Short answer: 6% to 8% on Zepbound, against 2% on placebo
In the pooled trials behind the current Zepbound prescribing information, injection site reactions were reported by 6% of patients on 5 mg, 8% on 10 mg and 8% on 15 mg, against 2% on placebo. Most are local: redness, itching, bruising, pain or a small rash confined to the spot you injected. The label counts those separately from hypersensitivity reactions, which are the body-wide kind. Which of the two you are looking at is what decides whether this is a rotation problem or a phone call.
Every page currently ranking for this question quotes either one number or a range, and those ranges disagree with each other and with the label. So the figures below are read straight from the ZEPBOUND prescribing information on DailyMed, with the group sizes attached, alongside the trial that reported a much higher rate and the reason it did.
What the Zepbound label reports, dose by dose
| Reaction, as the label names it | Placebo (N=958) | 5 mg (N=630) | 10 mg (N=948) | 15 mg (N=941) |
|---|---|---|---|---|
| Injection site reactions | 2% | 6% | 8% | 8% |
| Hypersensitivity reactions | 3% | 5% | 5% | 5% |
| Nausea, for scale | 8% | 25% | 29% | 28% |
Two things in that table are easy to miss. First, the gap between 5 mg and 10 mg is two percentage points, and between 10 mg and 15 mg there is no gap at all. Injection site reactions do not climb steadily with dose the way the gastrointestinal reactions do. Second, hypersensitivity reactions are a separate row with a separate pattern, flat at 5% across all three doses.
It also matters what the label folds into that 8%. The footnote to the injection site reactions row says the figure "includes multiple related adverse event terms, such as injection site bruising, injection site erythema, injection site pruritus, injection site pain, injection site rash, injection site reaction." Erythema is redness and pruritus is itching. So 8% is not 8% of people developing a rash. It is 8% reporting any of those, a bruise included.
The same reaction, counted three different ways
If you have read three articles on this and seen three different numbers, this is why. The rate depends entirely on which population is being counted.
- 6% to 8% in the pooled Study 1 and Study 2 population, which is the figure the label carries.
- 11.1% against 1.0% on placebo in SURMOUNT-3, the trial with a twelve week intensive lifestyle lead-in, where injection site reaction was one of the events reported in at least 5% of either group. Same drug, different trial, a rate roughly a third higher than the label pool.
- 11.3% against 1% when the label splits the treated group by antibody status. Patients who developed anti-tirzepatide antibodies had injection site reactions at 11.3%, compared with 1% among those who did not.
That last split is the most informative one on this page. It says the single largest difference between people who get reactions and people who do not is something that happens inside the immune system, not something happening at the needle tip. It is also not something you can feel, check or change.
Telling a local reaction from the two things it is not
This is not a diagnosis and it does not replace one. It is simply how the prescribing information draws the lines, which is useful because the lines decide who needs to be told and how quickly.
| What you are seeing | Where the label puts it | What it points to |
|---|---|---|
| Redness, itching, a bruise, soreness or a small rash, staying inside the area you injected | Injection site reaction, section 6.1 | The common, local, expected kind. Worth logging and worth mentioning at your next appointment. |
| Itching, rash or hives appearing away from the injection site, or on parts of the body you did not inject | Hypersensitivity reaction, sections 5.6 and 6.1. The label notes the majority of these in trials were skin reactions such as rash and itching | Not a local reaction any more. Contact your prescriber rather than waiting it out. |
| Swelling of the face, lips, tongue or throat, trouble breathing, a rapid heartbeat, feeling faint | Severe hypersensitivity, section 5.6, which instructs discontinuation | Emergency care now, not a phone call. |
| Redness and warmth spreading outward, pus, or fever, getting worse rather than better after two or three days | Not one of the label's reaction categories | This is how a skin infection presents. Same day medical care. |
| A firm lump, or a dent or thickening in the skin, at a spot you keep reusing | Lipodystrophy and localized cutaneous amyloidosis, named in the Instructions for Use as the reason to rotate | A slow build-up rather than a reaction. Tell your prescriber, because it can affect how the dose is absorbed. |
Why the label tells you to rotate, and what that is actually protecting
Section 2 of the prescribing information gives the instruction in five words: "Rotate injection sites with each dose." The Instructions for Use is more specific about why. It tells patients to "change (rotate) your injection site within the area you choose for each dose to reduce your risk of getting lipodystrophy (pits in skin or thickened skin) and localized cutaneous amyloidosis (skin with lumps) at the injection sites."
