorders, or PDD), there has been a ten-fold increase, by most estimates, in the number of diagnoses being made, not only here but in other Western nations. Some believe that we are in the midst of an autism epidemic.

An “epidemic” suggests an environmental cause for autism, rather than a purely genetic cause; in other words, if we’re in the midst of an epidemic, than something we eat or inhale or come in contact with, whether a food or a germ or a toxin in the environment, is to blame. So some people have begun to question the genetic hypothesis for autism and started searching for a toxin or other environmental trigger. In this camp are many parents of autistic children, who understandably have the sense that something has “happened” to their child. As Amy Carson, founder of Moms Against Mercury, writes on her group’s web site: “My son was born a healthy child. As time went on and the more he was vaccinated, the more he started to change.”

Carson and others turned their attention to vaccines. Over the same period of time that autism rates were soaring, the childhood vaccine schedule was also increasing, with more and more vaccines being added to the routine program. And many vaccines happen to be given around the same time that the symptoms of autism first become apparent. Two particular items came under fire: the first was the mumps, measles and rubella vaccine (MMR), and the second was thimerosal, a mercury-based preservative used in some vaccines at the time (and since removed from all vaccines given to American children, but much more on this later).

In 1998, the researcher Andrew Wakefield and some colleagues published a study in the Lancet , a prestigious English medical journal, that claimed to show a connection between the MMR vaccine and autism. Wakefield’s theory was that the MMR vaccine, which contains a live virus, can cause in susceptible children a chronic measles infection. This in turn leads to gastrointestinal disturbances, including what he calls a “leaky gut” syndrome. This then allows for certain toxins and chemicals, including those from bread and dairy that are normally broken down by the gut, to enter the bloodstream, where they can get access to and damage the developing brain.

Although the study was small, and the evidence was considered preliminary, this article sparked a firestorm. The vaccine-autism movement was soon in full swing.

The case against thimerosal has been well documented in Evidence of Harm , a new book by journalist David Kirby (although Kirby makes it clear that he’s not taking sides in the debate, just reporting about it). The alleged link between thimerosal and autism has also been publicized, with approval, by the environmental lawyer Robert F. Kennedy Jr., in the online magazine Salon and elsewhere. The radio talk show host Don Imus has used his radio show to rage against what he believes to be a horrific crime against humanity. And many parent activist groups have formed to advocate specifically for the mercury-causes-autism hypothesis, or to lobby against thimerosal. They have refined their argument to make the following claims:

During the 1990s, the vaccine schedule was increased to include many vaccines that use thimerosal as a preservative to prevent the growth of bacterial contaminants. Thimerosal contains ethylmercury, a known neurotoxin. When the total doses of ethylmercury for all vaccines in the schedule are added together, in some cases they exceed by many times the Environmental Protection Agency (EPA) or Food and Drug Administration (FDA) safety limits for mercury. The developing brains of infants, they argue, are probably more susceptible to the effects of mercury. All of which spells danger.

The specter of mercury can be compelling. As Carson, of Moms Against Mercury, writes: “I was outraged that I was not told that the most powerful neurotoxin was going to be injected in my newborn child. It has devastated and changed our lives forever.” She and others believe there is a compelling overlap between autism and the signs and symptoms of known mercury toxicity: speech delay, sensory hypersensitivity, and motor symptoms. They point to toxicological studies that suggest autistic children may have been exposed to more mercury than their non-autistic peers (although scientists feel the validity and implications of these studies are still unclear). They also claim that autistic children may also be especially susceptible to mercury, or may have impaired mechanisms for clearing mercury from their system. Some even link thimerosal and MMR, arguing that mercury impairs the immune system, allowing the live measles virus to cause a damaging infection.

