A new study is drawing attention to lesser-known side effects tied to popular GLP-1 medications used for weight loss and diabetes, after researchers analyzed hundreds of thousands of online discussions from users sharing their experiences.
According to researchers at the University of Pennsylvania, more than 400,000 Reddit posts from nearly 70,000 users were reviewed using artificial intelligence over a five-year period. The findings, first reported by Medical Xpress, were published last week in the journal Nature.
GLP-1 drugs, which include widely discussed medications prescribed for diabetes management and weight loss, are already known to cause gastrointestinal symptoms such as nausea, vomiting, diarrhea and headaches. Those side effects have been widely documented by medical sources including the Mayo Clinic.
However, the new study identified additional symptoms that users frequently discussed online. Researchers said those included “reproductive symptoms” and “temperature-related complaints.”
The reproductive issues mentioned in the study included “menstrual changes such as intermenstrual bleeding, heavy bleeding, and irregular cycles,” according to the university. Researchers also noted reports involving chills, feeling cold, hot flashes and fever-like symptoms.
Nearly 4 percent of users referenced what the study described as “menstrual irregularities,” according to Neil Sehgal, the study’s first author. He said the percentage could be higher in a sample made up exclusively of women and called it “a signal worth investigating.”
Researchers said approximately 44 percent of users reported at least one side effect. The most common complaint was nausea, with 37 percent of users mentioning it. Senior author Sharath Chandra Guntuku said that trend “shows that the method is picking up a real signal.”
Still, researchers cautioned against treating the findings as proof of direct causation. Sehgal added, “We can’t say that GLP-1s are actually causing these symptoms.”
The study highlights how large-scale social media analysis is increasingly being used to detect health trends, especially when patients discuss symptoms that may not always be raised in clinical settings.
