Hair loss more common with GLP-1s than other diabetes drugs

Clinical takeaway: GLP-1 receptor agonists may carry a small added risk of hair loss, most often reversible shedding tied to rapid weight loss. Mentioning it before starting therapy can help to prevent a later surprise and support longer-term adherence.

Patients are likely to notice hair loss, and they can be counted upon to act on it. It can dent confidence and quietly drive people off treatments that are working otherwise. The authors use alopecia as synonymous with hair loss, and the study tracks this as one broad outcome before splitting it into subtypes. The relevant clinical distinction is between the non-scarring kind, a diffuse shedding tied to rapid weight loss that usually grows back with weight stabilization, as opposed to the scarring or autoimmune forms that do not.

GLP-1 receptor agonists are now prescribed far beyond diabetes, and patients increasingly ask about the tradeoffs involved for weight loss. Hair loss has surfaced in case reports and FDA adverse-event filings, but those signals lacked comparison groups and controls. A new study emulating a randomized trial puts the question on firmer footing, measuring alopecia against two other glucose-lowering drug classes.

Users of GLP-1s had a 68% higher risk of hair loss than those on DPP-4 inhibitors, though the absolute numbers are small. That translates to roughly 7 cases per 1,000 person-years on treatment, versus about 4 cases in DPP-4 inhibitor users. The added risk was almost entirely for non-scarring alopecia, up 72%, with no clear rise in the autoimmune or pattern-balding types. The comparison against SGLT-2 inhibitors was weaker, a 37% higher risk that faded under testing. But SGLT-2 inhibitors also cause modest weight loss, so a smaller gap fits the theory that shedding tracks weight loss rather than the drug itself.

Hair loss tended to appear within the first year of treatment, peaking just before the 12-month mark, so the highest-risk window is early and finite rather than open-ended. The study did not track what happened after patients stopped, but non-scarring shedding of this kind typically reverses once the trigger settles, which is the basis for reassuring patients that it is usually temporary.

The study emulated a randomized trial from Penn Medicine records, comparing new GLP-1 users with new SGLT-2 or DPP-4 inhibitor users, about 12,000 per group, with weighting to balance the groups. Being observational, it shows association, not causation, so the researchers stress-tested each comparison for hidden bias.

What the study cannot say is how bad the shedding gets, how long it lasts, or how often it fully reverses, since it was based on diagnostic codes and not scalp exams or patient-reported detail. The study was also based on type 2 diabetes use, leaving people taking GLP-1s for weight alone unstudied. FDA has said it is evaluating hair loss as a potential side effect, and studies that follow patients with actual dermatology assessments would fill in the severity and recovery picture.

"Our findings extend previous anecdotal safety signals and provide more systematic evidence to inform clinical awareness of this potential adverse effect," the authors conclude. "In addition, these results highlight the need for further research to elucidate potential mechanistic links between GLP-1 receptor agonist therapy, weight loss, and hair loss, and to identify patient level risk factors that may increase susceptibility to hair loss."

Source: Tang H, et al. (2026 Jul 22) BMJ. Risk of hair loss associated with glucagon-like peptide-1 receptor agonists in adults with type 2 diabetes: target trial emulation