GLP-1 and Fatigue: the Label Says 11%, the Placebo Arm Says 5%, and One Phase 3 Trial Reports Less Fatigue on the Drug

Ask "does Ozempic make you tired?" and the answers divide into two kinds. One quotes the prescribing information: about 11% of adults on Wegovy 2.4 mg reported fatigue in the weight-reduction trials. The other says fatigue is not really the drug, it is eating far less. Both are guesses at the same question, and the label answers it directly in the column printed beside the 11%: 5% of the people on placebo in those same trials reported fatigue too.

This is the thirteenth piece in our evidence series, after which effects have enough evidence to calibrate a model, the corpus-derived pooled rates, what patients themselves report, gallstones end to end, what is inside a personalised risk number, hair loss end to end, why trial vomiting rates run from 6.5% to 58%, constipation read against the placebo arm, headache read against the placebo arm, abdominal pain, where the label and the registry disagree by a factor of two, nausea, where the label's 44% holds up and the largest trial reads 18% and diarrhoea, where the three trials behind the label's number add up exactly. Like those, every figure below is quoted from a primary source with the arm and the number of participants it belongs to, and every one of them was read on 18 September 2026 — from the current US prescribing information, from each trial's own results table on ClinicalTrials.gov, or from our own live production API.

What the label says, and what its footnote says

The current US prescribing information for Wegovy (semaglutide), revised June 2026, lists fatigue among the *"most common adverse reactions (incidence ≥5%) in adults or pediatric patients aged 12 years and older"*. Its adverse-reaction tables each give a drug column and a placebo column:

Table 3 is the source of the 11%, and its footnote is the part that matters here: the row *"includes fatigue and asthenia"* — two separate adverse-event terms, counted as one. A person who reported either appears once in the 11%. Table 4 carries the identical footnote. Table 5, the adolescent table, lists every reaction reported in at least 3% of drug-treated patients *and* more often than placebo — and fatigue is not in it, so in that 201-person trial fatigue either did not reach 3% on the drug or was not more common than on placebo. (Quoted from the prescribing information via DailyMed, set ID ee06186f, SPL version 19, effective 18 June 2026, accessed 18 September 2026.) The label's oral form gets no separate table: it says only that in the 204-patient WEGOVY 25 mg tablet trial *"the types and frequency of common adverse reactions were similar to those listed in Table 3"*.

For comparison, the same Table 3 puts nausea at 44% against 16% on placebo, diarrhoea at 30% against 16%, vomiting at 24% against 6%, constipation at 24% against 11%, abdominal pain at 20% against 10%, headache at 14% against 10% and dizziness at 8% against 4%. Ranked by how large the placebo figure is as a share of the drug figure, that table reads: headache 71% (10 of 14), diarrhoea 53% (16 of 30), abdominal pain and dizziness 50%, constipation 46% (11 of 24), fatigue 45% (5 of 11), nausea 36% (16 of 44) and vomiting 25% (6 of 24). Fatigue sits in the middle of that range — much closer to its own placebo arm than nausea or vomiting, not as close as headache.

So the label's own answer to "is the tiredness the drug?" is: about half of the fatigue reported on the drug was also reported on placebo, and the drug adds roughly six points on top.

What the trial registries say, arm by arm

Our database cited 12 distinct studies for a stated fatigue rate when this article was written at 13:10 BST on 18 September 2026. Seven of those 12 post a placebo arm of at least 20 people; every row below was read from that trial's own ClinicalTrials.gov adverse-events module on 18 September 2026. The counts are the registry's; the percentages and the differences are our division and subtraction of them.

The other five sources in our fatigue base are absent from that table for two different reasons. One is placebo-controlled but its placebo arm is too small to read a difference from: a 23-person study of semaglutide and the immune system in Alzheimer's disease (NCT05891496), reporting 0 fatigue of 12 on placebo against 2 of 11 and 1 of 22 across two drug periods. Four post no placebo arm at all: the long-term safety study ACHIEVE-J (NCT06010004, 0.8%, 0.7% and 2.2% across its three orforglipron doses); SURPASS-SWITCH (NCT05564039), where tirzepatide 15 mg or maximum tolerated dose reads 8/139 = 5.8% against 4/143 = 2.8% on dulaglutide, an active comparator rather than a placebo; a drug-interaction study of orforglipron with carbamazepine (NCT06370728, 0/30 on orforglipron alone and 0/28 on the combination — its carbamazepine-only period, 1/29, was withheld from our store on 16 September because carbamazepine is not a GLP-1); and a Phase 1 bioequivalence study of orforglipron tablets against capsules (NCT06440980), which is the subject of the next section.

