Adults with type 2 diabetes who started a GLP-1 receptor agonist were diagnosed with hair loss 37% more often than similar patients who started an SGLT-2 inhibitor, according to a study published July 22 in The BMJ that used electronic health records from Penn Medicine. The US Food and Drug Administration is now formally evaluating the finding as a safety signal, and the study authors say the absolute risk of hair loss remains low.
Key takeaways
- GLP-1 starters had a 37% higher rate of hair loss diagnoses than SGLT-2 inhibitor starters and a 68% higher rate than DPP-4 inhibitor starters.
- The signal was specific to non-scarring alopecia, the reversible kind, not the permanent scarring kind.
- These are percentage increases on a small baseline, so the number of people affected is small, and the FDA review is a look at the question, not a verdict.
What the study found
The researchers used a method called target trial emulation, which takes real-world records and analyzes them the way a randomized trial would be analyzed. They looked at adults over 18 with type 2 diabetes who started one of three drug classes at Penn Medicine between January 2019 and September 2024.
In the first comparison, 12,004 people starting a GLP-1 receptor agonist were compared with 15,221 people starting an SGLT-2 inhibitor. New hair loss diagnoses were more common in the GLP-1 group, with a hazard ratio of 1.37 (95% confidence interval 1.08 to 1.73).
The second comparison matched 11,964 GLP-1 starters against 11,238 people starting a DPP-4 inhibitor, an older class of diabetes pills. There the hazard ratio was 1.68 (1.28 to 2.20).
When the team broke hair loss into subtypes, the link held only for non-scarring alopecia, with hazard ratios of 1.53 (1.18 to 1.97) versus SGLT-2 inhibitors and 1.72 (1.28 to 2.31) versus DPP-4 inhibitors. The results held up across sensitivity and subgroup analyses, though the authors note the effect shrank after they calibrated against negative control outcomes, a check for hidden bias.
Dr. Kumar’s take
The subtype detail is the most useful line in the paper, and most coverage will skip past it. Non-scarring alopecia includes telogen effluvium, a shedding phase where the follicle itself stays alive and hair grows back once the trigger passes. Rapid weight loss and sharp caloric restriction are classic triggers, no matter how the weight comes off. Crash diets do it. Bariatric surgery does it. So the mechanism I would bet on is not something toxic in semaglutide reaching the scalp. It is the speed of the weight loss the drug produces.
That distinction changes the fix. If the problem is pace, then a slower dose increase, adequate protein, and staying on the drug long enough for weight to stabilize are all reasonable responses. If the problem were direct follicle damage, none of that would help.
I would also be careful with the size of the number. GLP-1 drugs are the most watched prescriptions in medicine right now. Patients on them see their clinicians more often, ask more questions, and get referred to dermatology more readily than someone quietly refilling an older pill like a DPP-4 inhibitor. Diagnosis codes only exist when somebody writes them down. Some of that 37% and 68% gap is likely extra looking, not extra shedding, and the fact that the effect faded during negative control calibration is a hint in that direction. This is the same problem that makes head-to-head trials so valuable, the way a randomized trial was able to settle whether pantoprazole raises kidney risk.
Finally, a 37% relative increase on a low baseline rate is still a low rate. Alopecia diagnoses are uncommon in this population, and raising an uncommon event by a third leaves it uncommon. That is a different fact than “GLP-1 drugs cause hair loss,” which is how this will get headlined.
What it means for you
If you are on a GLP-1 drug and noticing more hair in the shower drain, tell your prescriber, but do not stop the medication on your own. Cardiovascular and metabolic benefits are well established, and the evidence on early treatment is strong enough that starting tirzepatide early outperformed stacking other diabetes drugs. Shedding that follows fast weight loss usually settles within a few months.
If you are deciding whether to start, this finding is worth knowing but is not a reason to say no. Ask about a slower dose increase, keep protein and iron intake adequate, and get thyroid and iron levels checked if shedding starts, since those are common and treatable causes that have nothing to do with the drug.
And if your goal is prevention rather than treatment, the lifestyle route remains real: a Mediterranean diet plus exercise cut diabetes risk by 31% in a randomized trial.