GLP-1 Medications and Binge Eating Disorder: The Emerging Evidence

The clinical picture of binge eating disorder has been recognised since the 1950s, formally codified in the DSM-5 in 2013, and consistently underdiagnosed since. Lifetime prevalence in the United States runs to roughly 2.8% — higher than anorexia nervosa and bulimia nervosa combined — and the condition is associated with substantial cardiometabolic morbidity beyond what BMI alone would predict. The first-line treatments, cognitive behavioural therapy and lisdexamfetamine, work for many patients but not all.

Over the last decade, a set of small but increasingly rigorous studies has examined whether GLP-1 receptor agonists — designed for glycaemic control and now licensed for obesity — also reduce binge eating episodes in patients who meet diagnostic criteria for BED. The data is preliminary. The mechanistic rationale is strong. The clinical implications are beginning to interest a wider audience than obesity medicine specialists.

What binge eating disorder actually is

James Hudson, a psychiatric epidemiologist at McLean Hospital and Harvard Medical School, led the national prevalence work that established BED as the most common eating disorder in adults. The diagnostic criteria require recurrent episodes of eating an objectively large amount of food in a discrete period, accompanied by a sense of loss of control. The episodes are not followed by the compensatory purging that defines bulimia nervosa. They occur at least once a week for three months. They are accompanied by marked distress.

The behavioural phenotype overlaps with addiction in several ways. Patients describe a build-up of urge, a narrowing of attention onto food, a loss of conscious agency during the episode, and a post-episode period of regret and shame. Imaging research consistently shows altered reward-circuit responses to food cues in patients with BED — patterns that resemble, in their basic geometry, the responses seen in substance use disorders.

BED occurs across BMI categories. Roughly half of patients with BED have obesity, but a meaningful minority have normal weight, and the disorder is not synonymous with obesity. The behavioural pattern, not the BMI, defines the condition.

Why GLP-1 medications became interesting for BED

The mechanistic rationale for testing GLP-1 receptor agonists in BED rests on two converging observations. First, GLP-1 receptors are densely expressed in the brain's reward circuitry — the ventral tegmental area, the nucleus accumbens, portions of the prefrontal cortex — in addition to the hypothalamic and brainstem regions that govern homeostatic appetite. Activation in those reward regions attenuates the dopaminergic response to food cues, a mechanism we explore in more depth in our piece on how GLP-1 medications quiet food cravings.

Second, patients on GLP-1 medications for obesity have reported, in unsolicited fashion, marked reductions in the urge-driven, loss-of-control eating that characterises BED — even when they were not formally diagnosed. This pattern surfaced in clinical practice and in patient communities before it surfaced in the trial literature, which is often how off-label signals first appear. Our editorial on emotional eating reductions on GLP-1 covers some of that experiential reporting.

The pharmacology and the patient signal pointed in the same direction. The question was whether controlled trials would replicate it.

What the trials so far have shown

The first dedicated trial of a GLP-1 medication for binge eating appeared in 2015. Robert and colleagues published a pilot study of exenatide — an earlier-generation GLP-1 agonist — in patients with BED and obesity. The study was small and not designed for definitive efficacy claims, but it demonstrated reductions in binge frequency, in eating pathology scores, and in self-reported loss of control. The signal was modest but consistent.

Liraglutide, the next-generation GLP-1 agonist licensed for obesity at 3.0mg daily, generated more data through the late 2010s. Da Porto and colleagues in 2020 reported on liraglutide in patients with obesity and binge eating features, observing reductions in binge episodes that paralleled weight loss but appeared partially independent of it — the binge frequency reductions emerged earlier and at lower doses than would be expected from weight loss alone.

Semaglutide, at the doses now licensed for obesity, has begun to generate dedicated BED literature. A 2024 review by Goldschmidt and colleagues synthesised the available semaglutide BED evidence — including secondary analyses from obesity trials in which BED features were measured — and characterised the effect on binge frequency as clinically meaningful in the patients studied. The review was careful to flag the methodological limitations: most studies are still small, BED-specific endpoints are often secondary, and head-to-head comparisons with established BED treatments (CBT, lisdexamfetamine) have not yet been published.

