BMJ study finds higher hair loss risk with GLP-1 drugs including semaglutide and tirzepatide

A study published July 22 in The BMJ reports that adults with type 2 diabetes who started a GLP-1 receptor agonist were more likely to be diagnosed with hair loss than adults who started an SGLT-2 inhibitor or a DPP-4 inhibitor ^[1]. The researchers describe the absolute risk as low ^[1].

The work is a target trial emulation, an observational design that mimics a randomized trial, using electronic health records from Penn Medicine from January 2019 through September 2024 ^[1]. The exposed group included people who started exenatide, liraglutide, dulaglutide, semaglutide, lixisenatide or tirzepatide, so tirzepatide was grouped with the GLP-1 drugs for the analysis ^[1]. Patients who already had an alopecia diagnosis, type 1 diabetes, or end-stage kidney disease or dialysis were excluded ^[1].

The study compared 12,004 GLP-1 initiators with 15,221 SGLT-2 inhibitor initiators, and 11,964 GLP-1 initiators with 11,238 DPP-4 inhibitor initiators ^[1]. After weighting to balance baseline characteristics, GLP-1 use was linked to a higher risk of alopecia than SGLT-2 use (hazard ratio 1.37, 95% confidence interval 1.08 to 1.73) and than DPP-4 use (1.68, 1.28 to 2.20) ^[1].

The signal was specific to non-scarring alopecia, the type that does not destroy hair follicles ^[1]. For that subtype, hazard ratios were 1.53 (1.18 to 1.97) compared with SGLT-2 inhibitors and 1.72 (1.28 to 2.31) compared with DPP-4 inhibitors ^[1]. Results held across sensitivity and subgroup analyses but were weaker after negative control outcome calibration, a check the authors used to gauge how much unmeasured confounding could explain ^[1].

Weight loss and changes in nutritional status are established risk factors for telogen effluvium, a common and usually temporary form of non-scarring hair loss, which the authors offer as a plausible mechanism ^[1]. They also note that randomized trials have not routinely captured hair loss as an adverse event ^[1].

The study has limits. It is observational, so it can show an association but not prove that the drugs cause hair loss ^[1]. Hair loss was identified through diagnostic codes in one health system's records, and the authors say the absolute risk is low ^[1]. The published results do not give separate estimates for individual drugs such as semaglutide or tirzepatide, because the class was analyzed together, and whether the finding applies to people who take these drugs for weight loss without diabetes has not been reported ^[1].

Why it matters for patients

Hair loss does not typically cause physical harm, but the authors write that it can affect self-esteem, quality of life and adherence to treatment ^[1]. For adults with type 2 diabetes choosing among GLP-1 receptor agonists, SGLT-2 inhibitors and DPP-4 inhibitors, the study adds one more item to the list of possible side effects, alongside the low absolute risk the authors emphasize ^[1].

The study does not say that stopping a GLP-1 drug reverses hair loss, how long the increased risk lasts, or which patients are most likely to be affected; none of that has been reported ^[1]. Nor does it report whether the hair loss was mild or severe, since cases were identified from diagnostic codes rather than examined directly ^[1].

The FDA has indicated that it is evaluating the hair loss signal ^[1]. No agency conclusion, label change or review timeline has been reported ^[1].

What happens next

The BMJ published the study on July 22, 2026 ^[1]. The authors say awareness of this potential effect may help inform treatment decisions ^[1]. Whether the FDA's evaluation leads to any regulatory action has not been disclosed ^[1].