Self-Reported Side Effects of Semaglutide and Tirzepatide in Online Communities

Neil K. R. Sehgal, Jena Shaw Tronieri, Lyle Ungar, Sharath Chandra Guntuku

Abstract

Social media can reveal patient experiences with glucagon-like peptide-1 receptor agonists (GLP-1 RAs) that extend beyond clinical trial data. We analyzed 410,198 Reddit posts (May 2019–June 2025) mentioning semaglutide or tirzepatide. A total of 67,008 users self-reported using these medications, and 43.5% described at least one side effect. Gastrointestinal symptoms predominated, including nausea (36.9%), fatigue (16.7%), vomiting (16.3%), constipation (15.3%), and diarrhea (12.6%). Notably, reproductive symptoms (e.g., menstrual irregularities) and temperature-related complaints (e.g., chills, hot flashes) emerged as unrecognized potential effects. These findings highlight patient concerns not well captured in current labeling or trials. Large-scale social media analysis can complement traditional pharmacovigilance by detecting emerging safety signals and expanding understanding of the real-world safety profile of GLP-1 RAs.

Introduction

While clinical trials are the gold standard for drug approval, they often lack the scale and diversity of the general population. Once drugs like semaglutide (Ozempic/Wegovy) and tirzepatide (Mounjaro/Zepbound) reach millions of users, "real-world" side effects emerge that weren't captured in controlled environments. Patients often turn to online communities to discuss sensitive or "minor" symptoms they might not mention to a doctor. This study addresses the gap in traditional pharmacovigilance by using social media as a massive, unprompted sensor for drug safety, aiming to identify symptoms that haven't yet made it onto official drug labels.

Methodology

The researchers developed a high-throughput Natural Language Processing (NLP) pipeline using OpenAI’s "mini" models to process a massive corpus of text from nine large subreddits focused on GLP-1 RAs and weight management.

The pipeline involved three main stages:

  1. Identification: A GPT-4o-mini classifier filtered 410,198 posts to find users who explicitly disclosed personal use of the medications, achieving 86% sensitivity and 96% specificity.
  2. Extraction & Normalization: A second classifier (GPT-4.1-mini) extracted side effect mentions and mapped them to MedDRA Preferred Terms (PTs), the industry standard for medical coding.
  3. Validation: Manual annotation of samples confirmed a high sensitivity of 97% for side effect identification, ensuring the model accurately captured patient experiences without hallucinating symptoms.

The final cohort included 67,008 unique users identified as medication users, with data spanning from May 2019 to mid-2025.

Results: The Real-World Safety Profile

Of the self-disclosed users, 43.5% (29,172 individuals) reported at least one side effect, with an average of 2.7 symptoms per user. Gastrointestinal (GI) disorders were the most prevalent, reported by 65.8% of those experiencing side effects.

Table 1: Prevalence of Gastrointestinal, General, and Nervous System Disorders.

As shown in the data, nausea was the single most common complaint (36.9%). Fatigue emerged as a significant "general disorder," reported by 16.7% of users. This is particularly noteworthy as fatigue has met reporting thresholds in relatively few clinical trials, suggesting a discrepancy between clinical observation and patient experience.

Emerging Safety Signals

Beyond the expected GI issues, the study identified several "hidden" side effects that warrant further investigation.

Table 2: Prevalence of Metabolism, Psychiatric, Skin, Musculoskeletal, and Reproductive Disorders.

Reproductive and Temperature Regulation

Nearly 4% of users reporting side effects mentioned reproductive issues, such as menstrual irregularities and heavy bleeding (3.8% total for the category). Additionally, temperature-related complaints like chills (1.1%) and hot flashes (0.5%) were documented. The authors suggest these may be linked to GLP-1 receptors in the hypothalamus, which regulates both the menstrual cycle and thermogenesis.

Psychiatric and Neurological Observations

Anxiety (4.2%), depression (2.8%), and insomnia (3.1%) were also reported. In the neurological category, headaches (6.1%) and dizziness (5.0%) were the most frequent.

Drug Comparison

In a sub-analysis of users taking only one drug, semaglutide users reported higher rates of nausea (39.4% vs 28.6%) and vomiting (18.0% vs 11.1%) compared to those on tirzepatide.

Discussion and Limitations

This study demonstrates that LLMs can transform messy, anecdotal social media data into structured "signals" to inform future clinical trials. However, several limitations exist:

  • Selection Bias: Reddit users are not representative of the global population; they tend to be younger and primarily US-based.
  • Causality: As observational data, it is impossible to prove the drug caused the symptoms. Factors like rapid weight loss or underlying conditions (e.g., PCOS) could drive changes in menstruation.
  • Data Gaps: Social media lacks critical metadata like exact dosage or duration of treatment.

Conclusion

Large-scale social media analysis serves as a powerful "hypothesis generator" for pharmacovigilance. By capturing the unprompted voices of tens of thousands of patients, researchers can identify emerging safety signals—like reproductive and temperature regulation issues—that might otherwise remain overlooked in traditional clinical settings.

Summary updated

Abstract

ABSTRACT Social media can reveal patient experiences with glucagon-like peptide-1 receptor agonists (GLP-1 RAs) that extend beyond clinical trial data. We analyzed 410,198 Reddit posts (May 2019–June 2025) mentioning semaglutide or tirzepatide. A total of 67,008 users self-reported using these medications, and 43.5% described at least one side effect. Gastrointestinal symptoms predominated, including nausea (36.9%), fatigue (16.7%), vomiting (16.3%), constipation (15.3%), and diarrhea (12.6%). Notably, reproductive symptoms (e.g., menstrual irregularities) and temperature-related complaints (e.g., chills, hot flashes) emerged as unrecognized potential effects. These findings highlight patient concerns not well captured in current labeling or trials. Large-scale social media analysis can complement traditional pharmacovigilance by detecting emerging safety signals and expanding understanding of the real-world safety profile of GLP-1 RAs.
BibTeX
@misc{sehgal2026self,
  author = {Neil K. R. Sehgal and Jena Shaw Tronieri and Lyle Ungar and Sharath Chandra Guntuku},
  title = {{Self-Reported Side Effects of Semaglutide and Tirzepatide in Online Communities}},
  year = {2026},
  eprint = {2603.12341},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2603.12341},
}
Figure