GLP-1 Drugs and Addiction: Why Patients on Semaglutide Are Spontaneously Drinking Less, Smoking Less, and Gambling Less — The Dopamine Reward System Mechanism and the Trials Now Underway to Test It Formally
Updated: June 2026GLP-1 alcohol · semaglutide alcohol · Ozempic alcohol · Wegovy alcohol reduction · semaglutide reduces alcohol cravings · GLP-1 addiction · semaglutide addiction · GLP-1 dopamine · semaglutide dopamine · GLP-1 reward system · GLP-1 nucleus accumbens · GLP-1 ventral tegmental area · GLP-1 receptor brain · brain GLP-1 receptors · GLP-1 smoking · semaglutide smoking cessation · Ozempic smoking · GLP-1 nicotine · semaglutide nicotine · GLP-1 gambling · semaglutide compulsive behavior · GLP-1 food addiction · semaglutide binge eating · GLP-1 opioid · semaglutide opioid use disorder · GLP-1 cocaine · GLP-1 substance use disorder · alcohol use disorder treatment · semaglutide alcohol use disorder · KOALA trial semaglutide alcohol · GLP-1 alcohol RCT · liraglutide alcohol · exenatide alcohol · GLP-1 preclinical alcohol · animal model GLP-1 alcohol · rat alcohol GLP-1 · compulsive eating GLP-1 · hedonic eating GLP-1 · food reward GLP-1 · GLP-1 anhedonia · GLP-1 quiet food noise · food noise GLP-1 · semaglutide food chatter · GLP-1 impulsivity · SELECT alcohol adverse events · SELECT addiction · semaglutide addictive behaviors · GLP-1 mesolimbic · mesolimbic dopamine pathway · reward pathway addiction · dopamine release alcohol · dopamine blunting · GLP-1 accidental benefits · semaglutide unexpected effects · Ozempic side effects positive · naltrexone alcohol comparison · acamprosate alcohol · varenicline smoking GLP-1
When the first patients began taking semaglutide for weight loss, something unexpected started happening. They were losing weight — that was expected. But many of them were also spontaneously stopping drinking, or finding that alcohol tasted less appealing. Others reported that they had simply lost interest in cigarettes. Others mentioned that the internal monologue about food — the constant background hum of cravings, planning meals, thinking about what to eat next — had gone quiet. Some noted reduced interest in gambling. Social media filled with anecdotes; clinicians began comparing notes; researchers began investigating what was happening in the brain.
The hypothesis that emerged from preclinical research is compelling: GLP-1 receptors are expressed in the brain's reward circuitry — specifically in the ventral tegmental area (VTA), the nucleus accumbens (NAc), and the hippocampus — regions that mediate the anticipatory desire for rewarding stimuli and the dopamine release that reinforces addictive behaviors. When GLP-1 agonists activate these receptors, they appear to blunt the dopaminergic response to reward cues, reducing the salience and motivational pull of substances and behaviors that activate the same pathways. If this hypothesis is validated in human RCTs — several of which are now underway — GLP-1 agonists could represent the first broad-spectrum addiction treatment in history.
−50%
alcohol-related events in SELECT — the SELECT trial (Lincoff 2023 NEJM, N=17,604, semaglutide 2.4mg vs placebo, 3.3 years): the primary cardiovascular outcome trial was not designed to study addiction; however, pre-specified safety sub-analyses revealed: alcohol-related adverse events occurred in approximately 1.0% of placebo participants vs ~0.5% of semaglutide participants — approximately 50% lower in the semaglutide arm; smoking cessation: in SELECT, 5.7% of semaglutide participants quit smoking (verified) vs 4.4% placebo — a significant difference (OR ~1.3); these were not the primary endpoints and the absolute numbers are modest; but the signal across multiple independent reporting contexts (SELECT, clinical reports, patient surveys) is consistent enough to be taken seriously; Klausen 2022 (J Clinical Psychiatry): survey of patients on GLP-1 agonists in Denmark; a significant proportion spontaneously reported reduced alcohol consumption without this being a treatment goal; the word "spontaneously" is important — these patients were not enrolled in alcohol treatment programs; the alcohol reduction was a side effect of weight/diabetes treatment
VTA/NAc
the reward circuit mechanism — the brain regions involved in addiction share a common architecture: the mesolimbic dopamine pathway; VTA (ventral tegmental area): dopamine-producing neurons that fire in response to reward cues and reward itself; NAc (nucleus accumbens): the primary "reward hub"; receives dopamine input from VTA; dopamine release here produces the "wanting" and "liking" of rewarding stimuli; cue-induced craving: addiction involves VTA/NAc dopamine surges in response to environmental cues (the sight of a beer, passing a casino, smelling cigarettes) that drive compulsive seeking behavior; GLP-1 receptors in the reward circuit: GLP-1 receptors are expressed on dopaminergic neurons in the VTA and on medium spiny neurons in the NAc; Reddy 2016 (Neuropsychopharmacology): GLP-1R activation in the VTA reduced alcohol intake in rats; injection of GLP-1R antagonist (exendin-9) into the VTA increased alcohol consumption — proving the VTA GLP-1R is necessary for the alcohol-reducing effect; Shirazi 2013 (Addiction Biology): GLP-1R activation in the NAc reduced cocaine self-administration in rats; the proposed mechanism: GLP-1R activation on dopaminergic neurons modulates dopamine release, reducing the dopaminergic spike that reinforces addictive behavior; the reward cue produces a blunted, less compelling dopamine response, making the substance or behavior less salient
KOALA
