Does Zepbound Cause Insomnia? Sleep Disturbance, SURMOUNT-OSA, and the FDA Label

At a glance

• Insomnia is not on the Zepbound label. The FDA prescribing information lists nausea, diarrhea, vomiting, constipation, abdominal pain, dyspepsia, injection-site reactions, fatigue, hypersensitivity reactions, eructation, hair loss, and GERD at the ≥5% reporting threshold — no sleep-related adverse event appears.
• The pooled trial population was 3,477 patients on Zepbound (5, 10, or 15 mg) vs 958 on placebo across SURMOUNT-1 and SURMOUNT-2. A real insomnia signal at ≥5% incidence in that sample would have been detected.
• SURMOUNT-OSA showed tirzepatide IMPROVES sleep. In adults with obesity and moderate-to-severe obstructive sleep apnea, tirzepatide reduced the apnea-hypopnea index by ~25-29 events per hour and improved sleep-quality patient-reported outcomes vs placebo.
• Patient-reported insomnia is real but most plausibly indirect. Three mechanisms account for most reports: GI discomfort fragmenting sleep, sharp evening calorie reduction producing wake-from-hunger episodes, and shifted meal timing disrupting circadian alignment.
• Class-level pharmacovigilance is mixed. A 2025 disproportionality analysis of FAERS, CVAROD, and DAEN found an insomnia signal for dulaglutide but not for tirzepatide or semaglutide — consistent with the tirzepatide label being silent on sleep AEs.
• Red flags warrant prescriber contact. Severe insomnia >2 weeks, new or worsening depressed mood, any new suicidal thoughts, or sleep disruption that does not respond to addressing the three indirect mechanisms all warrant a prompt prescriber call.
• Most disruption is time-limited. The patient-reported pattern peaks during the first 4 weeks and during the 5-to-7-day window after each dose escalation, then attenuates as the patient adapts.

What the Zepbound FDA label says about insomnia

The Zepbound prescribing information (DailyMed SetID 487cd7e7-434c-4925-99fa-aa80b1cc776b)^[1] reports adverse reactions from a pool of two randomized, placebo-controlled trials — SURMOUNT-1 (NCT04184622) and SURMOUNT-2 (NCT04657003) — that enrolled 3,477 adult patients on Zepbound at 5 mg, 10 mg, or 15 mg weekly versus 958 on placebo, treated for up to 72 weeks. The Section 6.1 Adverse Reactions table enumerates events occurring at ≥5% incidence with Zepbound.

The full list of labeled ≥5% adverse reactions: nausea, diarrhea, vomiting, constipation, abdominal pain, dyspepsia, injection-site reactions, fatigue, hypersensitivity reactions, eructation, hair loss, and gastroesophageal reflux disease. A verbatim search of the complete label for the strings “insomnia”, “somnolence”, “sleep disturbance”, “abnormal dreams”, and “dyssomnia” returns zero matches anywhere in Sections 4, 5, 6.1, 6.2 (postmarketing experience), 7 (drug interactions), 8 (use in specific populations), or 17 (patient counseling)^[1]. Insomnia is simply not on the Zepbound label.

The Section 14.2 Clinical Studies block on the OSA indication goes further: in the two SURMOUNT-OSA trials (Studies 5 and 6, N=469 total), Zepbound-treated patients showed improvement in sleep-related impairment compared to placebo, measured by the PROMIS Sleep-Related Impairment 8a instrument^[1]. The label is therefore not merely silent on sleep AEs — it documents a measured improvement in sleep-quality patient-reported outcomes in the population with the strongest baseline sleep pathology.

The §6.1 adverse-reactions table in numbers

The Zepbound ≥5% adverse-reactions table from the pooled SURMOUNT-1 + SURMOUNT-2 safety analysis is the cleanest snapshot of which side effects do clear the labeled-disclosure threshold. None of them are sleep-related. The dominant cluster is gastrointestinal:

The pooled placebo-controlled sample size (3,477 Zepbound, 958 placebo) is large enough that a real insomnia signal at the ≥5% reporting threshold would have been detected and enumerated. The label’s silence on sleep AEs is therefore a meaningful absence rather than a missed signal in an underpowered study.

SURMOUNT-OSA: tirzepatide IMPROVES sleep in OSA

The strongest randomized GLP-1-class evidence on objectively measured sleep is the SURMOUNT-OSA program, which randomized adults with obesity and moderate-to-severe obstructive sleep apnea (AHI ≥15) to tirzepatide or placebo over 52 weeks across two parallel trials (one with patients not on positive-airway-pressure therapy, one with patients already on PAP). The primary results, published by Malhotra and colleagues 2024 in the New England Journal of Medicine^[3], showed tirzepatide reduced the apnea-hypopnea index (AHI — the standard measure of breathing pauses per hour during sleep) by approximately 25-29 events per hour relative to placebo. The effect size is large enough that a substantial fraction of treated patients moved from severe to mild OSA or normalized entirely.

