Potential Link Between MMR Vaccination, Fever, Acetaminophen Use, and Autism

Let's just start with: There is not "zero credible evidence". Studies of various types point to the possibility. Post any studies we missed in the comments.

This article presents the evidence of the potential link between acetaminophen use in infancy and autism risk, highlighting various studies and findings. ​

Post-MMR Vaccine Fever Suppression and Autism Risk​

• The hypothesis linking acetaminophen use in infants to autism risk was proposed in the early 2000s. ​

• Dr. Anthony R. Torres published a hypothesis in 2003 suggesting that acetaminophen could increase autism risk in genetically susceptible children. ​

• A notable correlation was observed between the rise in autism diagnoses during the 1980s and increased acetaminophen use. ​

Parental Survey Case-Control Study ​

• The 2008 study by Schultz et al. ​ found a significant association between post-MMR vaccination acetaminophen use and autism diagnosis. ​

• Children who received acetaminophen after MMR vaccination had an odds ratio (OR) of ~6 for autism. ​

• The study faced limitations such as recall bias and non-random sampling, prompting calls for further research. ​

Ecological Population-Level Studies​

• Bauer and Kriebel (2013) found a correlation between prenatal acetaminophen use and higher autism rates across countries. ​

• A Danish cohort study (Frisch et al., 2015) indicated that circumcised boys had a 1.5 times greater risk of developing ASD. ​

• These studies suggest a potential link but cannot establish causation due to confounding factors. ​

Follow-up Observational Studies 2016–2020 ​

• Schultz et al. (2016) found that children with autism were less likely to use acetaminophen for fevers. ​

• Bittker and Bell (2020) reported a dose-dependent relationship between acetaminophen exposure in infancy and autism risk, particularly in males. ​

• The studies highlight complex dynamics in acetaminophen use and autism risk, with calls for further research to clarify findings. ​

Summary of Evidence and Potential Limitations

• Mixed evidence exists regarding the link between acetaminophen use and autism, with some studies showing strong associations.

• Limitations include reliance on retrospective data, potential biases, and inability to fully control for confounding factors. ​

• Further large, well-controlled studies are needed to confirm or refute the potential causal link between acetaminophen use and autism risk. ​

The idea that fever suppression with acetaminophen (paracetamol) in infancy might contribute to autism risk was first proposed in the early 2000s. In 2003, Dr. Anthony R. Torres published a hypothesis in BMC Pediatrics suggesting that using fever-reducing medications (like acetaminophen) in infants or very young children could increase autism spectrum disorder (ASD) risk in genetically susceptible individuals (Utah State University, 2010). This hypothesis was partly motivated by temporal trends: the sharp rise in ASD diagnoses during the 1980s coincided with a shift from infant aspirin (curtailed due to Reye’s syndrome concerns) to widespread acetaminophen use (Utah State University, 2010). Some observers noted that periods of reduced acetaminophen use appeared to coincide with plateaus in autism rates. For example, one analysis pointed out that the highly publicized Tylenol™ tampering incidents in 1982 and 1986 – which caused acetaminophen sales to temporarily drop – coincided with a flattening of ASD incidence in California (Bittker & Bell, 2020). These ecological observations, while intriguing, could only suggest a possible link and could not prove causation. They set the stage for formal studies to investigate whether post-vaccination fever treatment with acetaminophen (especially around the measles-mumps-rubella (MMR) vaccine at age ~1 year) might be associated with autism in children.

Parental Survey Case-Control Study (Schultz et al., 2008)

The first peer-reviewed study to directly examine this link was a case-control study published in 2008 in the journal Autism by Schultz and colleagues (Schultz et al., 2008). Researchers used an online parental survey (conducted in 2005–2006) to compare 83 children with an ASD diagnosis to 80 typically developing controls. Parents were asked about their use of fever/pain medications after the MMR vaccine. Key findings: Children who received acetaminophen after their MMR vaccination were significantly more likely to have autism than those who did not.

