# A Noisy Signal? Geographic Bias in FAERS Reports Linking Paracetamol to Autism Spectrum Disorder

## TL;DR

This study examines if FAERS reports linking paracetamol use during pregnancy to autism are due to geographic and reporting biases rather than a true causal link.

## Contribution

The study identifies geographic and temporal biases in FAERS reports linking paracetamol to autism, highlighting the role of reporting dynamics.

## Key findings

- Most ASD-related paracetamol reports in FAERS originated from the U.S. and were submitted by consumers.
- A sharp increase in reports after 2022 suggests a trend driven by public attention rather than a medical reality.
- Disproportionality metrics showed strong signals, but these may not reflect a true causal relationship.

## Abstract

Background/Objectives: Recent public and scientific discussions have raised concerns about a possible link between prenatal paracetamol exposure and autism spectrum disorder (ASD). However, pharmacovigilance-based evidence remains scarce, and the role of reporting bias has not been systematically assessed. This study aimed to characterize ASD-related adverse event reports involving paracetamol in the U.S. Food and Drug Administration’s Adverse Event Reporting System (FAERS) and to evaluate potential disproportionality signals, considering demographic, temporal, and geographic patterns. Methods: FAERS data from January 2010 to September 2025 were screened for reports listing paracetamol as the suspect drug and ASD-related Preferred Terms. After excluding duplicates and concomitant drugs, 1776 unique cases were analyzed. Patient demographics, reporter type, and country of origin were summarized descriptively. Disproportionality was calculated using four algorithms: Reporting Odds Ratio (ROR), Proportional Reporting Ratio (PRR), Information Component (IC), and Empirical Bayes Geometric Mean (EBGM). Results: Among 172,129 paracetamol-associated reports, 1776 (1.03%) included ASD-related terms. All were classified as serious and mostly submitted by consumers (98.6%). Gender was available in 47.7% of cases, showing male predominance (68.8%). Most reports referred to fetal exposure during pregnancy. Nearly all originated from the United States (98.4%). A marked rise was observed after 2022, with 562 reports in 2023 and 1051 in 2025. Disproportionality analyses revealed consistently elevated signals (ROR = 69.8, PRR = 69.2, IC025 = 5.60, EB05 = 48.3). Conclusions: The strong disproportionality signal likely reflects increased public attention and reporting dynamics rather than a causal association. Further integration of pharmacovigilance and epidemiologic data is warranted to clarify the clinical significance of these findings.