GLP-1 and hair loss: what labels and research actually show

A reported symptom deserves a precise explanation

Hair loss during GLP-1 treatment can be upsetting, and it deserves a clinical conversation. Current Wegovy and Zepbound prescribing information includes hair loss among reported adverse reactions. That gives the concern a place in the evidence; it does not show why one person is shedding or mean that every change in hair has the same cause. ^[1]^[2]

I would separate three questions: was hair loss reported during treatment, what type of hair loss is being observed, and what might explain it? The first can be addressed by a trial table. The second and third require individual assessment. Combining them into “the drug damages hair follicles” moves beyond what the reported event itself establishes.

This article uses current product labels, a pivotal trial and primary dermatology research. It does not recommend a hair treatment, supplement or medication change. Its purpose is to help you read the findings and describe the concern clearly to the clinician responsible for your care.

What the labels report

In the current Wegovy label’s pooled adult weight-management studies of the 2.4 mg weekly injection, the adverse-reaction table reports hair loss in 3% of treated participants and 1% with placebo. A separate paragraph provides a more precise treated figure of 3.3% and says the reactions were associated with weight reduction. The label also contains other populations and formulations; this figure should not be applied to every semaglutide product. ^[1]

Zepbound’s pooled Studies 1 and 2 table reports 5%, 4% and 5% in its three treatment groups versus 1% with placebo. Its hair-loss paragraph also describes an association with weight reduction. These are study observations, not your personal probability and not proof of a direct toxic effect on a follicle. ^[2]

Read the event and its limit together

Wegovy injection study poolTable: 3% versus 1% placebo. Applies to the specified adult 2.4 mg injection studies. ^[1]
Zepbound study poolTable: 4–5% versus 1% placebo. Applies to the specified pooled studies. ^[2]
MechanismA reported hair-loss event does not identify its subtype or prove a direct follicle effect.

The two rows are not a head-to-head comparison. Different participants, regimens and study pools prevent using them as a simple ranking of which medicine is “safer for hair.”

A pivotal weight-loss trial is not a mechanism study

SURMOUNT-1 randomized 2,539 adults with obesity, or overweight with a weight-related complication, without diabetes, to tirzepatide or placebo for 72 weeks. It established substantial average weight reduction in the studied setting. Lilly funded the trial. The study’s main question concerned weight outcomes and safety, not a dedicated investigation of hair-follicle biology. ^[3]

Randomization supports interpretation of differences between assigned groups. It does not automatically separate a drug’s direct effect from an effect mediated through changes in weight, intake or another part of treatment. Resolving that pathway would require measurements and analyses designed for it. A trial can therefore provide a meaningful adverse-event signal while leaving the explanation unsettled.

Our guide to reading GLP-1 trials explains why the study population, comparator and duration belong beside any headline number. Hair outcomes deserve the same discipline as weight outcomes.

Why telogen effluvium enters the discussion

Telogen effluvium is a pattern of diffuse shedding associated with disruption of the hair-growth cycle after a trigger. Weight loss is one context in which it has been observed. Kang and colleagues retrospectively studied 140 patients already diagnosed with weight-loss-associated telogen effluvium at one dermatology center. Their discussion proposes reduced caloric intake as a possible explanation. ^[4]

That design matters. It describes selected people who had the condition; it does not measure how often all people losing weight develop it. It was not a randomized GLP-1 medication trial. The authors identify incomplete recording of other potential causes and recall bias in reported weight change. Their hypothesis does not prove that rapid weight loss explains every GLP-1-associated hair event. ^[4]

The useful distinction is between a plausible pathway and an established diagnosis. A treatment timeline and a weight timeline may help the clinician investigate. Neither a fast nor a slow scale change lets a reader identify the hair-loss subtype from an article. Other explanations may need consideration, and the assessment should remain open.

The dermatology literature still has gaps

A 2025 JAAD paper by Burke and colleagues specifically examines GLP-1 medications and hair loss using a retrospective cohort design. The PubMed record verifies the publication but provides no abstract; accessible full findings were not available for this review. We therefore do not quote its risk estimate or turn secondhand summaries into a clinical conclusion. ^[5]

More generally, a retrospective association requires careful interpretation of who received treatment, how diagnoses were recorded and what differences remained between groups. An association can identify an important question without resolving causation. “Alopecia” is also a broad term: a general event report and a dermatologist’s specific diagnosis are different levels of information.

I would be cautious with claims that promise certainty in either direction. The labels support taking reported hair loss seriously. The available evidence does not support assuming that every episode is directly caused by the medication or dismissing every episode as harmless temporary shedding.

Bring observations, not a self-diagnosis

Prepare a short record of the exact product, when treatment began, when shedding became noticeable and any clinician-directed changes. Describe whether the concern is shedding, thinning or patches, and whether other symptoms appeared. This helps frame an assessment without asking you to decide the cause yourself.

Ask who will review the concern and whether an examination or referral is needed. Our ongoing-support guide can help clarify the response pathway. A portal that accepts messages is only useful when you know how a clinical question reaches the prescriber.

Keep treatment decisions with your qualified clinician. A careful response should acknowledge what the labels report, consider your history and explain the next step. A product upsell or a generic reassurance cannot perform that assessment.