Understanding Recent Government Announcements on Autism: Insights from Decades of Research

In light of recent government announcements regarding autism causes and treatments, many have reached out to us, as experts in the field, to provide clarity on what decades of research tell us about these developments.

First, we wholeheartedly support the increased attention and investment being directed toward childhood neurodevelopmental disorders. Autism is a common neurodevelopmental disability that warrants serious focus from both the medical and research communities.

At the same time, some of the statements released recently have caused confusion and, in some cases, are misleading. We want to address a few frequently asked questions to help clarify the science behind autism research.

1) Does Tylenol cause autism?

No.  To date, there are no studies proving that Tylenol causes autism. Some studies have reported an association between prenatal exposure to Tylenol (acetaminophen) and an increased likelihood of autism in children. These findings come from epidemiological research comparing rates of autism in children whose mothers used Tylenol during pregnancy versus those who did not.

However, it is important to understand what these studies can and cannot tell us. While they identify an association, they do not demonstrate that Tylenol causes autism.

This distinction is critical because Tylenol is typically taken to treat symptoms such as fever or pain, which themselves may be linked to underlying infections or health conditions that could endanger the mother or infant, and could influence autism risk. Additionally, genetic factors may play a role. Parents who carry genes associated with autism could also be more prone to illnesses that prompt Tylenol use during pregnancy. When researchers carefully account for these variables, such as in a recent study from JAMA in 2024 (Ahlqvist et al., 2024), the association between Tylenol and neurodevelopmental disorders no longer exists.

In summary, current evidence does not establish that Tylenol causes autism or increases its risk. At the same time, avoiding fever reduction may also carry risks for the fetus, so judicious use of Tylenol in consultation with appropriate medical professionals remains the standard of care.

2) Is leucovorin established as an effective treatment for autism?

Leucovorin, a form of folinic acid, has been proposed as a treatment for autism spectrum disorder (ASD) based on the idea that low folic acid levels might contribute to autism. However, supporting evidence is weak and inconclusive.

Most studies are small and methodologically limited. Many lacked placebo controls (the placebo effect, which is improvement seen simply from believing one is receiving treatment), which is known to be particularly strong in autism and related conditions. Therefore, gold-standard medication trials assign participants randomly to receive either the active medication or a similar compound with no active medical ingredients. These studies are “blinded” which means neither the participants nor the researchers measuring their outcomes know whether they received the placebo or the active medication. Therefore, greater improvements in health or behavior for the medication compared to the placebo group can be attributed to the medication itself rather than the patients’ or clinicians’ expectations.

One of the two placebo-controlled studies was a U.S.-based study involving 48 children with ASD, where researchers primarily found improvements in verbal communication. However, this study had several important drawbacks: it was small in terms of the number of patients enrolled, it was not registered on federal clinical trial databases, and was poorly controlled statistically. Of the many outcomes measured (e.g., communication, socialization, stereotyped movements, attention, and dozens of blood biomarkers), verbal communication was the only outcome where a significant difference was found between the leucovorin and placebo groups. When multiple comparisons like these are made, it’s necessary to adjust the statistical analysis to avoid false positives, which appears not to have been done in this case. 

The larger of the placebo-controlled studies was done in India, and rather than showing a strong difference between placebo and folinic acid treatment, it shows most convincingly that both the placebo and treatment groups showed significant improvement over the study. In other words, whether participants received folinic acid, or a non-medically-active pill (placebo), they showed improvements, indicating that results were driven by the placebo effect, rather than the medication itself. The raw data or supplemental data are not published, so it is hard to evaluate the statistics, which also appear flawed as they do not control for all of the multiple tests that were done. We have asked the authors for the raw data, which they do say is available upon request and will update this website once we have had a chance to review it.

Overall, the evidence does not support leucovorin as an effective ASD treatment. No consistent improvements in core social or communication behaviors have been shown. More evidence from gold-standard, randomized controlled trials would be needed prior to concluding that leucovorin is an established, effective treatment for autism. While it appears safe and tolerable, current studies are too small and flawed to draw firm conclusions. Moreover, autism is not a single condition, so rigorous research would be needed to determine the population most likely to benefit, if at all.