Thoughtful House

Acetaminophen (paracetamol) use, measles-mumps-rubella vaccination, and autistic disorder. Published in Autism. 2008 May;12(3):293-307

Increased use of acetaminophen / paracetamol (Tylenol, Panadol, etc.) after MMR vaccination is associated with a substantial increased risk of autism, report Schultz et al. from San Diego State University [Autism 2008;12:293-307]. In a study based on parental survey of 83 children with autism and 80 developmentally normal controls, use of acetaminophen after MMR was higher in children with autism vs. controls. There was a similar effect for ibuprofen use after MMR, although this involved smaller numbers (ibuprofen use was historically less prevalent), and the differences between cases and controls is not significant.

The authors' hypothesis is that previous studies supporting or rejecting an association between MMR and autism have failed to take differences in acetaminophen use between the case and control groups (at around the time of MMR) into account. While this is a reasonable hypothesis, the design chosen to investigate it is inadequate, as it does not allow separate analyses of the roles of acetaminophen and MMR. Nonetheless, the study allows one hypothesis to be examined in part. It is possible that acetaminophen use is neither necessary nor sufficient to cause autism, but that its use in a relevant time-frame acts as a marker for children who are at risk of adverse response to MMR; this leads to some testable predictions.

The first is that children with autism are at greater risk of side-effects following MMR. The second is that at least some such side-effects might be responsive to therapeutic doses of acetaminophen. A third prediction is that if adverse response to MMR is associated with general ill health (as one might expect from experience of these children), and if MMR-adverse response is the 'real' causal factor in any observed association with autism, then general ill health prior to MMR will also be associated with autism. Finally, as children whose autism is apparently associated with MMR tend to have a later/regressive onset after a period of normal development, it might predict that the effect of acetaminophen as a marker of increased risk would be particularly strong in children with regressive onset. All of these predictions are supported by the data. The authors found that children with autism had significantly more post-MMR sequelae than controls, and that some such responses are likely to prompt acetaminophen use. There were higher levels of illness prior to MMR in children with autism, and the consequent risk of acetaminophen use was particularly high in the regressive group.

Two issues argue against a role for acetaminophen use as a marker of MMR adverse response: low risk associated with ibuprofen use appears to indicate a specific association with acetaminophen, rather than adverse response to MMR per se, however, overall numbers of ibuprofen use were too small to draw conclusions about this. Additionally, when the analysis was confined to children who suffered an adverse reaction, the likelihood of acetaminophen use was even higher than in the full sample. Again, this suggests that adverse response alone is not a sufficient predictor of autism. However, there may be differences in the severity or duration of the reaction between cases and controls - e.g. the worse the MMR reaction, the greater the chances of acetaminophen use, which would explain both the differences in acetaminophen use and in autism risk between the groups.

A further caution should also be highlighted. In addition to the design issue outlined earlier, the selection of cases and controls (partly from websites generally 'sympathetic' to the notion of an autism-MMR link) may have resulted in selection bias that inflates the differences between cases and controls in a direction supportive of the MMR-autism link. Whether this would systematically skew the apparent role of acetaminophen use is not clear.

Design issues aside, the findings are consistent with clinical experience and parental stories, that "my autistic child was ill when vaccinated with MMR", "my child had an adverse reaction to MMR vaccine", and "my child had an adverse reaction to MMR vaccine and, shortly after, suffered regressive autism".

1. Acetaminophen (paracetamol) use, measles-mumps-rubella vaccination, and autistic disorder: the results of a parent survey. Autism 2008;12:293-307

Competing interests:

Return to Press Releases