GLP-1 Drugs and Eating Disorders: The Risk
GLP-1 Psychological Effects
57% reduction in binge eating scores
Semaglutide reduced Binge Eating Scale scores more than standard anti-binge medications in a retrospective cohort, but the same appetite suppression that treats binge eating may trigger or worsen restrictive eating disorders in vulnerable populations.
Richards et al., Obesity Pillars, 2023
Richards et al., Obesity Pillars, 2023
View as imageGLP-1 receptor agonists suppress appetite through central and peripheral pathways that reduce hunger, slow gastric emptying, and alter reward signaling in the brain. For the estimated 2.8 million Americans with binge eating disorder (BED), that mechanism looks like a potential treatment. For the millions with restrictive eating disorders or subclinical disordered eating patterns, the same mechanism looks like a catalyst for relapse. The clinical evidence supports both concerns, and neither side of this equation has been adequately studied. As the broader psychological effects of GLP-1 weight loss continue to emerge, the intersection with eating disorders is one of the most urgent gaps in the literature.
Bartel et al. (2024) published the first comprehensive clinical perspective on GLP-1 receptor agonist use in eating disorder populations, concluding that it is possible GLP-1 drugs "could exacerbate, or contribute to the development of, eating disorder pathology and negatively impact eating disorder treatment."[1] That assessment, from a team that included eating disorder specialists, captures the state of the evidence: preliminary, concerning, and insufficient for firm conclusions.
Key Takeaways
- Semaglutide produced greater reductions in Binge Eating Scale scores than lisdexamfetamine or topiramate in a retrospective cohort of BED patients (Richards et al., Obesity Pillars, 2023)
- A systematic review of GLP-1 drugs for binge eating found that liraglutide and dulaglutide reduced binge frequency, but all studies had small sample sizes and short follow-up periods (Aoun et al., J Clin Transl Endocrinol, 2024)
- No randomized controlled trial has tested any GLP-1 drug in patients with anorexia nervosa, bulimia nervosa, or avoidant/restrictive food intake disorder
- GLP-1 drugs alter reward-related brain areas in the insula, amygdala, and orbitofrontal cortex, the same regions implicated in both binge eating and restrictive eating pathology (van Bloemendaal et al., Diabetes, 2014)
- Oral semaglutide reduced daily energy intake by approximately 24% (about 530 kcal/day) over 20 weeks, a degree of suppression that could push vulnerable individuals below safe caloric thresholds (Gibbons et al., Diabetes Obes Metab, 2021)
- Most GLP-1 prescriptions occur without eating disorder screening, creating a blind spot in a population where obesity and eating disorders overlap by 30-40%
The Binge Eating Evidence
Binge eating disorder is the most common eating disorder in the United States, affecting roughly 2.8% of adults. Only one medication, lisdexamfetamine (Vyvanse), carries FDA approval for BED, and it is a Schedule II controlled substance with abuse potential. The pharmacological gap has pushed clinicians toward off-label use of GLP-1 drugs.
Richards et al. (2023) conducted the first direct comparison. In a retrospective cohort of patients with diagnosed BED, those prescribed semaglutide showed greater reductions in Binge Eating Scale (BES) scores compared to patients on lisdexamfetamine, topiramate, or combination therapy. Adding lisdexamfetamine or topiramate to semaglutide did not improve BES outcomes beyond semaglutide alone.[2] The study was retrospective and open-label, which limits causal claims, but the effect size was notable in a population where treatment options are limited.
Aoun et al. (2024) systematically reviewed GLP-1 receptor agonists as pharmacotherapy for binge eating across BED and bulimia nervosa populations. The review found that liraglutide and dulaglutide reduced binge frequency and associated comorbidities with a favorable psychiatric side effect profile. However, every included study had design limitations: small samples, short follow-up periods, and no assessment of whether GLP-1 drugs affected other eating disorder symptoms beyond binge frequency.[3]
Folling et al. (2025) added qualitative depth with interviews of patients taking liraglutide for obesity alongside diagnosed BED. Participants described reduced preoccupation with food and fewer binge episodes. But several participants also reported a new and unfamiliar disconnect from hunger cues that they found psychologically distressing, raising the question of whether eliminating hunger signals is the same as resolving binge eating pathology.[4]
The evidence for GLP-1 drugs treating binge eating is real but preliminary. No phase III trial has tested any GLP-1 drug with BED as a primary endpoint. The retrospective and case-series data that exists cannot distinguish whether GLP-1 drugs resolve the underlying psychological drivers of binge eating or simply suppress the appetite that enables it.