Specifically, among children age 5 or younger, post-MMR acetaminophen use was associated with an odds ratio (OR) of ~6 for ASD (adjusted OR 6.11, 95% CI 1.4–26) (Schultz et al., 2008). The association was strongest in certain subgroups: in children who had regressive autism (loss of developmental skills) the OR was ~4 (Schultz et al., 2008), and in children who experienced pronounced post-vaccination reactions (such as fever or other sequelae) the OR was even higher at 8.23 (95% CI 1.6–43) (Schultz et al., 2008). Notably, ibuprofen use after MMR was not associated with autism (OR ~0.89, non-significant) (Schultz et al., 2008), suggesting the effect was specific to acetaminophen rather than a general tendency to treat fevers. The authors concluded this preliminary study showed an association between acetaminophen use after MMR and autism, while emphasizing that it did not prove causation (Schultz et al., 2008). They theorized that in some children an MMR-induced fever and concurrent acetaminophen exposure might interact to increase ASD risk, possibly via impacts on the immune or metabolic systems (though mechanistic speculation was beyond the study’s scope).

Limitations: Being an online survey-based study, this research had important limitations. Parents were recalling events (vaccinations and medication use) from years prior, so recall bias is a concern (Good, 2009). The recruitment via the internet and autism parent groups could

also introduce selection bias – the cases and controls may not be fully representative of the general population (Good, 2009). Indeed, a 2009 letter to the editor by Cox and McDowell in Autism criticized the sampling methods as potentially biased and not random, and noted the lack of an a priori sample size calculation (Good, 2009). They also questioned whether parents could accurately remember, years later, if they gave acetaminophen after the MMR (Good, 2009). In a published reply, Schultz defended the methodology, arguing that asking parents of autistic children to recruit friends with neurotypical children helped make the control group comparable (Good, 2009). He acknowledged the sample might not be fully random, but pointed out that the association was strong enough to be statistically significant despite the modest sample size (Good, 2009). To mitigate recall issues, the team had also analyzed the subset of children aged 1–5 (closer in time to the survey) and still found a significant association (Good, 2009). Nonetheless, the authors and commentators agreed that further research in larger, preferably prospective populations would be needed to confirm or refute this link.

Ecological and Population-Level Studies

After 2008, researchers explored the acetaminophen–autism connection using other study designs. One approach was ecological studies looking at population-level correlations. In 2013, Bauer and Kriebel published an analysis in Environmental Health examining country-level autism rates in relation to indicators of acetaminophen exposure (Bauer & Kriebel, 2013). They reported that across several countries, higher prenatal acetaminophen use (among pregnant women) correlated with higher autism/ASD prevalence (Pearson r ~0.80), although data were limited to only 8 countries (Bauer & Kriebel, 2013). More strikingly, they found that male autism prevalence by country was almost perfectly correlated (r ≈ 0.98) with the rate of early infant male circumcision (Bauer & Kriebel, 2013). The authors interpreted the circumcision finding as a proxy for perinatal acetaminophen exposure, because since the mid-1990s it became common to give infant boys acetaminophen for post-circumcision pain relief. In fact, when they looked within the United States, states with higher circumcision rates had higher autism rates in boys (Bauer & Kriebel, 2013). These ecological correlations are intriguing: circumcision (with analgesic use) becoming common in the 1980s–90s overlaps with rising autism diagnoses.

However, ecological studies cannot control for all confounders – many other factors (cultural, genetic, etc.) differ between populations. The authors cautioned that their results do not prove causality, but they noted the biologic plausibility and called for more direct epidemiological study of acetaminophen’s role (Bauer & Kriebel, 2013). (Biochemical evidence, such as impaired sulfation metabolism in some autistic children, has been cited to support plausibility (Good, 2009), but mechanistic studies were beyond this analysis’ scope. See ASD Causality Model for one view.)