Promoters of the vaccine-autism hypothesis also claim that the Centers for Disease Control (CDC), which is responsible for monitoring vaccine safety, and the FDA, responsible for approving vaccines as safe and effective, are engaged in coverup. Robert Kennedy Jr. writes, “The story of how government health agencies colluded with Big Pharma to hide the risks of thimerosal from the public is a chilling case study of institutional arrogance, power and greed.”

There are numerous complex issues involved in this controversy. I will address each one in turn.

2: Is there a real autism epidemic?

There is no doubt that between 1990 and 2001 there was a dramatic increase in the number of diagnoses of autism and related disorders, about 10 times by most estimates. The scientific community accepts this fact, which has been demonstrated in numerous epidemiological studies. Many autism activists believe this increase represents a genuine, and disturbing, epidemic.

However, the scientific community is fairly united behind a different interpretation. Dr. Fred Volkmar, a child psychiatrist at Yale and world-renowned autism expert, points to two other important forces. First, the range of symptoms that are considered to be “autistic” has greatly expanded. Doctors and parents now speak of an “autism spectrum disorder” – you can be a little autistic, more autistic, less autistic.

Autism is not a specific disease; you can’t do a blood test or a CAT scan to see if it exists. It is, rather, a disorder: It is defined solely by the constellation of signs and symptoms that it displays. This means that the number of autistics can be greatly expanded or contracted by changing the criteria for diagnosis. During the 1990s, many milder forms of the disorder were being recognized and diagnosed, as was a broader list of possible manifestations, and in fact new diagnoses, such as Asperger’s syndrome, began to be considered on the autistic spectrum. So a wider net was being cast. Here’s an example to show what I mean: If you had a loose category called “intelligent,” and it was understood to mean people who could do calculus well, then only 1 percent of Americans might be “intelligent.” But if you changed the definition to include everyone who can carry on a decent conversation about politics, or everyone who can do algebra, then you’d diagnose a lot more people as “intelligent.” That makes a disorder very different from a disease like cancer: You either have cancer or you don’t, but autism isn’t quite so clear-cut.

In addition to changes in diagnoses, systems for surveillance were also being increased – more nets, with tighter weaves, were being cast. Because of the availability of special services and early-intervention programs to help autistic children, there was a huge effort to perform routine screenings of all children by qualified professionals, which led to a great increase in the number of children seen by doctors looking for autism.

Finally, some health-care programs have provided an incentive to give children a diagnosis that is covered by insurance, and autism fits the bill. That doesn’t mean that parents are eager to have their children diagnosed as autistic, of course, but that some doctors might, to help the parents get services, be looking for a diagnosis for a child with developmental problems and might settle on a diagnosis of autism.

This issue, diagnosis, is absolutely critical to the question of vaccines and autism. If the increase is an artifact of expanded diagnosis and increased surveillance, then there is no epidemic. This fact alone would kill the vaccine-autism hypothesis, which is based largely on the correlation of increasing vaccines and increasing autism.

Defenders of a link have rejected this very important argument. Kennedy, for example, has wondered, if earlier cases of autism were simply going undiagnosed, where are all the autistics who are now in their twenties? But this observation is a bit too simple. First, many of the previously undiagnosed cases were toward the milder end of the spectrum. Also, and perhaps more important, autism is known, in many cases, to improve with age (especially in those towards the milder end of the spectrum). Finally, there is no reason to believe that adults with autism undiagnosed as children should somehow be apparent to casual observation. In other words, we still have good reason to believe that increased surveillance and more liberal diagnosis accounts for much, potentially all, of the spike in diagnoses.

Those who believe in the vaccine-autism link have a number of studies they point to. None, however, is very convincing. For example, a study recently published (but not peer-reviewed) by the MIND institute and authored by Dr. Robert Byrd concluded that the rising rate of autism diagnosis in California was not due to an influx of cases into the state, a change in diagnosis criteria, or a previous mislabeling of autism as mental retardation. This study is widely used to dismiss the “increased surveillance” explanation. But this study did not look at the effects of increased awareness, and therefore reporting, of autism or programs to increase surveillance. Experts also seriously doubt the credibility of this analysis, citing specifically that it is not peer-reviewed.