The Restrictive Eating Disorder Risk
The evidence on the other side of this equation is thinner and more alarming. There are no clinical trials and no prospective cohort studies. What exists are case reports, clinical observations, and mechanistic reasoning.
Bartel et al. (2024) identified several pathways through which GLP-1 drugs could worsen restrictive eating disorders. The appetite suppression that reduces food intake by 16-39% does not distinguish between someone with a healthy relationship with food and someone in fragile recovery from anorexia nervosa. In a patient with active or remitted restrictive patterns, the drug-induced absence of hunger provides what Bartel's team described as a pharmacological reinforcement of restriction, making it easier to eat dangerously little while experiencing less physical distress.[1]
Published case reports describe patients with childhood histories of anorexia nervosa who relapsed after being prescribed GLP-1 drugs for weight management or type 2 diabetes. In one case, a patient prescribed tirzepatide for metabolic indications returned to restrictive patterns, over-exercise, and covert continuation of the medication after her physician discontinued it. In another, a patient required hospitalization for rapid weight loss after starting semaglutide, despite having been in eating disorder recovery for over a decade.
These are individual cases, and they cannot establish a general risk rate. But they illustrate a mechanism that is biologically coherent: GLP-1 drugs remove the physiological barriers to restriction (hunger, food preoccupation) while leaving the psychological drivers of restriction (body dissatisfaction, need for control, fear of weight gain) intact. For individuals with a latent vulnerability to restrictive eating, the drug may lower the activation energy for relapse. The connection to how GLP-1 drugs quiet "food noise" is direct. What is experienced as relief from intrusive food thoughts by one patient may function as dangerous suppression of protective hunger signals in another.
How GLP-1 Drugs Alter the Brain's Relationship With Food
Understanding the eating disorder risk requires understanding how these drugs change brain function, not just appetite.
Van Bloemendaal et al. (2014) used functional neuroimaging to show that GLP-1 receptor activation modulates activity in the insula, amygdala, putamen, and orbitofrontal cortex in response to food stimuli. These are reward and salience processing regions, not simple hunger circuits.[5] The implication is that GLP-1 drugs do not merely reduce the feeling of hunger. They alter the motivational significance of food at a neural level.
Knakker et al. (2024) demonstrated this more precisely with the GLP-1 receptor agonist exenatide. In a rodent model, exenatide uncoupled food intake from hedonic and anticipatory regulation, meaning animals still consumed food when hungry but showed reduced anticipatory excitement and reward-driven eating behavior.[6] This dissociation between metabolic need and hedonic drive is exactly what makes GLP-1 drugs potentially beneficial for binge eating and potentially dangerous for restrictive eating.
Chang et al. (2025) found that chronic semaglutide treatment enhanced the incentive motivational value of smaller food rewards in animal models, suggesting the drug recalibrates reward sensitivity rather than simply blunting it.[7] This recalibration could explain why some patients report feeling satisfied with much less food, a subjective state that is adaptive in obesity treatment but hazardous when the baseline is already insufficient intake.
Arillotta et al. (2024) expanded the frame by reviewing GLP-1 receptor agonists' potential impact on substance use, compulsive behavior, and eating disorders collectively. The review noted that the same GLP-1 receptor populations in the nucleus accumbens and ventral tegmental area that mediate reduced alcohol and substance craving also modulate food reward processing, suggesting a shared neurobiological substrate for GLP-1 effects on compulsive behaviors including binge eating.[8] The connection to GLP-1's effects on alcohol use shares this same neurobiological pathway.
The Screening Gap
The clinical problem is not just pharmacological. It is structural. Most GLP-1 drugs are prescribed in primary care, endocrinology, or weight management settings. Eating disorder screening is not standard in any of these contexts.
Obesity and eating disorders overlap substantially. Studies estimate that 30-40% of individuals seeking weight loss treatment meet criteria for binge eating disorder, and a significant minority have histories of restrictive eating disorders that may not be documented in their medical records. The result is a prescription pattern where GLP-1 drugs reach patients with active or latent eating disorders without any systematic assessment of that risk.
Gibbons et al. (2021) measured the appetite suppression effect of oral semaglutide: a 24% reduction in total daily energy intake, approximately 530 kcal/day, over 20 weeks.[9] Gabe et al. (2024) confirmed that this suppression was consistent across meal types and times of day, suggesting central rather than purely peripheral mechanisms.[10] For someone eating 2,000 calories per day without an eating disorder, a 530-calorie reduction is a tolerable and often desirable deficit. For someone eating 1,400 calories while in partial recovery from anorexia, the same proportional reduction drops intake to dangerous levels. If caloric intake on GLP-1 therapy is already a concern in the general population, it becomes acute in patients with restrictive histories.