Another notable population-level study was a Danish cohort study by Frisch et al. (2015) that indirectly supported this link. This study, published in the Journal of the Royal Society of Medicine, investigated whether ritual circumcision in infancy is associated with autism. In Denmark, circumcision is relatively uncommon (mostly in certain religious communities), allowing a comparison. In a cohort of over 340,000 boys followed to age 9, the study found that boys circumcised in infancy had a significantly higher risk of developing ASD by age 9 compared to uncircumcised boys (approximately 1.5 times greater risk, according to news reports of the findings) (Bailey, 2015). The association was particularly pronounced among non- Muslim families, for reasons not fully understood (Frisch & Simonsen, 2015). The authors hypothesized that pain and stress from the procedure (even with local anesthesia) might have lasting neurodevelopmental effects (Frisch & Simonsen, 2015). However, subsequent commentators offered an alternative explanation: since modern circumcisions often involve acetaminophen for post-operative analgesia, it is possible that the use of acetaminophen around the time of circumcision could be the factor increasing autism risk, rather than the pain itself (Bittker & Bell, 2020). In other words, the circumcision study’s findings are consistent with the acetaminophen hypothesis as well. It’s worth noting the Danish study could not directly measure medication use, so this interpretation is speculative. Still, taken together, these ecological and cohort data points add circumstantial evidence linking early-life acetaminophen exposure (often in the context of fevers or pain) with later ASD diagnoses.

(Aside: A 2009 clinical trial in The Lancet had shown that prophylactic paracetamol given to infants after routine vaccinations blunted the immune response to the vaccines (Utah State University, 2010. While not an autism study, it suggested that acetaminophen can modulate immune function in infants, a finding that some have pondered in relation to autism-immunity hypotheses. However, this is a mechanistic outcome and the trial did not address neurodevelopment.)

Follow-up Observational Studies (2016–2020)

Research in the late 2010s continued to examine this potential link through observational data, with mixed and nuanced findings. In 2016, Schultz et al. (the same lead author as the 2008 study) conducted a follow-up analysis using the U.S. National Database for Autism Research (NDAR). This study (published in an open-access journal Autism–Open Access, 2016) looked at data on fever medication use in 118 children with ASD and 79 neurotypical controls (all older children, mean age ~11 years) (Schultz & Gould, 2016). Notably, this database wasn’t focused on the MMR vaccine per se, but it recorded parents’ general practices for treating childhood fevers. Key findings: Even in this older cohort, there was a significant difference in how parents used acetaminophen vs. other antipyretics. Overall, acetaminophen use for fever was significantly less common in the ASD group than in the control group (p = 0.013) (Schultz & Gould, 2016). In particular, parents of children with autism were more likely to report “rarely or never” using acetaminophen for fevers, whereas parents of controls more often used acetaminophen as their first-choice fever reducer (Schultz & Gould, 2016). By contrast, ibuprofen use patterns were similar between groups (Schultz & Gould, 2016). In logistic regression, having autism was associated with markedly lower odds of being an acetaminophen user (OR ~0.17 for acetaminophen as first-line, 95% CI 0.05–0.60) (Schultz & Gould, 2016). At face value, this inverse association seems contradictory to the earlier findings – it suggests autistic children received less acetaminophen for fevers. The authors offered an interpretation: perhaps early acetaminophen exposure in infancy predisposes to ASD, and those same children later tend not to respond well to acetaminophen (leading parents to switch to other remedies)(Schultz & Gould, 2016). Supporting this, the study found many ASD parents reported having to switch to ibuprofen because acetaminophen “was not effective” for their child’s fevers (Schultz & Gould, 2016). Another intriguing finding was that children with ASD were significantly more likely to exhibit improved social interaction during fevers (so- called “fever effect”) than controls (18.6% of ASD kids vs 7.6% of controls showed social improvement when febrile) (Schultz & Gould, 2016). The authors speculated this could relate to the endocannabinoid system and how acetaminophen acts on it (Schultz & Gould, 2016). In summary, the 2016 study reinforced a difference between ASD and non-ASD children in relation to acetaminophen, but it did not simply repeat the 2008 result. Instead, it highlighted complex dynamics: early acetaminophen use might be a factor in autism risk, yet by later childhood those with ASD seem less likely to be using that drug. A limitation here is that this study was retrospective and not originally designed to assess the MMR-autism question, so it doesn’t provide a direct risk estimate for post-vaccine use; rather, it adds evidence of unusual acetaminophen response or usage patterns in the ASD population (Schultz & Gould, 2016).