The 2026 secondary outcomes publication in Nature Medicine^[4] extends the readout to patient-reported sleep quality, daytime sleepiness on the Epworth Sleepiness Scale, and cardiometabolic risk markers. Tirzepatide-treated patients reported improved sleep quality, reduced daytime sleepiness, and improvements across cardiometabolic markers that track with both weight loss and improved sleep architecture. The mechanism is intuitive: in obstructive sleep apnea, excess pharyngeal soft tissue collapses the upper airway during sleep, and weight loss reduces that tissue load. Tirzepatide produces ~20% mean body-weight loss at 52 weeks in this population, and the AHI improvements track the weight-loss magnitude.

For a deeper walk-through of the SURMOUNT-OSA primary and secondary results, see the dedicated SURMOUNT-OSA tirzepatide sleep apnea evidence review.

Why this matters for the Zepbound insomnia question: the closest rigorous evidence on tirzepatide and objectively measured sleep shows the drug improves sleep in the population with the strongest baseline sleep pathology. That does not refute patient-reported insomnia — the SURMOUNT-OSA population is selected for OSA, not for primary insomnia, and the mechanism (weight loss reducing pharyngeal soft-tissue load) is specific to obstructive breathing rather than to sleep initiation or maintenance generally. But it argues strongly against a simple “tirzepatide disrupts sleep” framing and supports the more nuanced reading that any patient-reported signal is driven by indirect mechanisms specific to the early-treatment window rather than by direct pharmacology.

Likely mechanisms behind patient-reported insomnia

Three indirect mechanisms account for most patient reports of sleep disruption on Zepbound. Each is biologically plausible, each is most pronounced during the first 4 weeks of treatment and the first week after each dose escalation, and each is addressable with practical adjustments rather than requiring drug discontinuation.

Mechanism 1: GI discomfort and nausea fragmenting sleep

Gastrointestinal adverse events are the dominant labeled tolerability burden on tirzepatide. SURMOUNT-1 (Jastreboff and colleagues 2022, NEJM)^[2] reported nausea in 29% of patients on the 10 mg dose and 28% on the 15 mg dose across the 72-week trial, with rates highest during the dose-escalation period. The cross-class anchor for the higher doses comes from STEP-1 (Wilding 2021, NEJM)^[5], which tested semaglutide 2.4 mg weekly and reported nausea in about 44% of patients vs 16% on placebo — the magnitude signal is consistent across the incretin class.

Active nausea and GI discomfort fragment sleep in straightforward ways: difficulty falling asleep while feeling queasy, mid-night awakenings to vomit or pass a bowel movement, and reduced ability to find a comfortable sleeping position when the abdomen is uncomfortable. The relationship is mechanically direct — this is not a subtle effect on sleep architecture but a straightforward arousal from somatic discomfort. For broader management of the GI cluster, see the GLP-1 side effects Q&A hub and the GLP-1 side-effect timeline tool that visualizes the week-by-week adaptation curve.

Mechanism 2: evening calorie reduction and wake-from-hunger

The therapeutic effect of tirzepatide on weight is mediated primarily through appetite suppression and reduced caloric intake. In SURMOUNT-1, the 15 mg dose produced ~22.5% mean body-weight loss at 72 weeks^[2] via reduced food intake without an exercise prescription. In practice, many patients respond to the appetite suppression by sharply reducing or skipping the evening meal — the meal where appetite is often lowest because the day’s earlier meals have already been satisfying in a way they were not pre-treatment.

Skipping the evening meal entirely can produce a wake-from-hunger pattern in the 2-to-4-AM window. Even in patients without diabetes, overnight glycemic dips into the low-normal range can trigger counter-regulatory responses (cortisol, epinephrine, growth hormone release) that fragment sleep and produce vivid dreams during awakening. In patients with type 2 diabetes on tirzepatide plus a sulfonylurea or insulin, the risk of true nocturnal hypoglycemia is more substantial and the Zepbound label flags the interaction in Section 5.7^[1].

The practical pattern: if you are skipping dinner entirely and waking at 2-3 AM, a small protein-forward evening snack (a handful of nuts, a hard-boiled egg, a few ounces of Greek yogurt) often restores continuous sleep without compromising the overall calorie deficit. The goal is glycemic stability across the overnight window, not added daily calories.