Those two named conditions are not what most people mean when they say injection site reaction, and none of the pages currently ranking for this query mention either of them. A reaction shows up within hours and fades. Lipodystrophy and cutaneous amyloidosis are what repeated use of the same small patch builds up over months, and because they change the tissue, they can change how much drug gets absorbed from that spot. That is the real argument for rotating, and it is a slower and quieter problem than an itchy patch.
The label lists the abdomen and the thigh as sites you can use yourself, with the back of the upper arm available if someone else gives the injection. It does not rank those sites against each other for reaction risk. It also adds one instruction that is easy to overlook if you take more than one injected medicine: "Do not inject ZEPBOUND in the same injection site used for other medicines."
The four week rotation test
If reactions keep happening, the question worth answering first is whether site reuse is driving them, because that is the one input you fully control. Four doses is enough to tell.
- For each of your next four weekly doses, write down the exact spot, not just "stomach." Left upper abdomen, right outer thigh, and so on. Vague records cannot answer this question.
- Beside each one, note whether a reaction appeared, on which day it appeared, and how long it stayed.
- Note your dose next to each entry too, and mark any week where the dose stepped up.
- After the fourth dose, look at the pattern. Do the reactions cluster on areas you returned to, or are they spread evenly across every site you used?
- Clustered on reused areas: widening the rotation is within your control and is the first thing to try.
Even across all sites, or tracking the dose increases: technique is probably not your variable. That is a question for your prescriber, with the four weeks of notes in hand.
The reason this test is usually not run is that it needs a record that survives four weeks, and memory does not. Keeping the site, the date, the dose and the symptom in one place is what Jabby is built to do: the body map shows which areas you have already used so you are not guessing, and side effects are logged against the dose they followed rather than floating loose. That turns the four weeks above into something you can actually read at the end of it, and hand to a prescriber.
Keep the record the test needs
Log the site, the dose and what followed, in one place, on your phone. Jabby maps which areas you have already used and keeps side effects attached to the dose they came after.
Download Jabby, free on the App StoreDoes the pen or the vial make a difference?
Not as far as anyone has published. The 6%, 8% and 8% figures come from trials that used the single dose pen. Asked directly whether reactions differ between the vial and the single dose pen, Eli Lilly's medical information service states plainly that there are no data available on injection site reactions with tirzepatide administered from a single dose vial. So any claim that one presentation reacts more than the other is not resting on published evidence.
There is one factual difference between the presentations that is worth knowing, though the label draws no conclusion from it. Single dose pens and single dose vials contain only sodium chloride, sodium phosphate dibasic heptahydrate and water alongside the tirzepatide. The multi-dose vial and the single-patient-use KwikPen, which hold four doses each, additionally contain benzyl alcohol, glycerin and phenol as preservatives. That difference is in section 11 of the label. No trial has compared reaction rates between them, so it is a fact to mention to a prescriber, not a conclusion to act on.
How long does one last?
The label does not say, and it is worth being straight about that. Trials of this kind record whether an event happened, not how many days it ran, so no duration figure exists in the prescribing information. The "one to three days" you will read on most pages covering this topic is clinical experience rather than a trial result, and it is quoted without a source on every page we checked.
What the label does tell you about timing is specific to a different set of reactions. It states that "the majority of nausea, vomiting, and/or diarrhea events occurred during dose escalation and decreased over time." It makes no equivalent statement about injection site reactions, so do not assume they follow the same fading pattern.
What is in your control, and what is not
Worth separating, because effort spent on the second list is wasted.
- In your control: which site you use, whether you return to the same patch, keeping Zepbound away from sites used for other injected medicines, and inspecting the solution before use. The label says it should be clear and colorless to slightly yellow, and not to use it if you see particles or discoloration.
- Not in your control: whether you develop anti-tirzepatide antibodies, which is the factor with by far the largest measured effect, and the dose your prescriber has you on.
- Commonly advised but not in the label: letting the pen come to room temperature before injecting. The prescribing information sets storage limits, including up to 21 days at room temperature for a single dose pen or vial and 30 days for a multi-dose vial or KwikPen, but it gives no instruction to warm a pen before use. It is a widely repeated tip, not a labelled step.
When to stop reading and call
- Hives, rash or itching spreading beyond the injection area.