Most recently, a larger scale retrospective survey-based case-control study by Bittker and Bell (2020) added new data. They surveyed parents of 1,515 children (via an internet questionnaire) to quantify postnatal acetaminophen exposure in the first 2 years of life, then analyzed the association with autism outcomes (Bittker & Bell, 2020). Published in Behavioral Sciences (MDPI) in 2020, this study included both ASD cases and neurotypical controls. Key findings: The number of acetaminophen doses given to a child under age 2 was positively associated with the odds of an ASD diagnosis in a dose-dependent manner (Bittker & Bell, 2020). After adjusting for confounders, the authors reported that each acetaminophen dose was associated with a small increase in ASD risk among male children (adjusted OR = 1.023 per dose, 95% CI 1.005–1.043) (Bittker & Bell, 2020). When considering heavy exposure, the association was more pronounced – for example, boys who received 60+ total doses by age 2 had substantially higher odds of ASD than those who received none (in one unadjusted analysis, OR exceeded 5) (Bittker & Bell, 2020). Interestingly, no significant association was seen in girls, which aligns with the hypothesis that boys might be more vulnerable (and also the fact that autism is more prevalent in males). The study also found that parents who “couldn’t recall” the exact number of doses (perhaps implying frequent use) were more likely to have a child with ASD (Bittker & Bell, 2020). Using their data, Bittker and Bell estimated that about 40% of ASD cases in males could be attributable to postnatal acetaminophen exposure (population attributable fraction) (Bittker & Bell, 2020). This is a very high proportion, though the authors caution that this is an observational estimate. Limitations: The data came from an online, self-selected sample (recruited via social media and autism forums), which may not be representative and is prone to recall bias (parents of older children had to remember infant medication use). Moreover, it’s hard to separate acetaminophen use from the underlying reasons for use (fever, illness, vaccination) as the cause of autism. The study tried to adjust for factors and also looked at ibuprofen (which was not significantly associated with ASD, consistent with the 2008 finding) (Bittker & Bell, 2020). While the results support a link, the authors acknowledged that causation is not proven. They pointed out that their findings, together with prior evidence, “suggest the possibility that postnatal acetaminophen is a significant contributor to ASD risk among males” (Bittker & Bell, 2020), and called for further research.

Summary of Evidence and Potential Limitations

In aggregate, the studies above provide mixed but concerning evidence about a potential link between MMR vaccination (and the fever it can induce), subsequent acetaminophen use, and autism risk in children:

● Survey-based case-control data (2008) showed a strong association between post- MMR acetaminophen use and autism diagnosis (Schultz et al., 2008). This was the first signal, but it relied on self-reported retrospective data and has been criticized for possible bias (Good, 2009). Its strength was the specific comparison to ibuprofen, which did not show an association (Schultz et al., 2008), lending some specificity to the finding.

● Follow-up correspondence (2009) highlighted limitations: recall inaccuracies and non- random sampling (Good, 2009). The study authors responded by noting the robustness of the association even in a subset of younger children, though they conceded the sample may not have been fully random (Good, 2009).

● Ecological and time-trend studies (2013) found population-level correlations between surrogates of acetaminophen exposure (e.g. circumcision rates, changes in medication recommendations) and autism prevalence (Bauer & Kriebel, 2013). These studies (e.g. Bauer 2013) are useful for hypothesis-generation but cannot account for individual confounders, so they cannot establish causality (Bauer & Kriebel, 2013). They do, however, align with the hypothesis and provide a broader context (e.g., suggesting the autism “epidemic” might be partially explained by increased acetaminophen use in the 1980s (Bittker & Bell, 2020).

● National cohort data (Frisch et al., 2015) indicated that boys undergoing infant circumcision had higher ASD risk, an observation possibly explained by acetaminophen analgesia during infancy(Frisch & Simonsen, 2015). This was a large, registry-based study with thorough adjustment for confounders like cultural background. Still, since medication use wasn’t directly recorded, the acetaminophen connection remains an inference. Alternative explanations (e.g., stress of pain) were proposed by the original authors (Frisch & Simonsen, 2015). Critics pointed out that pain alone seemed unlikely to cause autism, especially given that historically circumcisions without pain relief did not coincide with autism surges (Bittker & Bell, 2020).