Mechanism 3: shifted meal timing disrupting circadian alignment

Sleep and glucose homeostasis are tightly coupled through the circadian system. The 2022 narrative review of sleep and circadian rhythm disturbances in diabetes (Rutters and Nefs, Diabetes, Metabolic Syndrome and Obesity)^[7] summarizes how meal timing acts as a peripheral circadian zeitgeber: meals consumed at consistent times synchronize hepatic and adipose-tissue circadian clocks with the central suprachiasmatic clock that drives sleep-wake rhythms. Disrupting the meal schedule disrupts the alignment.

Tirzepatide commonly shifts the meal pattern away from the pre-treatment baseline. Patients describe eating less in the evening, eating earlier in the day, occasionally skipping breakfast (because appetite has not yet recovered from the previous evening), and consolidating intake into one or two eating windows rather than three regular meals. Any of these shifts changes the timing signal the peripheral clocks receive and can produce a transient period of circadian misalignment until a new stable pattern is established — experienced as difficulty falling asleep at the usual time, waking unusually early, or daytime sleepiness without obvious cause.

The mitigation is to anchor at least one regular eating time each day — ideally a consistent morning meal, even a small one — rather than allowing all meal timing to drift with appetite. Consistent meal anchors are a well-established chronobiology lever for re-aligning peripheral clocks.

How Zepbound compares to Wegovy, Ozempic, and Mounjaro on sleep AEs

The picture is consistent across the GLP-1 / GIP-GLP-1 class labels: insomnia is not enumerated at the ≥5% reporting threshold on Zepbound, Mounjaro (also tirzepatide), Wegovy (semaglutide 2.4 mg), or Ozempic (semaglutide up to 2.0 mg). The Zepbound and Mounjaro labels share the SURMOUNT-1 and SURPASS adverse-reaction datasets and the silence on sleep AEs is identical. For a deeper read on the semaglutide-specific patient-reported insomnia question and how the same indirect mechanisms apply, see Ozempic insomnia and sleep disturbance evidence. For the broader trial-documented adverse-event picture across the class, see What the trials actually showed: GLP-1 side effects and the side-by-side Zepbound vs Wegovy side-effects comparison.

The 2025 disproportionality analysis by Katranski and colleagues in International Journal of Clinical Pharmacy^[6] examined psychiatric adverse-event reports across the US FAERS, Canadian CVAROD, and Australian DAEN pharmacovigilance databases for all approved GLP-1 analogues. The analysis found a significant insomnia signal for dulaglutide (FAERS reporting odds ratio 2.93, 95% CI 2.35-3.66) but did not identify a corresponding insomnia signal for tirzepatide or semaglutide in its published results. That is consistent with the Zepbound label being silent on sleep AEs and supports the framing that tirzepatide is not a direct insomnia-causing drug at the pharmacological level. The dulaglutide signal is worth noting for patients who have been on Trulicity and are switching to or from Zepbound — the molecule-specific pharmacology differs.

When insomnia is a red flag on Zepbound

The Zepbound label Section 5.8 carries the GLP-1-class depression and suicidal-behavior monitoring guidance^[1]: prescribers and patients are instructed to monitor for the emergence or worsening of depression, suicidal thoughts, and unusual changes in mood or behavior. New or worsening insomnia is a non-specific symptom but can be one of several prodromal mood-symptom signals, and the threshold for prescriber contact should be low when sleep disruption is accompanied by any of the following:

• New or worsening depressed mood, hopelessness, or loss of interest — particularly if these symptoms emerged after starting or escalating the dose.
• Any new suicidal thoughts, plans, or self-harm urges — this is an immediate prescriber contact and, in the presence of plan or intent, an immediate emergency-services contact (call or text 988 in the US for the Suicide and Crisis Lifeline).
• Severe insomnia for more than 2 weeks — defined as taking more than 30 minutes to fall asleep on most nights, or being awake for more than 30 minutes mid-night on most nights, with daytime functioning impairment.
• Sleep disruption that does not respond to addressing the three indirect mechanisms above (GI symptoms managed, regular evening protein-forward snack, consistent meal-time anchor) over a 2-to-4-week observation window.
• New or worsening daytime sleepiness in the OSA-risk profile (obesity, snoring history, witnessed apneas) — this warrants a sleep-study referral regardless of the Zepbound context, since untreated OSA is one of the more common reversible causes of fragmented sleep. The SURMOUNT-OSA data mean Zepbound itself is a candidate treatment if OSA is confirmed.
• Symptoms consistent with acute pancreatitis or gallbladder disease — severe persistent abdominal pain (with or without radiation to the back) that disrupts sleep is a Section 5.5 and 5.6 red flag and requires immediate prescriber contact and discontinuation pending evaluation.

