Autism, vaccines, and thimerasol: A review of the latest Pediatrics study.

By Nestor Lopez-Duran PhD

By now most of you are likely familiar with the background on the controversy regarding the proposed link between vaccines and autism. Briefly, there are a significant number of parents who believe that vaccines cause, or at least play a role in, autism and a growing research library of studies that find no link between vaccines and autism. A significant portion of the anti-vaccine public is skeptical of the previous research for various reasons, one of which is that previous studies have not been able to address all the possible questions and theories regarding the proposed autism-vaccines link.

For example, there is the issue of vaccine and ASD specificity. That is, some have argued that the reason why past studies did not find a link between vaccines and autism is because they have examined all types of ASDs as a single group. Instead, they suggest that vaccines may cause only one type of ASD (e.g., regressive autism, or simply autism). Another issue is that previous studies have not been able to truly examine whether vaccines provided at a specific age cause autism. The criticism is that by grou ping cases of vaccines provided at all ages together, you may miss a vaccine-autism link that occurs at a specific age (e.g., receiving a strong cocktail of vaccines thimerasol containing vaccines during the first month of life). One final criticism is that previous studies have not been able to accurately account for differences in the amount of thimerasol that the kids were exposed to at specific ages. So it may be that vaccines cause autism only when the child is exposed to very high levels of thimerasol.

Fortunately, the latest issue of the prestigious journal Pediatrics published one of the largest case-control studies to date examining the possible link between thimerasol exposure and autism. The study answered some of these lingering questions.

This study was conducted by a team of scientists from Harvard, UCLA, the Center for Disease Control, and a number of other organizations. To conduct the study, the researchers first examined the records of kids enrolled in 3 managed care organizations. They started with a pool of close to 71,000 children born between 1994 and 1999. This is key since thimerasol was mostly eliminated from child vaccines after 1999 but they were commonly used then (see here for a more extensive discussion on changes in autism prevalence during the 1990s and what it could say about the vaccine-link controversy). The authors identified all kids who had a diagnosis of and ASD in their record. A total 802 probable ASD cases were identified. Of these, 386 agreed to undergo a full ASD evaluation (via ADOS and ADI) to confirm and clarify the diagnosis. A total of 256 ASD cases were confirmed and included in the study. Of these cases, 187 received a diagnosis of autism and 49 received a diagnosis of ASD with regression.

Then the authors took each of the 256 cases and matched them with 3 kids from the 70,000 kids who didnt have a diagnosis of autism in their records. To make sure that these matched non-ASD controls in fact did not have an ASD, the kids were also evaluated. What do I mean by matched? For each child with an ASD, the researchers identified 3 kids who did not have an ASD but who were identical in a number of other factors, such as year of birth, gender, birth weight, age of biological mother and fathers, parental education, etc.

So at the end, the researchers were left with two groups of kids who were identical to each other with one main difference: one group had a diagnosis of ASD and the other group did not. The assumption is that if thimerasol plays a direct role in autism, then there should be a difference between these two groups in regard to their history of thimerasol exposure. For example, higher thimerasol exposure should increase the risk that the child would be in the group with the ASD diagnoses.

But how did they determine the level of thimerasol exposure? The authors actually obtained the records of all vaccinations that the kids received during early childhood. They then conducted a search of the actual content of each vaccine by matching the vaccine lot number with the manufacturers record. This allowed them to calculate the total exposure to thimerasol in all vaccines received while controlling for the kids body weight. Also, since they had the records of when the child received the vaccines, they could calculate the amount of thimerasol exposure for specific date ranges including:

  • prenatal (due to maternal flu shots)
  • birth to 1 month
  • birth to 7 months
  • birth to 20 months

Then they used statistical models to determine if increases in thimerasol exposure during any of the time point increase the probability of being in the ASD group, being in the Autism group, being in the ASD with regression group, or being in the No ASD group. That is, can we predict if the child had an ASD diagnosis based on the level of exposure to thimerasol?

What did they find?

  1. Exposure to higher levels of thimerasol during the prenatal period did not increase the risk of having any of the ASD diagnoses.
  2. Exposure to higher levels of thimerasol from birth to 1 months did not increase the risk of having any of the ASD diagnoses.
  3. Exposure to higher levels of thimerasol from birth to 7 months did not increase the risk of having any of the ASD diagnoses.
  4. Exposure to higher levels of thimerasol from birth to 20 months did not increase the risk of having any of the ASD diagnoses.

That is, those with ASD, Autism, or ASD with regression diagnoses were not exposed to any greater levels of thimerasol than those without ASD diagnosis. Actually the authors found an unexpected finding:


When controlling for a number of covariates, being exposed to very high levels of thimerasol when compared to being exposed to very low levels (as measured by the difference between the 10th and the 90th percentile in thimerasol exposure) actually reduced the odds of being in all of the ASD groups. That is, going from the 10th to the 90th percentile in thimerasol exposure (from very little to a lot of exposure) reduced the probability of having an ASD diagnosis. The authors indicated that they are not aware of any biological mechanism that could explain such association. However, the authors found that those with an ASD diagnosis were more likely than the control cases to receive thimerosol-free HepB and thimerasol-free combined DTaP-Hib, HepB-Hib vaccines resulting in slightly lower commutative exposure to thimerasol.

In sum, in this case controlled study, higher levels of thimerasol exposure among a cohort of kids born in the 1990s was not associated with a higher risk of having an autism diagnosis. Thimerasol exposure was not associated with higher risk for autism even when considering only regressive type, basic autism, or any ASD. Thimerasol exposure was also not associated with higher risk for autism even when considering the timing of the commutative exposure, including prenatal exposure.

Note1: Yes, I understand th this study addresses some specific criticisms of previous studies, but it does not address all the possible theories presented by those who believe that vaccines cause autism (e.g., genetic interaction).

Note2: I have no association, financial or otherwise, with any vaccine manufacturer.
The Reference: Price, C., Thompson, W., Goodson, B., Weintraub, E., Croen, L., Hinrichsen, V., Marcy, M., Robertson, A., Eriksen, E., Lewis, E., Bernal, P., Shay, D., Davis, R., & DeStefano, F. (2010). Prenatal and Infant Exposure to Thimerosal From Vaccines and Immunoglobulins and Risk of Autism PEDIATRICS, 126 (4), 656-664 DOI: 10.1542/peds.2010-0309