When you spend some time dispassionately comparing thimerosal exposure from vaccines and autism rates, one fact becomes quickly apparent: there is no correlation whatsoever between the exposure of children to thimerosal and rates of autism. This is true in the United States, in Canada, in the United Kingdom, and several other European countries. If the thimerosal theory of autism were true, you would expect exactly the opposite pattern.
Experience in the United States
Thimerosal was developed in the United States and used extensively here. Changes in thimerosal levels were also most dramatic here. So, if thimerosal were responsible for the autism ‘epidemic’, that fact should be easiest to glean by studying this country.
A Quick History of US Thimerosal Exposure
Vaccinations recommended for universal childhood administration in the US have varied greatly over time. In 1900, the smallpox vaccine was the only one administered to children. By the 1960s, children routinely received five vaccines, for protection against diphtheria, pertussis, tetanus, polio, and smallpox, with as many as eight shots by two years of age because certain vaccines required multiple inoculations. In the mid-1960s, the MMR vaccine was recommended for all infants in the US (a booster was added in the mid-1990s). And, starting in the early 1990’s, additional vaccinations were introduced to protect against other contagious diseases, including Haemophilus influenza type B (Hib B), Hepatitis B, and chicken pox (varicella).
Contrary to much public belief, only a limited number of these vaccinations have ever contained thimerosal and its active ingredient ethyl mercury. Before 1928, when thimerosal was developed, the average dose of ethyl mercury from vaccines was 0 micrograms. Between 1928 and 1991 (when additional vaccinations were introduced containing thimerosal), the only vaccinations containing thimerosal were those for diphtheria, pertussis, and tetanus which were originally administered in separate shots but which in the 1950s were combined into a single inoculation that needed three doses to reach maximum effectiveness. A course of immunization with the DPTvaccine consisted of three doses starting at two months, with an interval of one month between each dose. Each dose of the standard DPT vaccine contained 50 micrograms of thimerosal, which included 25 micrograms of ethyl mercury (a microgram (mcg) is one millionth of a gram). Therefore, before 1991, by 16 weeks total thimerosal exposure was no more than 150 micrograms, of which 75 micrograms were ethyl mercury.
The MMR vaccine which was introduced in the mid-1960’s never contained thimerosal. Neither did the vaccines for smallpox and polio. In 1991, Hib B and Hepatitis B vaccines became part of the recommended US vaccine schedule; most formulations of these contained thimerosal. The addition of these new vaccines increased ethyl mercury levels to a total of 187.5 micrograms in the first 7 months of life, 75 mcg from 3 doses of DTP, 75 mcg from 3 doses of HIB, and 37.5 mcg from 3 doses of HepB. This 187.5 mcg maximum dose of ethyl mercury within the first 7 months (375 mcgs of thimerosal) lasted for about 10 years (researchers focus on the first 7 months of life, because this is when the greatest concentration of vaccines is given, and the highest thimerosal exposure occurs).
In 1998, prior to any hypothesis that thimerosal might cause autism, the US government ordered a review of the total thimerosal exposure in vaccines. As discussed below in detail, at this time little was known about the effects of ethyl mercury in humans or animals, due to few studies having been done. However, the thimerosal review had been ordered and researchers used the best proxy they could find for which they had solid information – methyl mercury. It was strongly suspected but not known for sure that methyl mercury was significantly more problematic than ethyl mercury based upon the body’s ability to eliminate the two sources of mercury – methyl mercury cleared much more slowly. Nonetheless, methyl mercury standards were used to be conservative.
When the FDA calculated in 1999 that some children could receive thimerosal / ethyl mercury exposures that exceeded the conservative standards that had recently been adopted, in an abundance of caution the American Academy of Pediatrics recommended that thimerosal be removed from routine childhood vaccines in the U.S. By 2000, new lots of all Hib B and Hepatitis B vaccines in the US contained at most trace amounts of thimerosal. By March, 2001, all vaccines recommended for children (at that time) became available in formulations with at most trace amounts of thimerosal, including DTP. By 2001, total exposure (in the first 7 months of life and thereafter) had plummeted to trace levels.
In 2002, inactivated flu vaccine, frequently administered from multi-dose vials that contain thimerosal, was newly recommended for children aged 6 to 24 months in the US. This flu vaccine contained up to 50 mcg of thimerosal (25 mcgs ethyl mercury). Since less than 20% of children with ages between 6 and 24 months actually got a flu shot, the actual level of ethyl mercury from this shot on a population basis was never that great. And, by 2008, a greatly increasing percentage of flu vaccine shots given to children aged 6-24 months were thimerosal free.
