The US prescribing information for phentermine does not list hair loss or alopecia anywhere, and no randomised trial or cohort study has put a rate on it. The weight loss drug class that does have numbers is the GLP-1 receptor agonists: in Wegovy labelling, hair loss was reported by 3.3% of adults on semaglutide 2.4 mg against 1% on placebo, and by 5.8% at the 7.2 mg dose.
A 2026 meta-analysis of nine trials in 4114 GLP-1 users put the pooled risk ratio at 3.25 against placebo (PMID 42155605). What the evidence points at, consistently, is the weight loss rather than the molecule. Here is what each source actually recorded.
- Phentermine labelling lists cardiovascular, CNS, gastrointestinal, allergic and endocrine reactions, and no hair loss.
- Wegovy labelling records hair loss in 3.3% on 2.4 mg and 5.8% on 7.2 mg, against about 1% on placebo.
- The effect is heavily skewed by sex: 8.4% of women against 0.2% of men in the 7.2 mg trials.
- A 2026 meta-analysis of 4114 GLP-1 users found a pooled risk ratio of 3.25, with an absolute event rate of 3.9%.
- A matched cohort found telogen effluvium elevated, alopecia areata not, and lower follow-up BMI in the users who developed hair loss.
What phentermine's label actually lists
The US prescribing information for Adipex-P, the most widely prescribed phentermine product, lists its adverse reactions by body system:
- Cardiovascular: primary pulmonary hypertension and regurgitant cardiac valvular disease, palpitation, tachycardia, elevated blood pressure, ischaemic events
- Central nervous system: overstimulation, restlessness, dizziness, insomnia, euphoria, dysphoria, tremor, headache, psychosis
- Gastrointestinal: dryness of the mouth, unpleasant taste, diarrhoea, constipation, other gastrointestinal disturbances
- Allergic: urticaria
- Endocrine: impotence, changes in libido
Alopecia and hair loss appear nowhere in that document. Some drug reference databases list hair loss under phentermine with the frequency marked as not defined, which is a different claim: it means the effect has been reported somewhere without a denominator attached. No randomised trial or matched cohort study has quantified a rate.
That is not the same as proof of absence. It does mean that anyone quoting a phentermine hair loss percentage is quoting something that has not been measured.
What semaglutide's label records, with a dose gradient
Semaglutide is the opposite case, because the numbers exist and are specific. From the Wegovy US prescribing information:
| Population and comparison | Hair loss on treatment | Placebo |
|---|---|---|
| Adults, pooled weight-management studies, 2.4 mg | 3.3% (4% of women, 0.9% of men) | 1% (2% of women, 0 men) |
| Adults, 7.2 mg trials | 5.8% (8.4% of women, 0.2% of men) | 1.0% |
| Adults, 7.2 mg trials, 2.4 mg comparator arm | 3.3% (5.4% of women, 0 men) | 1.0% |
| Pediatric trial, 2.4 mg | 4% | 0% |
Two gradients run through that table. Hair loss roughly doubles from the 2.4 mg dose to the 7.2 mg dose, and it is overwhelmingly reported by women: 8.4% against 0.2% in the higher-dose trials. The label does not leave the interpretation open either. It states that hair loss adverse reactions in treated patients "were associated with weight reduction".
The Ozempic labelling, which covers semaglutide at a 2 mg ceiling rather than 2.4 mg, lists alopecia only under postmarketing experience, where reports arrive without a denominator and no rate can be calculated.
Neither of our existing rundowns of semaglutide side effects or tirzepatide side effects covers this one, because it sits well below the gastrointestinal events that dominate those frequency tables.
The pooled trial evidence
A 2026 systematic review and meta-analysis searched four databases to August 2025 and pooled nine interventional studies, seven of them randomised, covering 4114 GLP-1 receptor agonist users.
The pooled risk ratio for hair loss against placebo was 3.252 (95% CI 1.437 to 7.358). Restricted to the randomised trials in people with overweight or obesity, it was 3.587 (95% CI 2.100 to 6.124). A single-arm analysis put the event rate among GLP-1 users at 3.9%, which lines up closely with the 3.3% in Wegovy's own pooled table (PMID 42155605).
So the relative risk is roughly tripled and the absolute risk stays in the low single digits. Both halves of that sentence matter: a three-fold increase on a 1% baseline is still an uncommon event.
Which kind of hair loss, and the clue hidden in the data
A propensity-score-matched cohort study using TriNetX records from 2018 to 2025 compared adults starting semaglutide or tirzepatide with matched non-users, controlling for demographics, comorbidities and other medications associated with hair loss (PMID 42647098).
Both drugs were associated with higher odds of nonscarring alopecia in the type 2 diabetes and the BMI cohorts. Telogen effluvium specifically was elevated with semaglutide at three and five years and with tirzepatide at every follow-up window. Alopecia areata, included deliberately as a negative control because it is autoimmune rather than stress-driven, showed no association at all.
The most informative finding is the one buried in the secondary analysis: among GLP-1 users, follow-up BMI was lower in the patients who developed hair loss than in those who did not. The people losing the most weight were the people losing hair.
