There are two women who could be sitting across from you right now, both on semaglutide, losing weight. One of them is, for the first time in years, free. The other one is restricting again and calling it success.

“I felt untouchable,” one patient said. “Like I didn’t need anything. Food was just irrelevant. And it wasn’t even hard. I wasn’t white-knuckling it. It just happened. For the first time, I wasn’t negotiating with myself every hour. I didn’t know that was possible.” She lost weight, and something larger shifted with it. She wasn’t just smaller but finally in control of the thing that had underwritten all the other competence. If her body was obedient, the rest of the performance could hold.

But what she felt was not relief. It was mastery, delivered by prescription, over the appetite she had spent years trying to outmaneuver. The medication was not treating her relationship with food so much as accelerating where it was already headed.

The Psychology of Control, and Why It Matters Here

Hilde Bruch, whose work on anorexia nervosa remains foundational, defined the disorder as a “struggle for control, for a sense of identity, competence, and effectiveness.” For many of the women who end up on GLP-1 medications, that struggle looks nothing like failure. It looks like a career and a household held together with precision, where control over food is not compensation for what isn’t working but the thing that keeps everything else working. The restriction is load-bearing.

That framework has been debated and refined over the decades, and the research is clear that not all eating disorders map onto the control model in the same way. But for many women with histories of restriction, the clinical reality is recognizable: the food behaviors are doing psychological work that has nothing to do with food.

Which is why a medication that makes not-eating effortless is not a neutral intervention for this population. It bypasses the psychology and removes the friction that was, until now, the only thing slowing it down. And friction, for someone in recovery from restriction, is often what treatment spent years building back in.

What the Medication Actually Does

GLP-1 medications change how the brain processes hunger and the rewarding properties of food at a neurological level—a real change in brain chemistry, not willpower. The quieting that people describe, the sense that food has stopped pulling at them, emerges from shifts in the hypothalamus and the mesolimbic reward system, the same circuitry involved in addiction, compulsion, and craving.

The same neurological shift reads entirely differently depending on a person’s history with food. For someone whose eating has been chaotic and compulsive, the quiet can genuinely be relief. For someone whose eating has been organized around restriction, that same quiet carries a different charge; it amplifies what was already there.

For binge eating disorder, the early research is cautiously encouraging. Multiple systematic reviews have found reductions in binge eating behaviors with semaglutide and liraglutide. The effect makes neurological sense: Binge eating is significantly a reward-circuitry phenomenon, and GLP-1 medications dampen reward-driven responses to food. People who have spent years in the exhausting cycle of losing control and trying to compensate describe what the medication offers as genuinely transformative—not just less eating, but less struggle.

For restrictive eating disorders, the same mechanism runs in the opposite direction. A 2025 case involved a woman with a childhood history of anorexia who was prescribed tirzepatide for metabolic reasons. The medication helped her blood sugar. It also appeared to reactivate dormant restrictive patterns; she began over-exercising and continued dosing in secret after her physician stopped the prescription. A similar case was documented in 2024. In both instances, clinicians noted that the GLP-1 seemed to activate a dormant eating disorder rather than treat an existing one. The appetite suppression that is clinically unremarkable in most patients became, in someone whose nervous system already knew how to use not-eating, a tool she did not need more of.

The Structural Problem

Here is the overlap nobody planned for. Most people with active or historical eating disorders are not underweight. They are in average or larger bodies, the same population GLP-1 prescriptions are aimed at. A systematic review of more than 94,000 adults seeking obesity treatment found that up to 26 percent reported moderate-to-severe binge eating. The numbers for restrictive histories are harder to capture precisely because subclinical restriction is chronic, normalized, and rarely reported. But clinicians who work in this space know the profile: the woman who has never met diagnostic criteria and has been managing her relationship with food through vigilance for most of her adult life.

Eating Disorders Essential Reads

That woman can complete a telehealth intake questionnaire, receive a prescription, and begin a medication that interacts profoundly with her psychology, without anyone asking a single question about her history with food. The intake forms used by many telehealth platforms were designed to assess metabolic eligibility, not psychological history. The drug label for semaglutide does not include a warning for restrictive eating disorders. Required screening does not exist. No structured follow-up is designed to catch what happens when appetite suppression meets a nervous system that was already organized around not-eating.

The Clinical Evidence

In April 2026, physician Amanda Banks published a piece in the New England Journal of Medicine raising formal concern about GLP-1 medications and eating disorders. Banks noted that the proportion of GLP-1 prescriptions written for people who were not diabetic, obese, or overweight had increased from 4.5 percent in 2018 to 17 percent in 2023. She called it “troubling” how easy it had become to obtain a weight loss prescription and argued for consensus recommendations that would protect people with existing or emerging eating disorders. A 2023 U.S. Food and Drug Administration analysis found misuse reports for semaglutide were four times higher than for other GLP-1 drugs, rates comparable to phentermine-topiramate, a stimulant combination drug with its own history of misuse in eating disorder populations.

What these signals point to is a gap between what the medication does in a body with an eating disorder history and what the intake process currently asks. The prescribing culture has responded to a public health crisis in obesity. It has not yet built the infrastructure to catch a different kind of vulnerability in the same population.