If your feed looked anything like mine this week, you saw the headlines. The FDA is requiring a new warning on GLP-1 medications over a rare eye condition linked to sudden vision loss.
Some people are panicking and talking about quitting. Others are using the news as a cudgel against anyone taking these drugs.
I take one of these medications myself, so I read the news the way you probably did, with a knot in my stomach. Then I read the FDA’s actual safety communication, and I think the real picture deserves a calmer, more complete telling than the dramatic, clickbait headlines gave it.
What the FDA Actually Said
On November 9th, the FDA announced it is requiring the makers of every GLP-1 receptor agonist, a class that includes Ozempic, Wegovy, Mounjaro, Zepbound, and others, to add information about a potential increased risk of non-arteritic anterior ischemic optic neuropathy, or NAION, to the Warnings and Precautions section of their labels.
The word I don’t want you to miss here is “potential.” That’s the FDA’s own language. Keep it in mind. I’ll be getting back to that later.
What NAION Is, and Who’s at Risk
NAION is a rare condition in which blood flow to the front of the optic nerve drops, causing sudden, painless vision loss in one eye. People sometimes notice it when they wake up, as blurry or missing areas of vision, a blind spot, or washed-out color. It can be permanent, and it is as serious as it sounds.
The FDA says researchers don’t completely understand what causes it, but known risk factors include smoking, diabetes, high blood pressure, sleep apnea, anemia, and kidney disease. If it happens in one eye, the risk to the other eye goes up.
That list of risk factors matters because several are common in people who take GLP-1s, which is one reason it’s been hard to untangle whether the medication itself is playing a role.
It’s also worth noticing that a few of those risk factors, like type 2 diabetes, high blood pressure, and sleep apnea, are conditions these drugs are used to treat.
Not a Brand-New Discovery
This is also not a brand-new discovery. A research team first flagged a possible link with semaglutide in a paper published in JAMA Ophthalmology in July 2024. Europe’s medicines regulator reviewed the evidence and concluded in June 2025 that NAION is a “very rare” side effect of semaglutide, affecting up to 1 in 10,000 people taking it, and ordered the product information updated.
In other words, the label already exists in Europe, and the U.S. is catching up. What’s new this week is that the FDA is making it official here, and extending it to the entire class.
How Rare Is “Rare”?
So how rare is rare? The FDA did its own analysis, pooling 126 randomized clinical trials with more than 134,000 participants. Across those trials, researchers found just 15 probable NAION cases: 10 in people taking a GLP-1 and 5 in people taking a placebo. When the numbers were adjusted for time on treatment, that works out to about 6.4 cases per 100,000 person-years on GLP-1s versus 3.8 per 100,000 person-years in the comparison groups.
I was unfamiliar with the term person-years. If you are too, it’s simply a way to add up the total time everyone in a study was observed. This lets researchers compare groups fairly evenly when people are followed for different lengths of time.
For context, the FDA says that the background rate of NAION in the general population is estimated at 2.3 to 10.2 cases per 100,000 people per year among adults over 50. Put another way, the trial data suggest roughly two to three additional cases per 100,000 people treated per year.
The estimates vary depending on how they’re measured.
The European review leaned heavily on large observational studies in people with type 2 diabetes. They estimated about a twofold increase in risk. That works out to roughly one additional case per 10,000 person-years of treatment. That’s a bigger number than the FDA’s trial figure, but it’s still firmly in “very rare” territory.
Why the FDA Says “Potential”
Remember when I asked you to make note of the word “potential”? That’s because the evidence, while consistent enough to take a closer look, doesn’t establish cause and effect. The agency’s review of animal data didn’t find a mechanism that would explain how these drugs could cause NAION.
The clinical trial cases were so few that a meta-analysis was needed just to see a signal. And in the FDA’s own large study of real-world insurance data in adults with type 2 diabetes, it was a mixed picture. Compared with SGLT-2 inhibitors, another class of diabetes drug, GLP-1 users didn’t have a statistically significant increase in NAION, while compared with DPP-4 inhibitors they did.
The baseline rate in that population was strikingly higher than in the general public, around 26 to 32 cases per 100,000 person-years no matter which diabetes drug people took, which underlines how much diabetes itself matters.
None of that means there’s nothing here to pay attention to. The FDA also noted that well-designed observational studies have consistently found an increased risk compared with other diabetes drugs, and that’s why it acted.
A more balanced assessment is that this finding is real enough to warrant a warning, small enough to be very rare, and not yet proven to be caused by the drugs themselves.
The Other Side of the Ledger
Now for the part that seems to have been completely lost in all the fear-mongering. The SELECT trial followed more than 17,600 adults with obesity and existing heart disease, none of whom had diabetes.
Over an average of about three and a half years, 6.5% of people on semaglutide had a heart attack, stroke, or cardiovascular death, compared with 8.0% on placebo, a 20% relative reduction, or about 15 fewer major events for every 1,000 people treated.
Nonfatal heart attacks dropped from 3.7% to 2.7%, and deaths from any cause dropped from 5.2% to 4.3%. A later analysis suggested the heart benefit was largely independent of how much weight people lost.
In SURMOUNT-1, a trial of people with prediabetes and obesity, 13.7% of those on placebo developed type 2 diabetes over three years compared with 2.4% of those on tirzepatide. That’s a stunning 94% reduction in risk. And in an analysis of the STEP trials, semaglutide cut levels of C-reactive protein, a key marker of inflammation, by roughly 40%.
In other words, GLPs are linked to real, major reductions in risks for some life-threatening… and even life-ending… diseases.
A Comparison With Limits
That being said, I want to be VERY careful about this comparison, because it’s far from perfect. Researchers measured those benefits in specific groups of people, and each person’s own balance of benefit and risk will look different. Someone who has already had NAION in one eye is in a different position than someone who hasn’t. And someone with several risk factors may weigh things differently than someone with none.
That’s exactly why this label change exists. It forces us to make sure conversations between patients and their doctors include this.
Worried? Here’s What to Do
Here’s what the FDA says to actually do. If you ever experience sudden, painless vision loss or changes in one eye, blurry or missing areas, or color distortion, contact a health care professional right away. If NAION is suspected, you’ll be referred to an eye doctor.
Obviously, if you’ve ever had NAION and are on a GLP-1, you should have a chat with your doctor right away about whether to continue taking the med. And if you’re simply worried about the risk, go ahead and bring it up at your next appointment, especially if you have additional risk factors like diabetes, high blood pressure, or sleep apnea.
The FDA isn’t telling people to stop taking their medication on their own.
Two Things Can Be True at Once
I’m not a doctor, and nothing I ponder or discuss here could or should replace a conversation with your doctor about the risks versus rewards of taking a GLP-1 medication. But I’d gently push back on both of the reactions flooding my feed.
The warning isn’t proof these drugs are blinding people, and anyone claiming it is hasn’t read the data. It also isn’t nothing.
For the very small number of people it happens to, it’s life-changing. So the people raising it aren’t wrong to take it seriously. A rare risk and a substantial benefit can both be true at the same time. The honest way to handle that is the way we handle every other medication, with accurate numbers, a clear head, and a good conversation with the person who knows your health history.
