Ozempic Dreams and Nightmares: What the Evidence Shows

No study has measured dreams in people taking a GLP-1 drug. Abnormal dreams and nightmares do not appear on the Ozempic, Wegovy, Mounjaro, Zepbound or Saxenda labels.^[1][4] FDA’s adverse event database does hold reports of both, a few dozen per drug, and one analysis that tested those terms found no reporting signal.^[6] So the honest answer to “does Ozempic cause weird dreams” is that nobody knows, and the evidence that exists does not point to a direct effect. What is documented are things that break up sleep: nausea and reflux, and, in people on insulin or a sulfonylurea, low blood sugar at night, which can itself bring nightmares.^[2][11]

What the labels say about dreams

Nothing. Across the full prescribing information and patient leaflets of all five injectable GLP-1 brands sold for diabetes or weight loss, the words “dream” and “nightmare” do not appear once. Four of the five do not mention insomnia either.

Those tables have a cutoff. The Wegovy, Zepbound and Saxenda weight-loss tables list events seen in at least 2% of people on the drug and more often than on placebo; the Ozempic and Mounjaro diabetes tables start at 5%.^[2][3] The weight-loss tables alone draw on more than 2,000 people per drug. A common, drug-related change in dreaming would have had a fair chance to clear those bars. It did not. That leaves room for an uncommon effect, or one participants never thought to mention to trial staff.

Saxenda is the exception on sleep in general: insomnia was reported by 2.4% of people on liraglutide against 1.7% on placebo.^[5] That is a gap of under one percentage point, and it is about falling or staying asleep, not about dream content.

Reports to FDA: how many, and what they mean

FDA’s Adverse Event Reporting System (FAERS) collects voluntary reports from patients, doctors and manufacturers. Anyone can file one, for any suspicion, and the database has no count of how many people took the drug without a problem. Counts show what people noticed and reported. They cannot give a rate, and they cannot show cause.

Put another way, nightmares appear on roughly one semaglutide report in 800 and one tirzepatide report in 2,200. Sleep paralysis appears on a handful. Insomnia is reported at least nine times as often as either dream term. These figures cover only the reports FDA could match to a named product, so the true totals are higher, but the proportions are what matter here.

A raw count says nothing on its own, because every widely used drug collects reports of almost everything. The question a pharmacovigilance study asks is whether a term is reported more often with one drug than with all others in the same database. That comparison is called disproportionality. A positive result is a reporting signal, a reason to look closer. It is not a rate and does not prove the drug caused the event.

What the signal studies found

Three published FAERS analyses looked at psychiatric or sleep-related reports with GLP-1 drugs. None found a signal for dreams.

• US, Canadian and Australian databases, 2025. This study counted 49 nightmare and 38 abnormal-dream reports for semaglutide in FAERS, and 12 and 18 for tirzepatide. Both terms met the authors’ threshold for analysis. Neither appears among the statistically significant results, which included depressive symptoms and suicidal ideation with semaglutide. Insomnia-type sleep disorder did produce a signal for tirzepatide and dulaglutide.^[6] Reporting signals only, not rates.
• FAERS 2004 to early 2023. Of 181,238 reports for GLP-1 drugs, 8,240 involved a psychiatric event. Eight categories showed a reporting signal, including insomnia-type sleep disorder. Dreams and nightmares were not among them.^[7] Again a reporting signal, not proof of cause.
• Tirzepatide, 2022 to early 2025. Across 67,305 reports, 120 naming an insomnia-type sleep disorder were enough to produce a signal. The author’s explanation was malnutrition from eating very little, which is a hypothesis, not a finding.^[8] No dream term was flagged.

So the record from reporting databases runs one way. Sleep disruption of the insomnia kind does stand out in reports. Dream content does not. For mood signals from the same databases and how later studies handled them, see what happened with the GLP-1 suicidal ideation question.

The case reports

No published case report describes a GLP-1 drug causing vivid dreams, nightmares or sleep paralysis. The one case report that mentions both a GLP-1 and nightmares describes a 45-year-old woman with several long-standing sleep disorders, including nightmare disorder, night terrors and sleep-related eating. After she started a GLP-1 for obesity and lost substantial weight, her night eating stopped. The authors were explicit that other treatment changes happened at the same time and “causality cannot be inferred.”^[10] The report does not say the drug changed her dreams in either direction.

