GLP-1s and sleep: insomnia, vivid dreams and the sleep apnea approval

Reviewed 20 sourcesPublished by PepVault

Insomnia is on Saxenda's label only, 2.4% vs 1.7% on placebo. What the labels, trials and reports show for Ozempic, Wegovy, Mounjaro and Zepbound, dreams too.

Brown-haired man sleeps on his side in white linen in morning sun, a fern on the nightstand
Last reviewed
Sources
20
Questions
7
Compounds
3

The short answer

Insomnia is not listed for Ozempic, Wegovy, Mounjaro or Zepbound; only Saxenda's label lists it, at 2.4% against 1.7% on placebo. Across 86 trials and 133,378 people, sleep problems were no more common on a GLP-1 than on placebo or another drug. No label or trial gives a rate for vivid dreams. The routes the labels support for a bad night are nausea, reflux and low blood sugar.

  • Saxenda is the only GLP-1 label that lists insomnia: 2.4% vs 1.7% on placebo.
  • In pooled Wegovy trials, sleep complaints were 3.4% on the drug and 3.6% on placebo.
  • No label or trial has measured vivid dreams; in diabetes, nightmares can be a sign of low blood sugar at night.
  • Reports of sleep and mood complaints cluster in the first weeks, when nausea and reflux peak during the dose steps.

Every number checked at its primary source · · 20 sources · How we research

Evidence: Insomnia rests on drug labels, placebo-controlled trials and a meta-analysis of 86 trials; vivid dreams have no label or trial data, only user reports and the maker's statement.

Brown-haired man sleeps on his side in white linen in morning sun, a fern on the nightstand

Does Ozempic, Wegovy, Mounjaro or Zepbound cause insomnia?

Not according to their labels. The word insomnia appears nowhere in the current Ozempic, Wegovy, Mounjaro or Zepbound labels, patient leaflets included1234. A label's side-effect table lists each adverse reaction (an unwanted effect seen in the trials) that happened more often on the drug than on placebo (a dummy injection), above a cut-off: 2% for Wegovy, Zepbound and Saxenda, 5% for Ozempic and Mounjaro. The one label that lists insomnia is Saxenda's (liraglutide, an older daily GLP-1): 2.4% of 3,384 people against 1.7% of 1,941 on placebo, a gap of less than one person in a hundred5.

The labels do count the effects that can spoil a night: reflux and fatigue, and in diabetes, low blood sugar. The table puts the first two side by side, drug against placebo12345.

LabelInsomniaReflux (GERD)Fatigue
Ozempic (semaglutide, diabetes)Not listed1.5–1.9% vs 0%Listed above 0.4%, no placebo rate
Wegovy 2.4 mg (semaglutide)Not listed5% vs 3%11% vs 5%
Mounjaro (tirzepatide, diabetes)Not listed1.7–2.5% vs 0.4%Not in the placebo table
Zepbound (tirzepatide)Not listed4–5% vs 2%5–7% vs 3%
Saxenda (liraglutide)2.4% vs 1.7%4.7% vs 1.7%7.5% vs 4.6%

Ranges span the label's dose groups. Wegovy's and Zepbound's fatigue rows include weakness (asthenia). Mounjaro's 10% fatigue comes from a heart-outcomes trial with no placebo group, so it is left out3.

PepVault's daily check-in on Today takes a tap for nausea and energy, with chips for heartburn and fatigue, and a sleep rating when Recovery is on, so a rough night is logged on the day of the shot it followed. Log it in PepVault

A smiling woman in a blue top leaning on her kitchen counter with her phone and a notebook

PepVault for iPhone

Keep your protocol in PepVault

PepVault's daily check-in on Today takes a tap for nausea and energy, with chips for heartburn and fatigue, and a sleep rating when Recovery is on, so a rough night is logged on the day of the shot it followed.

Download on the App Store

7 days free on the yearly plan for new subscribers.

What did the trials count for sleep?

Trials record every complaint and file it under MedDRA, the dictionary regulators use to code side effects. In a pooled look at the Wegovy trials STEP 1, 2 and 3 (68 weeks, 3,377 adults), complaints in MedDRA's sleep group, the class that holds insomnia, came up in 3.4% of people on semaglutide 2.4 mg and 3.6% on placebo6. In the two-year STEP 5 trial it was 3.9% against 3.3%6. The authors call sleep disorders broadly similar between the groups6.

The widest view is a 2026 meta-analysis (a study that pools other studies) of 86 randomized trials with 133,378 people, GLP-1 drugs against placebo or another treatment. Sleep disorders were no more common on the GLP-1s: a risk ratio of 0.98, where 1.0 means equal, with a range from 0.70 to 1.377. For mental-health problems as a whole, the picture did not change with the drug, the dose or the length of treatment7.

