Joint and muscle pain are not in the GLP-1 labels. The nine current US prescribing labels read for this page on 22 October 2026 contain neither "arthralgia" (joint pain) nor "myalgia" (muscle pain). Two labels record something nearby — back pain on Victoza, and pain in the arms or legs in Saxenda's teenage trial — and the one GLP-1 trial that set out to measure joint pain found it fell further on the drug than on placebo. This page sets out that record, then what adverse-event reports and the case literature add. It covers aches as a possible side effect; the weight-loss arthritis trial itself is summarised here and the muscle-mass question is on GLP-1 drugs and muscle loss.
The label record
| Label (version read) | Joint pain (arthralgia) | Muscle pain (myalgia) | Related line, if any |
|---|---|---|---|
| Zepbound (v40, eff. 2026-08-28) | Not listed | Not listed | — |
| Mounjaro (v40, eff. 2026-08-27) | Not listed | Not listed | — |
| Wegovy injection and tablet (v19, eff. 2026-06-18) | Not listed | Not listed | — |
| Ozempic (v20, eff. 2026-06-01) | Not listed | Not listed | — |
| Semaglutide tablets (v14, eff. 2026-01-30) | Not listed | Not listed | — |
| Victoza (v31, eff. 2025-10-14) | Not listed | Not listed | Back pain: 4% (1.2 mg), 5% (1.8 mg) vs 3% placebo |
| Saxenda (v22, eff. 2026-02-25) | Not listed | Not listed | Teenagers: pain in extremity 4% vs 2.4%; blood creatine kinase raised 3.2% vs 2.4% |
| Trulicity (v62, eff. 2026-06-16) | Not listed | Not listed | — |
| Foundayo (v10, eff. 2026-07-29) | Not listed | Not listed | — |
Each label's full text searched for arthralgia, myalgia, back pain, muscle spasm and pain in extremity, DailyMed, 22 October 2026.
Victoza's back-pain line comes from Table 1, which lists reactions reported by at least 5% of adults on Victoza and more often than on placebo, pooled across 645 adults on 1.2 mg, 1,024 on 1.8 mg and 661 on placebo. Back pain sits at the bottom of that table, two percentage points above placebo at the higher dose. Saxenda, the same molecule at up to 3 mg for weight management, does not list back pain in its adult table.
Saxenda's paediatric lines come from a 56-week trial of 251 teenagers. At those sizes, 4% against 2.4% is about five children against three (our arithmetic from the label percentages) — enough to meet the table's 3% threshold, too few to say much more. Creatine kinase is an enzyme released from muscle; the label does not report symptoms alongside the raised readings.
What an absence means. A label lists reactions that crossed a threshold in its trials — commonly 2% or 5%, and more often than placebo. Joint pain is common in the populations these drugs are tested in, so it happens in both arms of a trial; it did not separate from placebo clearly enough to be tabled on eight of nine labels. That is evidence that any excess is small in trials, not proof that no one's joints hurt because of the drug.
Adverse-event reports
FDA's adverse event database, through openFDA (data updated 30 July 2026, queried 22 October 2026), holds these counts among reports naming each molecule:
| MedDRA term | Tirzepatide (160,663 reports) | Semaglutide (73,001) | Liraglutide (50,705) | Dulaglutide (88,160) |
|---|---|---|---|---|
| Arthralgia | 1,145 | 1,224 | 618 | 669 |
| Back pain | 827 | 944 | 690 | 640 |
| Myalgia | 812 | 484 | 299 | 274 |
| Pain in extremity | 691 | 817 | 634 | 677 |
openFDA drug/event endpoint, patient.reaction.reactionmeddrapt.exact, 22 October 2026.
These are reports, not rates. A report can list several drugs and several symptoms, reporting follows publicity and use, and the number of people who took each drug without reporting anything is unknown. The table shows that joint and muscle pain are being reported to FDA in people taking every GLP-1 molecule; it cannot show whether the drugs cause them or how often.
