[@PeterAttiaMD] 367 - Tylenol, pregnancy, and autism: What recent studies show and how to interpret the data

Link: https://youtu.be/GWVM_5knTXg

Short Summary

This Drive podcast episode explores the controversy surrounding the alleged link between acetaminophen (Tylenol) use during pregnancy and autism in children. While acknowledging a small statistical association in some studies, the host argues the evidence for a causal relationship is weak, highlighting issues with study designs, confounding variables, and lack of biological plausibility. Ultimately, the podcast emphasizes the importance of critical thinking, considering individual circumstances, and weighing the risks and benefits before making decisions about medication use during pregnancy, rather than relying solely on sensationalized headlines.

Key Quotes

Here are 4 direct quotes from the provided transcript that I found particularly insightful:

1. "Complex conditions usually don't have simple explanations."

2. "As humans, we are not wired to think scientifically... We are not wired for critical and scientific thought."

3. "...a thousand SARS makes not a pearl necklace."

4. "I think the probability that if a woman takes Tylenol during pregnancy, it's going to increase the probability that her child has autism is very low."

Detailed Summary

Okay, here's a detailed summary of the YouTube video transcript in bullet points, covering the key topics, arguments, and information:

I. Introduction and Context

• Topic: Controversy surrounding acetaminophen (Tylenol) use during pregnancy and alleged links to autism.
• Goal of the Podcast: Provide a framework for critically evaluating such claims, going beyond a simple "yes" or "no" answer.
• Audience: The episode is made available to everyone, unlike regular "Ask Me Anything" (AMA) sessions, due to the broad interest and importance of the topic.
• Initial Observations (Groundwork):
  • Autism rates have risen dramatically, suggesting a need for investigation.
  • Complex conditions like autism are unlikely to have a single, simple cause. Resist the temptation to assume a single cause.
  • Science should be apolitical, but unfortunately, topics like nutrition and autism have become politicized. The podcast aims for an objective examination of the evidence.
  • Humans are not naturally wired for scientific thinking. Scientific method and critical thought are inventions that need to be learned.
  • The Bradford Hill criteria will be used as a framework for evaluating the evidence.
  • The bar for medication use during pregnancy is very high. The general mantra is to avoid all medications and supplements beyond essential ones (prenatal vitamins, thyroid hormone).

II. FDA Pregnancy Drug Categories

• Historical System (ABC DX):
  • Category A: No demonstrated risk in well-controlled human studies (2-5% of drugs, e.g., thyroid hormone, prenatal vitamins).
  • Category B: No evidence of risk in humans, but animal data might show signals (15-25%).
  • Category C: Risk can't be ruled out. No evidence of risk, but no evidence of safety (60-75%).
  • Category D: Positive evidence of human fetal risk, used only if benefit to the mother outweighs the risk (5-8%, e.g., some seizure medications).
  • Category X: Proven to cause significant harm to the fetus, regardless of benefit to the mother (1-3%, e.g., statins, methotrexate).
• Acetaminophen: Currently in Category B. There has been debate about whether it belongs in Category C.
• Ibuprofen: Category B in the first two trimesters, but Category D in the third trimester due to premature closure of a fetal blood vessel.
• Important Consideration: Always evaluate the alternative choices if a medication is avoided. What is the "switching cost?"

III. Structured Approach to Evaluating Exposure/Condition Associations

1. Confirm Statistical Association: Verify that a statistically significant association exists between the exposure (acetaminophen) and the condition (autism).
2. Determine Likelihood of Causality: This is the most challenging aspect. Use sensitivity analyses and falsification tests. The goal is not to prove causality, but to determine its likelihood. It deals in probabilities, not proofs. Bradford Hill criteria help in this assessment.
3. Understand Effect Size: Even if an association is causal, the effect size might be so small that it doesn't significantly impact behavior.

IV. The Importance of Scientific Openness

• Science is about constantly updating priors as new evidence emerges. Changing views based on new data is a strength, not a weakness.

V. What Are the Claims Being Made About Acetaminophen and Autism?

• Maternal use of acetaminophen during pregnancy is associated with an increased risk of autism in the child.
• Some have prompted the government to ask the FDA to issue warnings and change product labels.
• Both the FDA and the scientific community agree that there is not yet evidence to assert a causal relationship.