Why some people may notice more dreams

If dreams feel more vivid after starting a GLP-1, there are documented effects that could plausibly explain it. Each step below is supported on its own. Joining them into an explanation for GLP-1 users is our inference, and no study has tested it.

Waking up more means remembering more

Everyone dreams most nights. Whether you remember it depends partly on whether you wake up soon after. In a sleep-lab study, people who often recall dreams spent more time awake during the night and had more awakenings than people who rarely recall them.^[12] Anything that fragments sleep can raise dream recall, with no change in the dreams themselves.

Nausea, reflux and burping at night

Stomach complaints top every one of these labels. On Wegovy 2.4 mg, 44% of trial participants had nausea and 24% had vomiting, against 16% and 6% on placebo.^[2] On Zepbound, 25% to 29% reported nausea depending on dose, against 8% on placebo, and reflux was twice as common.^[4] Our guide to burping, reflux and gas on a GLP-1 covers what helps. For when these symptoms usually ease, see how long GLP-1 side effects last.

Surveys of dreamers link stomach trouble with bad dreams. In an online survey of 1,082 adults, gastric symptoms and eating late in the evening predicted nightmares and negative dream content.^[14] That is a correlation in people not taking GLP-1s. It fits the idea that an upset stomach at night can color dreams, but it does not show that this happens on these drugs.

Eating less in the evening

GLP-1s reduce appetite, which can mean a smaller or skipped evening meal. The same research group found, in two separate surveys, that longer gaps between meals and less evening eating went with more vivid dreams and higher dream recall, not with nightmares.^[13][14] These were students and online volunteers answering questionnaires, so the finding is weak. It is the closest published evidence to the “my dreams got vivid when I stopped eating at night” experience, and it points to vivid dreams being a side note of eating patterns rather than a drug effect.

Low blood sugar during sleep

This one has a clear mechanism. NIDDK, the federal diabetes institute, lists nightmares, crying out in sleep, and sweating enough to dampen sheets among the symptoms of low blood glucose while asleep.^[11] Used alone, a GLP-1 seldom drives glucose too low, although the Zepbound label notes cases in people without diabetes. Add insulin or a sulfonylurea and the odds go up. The Ozempic and Mounjaro labels both warn of a higher risk of hypoglycemia, including severe episodes, in that combination.^[1][3] In the Zepbound trial in people with type 2 diabetes, 10.3% of those also taking a sulfonylurea had blood sugar below 54 mg/dL, against 2.1% of those who were not.^[4]

Nightmares on insulin or a sulfonylurea, especially with night sweats or waking with a headache or confusion, are a reason to check blood sugar rather than to blame the dreams on the GLP-1. Our guide to low blood sugar risk on GLP-1s explains which combinations matter and what the labels ask prescribers to do.

Sleep paralysis

Sleep paralysis is a brief spell of being awake but unable to move, at the edge of sleep. FAERS holds six reports naming it with semaglutide and eight with tirzepatide, out of more than 230,000 reports for the two drugs combined.^[9] No published study or case report links it to a GLP-1. Nothing in the record suggests these drugs cause it.

What to do about it

• Check the timing. If vivid dreams began with a dose increase and nights are broken by nausea or reflux, the stomach is the likelier cause, and treating it is the first step. The reflux guide linked above covers the practical options.
• On insulin or a sulfonylurea, check your glucose. Nightmares with sweating or a morning headache can be night-time hypoglycemia. Tell your prescriber; the labels say those doses may need to come down.^[1][3]
• Watch your mood, not only your dreams. Persistent distressing dreams alongside low mood, anxiety or thoughts of self-harm should go to a clinician promptly.
• Physically acting out dreams, such as shouting, punching or leaving the bed, is not a known GLP-1 effect and should be raised with a doctor.
• Report it if you want it counted. FAERS only knows what is sent in. Your prescriber or pharmacist can file a report, and so can you.

If your sleep problem is daytime sleepiness instead, that has its own evidence: GLP-1s and daytime sleepiness. Zepbound is also approved to treat obstructive sleep apnea in adults with obesity, covered in Zepbound for sleep apnea.