GLP-1 Medications and Substance Use: A New Tool for Psychiatry?

Key Points

• A landmark trial showed semaglutide significantly reduced alcohol consumption in people with alcohol use disorder
• Large registry studies found 36% reduction in AUD-related hospitalizations with semaglutide use
• GLP-1s appear to work by modulating brain reward pathways involved in addiction
• Preliminary evidence suggests potential benefits for nicotine, opioids, and other substances
• This represents a genuinely new mechanism for treating addiction that doesn’t exist in current medications
• Clinical use is off-label but increasingly common based on compelling early data

This might be the most exciting development in addiction psychiatry in decades.

I don’t say that lightly. Addiction treatment has been frustratingly limited for as long as I’ve been practicing. We have a few medications that help somewhat for alcohol use disorder. Even fewer for opioid use disorder. Almost nothing effective for stimulant use disorders. Relapse rates remain stubbornly high despite our best efforts.

Then data started emerging that GLP-1 receptor agonists, medications developed for diabetes and obesity, might reduce alcohol consumption. Initial reports were interesting but preliminary. Then came a randomized controlled trial with medium-to-large effect sizes. Then large-scale real-world data showing substantial reductions in hospitalizations. Then preclinical research clarifying plausible mechanisms.

This isn’t just interesting. It’s potentially practice-changing.

I’ve started using low-dose GLP-1 medications for carefully selected patients with alcohol use disorder, usually those who also have obesity or metabolic concerns. The results I’m seeing match what the research suggests. Not everyone responds, but for some patients, the reduction in alcohol craving and consumption is dramatic.

Let me walk you through what we know, what the mechanisms might be, and how to think about this emerging application of GLP-1 medications.

The Landmark Trial: Semaglutide for Alcohol Use Disorder

Let’s start with the study that really got everyone’s attention.

Published in JAMA Psychiatry in 2025, this was a phase 2, double-blind, randomized controlled trial. Researchers enrolled 48 non-treatment-seeking adults with DSM-5 alcohol use disorder. Half received once-weekly subcutaneous semaglutide (starting at 0.25 mg and titrating up to 1.0 mg over 9 weeks). Half received placebo.

The primary outcome was laboratory alcohol self-administration. Basically, they brought people into the lab and gave them access to alcoholic drinks to see how much they would consume in a controlled setting. Secondary measures included real-world drinking patterns tracked through daily diaries and weekly craving assessments.

The results were striking.

Semaglutide significantly reduced the amount of alcohol consumed in the laboratory setting compared to placebo. The effect size was medium to large. People on semaglutide drank less, reached lower peak blood alcohol concentrations, and reported less desire to drink.

In the real world, semaglutide led to significant reductions in drinks per drinking day, heavy drinking days, and weekly alcohol craving. The proportion of participants with zero heavy drinking days increased significantly during the higher-dose treatment period.

Here’s what particularly caught my attention. Among participants who also smoked cigarettes, semaglutide reduced cigarettes per day as well. This suggests the medication might be affecting reward pathways broadly, not just alcohol-specific mechanisms.

The effect sizes compare favorably to existing FDA-approved medications for alcohol use disorder like naltrexone and acamprosate. That’s remarkable for a medication not designed for this purpose.

Important context about the study:

The sample size was small (48 participants). This was a phase 2 trial meant to establish preliminary efficacy, not a definitive large-scale study.

The treatment duration was relatively short (9 weeks). We don’t know about longer-term effects yet.

Participants were “non-treatment-seeking,” meaning they weren’t actively trying to stop drinking or in formal addiction treatment. This is actually useful because it suggests the medication effect isn’t dependent on high motivation to change.

Most participants had overweight or obesity. The medication helped with both alcohol use and weight. Whether it works as well in people without obesity isn’t clear yet.

Adverse effects were mild and consistent with known GLP-1RA profiles. Mostly GI symptoms that were manageable. No serious adverse events or concerning interactions with alcohol.

Despite the limitations, this is strong preliminary evidence. Medium to large effect sizes in a randomized controlled trial provide much better evidence than case reports or observational data alone. It justifies larger trials and careful clinical use while we wait for more definitive data.

