Preprint links semaglutide to higher hair loss rate in women
A non-peer-reviewed claims study found women on semaglutide were diagnosed with hair loss about twice as often as women on another weight-loss drug.
A preprint posted March 6, 2025 reports that women taking semaglutide for weight loss were diagnosed with hair loss roughly twice as often as women taking bupropion-naltrexone, an older weight-loss drug combination [1]. The paper appears on medRxiv and has not been peer reviewed; its own disclaimer states it "should not be used to guide clinical practice" [1].
The authors are affiliated with the University of British Columbia, Epilytics and StatExpert Ltd, all in Canada [1]. They ran a retrospective cohort study using a large health claims database and a random sample of 16 million patients [1]. They identified new users of semaglutide and new users of bupropion-naltrexone, and excluded anyone with a diabetes diagnosis or antihyperglycemic use before entering the study group [1]. Semaglutide is the molecule in Ozempic, Wegovy and Rybelsus.
What the study found
The cohort included 1,926 semaglutide users and 1,348 users of bupropion-naltrexone [1]. Participants were followed from their first prescription for either drug until a hair loss diagnosis [1]. The analysis adjusted for age, sex, geographic location, depression, steroid use, hypothyroidism, polycystic ovary syndrome and anemia [1].
The adjusted hazard ratio for hair loss was 1.52 for all semaglutide users compared with bupropion-naltrexone, with a 95% confidence interval of 0.86 to 2.69 [1]. For men the adjusted hazard ratio was 0.86, with a wide confidence interval of 0.05 to 14.49 [1]. For women it was 2.08, with a confidence interval of 1.17 to 3.72 [1].
A hazard ratio compares how often a diagnosis happens in one group versus another. The women's figure of 2.08 means hair loss was diagnosed about twice as often in that group. Because the overall figure's confidence interval runs from below 1.0 to well above it, that overall result is not statistically significant; the women's confidence interval stays above 1.0. The number of hair loss cases in each group is not stated in the abstract. The authors concluded their results "demonstrate an increased risk of hair loss with semaglutide in women" and that future studies are needed [1].
The paper's revision notes say the authors replaced the term "alopecia areata" with "hair loss" throughout, added the specific ICD-9 and ICD-10 diagnosis codes used to identify cases, and state that "the results of the study changed when including different diagnosis codes" [1]. Only semaglutide was studied as a drug of interest; liraglutide was excluded [1].
One author, Dr. Mahyar Etminan, discloses that he has previously consulted on the Ozempic litigation [1]. The study was funded by internal research funds from the University of British Columbia's Department of Ophthalmology and Visual Sciences [1].
Why it matters for patients
Hair loss is a visible, distressing side effect, and this is one of the first large database studies to look specifically at it with semaglutide [1]. The design has an important strength: it compares semaglutide against another weight-loss drug rather than against no treatment, and it adjusts for thyroid problems, anemia, PCOS and steroid use, all of which can cause hair loss on their own [1].
It does not show that semaglutide causes hair loss. It shows an association in one claims database, in one non-peer-reviewed analysis, and only in women. Several things a patient might want to know have not been reported: the absolute number of people affected, whether hair regrew after stopping the drug, whether risk depends on dose or how long someone takes it, and whether the same signal appears with tirzepatide, the molecule in Mounjaro and Zepbound. Nothing in this paper addresses those questions.
What happens next
The preprint has not been peer reviewed, and no journal publication or review date has been disclosed [1]. The authors write that future studies are required to clarify the association [1]. The data are available from the authors on reasonable request [1]. Until the work is reviewed and replicated, this finding stands as a signal to follow, not an established risk.
Sources
Semaglutides.org provides information only, not medical advice. Talk to a licensed provider about your own care before making changes. Some telehealth links are affiliate links and are labeled where they appear.