Sleep on a GLP-1

The rare case where the claim is on the label

Most of what circulates about these drugs and sleep is inference. This part is not.

The current Zepbound label carries two indications. One is the familiar one, weight. The other reads: "to treat moderate to severe obstructive sleep apnea (OSA) in adults with obesity." For that indication the recommended maintenance dosage is 10 mg or 15 mg, injected subcutaneously once weekly¹.

That is a regulator agreeing that a weight-loss drug treats a sleep disorder, which is a genuinely unusual sentence, and it is worth being precise about how narrow it is. It is tirzepatide only. The semaglutide label — Wegovy, which now comes as an injection and as tablets — lists cardiovascular risk reduction, weight reduction, and metabolic dysfunction-associated steatohepatitis. There is no sleep apnoea indication on it².

So if you are on semaglutide and sleeping better, that is plausible and it is not a labelled effect. If you are on tirzepatide, there is a trial behind it, and it is worth reading properly.

What the trials found, placebo arm included

The indication rests on a master protocol containing two 52-week randomised, double-blind, placebo-controlled trials — Study 5 and Study 6 in the label, published together as SURMOUNT-OSA³. Between them they enrolled 469 adults with moderate-to-severe apnoea and obesity. Study 5 enrolled 234 people who were unable or unwilling to use positive airway pressure. Study 6 enrolled 235 people who were already on it.

These were not people with borderline numbers. Mean apnoea-hypopnoea index at baseline was 51.5 events per hour in Study 5 and 49.5 in Study 6 — an interrupted breath roughly every seventy seconds, all night. Around two-thirds of each cohort had severe apnoea¹.

Read the placebo column before the drug column. It moves — five events an hour, and one placebo patient in five halved their AHI — which is the reason uncontrolled before-and-after stories about this are worthless.

Then read the gap. Roughly four in ten in Study 5, and half in Study 6, reached remission or mild non-symptomatic disease, against 15.9% and 14.3% of the placebo groups respectively. Those are big numbers for a condition whose only reliable treatment until now was a machine you have to wear.

The label also notes that the reduction in AHI held irrespective of age, sex, ethnicity, baseline BMI and baseline severity¹.

Two-thirds of them were men

Which matters here more than it usually does, because sleep apnoea is the condition medicine most reliably fails to see in women.

Study 5 was 67% male. Study 6 was 72% male¹. That is not a flaw in the trials — it reflects who gets diagnosed — but it means the evidence base is built on the presentation that gets caught.

And the presentation differs. In a retrospective series of 514 patients diagnosed at one sleep laboratory in India, the 147 women were significantly older at diagnosis than the 367 men (56.0 versus 50.2 years) and significantly heavier (mean BMI 35.3 versus 29.6). Morning headache, nocturia, fatigability and depression were all more common in the women; hypersomnia — the classic falling-asleep-in-the-afternoon symptom — was more common in the men⁴.

A structured review of sex inequities in sleep-disordered breathing puts the consequence plainly: women often present with lower AHI values while reporting a substantial symptom burden, and conventional screening instruments and AHI-centred diagnostic criteria show reduced sensitivity in women, contributing to delayed referral and underdiagnosis⁵.

So a woman in her forties who is exhausted, waking with a headache, up twice in the night and low is describing textbook apnoea and will frequently be assessed for something else. The risk also rises around the menopause: the sex gap in prevalence narrows after it, which is one of the clearest signals that hormonal status is doing real work in this condition⁶.

If you are on a GLP-1 partly because perimenopause changed your appetite, you are in exactly the demographic this evidence keeps missing.

The tiredness that is not apnoea

Not every sleep complaint on these drugs is a breathing complaint, and one of them belongs to the medication itself.

Fatigue is a listed adverse reaction in the adult weight-reduction trials — reported by 3% on placebo (n=958) against 5%, 6% and 7% on tirzepatide 5 mg, 10 mg and 15 mg respectively¹. The label groups asthenia, lethargy and malaise into that figure. It is a real signal, it is modest, and it is dose-related.

On the other side of the ledger, in Studies 5 and 6 patients on tirzepatide showed improvement in sleep-related impairment compared with placebo, measured on the PROMIS Sleep-Related Impairment 8a short form¹ — a patient-reported instrument about how sleep affected the day, not a machine reading.

Both things can be true at once: daytime function improving while a low-grade tiredness in the first weeks belongs to the drug. Which is worth knowing, because the wrong attribution sends people either to the wrong doctor or to no doctor at all.

★ The line that matters most: do not stop your machine

If you use CPAP, do not stop because the scale has moved. Not for a week, not to test it.

The label is unusually direct about the limits of what was studied: "The clinical studies for OSA did not evaluate the timing or appropriateness of PAP discontinuation in patients who were previously compliant with PAP therapy."¹ In Study 6 patients stayed on their therapy and were instructed to suspend it for seven days before the primary endpoint was assessed. Nobody has tested stopping.

Coming off is a decision that needs a repeat sleep study and the clinician who prescribed the machine, in that order.

The other half of that is the women who never got as far as a machine. In a qualitative study of 27 women who had started CPAP, mean age 51, the support the women in the study described receiving — from equipment suppliers, clinicians, friends, family and social media alike — was inconsistent, and what separated those who succeeded from those who did not was persistence and self-advocacy⁷. The authors open by noting that women are less adherent to CPAP than men and that the reasons remain unclear.

Self-advocacy is a heavy thing to require of someone whose presenting complaint is that she is too tired to think. But it is what the evidence describes, and it is the reason to be specific in the appointment.

Ask for a sleep study if you snore, wake unrefreshed, wake with headaches, are up in the night to urinate, or have been told you stop breathing — particularly if you are perimenopausal. Bring the symptoms, not the conclusion. And get urgent help rather than an appointment if you are falling asleep while driving.

Where this leaves it

The unusual thing about this corner of the GLP-1 story is that the evidence is stronger than the conversation around it.

There is a regulator-approved indication, two randomised trials with placebo arms that moved, and remission rates that would be headline news for a condition with fewer jokes attached. And there is a diagnostic pipeline that was built around men, which means a large number of the women this would help have not been offered the test that would find out.

That is not a drug problem. It is a referral problem, and it is older than any of these medications. The medication has just made it more expensive to keep ignoring.