Zepbound at 5, 10 and 15 mg: what the label's own trials report
In the trial printed in Zepbound’s label, adults lost 15.0% of their body weight on 5 mg, 19.5% on 10 mg and 20.9% on 15 mg after 72 weeks, against 3.1% on placebo. The label approves all three as maintenance doses and names none of them as best: it says to choose on response and tolerability, and the side effects climb with the dose.
Reviewed by Ellis Brandt. Sources checked September 2026.
This guide draws on the cited FDA-approved prescribing information and other FDA sources. It is not medical advice; discuss treatment with your clinician. See the medical disclaimer.
What each dose did at 72 weeks
Study 1 is the trial most of these numbers come from: 2,539 adults with obesity, or with overweight and at least one weight-related condition, randomized to 5 mg, 10 mg, 15 mg or placebo for 72 weeks, everyone on a reduced-calorie diet and 150 minutes a week of activity. People with type 2 diabetes were excluded. Mean weight at the start was 104.8 kg.
Drag the table sideways to read the remaining columns.
| Arm | Mean weight change | Lost 5% or more | Lost 10% or more | Lost 15% or more | Lost 20% or more |
|---|---|---|---|---|---|
| Placebo (N=643) | −3.1% | 34.5% | 18.8% | 8.8% | 3.1% |
| Zepbound 5 mg (N=630) | −15.0% | 85.1% | 68.5% | 48.0% | 30.0% |
| Zepbound 10 mg (N=636) | −19.5% | 88.9% | 78.1% | 66.6% | 50.1% |
| Zepbound 15 mg (N=630) | −20.9% | 90.9% | 83.5% | 70.6% | 56.7% |
The gap between doses is not even. Going from 5 mg to 10 mg moved the average by 4.5 percentage points; going from 10 mg to 15 mg moved it by 1.4. Where the doses separate most is at the deep end of the range: half the patients on 10 mg and 56.7% on 15 mg lost at least a fifth of their body weight, against 30.0% on 5 mg.
Percent of body weight lost, as a least-squares mean for each arm of one 72-week trial. The track runs to 25%.
Zepbound prescribing information, section 14.1, Table 2, Study 1 (2,539 adults, 72 weeks). Four arms of one trial, which is the comparison it was designed to make. Label version 40, effective 2026-08-28, read September 2026.
The second trial, in people with type 2 diabetes
Study 2 ran the same 72 weeks in 938 adults who had type 2 diabetes and a BMI of 27 or above, and it tested only the two higher doses. The results are smaller: 12.8% on 10 mg and 14.7% on 15 mg, against 3.2% on placebo. Of that group, 60.5% on 10 mg and 64.8% on 15 mg lost at least a tenth of their weight, against 9.5% on placebo.
Those figures belong to that trial and that population. The label prints them separately for a reason, and a reader with type 2 diabetes reading the Study 1 column is reading a number measured in people who did not have it.
What the higher doses cost
The same section reports what came with the bigger numbers. Severe gastrointestinal reactions rose across the three arms of the two pooled weight trials: 1.7% on 5 mg, 2.5% on 10 mg and 3.1% on 15 mg, against 1% on placebo. Stomach complaints in general did not, holding at 56% on all three doses against 30% on placebo, and more patients stopped treatment because of them as the dose rose: 1.9%, 3.3% and 4.3%.
Fatigue and hair loss also track the dose in this label’s tables, and both are read out in full on fatigue on the four labels and hair loss on the four labels. The stomach rates for all four products sit together on side effect rates, label by label.
What the label says about choosing a dose
The label does not name a best dose. Section 2.2 lists the maintenance dosage for weight reduction as “5 mg, 10 mg, or 15 mg, injected subcutaneously once weekly”, and for obstructive sleep apnea as 10 mg or 15 mg. The maximum for every indication is 15 mg once weekly.
On how to land on one of the three, section 2.1 gives one sentence: “Consider treatment response and tolerability when selecting the maintenance dosage. If patients do not tolerate a maintenance dosage, consider a lower maintenance dosage.” Nobody starts at a maintenance dose either. Treatment begins at 2.5 mg once weekly for four weeks, a dosage the label says is “for treatment initiation and is not approved as a maintenance dosage”, and the climb after that runs in 2.5 mg steps no closer together than four weeks. The full ladder, beside the other three products’, is on the dose escalation schedules.
Two more things the trials do not tell you. They ran for 72 weeks with at least 52 weeks at the maintenance dose, so nothing in them describes a first month or a first quarter — what the label does and does not measure early on is on Zepbound at 3 months. And every number here is an average across an arm: the label’s own distribution shows patients across the whole range, which is why it reports how many reached 5%, 10%, 15% and 20% rather than one figure.