A much better study suggests that actual autism rates are not increasing; the study therefore supports the “increased surveillance” hypothesis. From 1992 to 1995, Suniti Chakrabarti and Eric Fombonne studied the incidence of autism in Stafford, England. Ten years later, they repeated their exact methods of diagnosis and surveillance with a cohort of children born in 2002, at the peak of the alleged autism “epidemic.” They found no difference in the rates of either pervasive developmental delay or autism between the two groups. This study essentially controls for diagnostic and surveillance differences, and therefore is very powerful evidence that there is, in fact, no autism epidemic. The true incidence is flat over this critical period of time.

In other words, in Stafford, England, when you use the same diagnosis criteria in 1995 and again in 2005, you don’t find any more autism. You only find more autism if you change the definition or surveillance of autism.

To be clear, it is not possible for such studies to rule out a small signal in the noise. In fact, it is logically impossible to prove that something does not exist. Responsible scientists will always be modest in their conclusions, saying, for example, that while their study has not detected a rise in autism, there could always be a real increase in autism rates, one too small to be detected by the methods used. But there is no reason to believe that this is so. There are many things that could, in theory, exist: alien abductions, secret plots to assassinate Gov. Jodi Rell, powerful healing benefits from eating carpenter ants. But as a rule, we should only believe in them if we have evidence – not just unproven hypotheses.

3: Does the MMR vaccine cause autism?

Subsequent to the seminal article in the Lancet, many follow-up studies were performed to see if autism is truly correlated with the MMR vaccine. It is important to note that epidemiological studies cannot prove a cause and effect, that MMR causes autism. They can only show a correlation: When this goes up, so does that. However, if there is true causation, then epidemiological studies should show multiple correlations. For example, autism should go up as MMR vaccinations do, and it should go down when vaccinations go down; they should go up and down predictably over time, depending on when the vaccinations are given; the autism rate and severity should correspond to the size of the vaccine dose. The more these correlations hold up, the greater the case for a cause-and-effect relationship. Finally, of course, biological data should show how MMR might cause autism – in other words, there should be actual evidence for the “leaky gut” theory, or some other theory.

As the follow-up studies started being published, however, it became more and more clear that there was no link between MMR and autism. For example, a study in the British Medical Journal found that autism rates continued to climb in areas where MMR vaccination rates were not increasing. Another article there found no association with MMR and autism or GI (gut) disorders. Other studies showed no difference in diagnosis rate of autism either before or after MMR vaccine, or between vaccinated and unvaccinated children. Most recently, a study found there was no decrease in autism rates following removal of the MMR vaccine in Japan.

In May of 2004 the Institute of Medicine (IOM) reviewed all of the MMR-autism data available to date and concluded that there was no association and that the case is essentially closed – a conclusion confirmed by still later studies, such as the one in Japan.

Believers in the MMR-autism hypothesis largely dismiss these findings as biased. They also dismiss the findings of the larger and more powerful epidemiological studies. Bernard Rimland, who leads the Autism Research Institute, rejected the IOM report, writing that the evidence “does not exclude the possibility that MMR vaccine could contribute to ASD in a small number of children.” Rimland interpreted this as support for a link. Rather, it merely reflects the logical necessity I referred to above: It is impossible to prove a risk of zero.

In May 2004, 10 of Wakefield’s co-authors on his original paper withdrew their support for its conclusions. One author, Dr. Simon Murch, stated: “There is now unequivocal evidence that MMR is not a risk factor for autism – this statement is not spin or medical conspiracy, but reflects an unprecedented volume of medical study on a worldwide basis.” The editor of Lancet also announced that they withdrew their endorsement of the paper, and cited as part of the reason an undisclosed potential conflict of interest for Wakefield, namely that at the time of its publication he was conducting research for a group of parents of autistic children seeking to sue for damages from MMR vaccine producers.