In summary, thimerosal exposure through childhood vaccines has changed greatly over the last 20 years. Based on available estimates, the maximum ethyl mercury dose from recommended vaccines for children up to 7 months old varied from 75 mg in 1990 to 187.5 mg in 1999, to less than 17.9 mg in 2004 (and for most children, much less than that). Thimerosal levels are now 10+ times lower than at their peak.
Removal of Thimerosal in the United States
The most important study on thimerosal done to date was completed in early 2008 using data from the California Department of Developmental Services (“DDS”). The DDS administers a statewide system of regional centers that serve persons who are substantially disabled because of autism, mental retardation and other disabilities. It is probably the most complete dataset of developmental disorders in the US and therefore uniquely useful for this type of measurement. This study measured rates of prevalence of autism in California children on an absolute basis from 2001 to 2007 and tracked increases in prevalence in an attempt to determine if the almost complete removal of thimerosal in the US by 2001 caused a reduction in autism prevalence. The conclusion of the California study was the following:
The estimated prevalence of autism for children at each year of age from 3 to 12 years increased throughout the study period. The estimated prevalence of DDS clients aged 3 to 5 years with autism increased for each quarter from January 1995 through March 2007. Since 2004, the absolute increase and the rate of increase in DDS clients aged 3 to 5 years with autism were higher than those in DDS clients of the same ages with any eligible condition including autism.
To translate, autism rates went up quarter by quarter for each quarter during the study period and at a faster rate than for any other development conditions. The removal of thimerosal from US vaccines did not even cause a blip in the continuing increase in autism prevalence. While this increase in autism prevalence can be explained in part by factors such as the increasing recognition of the diagnosis, substitution of the autism diagnosis for other ailments like mental retardation and learning disorder, and parental pressures in seeking a diagnosis and services, the theory that autism is caused by thimerosal is in direct contradiction to the data returned – you would have at least expected a decline of some sort in prevalence instead of an acceleration of the existing trend.
Past Changes in Autism Rates in the US do not Vary Consistently with Thimerosal Exposure
The ability to statistically track any connection between autism and thimerosal would have been clearest when thimerosal levels were reduced by 90% or more between 1999 and 2002. This is a drastic change in exposure. However, this is not the only period when changes in thimerosal exposure occurred quickly, such that any connection between thimerosal and autism should show up in varying rates of autism.
For instance, ethyl mercury exposure from thimerosal went from 0 mcgs to 75 mcgs between 1928 and twenty or thirty years thereafter. While this is not a huge change on an annual basis, the percentage change is enormous, in fact infinite. If thimerosal is responsible for the autism epidemic as many claim, then it stands to reason that autism rates should have shot up in the 1940s and 1950s. There is no evidence that this occurred.
Also, thimerosal exposure stayed steady throughout most of the 2nd half of the 20th century until the early 1990s when several new vaccines containing thimerosal were introduced. But, there is much evidence that autism rates had already started to rise in the 1980s and before. This also does not make sense in a world where thimerosal causes autism – changes in autism rates should lag changes in thimerosal.
Moreover, in the last several decades, there is only one period of time when thimerosal exposure increased consistently and significantly in the US. That is the period between the birth years 1990 and 1994, when exposure more than doubled in a two year period. Under the thimerosal theory, one would expect a sharp rise in autism prevalence followed by a plateau that was maintained for almost a decade (until thimerosal was removed). In contrast, the prevalence of autism showed a smooth upward trend throughout the 1990s that has been maintained to date.
Experience in Other Western Countries
While the removal of thimerosal is a very expensive proposition, because it forces several common vaccines to be injected from costly single dose vials, several countries other than the US (and before the US) have been able to afford to do so. And, similar to the US experience after 2001, the removal of thimerosal from most vaccinations given children in these countries has had no effect on autism rates. This finding makes no sense under a theory that thimerosal is responsible for the autism ‘epidemic’.
Thimerosal was removed from mandated childhood vaccines in Sweden and Denmark in 1992 / 1993. It was removed in Canada in 1996. If thimerosal were a / the cause of autism, then there should have been a dramatic decrease in autism rates in these countries during the 1990’s. This was not the case as has been shown in numerous studies.
