There’s been repeated claims over the past two decades that vaccines are responsible for causing autism. (A detailed backgrounder may be found here.) The incidence of autism diagnoses has climbed over the past two decades, and some have pointed to vaccines as the culprit – specifically, the mercury preservative used in most products.
Thimerosal has been a preservative in vaccines for decades. It is metabolized into ethylmercury and excreted by the kidneys in the urine. This may be contrasted with methylmercury, the mercury form for which exposure limits have been established by the US Environmental Protection Agency. In 1999, it was observed that children given routine vaccinations could have received more ethylmercury than the safe limit established for methylmercury. As a precaution, manufacturers were asked to remove thimerosal from vaccines, and it was further recommended that studies be conducted to evaluate the relationship between receipt of ethylmercury, and autism.
Previous studies have failed to identify any relationship between thimerosal and autism. And after thimerosal was removed from the majority of vaccines, the incidence of ASD hasn’t decreased. However, there hasn’t been a detailed analysis of the specific prenatal and post-natal exposure to thimerosal-containing vaccines. That’s what this trial set out to answer.
This study by Price et al, entitled, Prenatal and Infant Exposure to Thimerosal From Vaccines and Immunoglobulins and Risk of Autism is a partner to a study by Thompson et al published in the New England Journal of Medicine in 2007, that examined the relationship between thimerosal-containing vaccines and adverse neuropsychological outcomes – but deliberately excluded ASD. Briefly, the Thompson study failed to find any relationship between the two.
Canadians brag a lot about their health care system, but this trial could probably have only been done in the USA, where there’s an electronic medical record to identify patients and and collect medical information. Three managed care organizations participated in a case-control study of 256 children with Autism Spectrum Disorder (ASD) and 752 controls that were matched for birth year and gender. Children were eligible to participate if they were born between 1994 and 1999, so children were 6 to 13 years old at the time of the study. ASD children were identified by searching databases for diagnoses relevant to ASD. Mothers were interviewed, and children were also directly assessed using standardized tools to confirm the diagnosis. Three groups were tracked: ASD; autistic disorder (AD) (a subset with more severe symptoms); and ASD with regression (AD-R) (children reported to have lost previously acquired language skills). Assessors were blinded with respect to the thimerosal exposure of the child and mother. Controls were screened via a maternal interview to ensure no children in the control group had undiagnosed ASD.
Each child’s history of thimerosal-containing injections (TCIs) was pulled from electronic immunization records. The mercury content of each vaccine received was determined by tracking the manufacturer and product lot, which was compared against manufacturer records. The mother’s receipt of immunoglobulins, tetanus, and diptheria-tetanus vaccines was also obtained from medical records. This allowed a calculation of the exact amount of ethylmercury administered.
Multiple covariates (variables that could influence the outcome) were also tracked, including child and family characteristics (mother’s age at birth, education level, income, birth order, breastfeeding duration, etc.); prenatal mercury exposure from fish, cosmetics, or dental fillings; use of tobacco and other drugs; and multiple child birth factors (e.g., respiratory distress, etc).
After exclusion criteria and patient refusals were applied, 256 children met the criteria for ASD, including 187 with AD and 49 with AD-R. Characteristics were largely identical between the children with ASD and the 752 children in the control group.
Children with ASD, and controls, had similar exposure to ethyl mercury over sequential time periods (see Table 2 of the study). The variation in mercury exposure was considerable in each group, due to factors that included receipt of vaccine, different brands of vaccines, and administration of vaccines individually or as combined products. Here are the pivotal findings:
- Exposure to ethylmercury from TCIs either prenatally or in the first month of life was not associated with any form of ASD.
- In older groups, higher levels of ethylmercury exposure were associated with a decreased risk of each of ASD, AD, and AD-R.
There’s no known mechanism by which vaccines with ethylmercury would reduce the incidence of ASD. The authors looks at different reasons for the difference, but found none. It could be that this is simply a spurious finding reflective of the observation that the two groups don’t really differ.
This detailed analysis found no relationship between ASD and thimerosal exposure. Limitations to the analysis include the fact that this is an observational study. While painstaking measures were taken to control for confounders, it’s always possible there’s a unknown confounder.
A prospective, randomized trial (vaccinated versus unvaccinated) will never be conducted as it’s not ethically permissible. Nor is it even possible anymore, as thimerosal has been removed from virtually all vaccines. This study is “as good as it gets” with respect to what an observational trial can do to address the thimerosal – autism question. The results add to numerous other studies which also failed to identify a link between thimerosal-containing injections and autism. It’s persuasive and compelling evidence, that there’s no relationship to be found.
Will the publication of this study stop antivaccinationists from claiming that thimerosal in vaccines causes autism? Of course not. More data won’t be persuasive, because the idea that thimerosal causes autism was never based on good evidence evidence to begin with. But this study gives science advocates further evidence to reassure those unsure of the safety of vaccines that there is no relationship between thimerosal-containing vaccines and autism.
Price CS, Thompson WW, Goodson B, Weintraub ES, Croen LA, Hinrichsen VL, et al. (2010). Prenatal and Infant Exposure to Thimerosal From Vaccines and Immunoglobulins and Risk of Autism. Pediatrics, doi:10.1542/peds.2010-0309.