Telogen effluvium is what fits
Telogen effluvium is a diffuse shedding that happens when a physiological stress pushes an unusually large share of hair follicles out of their growth phase and into the resting phase at the same time. Because shed hairs do not fall until the resting phase ends, the shedding shows up roughly two to three months after the trigger rather than during it.
Rapid weight loss is a recognised trigger, and the clearest precedent comes from bariatric surgery, where the same pattern was documented long before GLP-1 drugs existed. A case report and review of post-surgical hair loss describes an acute telogen effluvium appearing within the first three months after surgery, and a separate, later-onset pattern around six months that is more often tied to nutritional deficiency (PMID 34055500).
That two-phase picture maps onto what the GLP-1 data shows: a shedding signal that scales with the amount of weight lost, is strongest in the groups losing most, and does not touch the autoimmune form of hair loss at all. The drug class appears to be the cause of the weight loss rather than a direct cause of the shedding.
What the evidence does not establish
- Whether it is reversible, and how fast. No GLP-1 trial has followed hair outcomes to resolution. The bariatric literature generally describes acute telogen effluvium as self-limiting, but that is a different population and a different trigger.
- Why the sex difference is so large. An 8.4% versus 0.2% split between women and men on 7.2 mg is too wide to be chance, and nothing in the published data explains it.
- Whether anything prevents it. Nutritional adequacy is the lever the bariatric literature discusses, since intake falls sharply on these drugs. Nobody has run a trial testing whether that changes hair outcomes on a GLP-1.
- Phentermine's rate, at all. There is no denominator to work with.
Any of this is worth raising with the prescriber or a dermatologist rather than acting on alone, particularly because patchy loss, scaling or scarring points somewhere other than telogen effluvium.
See the size of the loss the signal tracks
Put a starting weight against the published trial percentages and see what each one works out to.
Since the signal tracks the amount of weight lost rather than the drug itself, the useful context is the size of the expected loss. The GLP-1 weight loss calculator converts the published trial percentages into pounds from a given starting weight, and how much weight you can lose on Wegovy sets out the trial figures in full.
FAQ
Does phentermine cause hair loss?
The US prescribing information for phentermine does not list alopecia or hair loss among its adverse reactions, and no randomised trial or matched cohort has measured a rate. Some drug databases list it with the frequency marked as not defined, which means it has been reported without a denominator. Rapid weight loss itself is a recognised trigger for telogen effluvium regardless of which drug produced it.
How common is hair loss on Wegovy?
Wegovy labelling records hair loss in 3.3% of adults on 2.4 mg against 1% on placebo in pooled weight-management studies, and 5.8% on the 7.2 mg dose. In a pediatric trial it was 4% against no placebo-treated patients.
Is hair loss on a GLP-1 caused by the drug or the weight loss?
The evidence points at the weight loss. Wegovy labelling states that hair loss reactions were associated with weight reduction, a matched cohort study found lower follow-up BMI in the GLP-1 users who developed hair loss than in those who did not, and the subtype implicated is telogen effluvium, a known response to rapid physiological change.
What is telogen effluvium?
A diffuse shedding that happens when a stress pushes an unusually large share of hair follicles out of the growth phase into the resting phase together. Because the shed is delayed until the resting phase ends, it typically becomes visible two to three months after the trigger rather than at the time.
Why do women report it so much more often?
The size of the gap is clear and the explanation is not. In the 7.2 mg semaglutide trials, hair loss was reported by 8.4% of women and 0.2% of men; at 2.4 mg it was 4% of women and 0.9% of men. Nothing in the published data accounts for that difference.
Does hair grow back after stopping?
No GLP-1 trial has followed hair outcomes through to resolution, so there is no trial-based answer. The bariatric surgery literature, where the same acute telogen effluvium pattern was documented first, generally describes it as self-limiting, but that is a different population and trigger. Persistent, patchy or scarring loss is worth a dermatology assessment rather than an assumption.
Sources
- Cheng PL, Chang HC. Glucagon-like peptide-1 receptor agonists and hair loss: a systematic review and meta-analysis. Diabetes Res Clin Pract. 2026;237:113333. PMID: 42155605
- Katragadda R, Hayden J, Saltagi AK, et al. Nonscarring alopecia in adults treated with GLP-1s: a propensity score matched TriNetX cohort study. Laryngoscope. 2026. PMID: 42647098
- Cohen-Kurzrock RA, Cohen PR. Bariatric surgery-induced telogen effluvium (Bar SITE): case report and a review of hair loss following weight loss surgery. Cureus. 2021;13(4):e14617. PMID: 34055500
- Wegovy (semaglutide) injection, US prescribing information, Novo Nordisk, revised June 2026. Section 6.1 Adverse Reactions, hair loss subsection and Table 4.
- Adipex-P (phentermine hydrochloride) tablets, US prescribing information. Adverse reactions by body system.
- Ozempic (semaglutide) injection, US prescribing information, Novo Nordisk. Section 6.2 Postmarketing Experience, skin and subcutaneous tissue disorders.
This article is for information only and is not medical advice. Dosing, suitability and any change to treatment are decisions for your prescriber.