- Swelling of the face, lips, tongue or throat, any difficulty breathing, a racing heart or faintness. That is emergency care, immediately.
- Redness spreading outward with warmth, pus or fever, or a site that is worse on day three than it was on day one.
- A lump, dent or thickened patch that does not settle, especially at a site you reuse.
- Reactions severe enough that you are considering skipping a dose. That decision belongs with your prescriber, not with an article.
Frequently asked questions
What does a normal Zepbound injection site reaction look like?
The label's own list is the best description available: bruising, redness, itching, pain or a rash, confined to the spot injected. It stays local. Anything crossing that boundary moves into a different category.
Why is my Zepbound injection site itchy?
Itching, which the label records as pruritus, is one of the terms counted inside the 6% to 8% figure. On its own and staying local, it is the common form. Itching that spreads or turns up elsewhere on the body is counted by the label as a hypersensitivity reaction instead, and that is the version to report.
What is the lump at my injection site?
There are two different lumps and they behave differently. A firm swelling appearing within hours and settling over a few days falls within the reaction category. A persistent lump at a spot you keep reusing is what the Instructions for Use is warning about when it names localized cutaneous amyloidosis and lipodystrophy. Persistence is the distinguishing feature, and the second kind is worth showing to a prescriber.
Does a reaction mean I am allergic to Zepbound?
Not by itself. The label treats them as separate categories, and the numbers are separate too. Immediate hypersensitivity reactions, meaning within one day of a dose, occurred in 2.1% of Zepbound treated patients against 0.4% on placebo, which is a much smaller group than the 6% to 8% with local reactions. The distinguishing question is whether it stayed at the site.
Is the abdomen or the thigh less likely to react?
The label does not answer that. It lists the abdomen, the thigh and the back of the upper arm as the options, and instructs rotation, but does not rank them for reaction risk. Anyone stating one is safer than another is going beyond the published data.
Can I put ice or a cream on the site?
That belongs with your pharmacist or prescriber, who knows what else you take. This page deliberately does not give doses or product recommendations, and you should be wary of any page that does without knowing your history.
Should I stop Zepbound because of a reaction?
That is a prescriber's decision, and the label reserves the instruction to discontinue for severe hypersensitivity reactions specifically. Bring the four weeks of notes described above to that conversation rather than making the call alone.
Related reading
- Zepbound KwikPen needles: which pen needles fit, how many each pen uses, and the 30 day discard rule.
- Zepbound reviews: patient ratings, physician ratings and trial results set side by side.
- Does Zepbound cause headaches?: the same placebo-controlled reading applied to a different side effect question.
- Tracking tirzepatide injections: keeping site, dose and symptom together across a dose escalation.
- GLP-1 tracker app overview: how the body map and the dose history fit together.
- Semaglutide side effects: the same questions for Ozempic and Wegovy.
If you are on Zepbound and want the logging side handled rather than improvised, Jabby is built for exactly this: the site map, the dose history and the symptom log in one place, on your phone, so the pattern is readable when you need it. It is free on the App Store.
Sources
- Eli Lilly and Company. ZEPBOUND (tirzepatide) injection, prescribing information. DailyMed, SPL version 40, published 2 September 2026. Read 29 September 2026. Table 1 adverse reactions, section 5.6 hypersensitivity reactions, section 6.1 injection site reactions and antibody subgroup, section 11 description and excipients, section 14.1 trial identifiers, and the Instructions for Use.
- Eli Lilly and Company Medical Information. Are there any differences in injection site reactions with Zepbound (tirzepatide) administered from a vial versus the single-dose pen? Read 29 September 2026.
- Wadden TA, Chao AM, Machineni S, et al. Tirzepatide after intensive lifestyle intervention in adults with overweight or obesity: the SURMOUNT-3 phase 3 trial. Nat Med. 2023 Nov;29(11):2909-2918. PMID 37840095. DOI 10.1038/s41591-023-02597-w. ClinicalTrials.gov NCT04657016. Table 3, injection site reaction 32 of 287 (11.1%) on tirzepatide against 3 of 292 (1.0%) on placebo.
This page is information, not medical advice. It does not recommend a medicine, a dose, a treatment for a reaction, or a change to any of them, and it is not a substitute for examination by a clinician. Trial rates describe groups and cannot tell you what will happen to you. Decisions about starting, continuing, changing or stopping treatment belong to you and your prescriber, working from the approved prescribing information.