VI. Why Are So Many Things Being Linked to Autism?

• There's a strong desire to find the triggers of autism, leading to extensive research looking for associations between autism and countless variables.
• The Multiple Comparisons Problem: When you look at enough variables, you're bound to find statistically significant associations by chance alone. Example: psychic testing with coin flips. Testing many hypotheses increases the chance of false positives.
• Spurious Correlations: It's easy to find strong correlations between unrelated variables if you look hard enough. Example: margarine consumption and divorce rate in Maine. The speaker mentions a website with these examples.
• Impossible to Disprove: In epidemiology, it's virtually impossible to disprove a link due to the inability to conduct randomized trials for ethical and logistical reasons.

VII. The Recent Alarm and the Systematic Review

• The recent concern was triggered by a systematic review published in BMC Environmental Health.
• The publication was not a meta-analysis. It didn't pool the data from the studies to re-evaluate the overall association or perform any new statistical tests.
• It was a review of earlier research that collected relevant studies on the relationship between prenatal acetaminophen exposure and the risk of autism, ADHD, and some other neurodevelopmental disorders.

VIII. Breakdown of the Paper

• What the Paper Showed (according to the authors): The authors claimed that six observational studies consistently reported a positive association between prenatal acetaminophen use and autism spectrum disorder (ASD).
• Critiques of the Paper and Included Studies:
  • Inaccurate Reporting: Two of the six studies actually showed no significant association.
  • Dose-Response Issues: The dose response relationship was observed in four of the five studies they claimed.
  • Problems with the Xi Study (Smallest Study Showing Strongest Effect):
    • Used a single blood test at birth to assess acetaminophen levels, a poor indicator of overall exposure during pregnancy.
    • Acetaminophen is quickly eliminated from the body.
    • All cord blood samples contained detectable levels of acetaminphen, but they don't actually report how those levels differed across the tertiles, either averages or thresholds.
    • 70% of the samples had no detectable levels of acetaminophen metabolites, which would strongly suggest that the majority of participants had very minimal levels of acetaminophen exposure.
    • Biased participant inclusion: The study enrolled participants in 1998 and followed them for 20 years. People are more likely to stay in a longitudinal study if they have a personal interest in the results (e.g., their child has autism). The study also was disproportionately based on pre-term births. 11% of the kids had autism. But for context, the general population currently at that 5x increase is 3%.
• Overall Association: The overall association between acetaminophen use and autism is small, corresponding to about a 5% increase in relative risk between exposed and unexposed children.
• Xi study does not look like a smoking gun because of the concentration effect.

IX. The Swedish Cohort Study and the Japanese Cohort Study

• Description: Large prospective cohort study with just under 2.5 million Swedish children.
• Results: A small but statistically significant positive association between prenatal acetaminophen exposure and autism in the general cohort (hazard ratio of 1.05, or a 5% relative risk increase). The absolute risk increase was 0.09% increase at 10 years.
• Sibling Analysis: When sibling pairs discordant in acetaminophen exposure were examined, no significant difference in autism risk was found between exposure and lack of exposure.
• Conclusion: Authors concluded that the association is non-causal and the observed link in the full cohort was likely due to confounding factors. Once genetics and home environment are taken into account, acettomenophen is not casually related to autism.
• Criticism: Acetaminophen use in Sweden is much lower than in the US.
• Japanese Cohort Study: This confirmed what the Swedish cohort found.
• Conclusion: It is very difficult to make a strong case for causality when two such large studies show no effect.

X. Challenges of Establishing Causality in Observational Data

• Difficult to identify and account for all confounding variables.
• Analogy: ice cream consumption and drowning (confounding variable is heat).
• Randomized trials are needed to establish causality, but they are often not possible for ethical or logistical reasons.
• Epidemiology is often at the mercy of not seeing all confounders.

XI. Applying the Bradford Hill Criteria

• A set of nine principles used to assess whether an observed association is likely to reflect a true causal relationship.
  • Strength: Very weak (1.05, the threshold for interesting is 1.5)
  • Consistency: Reasonably consistent, but disappears with the sibling analysis.
  • Specificity: Very non-specific; many other variables are linked to autism.
  • Temporality: Yes, acetaminophen use precedes autism.
  • Biological Gradient: Modest.
  • Biological Plausibility: Weak.
  • Analogy: Provides evidence against a causal relationship.
  • Experiment: No meaningful data.
  • Coherence: Somewhat weak.