Practical sleep hygiene during Zepbound titration

For patients in the first 4 weeks of Zepbound or in the 5-to-7-day window after a dose escalation (2.5 → 5 → 7.5 → 10 → 12.5 → 15 mg), a few practical adjustments address most of the patient-reported sleep-disruption pattern without requiring drug discontinuation:

• Time the weekly injection earlier in the day rather than evening. Tirzepatide’s 5-day half-life makes within-week timing essentially flat at steady state, but the first 24-48 hours after the injection are when peak GI tolerability challenges occur. Injecting in the morning of a low-demand day (many patients pick a Saturday or Sunday morning) keeps the worst tolerability hours away from the sleep window.
• Eat a small, protein-forward evening snack if you skipped dinner. A handful of nuts, a hard-boiled egg, a small Greek yogurt, or a turkey roll-up 1-2 hours before bed addresses the wake-from-hunger mechanism. The total calorie addition is small (100-200 kcal) and does not meaningfully blunt the overall calorie deficit.
• Anchor one consistent meal time per day. A morning meal at a consistent time — even a small one — reinforces the peripheral circadian alignment that drifting meal patterns disrupt. Coffee plus a small protein-forward item works for most patients.
• Skip evening alcohol during the diagnostic period. Alcohol on top of altered gastric emptying is a common unrecognized driver of fragmented sleep on Zepbound. A 2-to-3 week alcohol-free observation is the highest-yield first intervention. See Can you drink alcohol while taking tirzepatide? for the full mechanistic walk-through.
• Hydrate during the day, not the evening. Front-loading water intake to the first 8-10 waking hours reduces mid-night urination while still addressing the dehydration risk flagged in Zepbound Section 5.3^[1].
• Standard sleep hygiene applies. Consistent bedtime, cool dark room, screens off 30-60 minutes before bed, no caffeine after early afternoon. None of these are Zepbound-specific, but the early-treatment window is a poor time to ignore the basics.
• Track fatigue separately from sleep quality. Daytime fatigue on Zepbound (a labeled ≥5% adverse reaction at 7% on the 15 mg dose^[1]) is a distinct phenomenon from sleep disruption and has its own mechanisms (caloric deficit, electrolyte shifts, dehydration). The Zepbound fatigue causes and management evidence review covers that side of the picture.

How this fits the broader Zepbound side-effect picture

Insomnia sits in an unusual position relative to the rest of the Zepbound side-effect cluster: the FDA label is silent on sleep AEs across Sections 4, 5, 6, 7, and 17, but the patient-reported signal in online communities is persistent enough that ignoring it would be dishonest. The most defensible framing is the one that respects both the trial evidence and the patient experience: the registrational data and the labeled-AE table provide no support for a direct pharmacological insomnia effect at clinically used doses, but the drug reliably produces three downstream changes (GI discomfort, evening calorie reduction, meal-timing shifts) each of which can plausibly disrupt sleep through mechanisms that the trial framework was not designed to capture.

SURMOUNT-OSA provides the affirmative counterweight: the closest randomized GLP-1-class sleep readout shows tirzepatide improves both objective sleep apnea severity and subjective sleep quality in the population with the strongest baseline sleep pathology. That is the most rigorous sleep-related evidence available on this drug and it points in the opposite direction from the patient-reported insomnia narrative.

Verdict

Insomnia is not listed anywhere in the Zepbound FDA prescribing information — not in Section 6.1 Adverse Reactions, not in Section 5 Warnings and Precautions, not in Section 17 Patient Counseling. The pooled SURMOUNT-1 + SURMOUNT-2 safety database (N=3,477 Zepbound vs 958 placebo) is large enough that a real signal at the ≥5% reporting threshold would have been detected. SURMOUNT-OSA, the closest randomized sleep-architecture readout, shows tirzepatide improves sleep in adults with obesity and moderate-to-severe obstructive sleep apnea. Patient reports of difficulty falling asleep and fragmented mid-night awakenings are real but most plausibly mediated by three indirect mechanisms: GI nausea and discomfort, sharp evening calorie reductions producing wake-from-hunger episodes, and shifted meal-timing patterns disrupting circadian alignment.

For most patients, sleep disruption on Zepbound is time-limited, peaks during the first 4 weeks and after dose escalations, and responds to practical adjustments: morning injections, a small evening protein-forward snack, consistent meal-time anchors, and skipping evening alcohol during a 2-to-3 week diagnostic period. Severe insomnia lasting more than 2 weeks, sleep disruption that does not respond to those adjustments, or any sleep symptom accompanied by new or worsening depressed mood warrants prompt prescriber contact under the Section 5.8 mood-monitoring guidance. Any new suicidal thoughts are an immediate prescriber contact and, in the US, an immediate 988 call or text.

This article is educational and does not constitute medical advice. Decisions about Zepbound dosing, timing, and management of sleep symptoms should be made with the prescribing clinician, particularly for patients with type 2 diabetes on insulin or a sulfonylurea, a history of mood disorder, or severe or persistent sleep disruption.