Sadly, the controversy led to decreased vaccination of children in England. There was an increase in measles, mumps, and rubella, each of which can, in rare circumstances, be fatal.

4: Does thimerosal cause autism?

There is little doubt, and no controversy, that mercury, the major component of thimerosal, is a powerful neurotoxin, or poison to the brain. However, toxicity is always a matter of dose. Everything is toxic in some dose; too much Vitamin C can kill you. So the real question is whether the amount of mercury given to children in vaccines containing thimerosal was enough to cause neurological damage. On this question, there is much controversy.

Proponents of the mercury hypothesis argue that the ethylmercury found in thimerosal was given in pulse doses (all at once) exceeding EPA limits. This load of mercury should be added to prenatal vaccine loads possibly given to mothers, and to other environmental sources of mercury, such as seafood. Furthermore, underweight or premature infants received a higher dose by weight than larger children. Some children, they argue, may have a specific inability to metabolize mercury, and perhaps these are the children who become autistic.

But wait. Ethylmercury, the form of mercury found in thimerosal, is not as toxic as methylmercury; the EPA limits were based upon the more toxic form, and had a built-in safety margin. Recent studies also show that ethylmercury is removed more quickly than methylmercury and probably does not build up in the body, so doses would not have a cumulative toxicity. Plus, it seems that children have a greater capacity to metabolize mercury than adults.

There are other data and arguments, too complex to explain here. The bottom line is this: Yes, there is reason to believe that thimerosal, in sufficient doses, could be toxic and cause neurological damage. But the best data we have so far indicates that thimerosal probably was not given in high enough dose to be neurotoxic, although there is room for reasonable doubt. So, given the uncertainty, the FDA did recommend the removal of thimerosal from childhood vaccines, and by January 2003 the removal was complete (according to the CDC, all lots of childhood vaccines containing thimerosal expired by January 2003), although it is still found is some flu vaccines and some multi-dose vaccine vials exported outside the United States.

In any case, the biological data can illuminate a possible mechanism of damage, but it can never, by itself, prove that thimerosal actually did cause autism. That evidence must come from epidemiological studies. In other words, we need not only a theory of how thimerosal might be able to cause autism, we need numbers to show that where there was thimerosal, there was autism. We always need both.

At the time of the IOM review in 2004, there were five published epidemiological studies on thimerosal and autism that showed no link. Together, they provide strong evidence that thimerosal does not cause autism. Subsequent to the IOM report, there has been one additional published study, for a total of six, from Great Britain, the United States, Sweden and Denmark, all showing a lack of correlation. These studies were meticulously reviewed by Sarah Parker and others, who published their conclusions in the prestigious journal Pediatrics in September 2004.

Parker also reviewed the only epidemiological studies that claim to show a link. They are all published by the same authors, the father-and-son team of Mark and David Geier. Parker and her coauthors concluded that these studies contained fatal methodological flaws rendering their conclusions either invalid or uninterpretable. Their final conclusion echoes the IOM report: “Studies do not demonstrate a link between thimerosal-containing vaccines and [autism], and the pharmacokinetics of ethylmercury make such an association less likely. Epidemiological studies that support a link demonstrated significant design flaws that invalidate their conclusions.”

To date there is not one well designed, peer-reviewed study that shows there is a link. Again, there still could be a link, but we have no reason to believe there is.

For the vaccine, autism-research and medical communities, the scientific case was all but closed. But this has not ended the public controversy. Parent groups such as Moms Against Mercury and Safe Minds have not accepted this consensus. Critics like Kennedy continue to argue that the studies debunking a link are not reliable.

5: Conspiracy Theories

If the scientific data shows, according to a solid consensus of scientific opinion, that there is no reliable evidence for an autism epidemic, and that there is no statistical correlation between MMR and GI symptoms or autism, or between thimerosal and autism, how do believers maintain their claim? Mostly by dismissing the scientific consensus as a result of bias, of conspiracy and of influence from the pharmaceutical industry.