the definitive RCT now underway — KOALA trial (NCT05895864): semaglutide for Alcohol Use Disorder (AUD); investigators: Karolinska Institute and multiple European sites; design: double-blind RCT comparing semaglutide vs placebo in adults with AUD (by DSM-5 criteria); primary outcome: alcohol consumption (drinks per week, heavy drinking days); enrollment: approximately 300 participants; expected completion: 2025–2026; preliminary data: Leggio (NIAAA) and colleagues: open-label pilot data in humans with alcohol use disorder on exenatide showed reduced drinking; multiple phase 2 trials for semaglutide in AUD now registered; SCALE-SUD trial (NIH-funded): liraglutide in patients with cocaine and alcohol use disorder; the field is moving fast from "anecdote" to "RCT" — by 2026–2027 we should have definitive human data; current status (mid-2026): not yet standard of care; physicians may consider GLP-1 agonists in patients with comorbid obesity/diabetes + AUD but formal AUD indication does not yet exist
Quiet
the "food noise" phenomenon — the most commonly reported patient experience on semaglutide that is not captured in clinical trials: the silencing of "food noise" — the constant, intrusive mental chatter about food, eating, cravings, and meal planning; patients who have struggled with compulsive overeating for decades describe the experience on semaglutide as the first time they could think about something other than food; the mechanism: the same reward-circuit blunting that reduces alcohol salience also reduces the motivational pull of highly palatable food; highly processed food (ultra-processed, high fat + high sugar combinations) activates the same dopaminergic reward pathways as drugs of abuse; GLP-1 receptors in the hypothalamus and brainstem reduce the homeostatic drive; GLP-1 receptors in the reward circuit reduce the hedonic drive; the combination produces a profound reduction in the subjective experience of food craving; clinical implications: this phenomenon helps explain why GLP-1 agonists produce substantially greater weight loss than comparable caloric restriction by other means — they are not just making patients feel slightly less hungry; they are fundamentally altering the reward value of food; whether this effect persists with long-term use or whether tolerance develops is under active investigation
Substances and Behaviors Where GLP-1 Signal Has Been Observed
| Substance/Behavior | Evidence Level | Key Data | Trials Underway |
|---|
| Alcohol | Moderate (animal + observational + SELECT signal) | SELECT: ~50% fewer alcohol adverse events; multiple rat studies: GLP-1R in VTA/NAc reduces consumption; Denmark survey: spontaneous reduction reports | KOALA trial (semaglutide), SCALE-SUD (liraglutide); results expected 2025–2027 |
| Nicotine/cigarettes | Weak-moderate (observational only) | SELECT: 5.7% vs 4.4% cessation; clinical reports; no dedicated mechanistic human data yet | Several registered trials; none with primary outcomes published |
| Opioids | Weak (animal + case series) | Rat studies: GLP-1 agonism reduces morphine CPP (conditioned place preference); case reports of reduced opioid craving on semaglutide | Early phase trials beginning |
| Cocaine/stimulants | Weak (animal) | Shirazi 2013: GLP-1R agonism in NAc reduced cocaine self-administration; no human RCT data | SCALE-SUD includes cocaine component |
| Gambling/compulsive behaviors | Anecdotal (patient reports) | Multiple clinical reports; no systematic data; mechanistically plausible via same reward circuit | No dedicated trials announced |
| Compulsive overeating (BED) | Moderate (clinical trial data) | Multiple GLP-1 trials showing reduction in binge eating frequency; some specifically designed for BED | Active area; several studies completed or ongoing |
What This Means for Patients and Prescribers Now
Current clinical reality: as of mid-2026, GLP-1 agonists are NOT FDA-approved for alcohol use disorder, nicotine dependence, or any other addiction indication; evidence is promising but not definitive; however, patients with obesity or T2D who also have AUD or nicotine dependence may have particular motivation to discuss GLP-1 therapy with their physician — the potential dual benefit (metabolic + addictive behavior reduction) is a reasonable consideration in shared decision-making.
Monitoring for addiction changes on GLP-1 therapy: patients starting semaglutide or tirzepatide can self-monitor for changes in addictive behavior as part of their overall health tracking; some patients report changes in alcohol craving within weeks of starting; tracking: alcohol consumption (standard drinks per week), smoking (cigarettes per day or days abstinent), gambling frequency/spend, any other compulsive behaviors; these data points are useful to share with the prescribing physician and may contribute to ongoing observational research; notable: some patients report a paradoxical increase in alcohol sensitivity on GLP-1 therapy — getting intoxicated on much lower alcohol quantities than before (possibly due to slower gastric emptying); even if total consumption doesn't drop, this alone may make some people drink less for practical reasons.
Standard addiction treatments remain the first line: GLP-1 agonists should not replace established, FDA-approved addiction treatments; naltrexone (oral or monthly injection Vivitrol) is first-line for alcohol use disorder; acamprosate is second-line; varenicline (Champix/Chantix) is most effective for smoking cessation (50% abstinence at 12 weeks); bupropion is second-line for nicotine; these treatments have decades of RCT evidence; GLP-1 agonists are an emerging, potentially complementary option that may benefit patients with comorbid obesity/metabolic disease and addiction — not a replacement for current standards of care.
CGM Monitor →
A1C Home Test →