The case literature
A PubMed search on 22 October 2026 for a named GLP-1 drug or the class together with "arthralgia" or "joint pain" in the title or abstract returned 13 records, most of them about other subjects (arthritis treatment reviews, a supplement case, a different diabetes drug class). The one case report describing joint pain as an apparent reaction is from 2014: a 42-year-old man with type 2 diabetes developed bilateral, symmetrical joint pain with raised inflammatory markers six months into liraglutide 1.8 mg; tests for rheumatoid and other causes were negative, and the symptoms cleared within a week of stopping the drug (Acta Diabetologica 2014). The authors noted the resemblance to polyarthritis described with DPP-4 inhibitors, another class of diabetes medicine. One case cannot establish frequency.
The trial that measured joint pain
STEP 9 is the only GLP-1 trial built around joint pain as an outcome, and it measured improvement, not harm. It randomised 407 adults with obesity (mean BMI 40.3) and moderate knee osteoarthritis with at least moderate pain, 2:1, to semaglutide 2.4 mg or placebo for 68 weeks, both with diet and activity counselling.
| Week 68 | Semaglutide 2.4 mg | Placebo |
|---|---|---|
| Body weight change | −13.7% | −3.2% |
| WOMAC knee-pain score change (0–100) | −41.7 | −27.5 |
| SF-36 physical function change | +12.0 | +6.5 |
| Stopped treatment for adverse events | 6.7% | 3.0% |
Bliddal et al., N Engl J Med 2024, NCT05064735. Funded by Novo Nordisk.
Two things limit what STEP 9 says about aches on GLP-1 drugs in general. Its participants already had painful knee arthritis, so it measured pain falling from a high baseline (mean 70.9), not new pain appearing. And the placebo group improved substantially too — 27.5 points — so a large share of the improvement came with taking part in the trial. Most people who stopped semaglutide did so for gastrointestinal reasons. A 2026 survey of 767 semaglutide users at Danish pharmacies adds a real-world signal in the same direction: among the 24% who reported cutting back other medicines, muscle and joint pain medicines were the most common category (30%). That is self-report from a survey, not a measured effect.
What this page cannot tell you
- How often joint or muscle pain happens on Zepbound, Wegovy or Ozempic. No label reports a rate, and reports cannot supply one.
- Whether lean-mass loss produces aches. No label, trial or study counted here links the two.
- Whether a particular person's pain is from the drug. Inflammatory arthritis, gout, injury and many other causes need ruling out by a clinician.
New joint swelling, pain in several joints at once, or pain with fever or a rash is worth a prompt clinical assessment whatever medicines someone is on. The class's other musculoskeletal record — bone density and fractures — is on GLP-1 drugs and bone loss, and the skin sensations some people describe as aching or burning are on skin sensitivity on GLP-1 drugs.
Sources
- DailyMed prescribing information read 22 October 2026: Zepbound (set ID 487cd7e7-434c-4925-99fa-aa80b1cc776b, v40), Mounjaro (d2d7da5d-ad07-4228-955f-cf7e355c8cc0, v40), Wegovy (ee06186f-2aa3-4990-a760-757579d8f77b, v19), Ozempic (adec4fd2-6858-4c99-91d4-531f5f2a2d79, v20), semaglutide tablets (27f15fac-7d98-4114-a2ec-92494a91da98, v14), Saxenda (v22, Table 3), Victoza (v31, Table 1), Trulicity (463050bd-2b1c-40f5-b3c3-0a04bb433309, v62), Foundayo (8ac446c5-feba-474f-a103-23facb9b5c62, v10).
- openFDA drug adverse event API, data last updated 2026-07-30, queried 22 October 2026.
- Bliddal H et al. Once-weekly semaglutide in persons with obesity and knee osteoarthritis. N Engl J Med 2024;391:1573–1583. PMID 39476339; NCT05064735.
- Ambrosio ML et al. GLP-1 receptor agonist-induced polyarthritis: a case report. Acta Diabetol 2014;51(4):673–674. PMID 24158775.
- Kjærgaard K et al. Treatment patterns and experienced effects of semaglutide for weight management among adult users in Denmark. Diabetes Obes Metab 2026;28(10):8955–8965. PMID 42443134.
- PubMed search, 22 October 2026: named GLP-1 drug or class AND (arthralgia OR "joint pain") [tiab] — 13 records.