The difference, not the headline

Take the difference column seriously and three things stand out.

In the adult weight-management trials of injectable semaglutide 2.4 mg, the gap is four to six points: Table 3 (11 − 5 = 6), STEP 1 (8.0 − 4.3 = 3.7), STEP 2's 2.4 mg arm (6.9 − 1.0 = 6.0) and STEP 3 (12.8 − 7.4 = 5.4). Those four readings are not four independent trials — Table 3 is the label's pool of three weight-reduction trials, and STEP 1, STEP 2 and STEP 3 are trials inside that programme with their own registry records — so the point is that the difference survives being computed at both the pooled and the single-trial level, not that four separate studies agree. Note also how much the raw drug figure moves while the difference stays put: 6.9% in STEP 2, 8.0% in STEP 1, 12.8% in STEP 3, same molecule at the same maintenance dose. The placebo arms move with them, 1.0%, 4.3% and 7.4%. The drug figure on its own tells you more about the trial and its population than about the drug.

In the orforglipron Phase 3 trials the gap is close to zero, and in one of them it is negative. ATTAIN-2's three doses sit 0.2 to 3.2 points above their own placebo. In ACHIEVE-1 all three doses sit *below* placebo — 2.1%, 0.7% and 2.8% against 3.6% — which on 138 to 143 people per arm means a handful of reports either way, but it is the registry's own number and we are not going to leave it out because it is inconvenient. ACHIEVE-J posts no placebo arm and its three doses read 0.8% to 2.2%, in the same range.

One row sits far above the rest, and it is the one most often quoted back as "GLP-1 fatigue". The 120-person STABLE Wt Loss study reports 31.9% against 16.7% — a 15-point gap, the largest in the table, in a Phase 4 trial explicitly studying appetite and eating behaviour, with 72 people on the drug and 48 on placebo. It is also the trial whose placebo arm reports the most fatigue of any in our base, at 16.7%. Both halves of that row are worth carrying: the drug arm is the highest fatigue figure anyone can cite from a registry record in our corpus, and its own control arm, at 16.7%, reports more fatigue than the highest *drug* arm of every one of the other six placebo-controlled studies in the table — the next highest is STEP 3 at 12.8%.

So the honest one-line answer to "how much fatigue does the drug itself add?" is: in every placebo-controlled trial arm of more than 300 people that posts a fatigue rate, the drug arm sits 0.2 to 6.0 points above its own placebo arm; in smaller arms it runs from 2.9 points below placebo to 15.3 above. The label's six-point gap sits at the top of that first range.

What our own number counts, and the one thing a reader should know before quoting it

Our fatigue endpoint publishes a pooled clinical estimate of 6.9% (95% CI 2–25%), from 65 eligible stated rates across 12 distinct studies, source diversity "high", as read at 13:10 BST on 18 September 2026. It is lower than the label's 11% for two reasons, and the second one is more interesting than the first.

The first is the one this series has reported for every effect: the pool is the whole class, not one drug at one dose in one population. Our 12 studies span semaglutide, tirzepatide, dulaglutide, retatrutide and orforglipron, at Phase 1 through Phase 4, in populations from otherwise-healthy volunteers to people with type 2 diabetes.

The second is specific to fatigue, and it is the kind of thing a hostile reader should not have to find for us. 38 of those 65 stated rates — 58% — come from a single study: NCT06440980, a Phase 1 study whose official title is *"A Multiple-Dose Study to Investigate the Bioequivalence of Orforglipron (LY3502970) Capsules and Orforglipron Tablets in Participants With Obesity or Overweight Who Are Otherwise Healthy"*. It enrolled 533 people and posts fatigue for 38 separate dose-level and formulation arms, from 0.0% to 5.8%. Those 38 arms have 3,772 participants between them, 7.1 times the trial's own enrolment, because the same people appear in arm after arm as the study moves them through cohorts and dose levels.