Real-World Evidence: The Swedish Registry Study

The randomized trial gave us controlled experimental evidence. But what about real-world effectiveness?

A massive registry study from Sweden looked at this question using nationwide data. Researchers identified over 227,000 individuals with diagnosed alcohol use disorder. Within this cohort, they compared periods when people were taking semaglutide or liraglutide (for diabetes or obesity) to periods when they weren’t.

The findings were even more impressive than the randomized trial.

Semaglutide use was associated with a 36% reduction in risk of hospitalization for alcohol use disorder (adjusted hazard ratio 0.64, 95% CI 0.50-0.83). Liraglutide showed a 28% reduction (aHR 0.72, 95% CI 0.57-0.92).

The protective effects extended beyond alcohol. Both medications reduced risk of hospitalization for substance use disorders generally, suggesting benefits for multiple substance types.

Perhaps most striking, these effects were stronger than those observed with currently approved alcohol use disorder medications in the same population. Naltrexone, disulfiram, and acamprosate all showed smaller risk reductions than the GLP-1 medications.

Context and limitations:

This was an observational study, not a randomized trial. People weren’t randomly assigned to take GLP-1s or not. There could be confounding factors we can’t fully account for.

The outcomes were hospitalizations, which are severe events. We don’t know about effects on less severe drinking or quality of life.

People were taking GLP-1s for diabetes or obesity, not specifically for alcohol use disorder. This is both a limitation (we’re not studying the intended use) and a strength (shows real-world effectiveness in patients taking the medication for other reasons).

Still, having both randomized trial data and large-scale observational data pointing in the same direction is compelling. The convergence of evidence from different study designs strengthens confidence in the findings.

How Do GLP-1s Affect Alcohol Use? The Mechanisms

Understanding mechanisms helps clarify when these medications might work and how to use them most effectively.

GLP-1 receptors exist in brain reward pathways. This is key. GLP-1 receptors aren’t just in your gut and pancreas. They’re also in brain regions central to reward, motivation, and addiction. Places like the ventral tegmental area, nucleus accumbens, and prefrontal cortex.

When GLP-1 receptor agonists activate these brain receptors, they appear to modulate reward processing. They reduce the rewarding value of alcohol (and other substances) without just making you feel sick or sedated.

Preclinical research is extensive and consistent. In animal models, GLP-1 receptor agonists reduce alcohol intake, motivation to work for alcohol, and relapse-like drinking after periods of abstinence. These effects occur across different species and experimental paradigms.

The same pattern holds for other substances. Nicotine, cocaine, methamphetamine, opioids. GLP-1 medications reduce intake and seeking behavior in animal models of addiction.

This suggests a common mechanism affecting reward system function broadly, not substance-specific effects.

The mechanism appears specific to GLP-1 receptor activation. A clever study compared GLP-1 receptor agonists to DPP-4 inhibitors (another class of diabetes medications that increase GLP-1 but work differently).

GLP-1 receptor agonists reduced alcohol consumption in both animals and humans. DPP-4 inhibitors didn’t. This suggests it’s specifically the GLP-1 receptor activation that matters, not just generally improving metabolic health or increasing GLP-1 levels through other mechanisms.

Human neuroimaging studies are emerging. A few small studies have looked at brain responses to alcohol cues in people taking GLP-1 medications. They show reduced activation in reward-related brain regions when viewing alcohol-related images. Less neural reactivity to alcohol cues might translate to reduced craving and consumption.

Genetic studies support the connection. Research on genetic variants affecting GLP-1 signaling shows associations between these variants and alcohol consumption in large population studies. This suggests the GLP-1 system naturally influences alcohol use, and medications targeting this system could modify that influence.

The mechanistic story isn’t complete yet, but it’s coherent and supported by converging evidence from multiple approaches.

Beyond Alcohol: Evidence for Other Substances

While alcohol use disorder has the strongest evidence, preliminary data suggests GLP-1 medications might help with other substance use as well.

Nicotine: The randomized trial I mentioned showed reduced cigarette smoking in participants who smoked. Animal studies consistently show GLP-1 agonists reduce nicotine intake and seeking. A few small human studies suggest reduced smoking and possibly increased quit rates, though the data is very preliminary.

