The risk too many people are taking with Ozempic and other GLP-1s.
Sign up for the Slatest to get the most insightful analysis, criticism, and advice out there, delivered to your inbox daily.
Every week, millions of Americans fill prescriptions for GLP-1 receptor agonists, drugs like Ozempic and Wegovy. These injections—and, now, pills—have reshaped how medicine thinks about obesity and diabetes, as well as helped usher in a new era of hyperskinny celebrities and models. Many experts agree: GLP-1s are, when it comes to their medical applications, something of a miracle drug. But as a neuroscience Ph.D. candidate who studies how the gut and nervous system negotiate pain, I keep returning to a question I haven’t seen asked loudly enough: What happens to a body that already makes GLP-1 naturally when we flood it with pharmacological doses of the same signal?
GLP-1 is not a foreign molecule. It is a hormone your intestines release after every meal, a chemical signal between your gut and your brain that says Slow down, you’ve had enough. The drugs don’t introduce something new to the body—they connect with receptors that are already there. They make you feel full much sooner than you would otherwise by flooding the zone with the message You’re full! You’ve had enough! And in my field, when you chronically overstimulate a signaling pathway the nervous system depends on for normal function, things tend to go wrong in ways that may be quiet at first but get much louder later.
The evidence is beginning to catch up. Between 40 and 70 percent of patients on these drugs report gastrointestinal adverse effects, like acid reflux, chronic diarrhea, and/or constipation. Studies now show that patients on GLP-1 agonists face a nearly fourfold increased risk of gastroparesis (a chronic slowdown of the gut causing abdominal bloating and pain) and a ninefold increased risk of pancreatitis compared to people on other weight-loss medications. If those numbers are borne out in the real world, most people on GLP-1s may have some kind of undesirable GI effect from them. Yet current prescribing guidelines don’t recommend asking a basic question before initiating treatment: Does this patient actually have low endogenous GLP-1 levels? At a time when many doctors pride themselves on practicing precision medicine, that omission is striking and dangerous.
Consider a hypothetical woman in her late 40s who is just looking to lose 10 pounds before summer so that she can finally wear that bikini she bought a decade ago. She has no diabetes, no prior GI history—but she also has a body she doesn’t love after bringing two kids into the world. She’s prescribed semaglutide for weight loss by her primary care physician, but six months later, she is nauseated most mornings, bloated after small meals, and losing an amount of weight she didn’t intend to lose. Her doctor attributes it to the drug working. A gastroenterologist eventually diagnoses her with gastroparesis. Whether the medication caused it, accelerated a subclinical condition, or simply unmasked something already there is nearly impossible to determine. That ambiguity is precisely the problem.
GLP-1 receptor agonists slow gastric emptying. That slowing isn’t incidental—it’s the drug doing exactly what it was designed to do, because GLP-1 is the signal that makes it happen. But the enteric nervous system, which governs gut motility through a dense network of neurons that neuroscientists sometimes call the “second brain,” is not built to sustain that kind of chronic, pharmacological override. In patients who already produce normal or elevated levels of endogenous GLP-1, flooding those same receptors with long-acting agonists may push a finely tuned system past its adaptive range.
Proving that in any individual case remains elusive. The population-level signal, however, is hard to dismiss. A large cohort study found the gastroparesis and pancreatitis risk figures cited above among patients using GLP-1 agonists specifically for weight loss, not diabetes. And because obesity itself is a known driver of GI dysfunction, every data point carries an asterisk: Is this the drug or the disease it’s treating? The honest answer is that we don’t fully know. That uncertainty should be making prescribers more cautious, not less.
What makes this especially frustrating from a neuroscience perspective is that the gut-brain axis is one of the most pharmacologically sensitive systems in the body. We have decades of evidence showing that chronic manipulation of enteric signaling (i.e., telling different sections of the digestive tract what to do and when), whether through opioids, supplements, or antidepressants, produces adaptations that outlast the drug itself. GLP-1 receptors are expressed not just in the pancreas and stomach, but throughout the vagus nerve, the brain stem, and the hypothalamus. When we prescribe these drugs to patients whose endogenous GLP-1 system is functioning normally, we are not filling a deficit. We are overwriting a signal the body is already sending, and we’re doing it at scale, in millions of people, without a baseline measurement to tell us what we’re overwriting.
The guidelines have not caught up. Current prescribing criteria center on body mass index thresholds and glycemic targets, metabolic markers that tell us something about a patient’s need for weight loss but nothing about their underlying GLP-1 physiology. No major guideline recommends measuring endogenous GLP-1 levels before initiating treatment. What’s more, patients can order these drugs without ever setting foot in a doctor’s office, or even truly needing the assistance of these drugs from a clinical perspective.
So what would a more thoughtful approach look like? The proposal I want to put on the table is simple: Before prescribing a GLP-1 receptor agonist, measure the patient’s endogenous GLP-1 levels. This is not a radical idea. Medicine already requires baseline measurements before initiating treatments that manipulate endogenous hormone systems; we check thyroid levels before prescribing levothyroxine, testosterone and estrogen before starting hormone replacement, insulin secretion capacity before choosing between diabetes drug classes. The logic is the same here. If a patient’s body is already producing GLP-1 at normal or elevated levels, prescribing a long-acting agonist isn’t correcting a deficit; it may be amplifying a signal that’s already there. Are the potential effects of that a risk the prescriber and patient are willing to take on?
To be clear, the evidence linking GLP-1 agonists to gastroparesis and pancreatitis is suggestive, not definitive. I am not arguing that these drugs are dangerous and should be restricted. I am arguing that the question of who should receive them has not been asked with nearly enough precision, and that a baseline GLP-1 measurement is an obvious, low-cost starting point. It could be as simple as a blood-sample analysis in a lab, but we have yet to commercialize the currently available test for GLP-1s so that prescribers could truly do due diligence before prescribing them.
For researchers, this is a tractable problem. We need prospective studies that stratify GLP-1 agonist outcomes by baseline endogenous hormone levels, something no major clinical trial has done. We need validated instruments for diagnosing functional GI disorders in patients on these drugs, so that adverse events stop being absorbed into the noise of “expected side effects.” Perhaps that is simply your doctor assessing your risk for GI disorders before prescription, or maybe they take you off the drug for a few months to see if symptoms subside. Further, we need to consider the neuroscience seriously: GLP-1 receptors in the vagus nerve and brain stem are part of the same pain-modulating circuitry my own field studies in the context of visceral hypersensitivity and chronic GI pain.
For clinicians, the ask is more immediate: Pause before treating BMI as sufficient justification for a prescription. Ask whether your patient has any history of functional GI symptoms. The Food and Drug Administration’s own prescribing label already notes that semaglutide is not recommended in patients with severe gastroparesis, meaning that the agency acknowledges the risk. The question is whether that caution is traveling far enough upstream, before patients are symptomatic. I’m not sure that it is.
And for the guideline bodies: The precision-medicine framework applied to oncology and cardiology has not yet reached metabolic pharmacology. Endogenous GLP-1 testing is not currently standard, but it is not technically out of reach. The question is whether the field is willing to slow down long enough to ask what it’s missing.
GLP-1 receptor agonists are genuinely remarkable drugs. For patients with Type 2 diabetes, for people whose endogenous GLP-1 signaling is impaired, they represent a meaningful advance. I am not arguing against them. I am arguing that doctors should be able to pause and collect a little bit more information before prescribing them.
First Appeared on
Source link