The Stehr-Green study (2003) reported the incidence of autism in Sweden and in Denmark from 1987 to 1999. The results for both countries were similar. Autism incidence increased throughout the study period and continued to increase (although with some fluctuation) after elimination of thimerosal.
Madsen, Lauritsen et al (2003) used the same dataset as the Stehr-Green study but evaluated Denmark only but in greater detail. This study expanded the Denmark information to include 1961-70 (when the cumulative ethyl mercury dose was 200 micrograms in the first 15 months of life), 1970-92 (when the dose was 125 micrograms in the first 10 months of life), as well as 1992-2000 (when vaccines in Denmark did not contain thimerosal). The incidence of autism was stable until 1990 and thereafter increased throughout the study period. There was no association at all between thimerosal levels and autism.
Another Danish study used a different database. Hviid, Stellfild (2003) used the Danish Civil Registration System to compare rates of autism in children who received thimerosal containing vaccines (DPT only) to those who receive thimerosal free vaccines. The study concluded that rates of autism and other autistic-spectrum disorders did not differ significantly between children vaccinated with thimerosal-containing vaccine and children vaccinated with thimerosal-free vaccine. Also, the study found no evidence of an association between amount of thimerosal (in 25 microgram increments) and rates of autism spectrum disorders.
A 2006 Canadian study by Fombonne et al found no significant effect of thimerosal exposure on the steadily increasing rates of pervasive developmental disorders (“PDDs”) including autism in Canada for kids born between 1987 and 1998, including those born after thimerosal was removed from Canadian vaccines. In fact, the study showed a significantly reduced level of PDD in kids exposed to thimerosal (82.7 without thimerosal to 59.5 with it).
Thimerosal Variation from Country to Country Does not Explain Autism Rates
If the thimerosal hypothesis is true, there should also be big differences in autism prevalence between countries with different levels of thimerosal exposure. For instance, the US autism rate in the 1990’s (187.5 micrograms of ethyl mercury in the first 7 months) should have been much higher than that of Sweden and Denmark and Canada in the 1990’s (trace exposure to ethyl mercury). No significant difference in autism rates has been recorded between those countries and the US.
The UK was another country where levels of thimerosal exposure were lower than that in the US. This is because the only thimerosal containing vaccine used in the UK was the DTP vaccine. When the US thimerosal exposure peaked in the 1990’s, UK exposure was less than half of that in the US (75 micrograms to 187 micrograms of ethyl mercury from thimerosal). However, the UK has consistently recorded comparable if not higher rates compared to the US.
Also, if autism rates varied with thimerosal exposure, there should have very different rates of growth of the diagnosis of autism. The US growth rate in the early 1990’s should have greatly exceeded that of countries like Denmark (thimerosal removed in 1992) or the UK (thimerosal from Hep B or Hib never introduced). However, it simply did not. Also, the changes in US rates should have been severe right after the millennium, after thimerosal was largely removed from US vaccinations, compared to other countries like Canada and Sweden, where thimerosal had already been removed and therefore change did not occur. Again, the US rates did not change in comparison.
Pointless Arguing Over Minor Details
A common criticism by vaccine theorists of the logic underlying this section is that all thimerosal has not been removed – kids are still poisoned by mercury and that is causing autism. It is true that some kids still are exposed to minor quantities of thimerosal, for instance in the limited number of kids (less than 20%) who get an influenza vaccination, some of which contain thimerosal, which is now recommended for kids between 6 and 24 months. And, it is possible that a very small number of kids in the US have also been exposed to another vaccine that still contains thimerosal because of old vaccine stock still being used or vaccines being used in the US that are intended for use in other countries that still allow thimerosal or otherwise.
However, the endless online arguing on this point, whether a very small number of vaccines given to kids still contain thimerosal, is a phenomenal waste of time. This is because, as discussed in detail elsewhere in this paper, heavy metals work their harm by accumulating in the body in abnormally large quantities. It is the cumulative exposure to high levels of heavy metals that is potentially dangerous. If thimerosal exposure in children has been reduced by 90% plus, as is clear in countries like the US and Sweden, then a dramatic decrease in autism rates would have to been evident if thimerosal actually caused autism. And, that decrease did not occur, in the US, or in Canada, or in the UK. An endless debate on how much of the last 10% of the original exposure remains completely misses point. If elimination of 90%+ of the heavy metal did not reduce autism levels, then why should the elimination of the last few percentage points of the heavy metal have a different result?