If GLP-1s help with both alcohol and nicotine, that would be huge. These are the two most common substance use disorders, and we need better treatments for both.

Cannabis: Very limited human data, but some animal studies suggest GLP-1 agonists might reduce cannabis seeking. Given the increasing prevalence of cannabis use disorder as legalization expands, this deserves more research.

Opioids: Animal models show GLP-1 receptor agonists reduce opioid intake, seeking, and relapse-like behavior. Almost no human data yet. But mechanistically, if GLP-1s modulate reward pathways broadly, they might help with opioid use disorder too.

This would be incredibly valuable given the opioid epidemic and limited effective treatments.

Stimulants: Some preclinical evidence for reduced cocaine and methamphetamine seeking with GLP-1 agonists. Extremely limited human data. Stimulant use disorders are particularly difficult to treat with current options, so any new approach is worth investigating.

Behavioral addictions: A few case reports and small studies suggest possible benefits for gambling disorder and other behavioral addictions. This makes sense if GLP-1s are affecting reward system function generally. But the evidence is very preliminary.

The common thread across all these substances is that GLP-1 receptor agonists appear to reduce the rewarding value of the substance and the motivation to seek it. This is a fundamentally different mechanism than current addiction medications, which mostly work through blocking receptors (naltrexone), causing unpleasant reactions (disulfiram), or substitution (methadone, buprenorphine).

My Clinical Experience: What I’m Seeing

I’ve started using low-dose GLP-1 medications for carefully selected patients with alcohol use disorder, typically those who also have obesity or metabolic concerns. Always off-label with clear discussion about the evidence gaps. Always as part of comprehensive treatment including therapy and, when appropriate, other addiction medications.

Here’s what I’ve observed.

Some patients report dramatic reduction in alcohol craving. Not everyone, but when it works, it can be striking. “I just don’t think about drinking as much.” “The craving that used to build through the day isn’t there.” “I can have one drink and stop, which I’ve never been able to do.”

This reduction in craving seems to happen fairly quickly, often within a few weeks. It’s not complete elimination of desire to drink, but a meaningful decrease in intensity and frequency of cravings.

Drinking patterns change even without complete abstinence. Some patients achieve abstinence while on the medication. Others don’t stop drinking entirely but reduce significantly. Heavy drinking days decrease. Total weekly consumption drops. Binges become less severe or stop entirely.

For alcohol use disorder, harm reduction is valuable even when abstinence isn’t achieved. Reducing from 40 drinks per week to 10 drinks per week, or from daily blackout drinking to occasional moderate drinking, represents meaningful improvement.

The medication seems most helpful for reward-driven drinking. Patients describe less appeal of alcohol. It doesn’t taste as good or provide the same pleasure. This suggests the reward modulation mechanism is operating.

For patients whose drinking is primarily about managing withdrawal or physical dependence, GLP-1s alone aren’t sufficient. They need medical detoxification and medications that address the physical dependence aspect.

Combination with other treatments works well. I often use GLP-1s alongside naltrexone or other addiction medications. The mechanisms are different and potentially complementary. I haven’t seen concerning interactions.

All patients should be engaged in therapy. Medication alone is rarely sufficient for addiction recovery. But the reduced craving from a GLP-1 creates space for the therapeutic work to be more effective.

The weight loss is a bonus for many. Most of my patients with alcohol use disorder and obesity appreciate losing weight while addressing their drinking. The dual benefit feels encouraging and reinforces engagement with treatment.

Some patients are more motivated to take a medication for weight loss than for alcohol use disorder due to stigma around addiction. If a GLP-1 helps with both, that’s valuable regardless of which benefit they’re primarily seeking.

Low doses often sufficient. Similar to what I’ve described for other indications, I use relatively low doses. Often 0.5-1.0 mg of semaglutide weekly, sometimes less. The goal is reducing craving and reward-driven drinking, not maximally suppressing appetite.

Lower doses minimize side effects and cost while still providing benefit for craving and alcohol use.

Not everyone responds. Some patients don’t notice meaningful reduction in drinking or craving on a GLP-1. The medication just doesn’t help for them. That’s okay. We have other options, and we’re still learning about predictors of response.

Who Might Benefit? Patient Selection

Given that this is off-label use based on emerging evidence, thoughtful patient selection is important.