XII. Overall Assessment of Acetaminophen and Autism

• There is possibly some statistical association.
• The probability that the association is causal is very low. It is a very low probability event that a woman who takes Tylenol during pregnancy is going to increase the probability that her child has autism.

XIII. Explaining the Increase in Autism Rates

• Even assuming causality, acetaminophen is not enough to explain the fivefold increase in autism prevalence.
• Genetics: Account for 80-90% of autism risk (heritability). Genetics do not account for a time scale of this speed of increase.
• Increased Awareness and Expanded Diagnostic Definitions: Account for 40-60% of the increase. The diagnostic aperture has also been expanded. 20-30% of this came from increased awareness. A 2009 study found that roughly 26% of the increase in autism diagnoses in California between 92 and 2005 were attributable specifically to cases in which children had previously been diagnosed with mental retardation and were then subsequently screened for autism.
• Advancing Parental Age: Accounts for 5-15% of the increase.
• Other Factors: Maternal obesity, metabolic disease, pre-term birth, air pollution. We've seen a 38% increase in PM2.5s. Air pollution in the US is largely attributed to wildfires.
• The long and short of it is genetics play a much larger role in autism risk than all other variables combined and account for an estimated 80 to 90% of the interindividual variability and autism risk.

XIV. Is Acetaminophen in the Bucket of Environmental Factors?

• Yes, it is possible.
• It is impossible to disprove anything.
• A great number of the variables are either demonstrably or have a much stronger association that even if acetaminophen plays some casual role it is going to be very very low.
• There are many things we should be looking at before this.

XV. Advice to Pregnant Women About Acetaminophen Use

• As a general rule, women should stop taking medications when they get pregnant.
• Balance the risks and benefits of medication use against the health of the mother. The medical conditions that these medications are intended to treat can sometimes also create problems indirectly or directly.
• Examples: Thyroid hormone (essential), GLP1 Agonists (evaluate if benefit outweighs risk).
• Important Principle: The single most important thing for the healthy development of a fetus is a healthy environment in utero.
• In general, women should stop taking medications when they get pregnant. The medications aren't the only potential threat to the unborn child. The health of the mother is also important to the unborn child.
• Look at how one would have to think about this.
• If a woman has an elevated APOB, she should be taking a lipid-linging medication. But does she need to take that during pregnancy?
• Conversely when we think about something like thyroid hormone where we've established actually quite um safe use during pregnancy if a woman is requiring thyroid hormone because she has hypothyroid to withhold that from her during pregnancy would pose enormous risk to her and by extension to the child.

XVI. Walking Through The FDA Risk Categories

• Category A, Category B, Category C, Category D, Category X
• You should be talking about this with your with your doctor.

XVII. Balancing the Benefits of Acetaminophen

• What's the risk of not taking Tylenol during pregnancy?
• For minor aches and pains, consider skipping the Tylenol.
• For more intense pain, the mother's well-being is important for the child.
• Tylenol is also used to reduce fever. It is currently evidenced that the scales clearly tip in favor of using Tylenol since exposure to fever itself carries a number of known risk factors to a developing fetus.
• In fact, prenatal exposure to fever and maternal infection are also separate risk factors for autism and other neurodedevelopmental disorders.
• It could be that it's the infection that is the issue and the signal we're picking up and measuring is the acetaminophen use.

XVIII. The Importance of Critical Thinking

• People don't want to think and want to outsource thinking to somebody else.
• We want to help you think about this stuff.
• If you just want sound bites, you're never going to learn.
• We must be able to do the thinking about it.
• People use stupid vehicles like social media to get their information and they don't want to read the fine print.
• Encourage everyone to practice thinking.
• If you don't want to, that's fine. It is hard. If so, then I think you've sort of forfeited your right to have an opinion on it.

XIX. Conclusion

• Any potential risk for autism can't be considered in isolation.
• Minor aches and pains: err on the side of caution.
• When the pain interferes with the environment for the fetus: judicious use with physician oversight.
• For maternal fever, the balance leans towards acetaminophen.
• The strength of the associations is small and vanishes with statistical corrections.