Jalleh et al. (2024) reviewed the clinical consequences of delayed gastric emptying with GLP-1 receptor agonists and tirzepatide, documenting that the slowed gut transit contributes to early satiety and reduced meal size.[11] For patients with eating disorders, delayed gastric emptying is already a common complication of malnutrition. Adding a drug that further slows transit creates a compounding effect that is not captured in clinical trials that excluded patients with eating disorder histories.
What the Evidence Does Not Yet Answer
The published literature leaves several clinically important questions unresolved:
Does semaglutide cause eating disorders, or does it unmask latent ones? The case reports involve patients with prior eating disorder histories. Whether GLP-1 drugs can trigger de novo eating disorder pathology in individuals with no history or vulnerability is unknown.
What is the actual incidence of eating disorder exacerbation among GLP-1 users? Some clinical commentaries cite estimates as high as 35% of new eating disorder diagnoses occurring in patients on GLP-1 agonists, but this figure has not been validated in a controlled epidemiological study.
Do the BED benefits persist after discontinuation? The appetite suppression is drug-dependent. If binge eating returns when semaglutide is stopped, the drug is managing a symptom rather than treating the underlying pathology. No follow-up study has tracked BED outcomes after GLP-1 drug cessation.
Are adolescents at higher risk? GLP-1 drugs are increasingly prescribed to adolescents with obesity, a population with elevated rates of eating disorder onset. Eating disorders most commonly emerge between ages 12 and 25, and adolescent girls prescribed GLP-1 drugs for weight management fall squarely within this window. The Wilding et al. (2021) STEP 1 trial that established semaglutide's weight loss efficacy excluded patients with eating disorders, as have subsequent pediatric trials.[12] The evidence base for this drug class was built on populations where the eating disorder risk was systematically removed.
Does the route of administration matter? Oral semaglutide and injectable semaglutide achieve different pharmacokinetic profiles, and their effects on appetite suppression and gastric emptying differ in magnitude and timing. Whether these differences translate to different eating disorder risk profiles has not been studied.
What role does weight stigma play? Patients prescribed GLP-1 drugs often carry long histories of weight-related shame and failed diet attempts. The rapid weight loss from semaglutide or tirzepatide can generate intense positive social reinforcement, praise from family, compliments from colleagues, changed treatment by strangers, that powerfully rewards continued weight loss regardless of whether it has become medically excessive. This social dimension of eating disorder risk is entirely absent from the pharmacological literature.
These are not theoretical concerns. They are gaps in the safety data for a drug class prescribed to tens of millions of people.
The Dual Nature of Appetite Suppression
The central tension in this evidence is that the same pharmacological effect, suppression of appetite through GLP-1 receptor activation in the brain, can be therapeutic or harmful depending on the patient's psychological relationship with food.
For a patient with binge eating disorder, reduced food preoccupation and decreased hedonic eating may provide the neurobiological breathing room to develop healthier eating patterns. The Richards et al. data showing semaglutide outperforming lisdexamfetamine for BED is clinically meaningful, even with its limitations.[2]
For a patient with a restrictive eating disorder or history, the same reduced food preoccupation removes one of the body's primary defense mechanisms against starvation. Hunger is not just a sensation. It is a survival signal. Pharmacologically suppressing it in someone who already struggles to eat adequately is a categorically different intervention than suppressing it in someone who eats excessively.
This distinction is clear in concept. In clinical practice, where eating disorders are under-diagnosed, where patients do not always disclose histories of disordered eating, and where weight loss is treated as an uncomplicated good, the distinction collapses. The same prescription reaches both populations. Whether GLP-1 drugs affect suicidal ideation or depression adds further layers to the psychological risk profile that prescribers must consider.
The Bottom Line
GLP-1 receptor agonists show genuine promise for binge eating disorder, with semaglutide outperforming existing pharmacotherapy in early data. The same drugs carry a biologically coherent and clinically documented risk of worsening restrictive eating disorders, though the incidence and magnitude of that risk remain unquantified. The evidence base was built on clinical trials that systematically excluded patients with eating disorders, creating a safety blind spot that is only now becoming visible as prescriptions reach tens of millions of people, many of whom have unscreened eating disorder histories or vulnerabilities.