● Newer observational studies (2016, 2020) continued to find differences consistent with the hypothesis. Schultz et al. (2016) using clinical data noted autistic children were less likely to be given acetaminophen for fever, perhaps due to underlying differences in response (Schultz & Gould, 2016). Bittker & Bell (2020) found a dose-dependent relationship between infant acetaminophen use and ASD in males (Bittker & Bell, 2020). Both studies again found no similar association with ibuprofen, strengthening the case that acetaminophen may be the unique factor (Schultz et al., 2008). However, both were observational and relied on parental reports, with inherent limitations of confounding (e.g., underlying illnesses or vaccines could be the real triggers) and self-report bias.

In summary, clinical and epidemiological studies to date have observed a correlation between post-vaccination acetaminophen use and autism in several settings (Bittker & Bell, 2020). The key recurring theme is that acetaminophen exposure in infancy (often coincident with vaccinations or illnesses) shows up as a risk marker for ASD, whereas other analgesics (like ibuprofen) do not. Researchers have postulated mechanisms ranging from impacts on inflammatory processes to effects on the endocannabinoid system (Schultz & Gould, 2016), but no definitive biological mechanism has been confirmed (mechanistic studies were beyond the scope of these observational analyses). It is important to note that not all evidence is consistent – for example, large prospective studies on acetaminophen use in pregnancy have had mixed results (some showing a mild association with neurodevelopmental issues, others finding no significant link after controlling for genetics (Drexel University, 2024). The focus here, however, is on postnatal use around vaccination, where data are more sparse. All the studies highlighted call for caution and further research rather than definitive conclusions.

The limitations across this body of work include reliance on retrospective survey data, potential recall and selection biases, inability to fully rule out confounding factors (such as the fever or infection for which acetaminophen was given), and in the case of ecological studies, the ecological fallacy (population correlations may not hold at the individual level - but saying that it is possibly a problem does not mean they do not) (Bauer & Kriebel, 2013).

In conclusion, the epidemiological evidence accumulated thus far suggests a possible link between MMR vaccination with fever, subsequent acetaminophen use, and autism risk in susceptible children, but this link remains unproven and controversial. The association has been observed in multiple observational studies (Schultz et al., 2008), yet each study had important caveats. Researchers stress that further large, well-controlled studies (ideally prospective in design) are needed to confirm whether this correlation is indeed causal or if it reflects other underlying factors. Until then, the hypothesis that treating post-vaccination fever with acetaminophen might elevate autism risk is an intriguing one with some supportive data, but it is not settled science. As one review summarized, “other evidence, both abundant and robust, demonstrate the critical role of acetaminophen in the etiology of ASD”, whereas mainstream cohort analyses often find little risk – highlighting the ongoing debate in the scientific community (Jones et al., 2024).

Sources: Studies and reports cited above include peer-reviewed case-control research (Schultz et al., 2008), letters and replies in scientific journals (Good, 2009), an ecological analysis (Bauer & Kriebel, 2013), a Danish national cohort study (Frisch & Simonsen, 2015), and recent observational studies leveraging clinical databases and surveys (Schultz & Gould, 2016). All illustrate different pieces of the puzzle regarding MMR, fever control, and ASD risk. Each provides valuable data points, while also underscoring the need to interpret findings carefully given the observational designs and potential biases. The consensus so far is that more research is warranted to either validate or refute a causal link.

That said, it would be reckless and irresponsible to pretend this evidence does not exist, because if there is a causal link, societal strife and discord will risk totalitarian means to enforcing a flawed vaccine mandate program. That did not go well at all with COVID-19.

● Bittker, S. S., & Bell, K. R. (2020). Paracetamol (Acetaminophen) Use during Infancy and the Risk of Autism Spectrum Disorder: A Prospective Study. Behavioral Sciences, 10(8), 116. https://doi.org/10.3390/bs10080116

● Bauer, A. Z., & Kriebel, D. (2013). Prenatal and perinatal analgesic exposure and autism: An ecological link. Environmental Health, 12, 41. https://doi.org/10.1186/1476- 069X-12-41

● Frisch, M., Simonsen, J., & Melbye, M. (2015). Ritual circumcision and risk of autism spectrum disorder in 0- to 9-year-old boys: national cohort study in Denmark. Journal of the Royal Society of Medicine, 108(7), 266–279. https://doi.org/10.1177/0141076815585322

