# Political Affiliation Drives Beliefs on Tylenol-Autism Link Over Wording Survey experiment shows partisan divides outweigh linguistic nuances in public health perception. - **Source:** [PsyPost](https://www.psypost.org) - **Published:** 2026-11-09 - **Topic:** [science](https://brifing.si/c/science) - **Original article:** - **On Brifing:** https://brifing.si/s/f5f881af-03b3-45ab-9677-4141a9de512c ## Study Design and Methodology Researchers recruited 1,957 U.S. adults via the CloudResearch Connect platform to test how survey phrasing impacts health beliefs. Participants were randomly assigned to read claims using 'causes,' 'increases the risk,' or 'link' regarding Tylenol and autism. The study was preregistered to ensure analytical rigor and prevent bias during data collection. ## Minimal Impact of Phrasing The experiment revealed that specific wording had a surprisingly modest effect on overall agreement rates. While participants were slightly less likely to endorse the claim when framed as 'causes,' the difference between the three phrasings was not statistically significant. Most respondents maintained their positions regardless of the specific linguistic framing used. ## Partisan Divide Remains Dominant Political affiliation emerged as the strongest predictor of belief, far outweighing the influence of survey wording. Republicans were nearly ten times more likely to endorse the link between Tylenol and autism compared to Democrats. Independents also showed significantly higher odds of agreement than Democrats, highlighting a deep, persistent partisan polarization. ## Education and Causal Reasoning Educational attainment influenced how participants interpreted causal language. Respondents with a four-year college degree were 34 percent less likely to endorse the claim overall. Specifically, these graduates were more sensitive to the distinction between 'causes' and 'increases the risk,' whereas participants without a degree showed almost no difference across the three phrasings. ## Context of the Controversy The study followed a September 2025 press conference where President Donald Trump and HHS Secretary Robert F. Kennedy Jr. discussed acetaminophen use during pregnancy. This event triggered widespread public debate and a subsequent national poll by the Kaiser Family Foundation, which reported high levels of belief in a Tylenol-autism connection. ## Scientific Evidence vs Public Perception While some observational studies have noted statistical associations, the broader scientific consensus does not support a cause-and-effect relationship. A 2024 study of 2.5 million children in Sweden accounted for genetic factors and found no increased risk, contradicting the narrative that prenatal acetaminophen exposure directly causes autism in developing children. ## Limitations of Survey Data The researchers acknowledged that their sample was not nationally representative, as participants were recruited from an online platform. Additionally, the study examined a single, highly polarized health claim. These factors suggest that wording effects might function differently in less politically charged health topics, requiring further investigation into broader public health communication. ## Distinguishing Correlation and Causation Scientific communication often struggles to convey the difference between correlation and causation to the general public. Roughly one-third of university press releases have been found to use causal language for correlational findings. This study suggests that current survey methods fail to capture whether respondents truly believe in direct causation. ## Future Research Directions The authors propose developing new survey formats that explicitly separate awareness of statistical associations from beliefs in direct cause-and-effect. By refining these tools, policymakers and health communicators can better understand public sentiment. Such improvements are essential for addressing misinformation and ensuring that scientific findings are accurately interpreted by the general population. ## Implications for Health Policy The findings underscore that changing survey wording is insufficient to alter deeply held beliefs on polarized issues. Because political identity and education level drive perception more than linguistic precision, health authorities face significant challenges in correcting misinformation. Effective communication strategies must account for these underlying social factors rather than relying on phrasing. ## Sources covering this story - https://www.psypost.org/republicans-much-more-likely-to-think-tylenol-in-pregnancy-is-tied-to-autism-no-matter-how-the-question-is-asked/ — --- _Summary written by Brifing from the original reporting. Copyright in the original article belongs to its publisher; read the full piece at the link above._