Adults with type 2 diabetes who use GLP-1 receptor agonists face a higher risk of hair loss compared to those on other common diabetes medications, according to a large study published in the BMJ.
The research found that patients taking GLP-1 drugs like semaglutide and tirzepatide had a 37% higher risk of alopecia than those using SGLT-2 inhibitors, and a 68% higher risk compared to DPP-4 inhibitor users.
The absolute rates remained low, the team noted, but the difference was statistically significant.
Study design and patient groups
Researchers from the University of Pennsylvania Health System analyzed electronic health records for more than 23,000 patients who started GLP-1 receptor agonists between January 2019 and September 2024. They compared these patients against similar groups using SGLT-2 inhibitors and DPP-4 inhibitors.
The GLP-1 group skewed younger on average, with a mean age of 58 compared to 65 for SGLT-2 users and 67 for DPP-4 users. Body mass index was also higher in the cohort at 36.2 versus 32.3 and 31.3 in the comparison groups.
Related: New Degrader Technology Targets Cancer Cells Selectively
After adjusting for age, sex, ethnicity, pre-existing conditions, other medications, and BMI, the association held steady. Hair loss occurred at a rate of 6.91 per 1,000 person‑years among these agents compared to 5.04 for SGLT-2 users and 3.89 for DPP-4 users.
Types of hair loss and possible causes
The increased risk appeared specific to non‑scarring alopecia, where hair follicles remain intact and regrowth is theoretically possible. In this category, the risk jumped 53% relative to SGLT-2 inhibitors and 72% relative to DPP-4 inhibitors.
The study authors suggested rapid weight loss as a plausible explanation. Sudden weight reduction is known to trigger hair shedding and can lead to deficiencies in iron and zinc, both critical to the hair growth cycle. Hormonal changes triggered by the drugs may also play a role, though they said further studies are needed to confirm the exact mechanism.
The authors acknowledged gaps in the data. Details on severity, duration, and whether hair loss reversed after stopping treatment were not available. Unmeasured factors could also have influenced the results, they said.
Despite the limitations, the investigators described their study as rigorous and based on high‑quality data from a large representative cohort. Results held after additional analyses, they said, suggesting the findings are reliable.
Related: AI platform accelerates discovery of cancer drug candidates
The study was authored by Tang H, Zhang B, Lu Y, Zhang D, Liu R, Lu Y, Cotsarelis G, Asch DA, and Chen Y.
It appeared in the BMJ.
The citation reads BMJ. 2026 Jul 22;394:e100077.
For now, the team said awareness of this potential side effect may help doctors and patients weigh the benefits of GLP-1 medications against the risks when making treatment decisions.