● Schultz, S. T., Klonoff-Cohen, H. S., Wingard, D. L., Akshoomoff, N. A., Macera, C. A., Ji, M., & Bacher, C. (2008). Acetaminophen (Paracetamol) use, measles–mumps– rubella vaccination, and autistic disorder: the results of a parent survey. Autism, 12(3), 293–307. https://doi.org/10.1177/1362361307089518

● Schultz, S. T., Klonoff-Cohen, H. S., Wingard, D. L., Akshoomoff, N., Macera, C. A., Ji, M., & Bacher, C. (2016). Acetaminophen use for fever in children associated with autism spectrum disorder. Autism-Open Access, 6(5), 1–7. https://doi.org/10.4172/2165- 7890.1000196

● Cox, K. M., & McDowell, C. (2009). Caution required when interpreting results from internet-based case–control studies: a response to Schultz et al. (2008). Autism, 13(1), 123–126. https://doi.org/10.1177/1362361308099183

● Good, P. (2009). Did acetaminophen provoke the autism epidemic? Alternative Medicine Review, 14(4), 364–372. [Archived PDF source: pages.ucsd.edu/~mboyle/COGS163]

● Torres, A. R. (2003). Is fever suppression involved in the etiology of autism and neurodevelopmental disorders? BMC Pediatrics, 3, 9. https://doi.org/10.1186/1471- 2431-3-9

References

● Bailey, R. (2015, February 18). Does circumcision cause autism? Reason.Com. https://reason.com/2015/02/18/does-circumcision-cause-autism/

● Bauer, A. Z., & Kriebel, D. (2013). Prenatal and perinatal analgesic exposure and autism: An ecological link. Environmental Health, 12(1), 1–13. https://doi.org/10.1186/1476-069X-12-41

● Bittker, S. S., & Bell, K. R. (2020). Postnatal acetaminophen and potential risk of autism spectrum disorder among males. Behavioral Sciences, 10(1), 26. https://doi.org/10.3390/bs10010026

● Drexel University. (2024, April 9). No link between acetaminophen use during pregnancy and children’s autism risk. Drexel.Edu. https://drexel.edu/news/archive/2024/April/No- link-between-acetaminophen-use-during-pregnancy-and-child-risk-of-autism

● Frisch, M., & Simonsen, J. (2015). Ritual circumcision and risk of autism spectrum disorder in 0- to 9-year-old boys: National cohort study in Denmark - PMC. Journal of the Royal Society of Medicine, 108(7), 266–278. https://doi.org/10.1177/0141076814565942

● Good, P. (2009). Did acetaminophen provoke the autism epidemic? Alternative Medicine Review, 14(4), 364–372. https://pages.ucsd.edu/~mboyle/COGS163/pdf- files/Did%20Acetaminophen%20Provoke%20the%20Autism%20Epidemic-PeterGood- 2009.pdf#:~:text=Schultz%20et%20al%20used%20to,because%20they%20repre%02se nted%20the%20population

● Jones, I. J. P., Williamson, L., Konsoula, Z., Anderson, R., Reissner, K. J., & Parker, W. (2024). Evaluating the role of susceptibility inducing cofactors and of acetaminophen in the etiology of autism spectrum disorder - PMC. Life, 14(8). https://doi.org/10.3390/life14080918

● Schultz, S. T., & Gould, G. G. (2016). Acetaminophen use for fever in children associated with autism spectrum disorder - PMC. Autism-Open Access, 6(2). https://pmc.ncbi.nlm.nih.gov/articles/PMC5044872/#:~:text=for%20controls,037

● Schultz, S. T., Klonoff-Cohen, H. S., Wingard, D. L., Akshoomoff, N. A., Macera, C. A., & Ming Ji. (2008). Acetaminophen (paracetamol) use, measles-mumps-rubella vaccination, and autistic disorder. Autism, 12(3), 293–307. https://doi.org/10.1177/1362361307089518

● Utah State University. (2010, April 1). CPD researcher studies acetaminophen, autism. USU.Edu. https://www.usu.edu/today/story/cpd-researcher-studies-acetaminophen- autism