Why do so many people say a GLP-1 ruins their sleep?

Because people do report it, often, even where trials find no gap. In the FDA's reporting system, insomnia was the most reported mental-health complaint with GLP-1 drugs from 2004 to 2023, named 1,198 times across 8,240 reports of a mental-health complaint8. Common is not the same as linked. Plain insomnia did not pass that study's signal test; one narrow term did, insomnia put down to a medical condition (75 reports, about twice as often as with other drugs in the database)8. A 2026 analysis of 111,105 GLP-1 reports found 16 mental-health terms reported more often than with other diabetes and weight drugs, and none was a sleep term9. One older weekly drug, dulaglutide (Trulicity), did show an insomnia signal, reported about three times as often as with other drugs10.

Health-record studies, which follow real patients over time, point the other way. In US records of people with type 2 diabetes, semaglutide carried a 14% lower risk of a new insomnia diagnosis over a year than sitagliptin, a gap that did not pass the authors' correction for testing 22 outcomes, and no difference against empagliflozin or glipizide11. In adults with obesity, semaglutide carried a lower insomnia risk than naltrexone-bupropion (Contrave) over two years, while tirzepatide carried a 13% higher risk than semaglutide in people without diabetes12. Studies like these show links, not causes.

Social media tells the same story. A study of 165 popular TikTok side-effect videos found that every complaint raised in five or more videos was on the package insert, except insomnia15. Across Reddit, YouTube and TikTok comments, sleep was the most common mental-health topic, with users describing both worse and better nights16.

Commonly reported: trouble falling asleep, waking at night and restless evenings in the first weeks or after a dose step, and for others the opposite, deeper sleep without waking hungry. On PepVault's reports pages, which tally what people post on forums and in videos, 2 of 41 semaglutide users named worse sleep; for tirzepatide, 2 of 77 named better sleep and 1 worse.

Do GLP-1s cause vivid dreams or nightmares?

No study has measured it. Abnormal dreams and nightmares are in none of the five labels12345, no published trial reports a rate, and the two large studies of reports sent to the FDA did not flag them89. The only number comes from the maker: Novo Nordisk told Fox News in 2023 that abnormal dreams were reported at under 0.1 per 100 patient-years in its trials, too rare for the label, and that reports after launch did include them, at a frequency that cannot be estimated19.

One known cause does fit, in people with diabetes. Nightmares, crying out in sleep and waking with damp sheets are signs of low blood sugar during the night17. The labels say that risk rises when a GLP-1 is taken with insulin or a sulfonylurea, a pill that pushes the body to release insulin1234. On Zepbound in type 2 diabetes, 10.3% of people also on a sulfonylurea had a low reading, against 2.1% of those not on one4. Without diabetes it is rare: 0.3% on Zepbound against none on placebo, in a trial that did not track it closely4.

Why might you sleep worse in the first weeks?

Because that is when the stomach effects peak. Zepbound's label says most nausea, vomiting and diarrhea happened while the dose was being stepped up, and eased over time4; Wegovy's says the same of indigestion and reflux2. Reflux, acid rising into the food pipe, runs at 4 to 5% on Wegovy and Zepbound against 2 to 3% on placebo24, and the NIH's advice for reflux at night is to eat at least three hours before lying down18. Reports of mental-health complaints as a whole also cluster early: a median of 12 days after starting semaglutide and 10 days after tirzepatide, with nearly half in the first 30 days8.

Some users tie their nights to hunger, in both directions; no trial has measured it. These steps follow the labels and the NIH's advice on reflux and low blood sugar at night.

  1. Note which nights are bad and how many days after your shot they fall.
  2. If heartburn or a sour taste wakes you, finish your last meal at least three hours before you lie down18.
  3. If reflux keeps coming back, raise your head and upper back 6 to 8 inches with a foam wedge or extra pillows20.
  4. If you take insulin or a sulfonylurea and wake sweaty, shaky or from a nightmare, check your blood sugar then and tell your prescriber, who may lower those drugs117.
  5. Bring your notes to your next appointment if poor sleep lasts past the dose steps.

Log each bad night as a symptom in PepVault with a severity from 1 to 5 and a note; the log keeps the date, so you can see whether it follows your dose steps. Log it in PepVault

When to get help Waking confused or drenched in sweat while on insulin or a sulfonylurea, or weeks of sleeplessness with low mood, is a reason to call your prescriber soon.

Can a GLP-1 improve sleep?

In the trials that asked, yes. SURMOUNT-OSA was two 52-week trials of tirzepatide in people with obesity and sleep apnea, and both rated sleep on a standard questionnaire (PROMIS Sleep Disturbance, where a lower score means better sleep). Tirzepatide beat placebo by 2.0 points in the trial without breathing machines and by 3.9 points in the trial with them13. An earlier sleep study counted 3.1 points as a meaningful change for one person, but the authors say that bar was not built for comparing group averages, and that meaningful-change thresholds for these questionnaires in sleep apnea are still unsettled13.