Sources
Drag the table sideways to read the remaining columns.
| Label | Applicant on the label | SPL version | Label effective date | Where it was read |
|---|---|---|---|---|
| Zepbound (tirzepatide) | Eli Lilly and Company | 40 | 2026-08-28 | DailyMed, set id 487cd7e7-434c-4925-99fa-aa80b1cc776b |
Every percentage is from Table 2 or section 6.1 of the Zepbound label; the dosing sentences are sections 2.1 and 2.2. Every figure and quotation comes from the label’s own tables and text. Each label is the FDA-approved version that was current when it was read in September 2026, identified in the table by its version number and effective date.
See the providers
A dose is a prescribing decision, not a purchase. What sellers charge, and whether the price moves when the dose does, is on the boards.
- Tirzepatide providersprovider board
Compare tirzepatide plans, pharmacy details and cancellation terms.
- Semaglutide providersprovider board
Compare semaglutide plans, pharmacy details and cancellation terms.
More, read from the same labels
- First-year GLP-1 costsresearch guide
A low monthly headline does not tell you what a year of treatment will cost. Compare the same medication and service package, then separate the total you expect to spend from the amount you must pay before treatment begins.
- Cancellation and refundsresearch guide
Canceling future membership charges, stopping an upcoming medication order and requesting a refund are three different requests. Get a written answer for each one that applies to you.
- Insurance denial documentsresearch guide
The next step depends on why coverage was denied. A missing document, a formulary restriction and an excluded benefit are different problems. Start with the written decision before paying someone to submit another request.
- Savings and cash-pay eligibilityresearch guide
A manufacturer savings card and a self-pay offer can have different eligibility rules, even for the same medicine. Match the offer to your coverage and exact product before using its advertised price in your budget.
- Verify your GLP-1 pharmacyresearch guide
The website taking your payment, the clinician prescribing and the pharmacy dispensing may be different businesses. Identify each one. A recognizable brand or a reassuring badge does not tell you who will fill the prescription.
- Changing GLP-1 providersresearch guide
Before ending your current service, establish what the new practice needs, what it will charge and when it can evaluate you. Transferring your records does not guarantee that another clinician will continue the same prescription.
- Weight regain after stoppingresearch guide
Weight regain occurred after treatment withdrawal in studies of semaglutide and tirzepatide. The findings support planning follow-up and affordability before an interruption, but they do not predict exactly what will happen to one person.
- Semaglutide vs tirzepatideresearch guide
What the four FDA-approved labels say each molecule is, what each is approved to treat, and where the two differ.
- How they workresearch guide
The receptor each molecule switches on, what the labels say that does to appetite, blood sugar and the stomach, and why one dose lasts a week.
- Is Ozempic insulin?research guide
What the Ozempic label calls semaglutide, how that differs from insulin, and where insulin appears on these labels as a separate medicine.
- Zepbound vs Ozempicresearch guide
Four brand names, two molecules and two kinds of approval: which of these products are the same drug, and what each one is actually approved to do.
- Dose escalation schedulesresearch guide
The starting dose, the escalation interval and the maximum dose each label sets, and why they are not interchangeable.
- Side effect ratesresearch guide
What each label reports for nausea, vomiting, diarrhea and constipation — and the sentence every label prints above those numbers.
- Hair lossresearch guide
Which labels report hair loss, the rate against placebo where a trial measured it, and why two of them put no number on it.
- Fatigue and tirednessresearch guide
Which labels list fatigue, the rate against placebo where a trial measured it, and what each label counts as fatigue.
- Missed dose rulesresearch guide
The window each label gives for a late dose, and the minimum gap it requires when the injection day changes.
- Drug interactionsresearch guide
Insulin and sulfonylureas, oral medications, and the oral-contraceptive precaution that appears on two of these four labels and not the other two.
- Buying one onlineresearch guide
The agency's own warning signs for a telehealth GLP-1 seller, its 28-day vial rule, and what it says about drugs that arrive warm.
- Zepbound at 3 monthsresearch guide
What does the Zepbound label report about three-month weight loss and maintenance dosing?
- Injecting tirzepatideresearch guide
The injection sites, the rotation rule and the steps for the single-dose pen, the KwikPen and the vial and syringe, exactly as the FDA-approved Instructions for Use give them.
- Refrigeration and storageresearch guide
How cold each pen is kept, how many days it can spend at room temperature, and what the labels say about heat and freezing.
- Who should not take oneresearch guide
The boxed warning every one of these labels carries, the two contraindications, and which warnings appear on which product.
Week 72, Study 1
| Placebo | −3.1% |
|---|---|
| Zepbound 5 mg | −15.0% |
| Zepbound 10 mg | −19.5% |
| Zepbound 15 mg | −20.9% |
Mean weight change in one 72-week trial of 2,539 adults without type 2 diabetes. All three doses are approved maintenance dosages.