In the case of thimerosal and autism, believers argue that the FDA, CDC, IOM, the World Health Organization (WHO) and the American Academy of Pediatrics (AAP) are all involved in the cover-up: the FDA and CDC in order to hide their prior incompetence; the AAP because they cannot accept the horrible notion that vaccines could be harming children; the IOM, simply to please the CDC; and the WHO – well, who knows, but all of them are influenced by the powerful pharmaceutical industry, whose only motive is to protect their profits.

Josh Day, writing on the Health and Beyond website, captures the conspiracy attitude when he writes, referring to a CDC meeting at a site called Simsonwood, “In a chilling Wannsee conference-style meeting, top brass in the pharmaceutical empire, FDA and CDC lackeys, and their scientist dogs have concocted their own Final Solution in ‘handling’ the thimerosal problem. These men met and decided to cover up a study that proved a connection between thimerosal and autism, as well as other serious disorders.” In other words, the CDC are like Nazis.

Now, to be fair, there are legitimate factors that have helped foster conspiracy theories. The CDC, for example, has made many statements that suggested they were trying to “manage” the perception that vaccines are unsafe. They have also tried to hide preliminary evaluations of their data. Many of the experts at the CDC and FDA have ties to the pharmaceutical industry, have potential conflicts of interest, or even go back and forth between industry and government. These are all legitimate concerns, but they do not necessarily add up to a conspiracy. Let’s look closer at the situation.

The CDC is responsible for running, and therefore promoting, the vaccine program in the United States. They are also responsible for monitoring its safety. Some have argued that this is an inherent conflict of interest, and they have a point. For this reason, and in response to criticism, these two functions were recently separated at the CDC.

But the CDC has to deal with a very serious dilemma. Anything that serves to undermine public confidence in the safety of vaccines may decrease compliance and thereby increase the rate of preventable diseases, causing harm and even death. In other words, if people begin to doubt vaccines, children may die. The CDC’s job is to protect the public health, and that means doing what they can to quell false fears. In order to accomplish this, they have chosen to play their cards close to their vest: to monitor vaccine safety in secret and then only make concerns public when they have been confirmed. This will avoid countless false alarms, that would constantly send waves of fear about vaccine safety across the country.

The downside to this, however, is that the necessary secrecy fosters distrust, especially in the internet age, with well-meaning activists spreading rumors and exaggerations. The CDC must strike a delicate balance, but in the end theirs is probably a no-win situation. Believers in a CDC coverup interpret all of the CDC’s skittishness and statements about containment and perception as evidence of a conspiracy. However, all of this can also be interpreted as a sincere attempt to protect the public from an unwarranted fear that could lead to reduced vaccination and many, many sick children.

On close inspection, the conspiracy theories are not compelling, or even logical. But once the idea of a conspiracy is accepted, it can be used to dismiss all contradicting evidence, and to explain away the lack of confirming evidence. Conspiracy theorists quickly become insulated from any possible refutation creating a closed belief system reminiscent of cults.

Also, grand-conspiracy theorists take a very black and white view of the world. David Kirby’s book, Evidence of Harm , told largely from the point of view of activist parent groups, reflects this attitude. All of the thimerosal skeptics in his book are described as “aloof,” “cold,” “arrogant,” and “dismissive.” Believers are always charming, friendly, and sincere. Dark-suited pharmaceutical representatives are literally skulking in the background, talking into their cellphones in conspiratorial tones.