Read as "65 independent observations", that base would be badly misleading. It is not how we pool. Our pooled figure collapses each study to one entry — the unweighted mean of its posted arms, carried at the size of its largest arm — and then weights those entries by their stated sample size, multiplied by a discount for how confident our extraction was in reading each figure. So NCT06440980 contributes 1 of 12 entries, at 1.0% and n=215, not 38 observations; the entries with the most weight are STEP 1 (8.0%, n=1,306), STEP 2 (5.8%, n=403), STEP 3 (12.8%, n=407) and ATTAIN-2 (4.8%, n=331). The collapsing is why our 6.9% is not dragged to 1% by one Phase 1 study. But the 65 is a count of rows, and rows are not people: the honest reading of our fatigue base is twelve studies, four of which (NCT06440980 at 7.1 times enrolment, the carbamazepine interaction study at 2.9, the 23-person Alzheimer's study at 2.0, and STABLE Wt Loss at 1.6) post overlapping periods or cohorts of the same participants as separate arms. Those four supply 44 of the 65 stated rates between them, so two thirds of the row count in our fatigue base comes from studies that count some participants more than once. The other eight studies do not: each one's posted fatigue arms sum to its own enrolment or to within five of it (STEP 1 1,961 of 1,961, STEP 3 611 of 611, ACHIEVE-1 559 of 559, ACHIEVE-J 401 of 401, SURPASS-SWITCH 282 of 282, STEP 2 1,207 of 1,210, ATTAIN-2 1,608 of 1,613, retatrutide 337 of 338). Whether overlapping-period arms belong in a pooled incidence estimate at all is a methodology question we have open at the time of writing; for now they are in, collapsed to one entry per study, and the live endpoint names every source with its URL so you can take any of them out yourself.

There is a third reason our number is not the label's, and it runs the other way. The label's row groups fatigue and asthenia; we count only the term "fatigue". Registries post them separately, and two of our 12 studies post both: ACHIEVE-1 lists asthenia at 0.7% on placebo against 2.1%, 0.7% and 0.7% across its three doses, and ACHIEVE-J lists it at 0.0%, 0.0% and 0.7% (plus three "muscle fatigue" rows). Our base contains none of those asthenia rows. Grouping the two terms the way the label does would raise our figure, not lower it — though by less than simple addition suggests, because the label counts a person who reported either term once, and we cannot tell from a registry table how many people reported both. So our 6.9% and the label's 11% are not two measurements of one quantity that disagree: one is a class-wide pooled estimate of a single MedDRA term, the other is one drug's whole-trial incidence of two terms grouped. If you want the Wegovy 2.4 mg number, quote the label: 11% against 5% on placebo, N=2,116 and N=1,261, Table 3. If you want the whole class with its heterogeneity visible, quote ours, with its interval and its source count. Neither is the other.

What the FDA's adverse-event reports show

A different kind of number sits outside the incidence pool on purpose. FDA FAERS publishes the share of *spontaneous adverse-event reports* for each drug that mention fatigue: 6.1% of orforglipron reports (70 reports), 4.3% of semaglutide reports (3,563), 3.8% of liraglutide reports (1,656), 2.4% of dulaglutide reports (2,409), 2.3% of tirzepatide reports (3,179) and 1.9% of exenatide reports (1,759), as read from our own endpoint on 18 September 2026. A share of complaints is not a share of patients — it answers "of the people who filed a report about this drug, how many mentioned fatigue," not "how many people on the drug got it" — so our eligibility rules report these separately rather than average them in. The orforglipron figure rests on 70 reports of a drug that is not yet widely marketed, and should be read as a count rather than a rate.

What patients themselves report

Of the 26 distinct community reports in our corpus, 3 mention fatigue — an 11.5% reporting frequency. That is not an incidence rate and we never treat it as one: it is the share of people who chose to post about their GLP-1 experience who mentioned fatigue at all, and at three reports it is closer to a count than a rate. See that article for why the pool is small and cannot currently grow.

If you are on a GLP-1 now

This is educational content, not medical advice. The label lists fatigue among the most common adverse reactions, at about double the placebo rate, and the trial evidence above says the drug adds a few points on top of a background rate that is itself substantial. What the label does *not* say anywhere is that the fatigue is caused by eating less — that is a plausible mechanism, widely repeated, and no source in our corpus states a rate that tests it.

Two things in the patient information are worth knowing rather than paraphrasing. The label lists tiredness among the *"signs and symptoms of low blood sugar"*, a risk it says applies especially to people with type 2 diabetes who also take insulin or a sulfonylurea; and it separately warns that *"diarrhea, nausea, and vomiting may cause a loss of fluids (dehydration)"*, which can present as exhaustion. Fatigue that arrives suddenly, comes with confusion, sweating, shakiness or a fast heartbeat, or follows days of being unable to keep fluids down, is a reason to contact a clinician rather than to wait it out. Nothing here should be used to start, stop or change a medication or its dose without your prescriber.

Every figure in this article was read on 18 September 2026 from the cited primary source or from our live production API, and is stated as of that date. Query everything behind it — all 15 effects, both tracks, every source with its URL — at magistra.health/en/data-api, with the full eligibility methodology at magistra.health/en/methodology.