# GLP-1s and Daytime Sleepiness

By Nora Bissett, Pricing Editor (https://glpwatchdog.org/authors/nora-bissett) | Published: 2026-08-20 | Updated: 2026-10-27

Canonical page: https://glpwatchdog.org/research/glp1s-and-daytime-sleepiness

> A cohort found more diagnoses of excessive daytime sleepiness, at a 21% relative increase, which is 0.4% in absolute terms at one year. Four other sleep outcomes showed nothing.

## Key points

- Hypersomnolence HR 1.21 at one year, an absolute increase of 0.4%.
- No association with parasomnia, disturbed sleep, restless legs or narcolepsy.
- Iron deficiency rose at five years (HR 1.11): small, obvious, and actionable.
- The Parkinsonism figure rests on low event counts and is not reported here as a benefit.

A real-world cohort found GLP-1 users more likely to be diagnosed with **hypersomnolence** (excessive daytime sleepiness), with hazard ratios of 1.21 at one year and 1.23 at five. Before that sounds alarming, look at the absolute figures: roughly a **0.4%** increase at one year and **0.7%** at five.[1]

## What was and was not found

The signal was specific rather than general, which is the most informative part of the study.

*Sleep and related outcomes in a real-world cohort of GLP-1 users.[1]*

| Outcome | Finding |
| --- | --- |
| Hypersomnolence | Increased, HR 1.21 at 1 year and 1.23 at 5 (absolute +0.4% at 1 year, +0.7% at 5) |
| Iron deficiency | Increased at 5 years, HR 1.11 (95% CI 1.06–1.16) |
| Parasomnia | No significant association |
| Disturbed sleep | No significant association |
| Restless legs syndrome | No significant association |
| Narcolepsy / cataplexy | No significant association |

One sleep outcome moved and four did not. That is a narrower and more credible finding than “these drugs affect sleep”, because a broad disruption of sleep architecture would be expected to show up in more than one place. Dreams were not among the outcomes; what is known about those is in [GLP-1s, vivid dreams and nightmares](https://glpwatchdog.org/research/glp1-vivid-dreams-and-nightmares).

## Why the absolute numbers matter here more than usual

> A 21% relative increase on a rare diagnosis is four extra people per thousand.

Daytime sleepiness is also one of the most confounded symptoms it is possible to study in this population. Obesity causes it. Sleep apnea causes it, and is very common in the same people (our [sleep apnea article](https://glpwatchdog.org/research/zepbound-for-sleep-apnea) covers a trial in exactly that group). Depression causes it. Eating far less causes it. Any of those could produce a diagnosis code without the drug doing anything.

There is also a detection effect: people starting a new medication see clinicians more often, and symptoms that were always present get recorded for the first time.

## The iron finding is the practical one

Iron deficiency rose at five years, at a hazard ratio of 1.11 with a tight confidence interval.[1] That is a small effect and a mechanistically obvious one: eating substantially less, for years, reduces intake of everything, and iron is among the first things to run short.

It is also directly actionable, which the sleepiness finding is not. Iron studies are one of the tests in the monitoring panel we covered in [protein targets and the lab panel](https://glpwatchdog.org/research/protein-targets-and-the-lab-panel), and this is the sort of evidence that panel exists for. Iron deficiency also causes fatigue, so some of the sleepiness signal may simply be the iron finding wearing a different diagnosis code.

> **The Parkinsonism finding should not be reported as a benefit, and we are not reporting it as one.** The study found a hazard ratio of 0.26 (a 74% reduction) and states that event counts were low. A dramatic ratio built on a handful of events is exactly the shape we take apart in [when a hazard ratio is too good](https://glpwatchdog.org/research/when-a-hazard-ratio-is-too-good). The randomized trials in people who already have Parkinson’s are set out in [GLP-1 drugs and Parkinson’s](https://glpwatchdog.org/research/glp1-and-parkinsons).

## Frequently asked questions

### Do GLP-1 drugs make you sleepy?

A real-world cohort found more diagnoses of excessive daytime sleepiness, at hazard ratios of 1.21 at one year and 1.23 at five, but the absolute increase was around 0.4% at one year and 0.7% at five. Four other sleep outcomes showed no association at all.

### Could something else explain it?

Easily. Obesity, sleep apnea, depression and eating very little all cause daytime sleepiness, and all are common in this population. People starting a new drug also see clinicians more often, so long-standing symptoms get recorded for the first time.

### What about the iron finding?

Iron deficiency rose at five years with a hazard ratio of 1.11. That is small but mechanistically obvious, since eating much less for years reduces intake of everything. It is the actionable result here, and iron studies are already part of the recommended monitoring panel.

### Could the iron deficiency be causing the sleepiness?

Plausibly. Iron deficiency causes fatigue, so part of the sleepiness signal may be the iron finding appearing under a different diagnosis. The study does not separate them.

### Does this reduce Parkinson's risk?

No conclusion should be drawn. The study reported a hazard ratio of 0.26 while noting event counts were low, and a dramatic ratio built on very few events is the least reliable kind of finding.

## References

1. Kamel-Abusalha L, Afifi AM, Dawoud A, et al. Association of GLP-1 Receptor Agonist Use with Hypersomnolence: A Real-world Cohort Analysis. Journal of Diabetes and Metabolic Disorders. 2026. PMID: 41867417. https://pubmed.ncbi.nlm.nih.gov/41867417/

---

Cite as: GLP Watchdog, Nora Bissett, "GLP-1s and Daytime Sleepiness", published 2026-08-20, updated 2026-10-27. https://glpwatchdog.org/research/glp1s-and-daytime-sleepiness

Canonical HTML page: https://glpwatchdog.org/research/glp1s-and-daytime-sleepiness

Medical disclaimer: Nothing here is medical advice. We report on this subject; we do not treat anyone. Whether to begin a medication, supplement, or treatment, whether to adjust it, and whether to stop are questions for a clinician licensed to answer them, who knows your history.