The weakest aspect of the conspiracy theories, however, is that the motivation of the “villains” is simply not credible. The IOM, for example, has no motivation that anyone can point to for not giving an honest report of the data. Also, they based their report on published data – data any scientist in the world can review for herself. If they gave a biased report, their malfeasance would be transparent to their colleagues, and their reputations would be ruined. Further, if there were a thimerosal-caused autism epidemic, the truth would eventually come out; it could not be hidden from the world. When it did come out, the IOM committee members would be further disgraced. The same would be true of Sarah Parker and the other authors of the Pediatrics paper reviewing all the data. No scientist wants to be on the wrong side of history.

Officials at the CDC and FDA would also eventually be found out, and they know it. There have been regime changes at these institutions, giving the opportunity for newcomers to blame any wrongdoing on predecessors, yet no revelations have come. The AAP is a professional organization, charged with improving health-care for children, what possible motivation could they have for condemning millions under their watch to neurological damage?

Much of what is driving the conspiracy theories is based on a misunderstanding of the scientific process. Kirby reports on the argument that the CDC studies’ being altered in response to comments from reviewers was “suspicious.” But this alteration is almost universal practice – part of the peer-review process. Many parent groups cannot understand how preliminary data can show a link, and then later analysis show no link. But epidemiology and statistics are very complex enterprises. No study is perfect, and all involve choices that affect the outcome. Many rounds of analysis and peer-review are often required to achieve reliable results.

Finally, we must all be wary of “confirmation bias.” It is human nature to look for evidence that confirms what we already believe. Once someone believes in a link between mercury and autism, it’s easy to collect all the data that seems to support a link and become increasingly compelled by the shear volume of evidence. It takes critical thinking, and some experience, to fight against this basic human tendency and to put all the data into some objective perspective.

6: What should we believe?

We have to try to look at evidence as best we can. And there is now adequate scientific evidence to conclude that there is no epidemiological association between either MMR or thimerosal and autism. At present, the evidence strongly suggests that the autism “epidemic” is largely an artifact of increasing diagnosis. But there is room for the possibility of a real increase in this disease, and if that’s the case, then searching for the real cause or causes of autism is critical. There is also a strong and growing scientific consensus that autism is largely genetic – a fact that is incompatible with the vaccine-autism hypothesis.

Regardless of this evidence, many countries have removed thimerosal from childhood vaccines – the United States phased it out by early 2003. This removal will provide a final test of the thimerosal-autism hypothesis. If there is a real epidemic caused by thimerosal, then autism numbers should drop precipitously to, or near, 1990 levels, back before the increase in the vaccine schedule resulted in higher thimerosal doses.

Kennedy is already declaring victory on this count, citing “yet to be published” data by none other than Geier and Geier, the father-son authors of so many flawed studies, to suggest a drop in autism numbers. Kirby also gives a preliminary report of a trend downward in the California numbers, but he agrees it’s too early to tell (and autism experts feel these numbers are unreliable). It will probably take another two years before epidemiological studies conclusively show the true effect of removing thimerosal from American childhood vaccines.

Vaccines are a safe and effective public health measure, and they deserve broad public support. Because they are given to such a large segment of society, and are in fact mandatory in the United States for those attending school, it is vital that we have in place mechanisms to ensure both the safety of vaccines and public confidence in them. Zero risk is an impossible standard, as everything in life comes with some risk. It is more reasonable to consider risk versus benefit. What we can say about vaccines is that any potential risk is very low, and the benefits are both substantial and proven. In short, the benefits clearly far outweigh the risks.

Vaccine controversies, real or imagined, can do real harm to the public. There was a time in this country, before we had the vaccines we have now, when people regularly suffered, even died, from influenza, smallpox, measles and polio. Barely more than 50 years ago, going to the beach during a polio epidemic was dangerous and terrifying. Vaccines have changed our lives for the better, profoundly. Vaccines are also one of the great social justice achievements. Poor people suffer much more when they’re sick than rich people do; when a population manages to reduce or eliminate the incidence of a disease, it’s poor and oppressed people whose lives change most for the better. So while it’s always important to question our medical and scientific establishments, asking the hard questions, it’s also important not to throw out the great progress we have made.