Outside apnea the evidence is thinner. In a 2026 Turkish study of 78 adults starting semaglutide or tirzepatide, the share with poor sleep on the Pittsburgh Sleep Quality Index fell from 55.1% to 16.7% in three months14. It had no comparison group, so it shows a change after starting, not that the drug caused it.

What about sleep apnea?

Zepbound is the one GLP-1 drug approved for it: its label covers moderate to severe obstructive sleep apnea in adults with obesity4. No semaglutide brand carries the approval. Our guide to GLP-1s and sleep apnea covers the trial numbers, how fast it works and what it means for a CPAP machine.

Track it in the app

Sources

  1. Ozempic (semaglutide) injection, prescribing information. Novo Nordisk. DailyMed, revised June 2026. dailymed.nlm.nih.gov
  2. Wegovy (semaglutide) injection and tablets, prescribing information. Novo Nordisk. DailyMed, revised June 2026. dailymed.nlm.nih.gov
  3. Mounjaro (tirzepatide), prescribing information. Eli Lilly. DailyMed, revised August 2026. dailymed.nlm.nih.gov
  4. Zepbound (tirzepatide), prescribing information. Eli Lilly. DailyMed, revised August 2026. dailymed.nlm.nih.gov
  5. Saxenda (liraglutide), prescribing information. Novo Nordisk. DailyMed, revised February 2026. dailymed.nlm.nih.gov
  6. Wadden TA et al. Psychiatric safety of semaglutide for weight management in people without known major psychopathology: post hoc analysis of the STEP 1, 2, 3, and 5 trials. JAMA Intern Med. 2024. PubMed 39226070
  7. Han S et al. Association of GLP-1 receptor agonist use with risk of psychiatric disorders: a systematic review and meta-analysis of randomised controlled trials. Diabetes Obes Metab. 2026. PubMed 41852258
  8. Chen W et al. Psychiatric adverse events associated with GLP-1 receptor agonists: a real-world pharmacovigilance study based on the FDA Adverse Event Reporting System database. Front Endocrinol. 2024. PubMed 38390214
  9. Moon S et al. Psychiatric safety signals of GLP-1 receptor agonists: a FAERS-based pharmacovigilance study with explainable machine learning. Pharmaceuticals. 2026. PubMed 42356570
  10. Katranski J et al. Psychiatric adverse events linked to glucagon-like peptide 1 analogues: a disproportionality analysis in American, Canadian and Australian adverse event databases. Int J Clin Pharm. 2025. PubMed 40522403
  11. De Giorgi R et al. 12-month neurological and psychiatric outcomes of semaglutide use for type 2 diabetes: a propensity-score matched cohort study. EClinicalMedicine. 2024. PubMed 39764175
  12. Huang YN et al. Neuropsychiatric association of tirzepatide and semaglutide in obesity with and without type 2 diabetes. Commun Med. 2026. PubMed 42373755
  13. Kanu C et al. Effect of tirzepatide treatment on patient-reported outcomes among SURMOUNT-OSA participants with obstructive sleep apnea and obesity. Sleep Med. 2025. PubMed 40774158
  14. Kır Y et al. Prospective changes in sleep quality and depressive symptom burden following initiation of GLP-1-based therapies in patients with obesity: a real-world study. J Endocrinol Invest. 2026. PubMed 42611111
  15. Wantuch GA, Singleton J. TikTok's take on adverse effects for glucagon-like peptide-1 and gastric inhibitory polypeptide receptor agonists. J Am Pharm Assoc. 2025. PubMed 40049364
  16. Arillotta D et al. GLP-1 receptor agonists and related mental health issues; insights from a range of social media platforms using a mixed-methods approach. Brain Sci. 2023. PubMed 38002464
  17. Low blood glucose (hypoglycemia). National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK). niddk.nih.gov
  18. Eating, diet, and nutrition for GER and GERD. NIDDK. niddk.nih.gov
  19. Rudy M. Ozempic, Wegovy and all those crazy, vivid dreams: is there a connection? Fox News, May 2, 2023 (Novo Nordisk's statement). foxnews.com
  20. Treatment for GER and GERD. NIDDK. niddk.nih.gov

Newsletter · Weekly

The weekly peptide briefing.

New guides, the week’s peptide news, one tool. Free, one email a week, unsubscribe anytime.

We’ll email you a link to confirm. Your address is used for the briefing only. Privacy

PepVault for iPhone

Keep your protocol in PepVault

Log every dose, track every vial, and see your protocol come together.

Download on the App Store

7 days free on the yearly plan for new subscribers.

Explore the live demo