Susan McElroy at the Lindner Center of HOPE and the University of Cincinnati, who led much of the clinical trial work establishing lisdexamfetamine as an FDA-approved treatment for BED in 2015, has been measured in her commentary on the emerging GLP-1 BED literature. The mechanism is plausible, she has noted, the early signals are encouraging, and the field needs adequately powered randomised trials with BED-specific primary endpoints before practice should change.

Mechanism: why this might genuinely work

The mechanistic case rests on a particular interpretation of what binge eating is. The behavioural pattern is, in this framing, the visible output of a reward-circuit dysregulation in which the dopaminergic response to highly palatable food cues is amplified, the homeostatic satiety signals are attenuated, and the cognitive capacity to interrupt the eating sequence is overwhelmed. The episodes are not failures of motivation; they are episodes in which the reward signal becomes large enough that the behaviour follows it.

GLP-1 receptor activation, in this account, attenuates the reward signal at its source. The dopaminergic response to food cues drops. The urge that would otherwise build into a binge does not build to the same intensity. The cognitive capacity to interrupt the sequence is preserved because the demand on that capacity is reduced.

This is a mechanistically clean account, and the imaging and animal-model literature broadly support it. Whether it translates into the durability and effect size needed for BED treatment specifically — as opposed to obesity treatment with collateral benefits on binge frequency — remains the open question.

The important caveats

No GLP-1 medication is FDA-approved for BED. Prescribing for that indication is off-label. Insurance coverage is unlikely. Clinical guidelines for BED — those issued by the American Psychiatric Association, the World Federation of Societies of Biological Psychiatry — do not currently include GLP-1 medications among recommended treatments. Cognitive behavioural therapy, particularly the BED-adapted protocols developed by Christopher Fairburn at Oxford and Terence Wilson, remains the first-line psychological intervention. Lisdexamfetamine is the only FDA-approved pharmacotherapy.

For patients with BED at low or normal weight, the situation is more complex. The reduction in caloric intake that accompanies GLP-1 treatment is therapeutically welcome in patients with obesity, but in patients with normal-range BMI and BED, the weight loss may be unwanted or clinically inappropriate. The medications are calibrated for obesity, not for binge-frequency reduction without weight effect.

The history of eating-disorder treatment is also a reminder of why caution matters. The field has periodically embraced pharmacological interventions — fenfluramine, sibutramine — that subsequently proved unsafe or had unfavourable long-term profiles. Philip Mehler at the ACUTE Center for Eating Disorders has consistently argued for conservative incorporation of new pharmacotherapy into eating-disorder practice, particularly given the medical fragility of the patient population and the difficulty of distinguishing therapeutic from disordered eating-pattern change.

How clinicians are using the evidence now

In practice, the most defensible current position is roughly this: for patients with co-occurring BED and obesity, particularly those who have not responded to CBT and lisdexamfetamine, a trial of GLP-1 medication can be reasonable in consultation with a clinician experienced in both obesity medicine and eating-disorder treatment. For patients with BED at normal weight, the evidence does not yet support routine use. For patients in active eating-disorder treatment, integration with the existing therapeutic plan — rather than substitution for it — is the prevailing recommendation.

The mechanistic story is part of why some patients on GLP-1 medications describe a broader quieting of urge-driven eating patterns that they would not necessarily have labelled as BED. Our pieces on recognising emotional eating patterns and the neuroscience of food cravings cover that adjacent terrain in more depth.

Key takeaways

• Binge eating disorder affects roughly 2.8% of US adults over a lifetime — more than anorexia and bulimia combined — and remains substantially underdiagnosed.
• The mechanistic rationale for GLP-1 medications in BED is strong: receptor activation in the mesolimbic dopamine system attenuates the reward response to food cues that drives binge episodes.
• The trial evidence — exenatide (Robert 2015), liraglutide (Da Porto 2020), early semaglutide data — shows reductions in binge frequency that appear partially independent of weight loss, but studies remain small and BED-specific endpoints are often secondary.
• No GLP-1 medication is FDA-approved for BED; prescribing for the indication is off-label, and guidelines still place CBT and lisdexamfetamine as first-line.
• For patients with co-occurring BED and obesity who have not responded to established treatments, GLP-1 trials with experienced clinicians can be defensible; for normal-weight BED, the evidence does not yet support routine use.
• Integration with existing eating-disorder treatment — rather than substitution — is the prevailing clinical recommendation while the evidence base matures.