Good candidates in my experience:

Patients with alcohol use disorder and overweight/obesity. This is the population with the most evidence. The medication addresses both concerns.

Patients with alcohol use disorder who describe strong craving and reward-driven drinking. If the appeal and pleasure of alcohol are major drivers, GLP-1s’ effects on reward processing could be particularly useful.

Patients who haven’t responded adequately to existing alcohol use disorder medications. If naltrexone or acamprosate haven’t worked well, trying something with a different mechanism makes sense.

Patients stable enough for outpatient treatment. If someone needs medical detoxification, that’s the priority. Once they’re through withdrawal, a GLP-1 could help maintain gains.

Patients willing to engage with comprehensive treatment including therapy. The medication alone isn’t sufficient addiction treatment.

Patients with polysubstance use including alcohol. If someone uses alcohol plus nicotine, or alcohol plus cannabis, a medication that might help with multiple substances is appealing.

Patients where I’m more cautious:

Severe alcohol use disorder requiring medical detoxification. GLP-1s don’t treat alcohol withdrawal. Someone in acute withdrawal needs medical management first.

Patients with only alcohol use disorder without obesity. The evidence is strongest in people with obesity. Whether GLP-1s work as well in normal-weight individuals with AUD isn’t clear yet.

Patients who can’t afford the medication or access it consistently. Starting something that helps and then having to stop due to cost is problematic.

Patients with severe liver disease. GLP-1s are generally safe, but in someone with advanced cirrhosis, I’d want hepatology input before prescribing.

Patients with contraindications to GLP-1s generally (history of medullary thyroid cancer, pancreatitis, etc.).

Practical Considerations: Using GLP-1s for Substance Use

If we’re going to use these medications for substance use disorders, here’s how to do it thoughtfully.

This is off-label use. GLP-1 receptor agonists are not FDA-approved for any addiction treatment. Patients need to understand this is emerging off-label use based on preliminary but promising evidence.

Comprehensive addiction treatment is essential. Medication alone isn’t adequate addiction treatment. Patients need therapy, support systems, lifestyle changes, and often multiple interventions addressing different aspects of addiction.

The GLP-1 is one tool supporting recovery, not the entire treatment.

Start low and titrate slowly. I use the same low-dose approach I’ve described for other indications. Start with 0.25 mg semaglutide weekly, increase to 0.5 mg after a month, potentially increase to 1.0 mg if needed and tolerated.

Many patients notice benefit at lower doses. We don’t need maximum doses to see effects on craving and consumption.

Monitor drinking patterns carefully. I track drinks per week, heavy drinking days, craving intensity, consequences of drinking. We’re looking for meaningful reduction even if abstinence isn’t achieved.

I also monitor for increased consumption of other substances. If someone stops drinking but starts using more of something else, that’s important information.

Monitor mental health closely. Addiction and mental health are often intertwined. Addressing the addiction affects mood, and vice versa. Regular check-ins about depression, anxiety, suicidal thoughts, overall wellbeing are essential.

Coordinate with other providers. If the patient has an addiction counselor, therapist, or is in a treatment program, communication and coordination improve outcomes. We’re all working toward the same goals.

Address cost and access proactively. The same cost issues we discussed for other indications apply here. Work with insurance, consider patient assistance programs, sometimes use compounded versions when necessary with appropriate caution.

Plan for the long term. How long should someone with alcohol use disorder take a GLP-1? We don’t know yet. Some might need long-term treatment to maintain recovery. Others might be able to taper off after establishing stable sobriety and building recovery skills.

These decisions should be individualized based on the person’s response, recovery stability, and preferences.

Comparison to Existing Alcohol Use Disorder Medications

How do GLP-1s compare to medications we currently have for alcohol use disorder?

Naltrexone blocks opioid receptors and reduces the rewarding effects of alcohol. It works through a different mechanism than GLP-1s. Effect sizes in clinical trials are modest. Many patients don’t respond. Combining naltrexone with a GLP-1 is reasonable and potentially synergistic.

Acamprosate modulates glutamate neurotransmission and may reduce cravings. Again, different mechanism. Modest effect sizes. Requires three-times-daily dosing which affects adherence. Some patients benefit. Could potentially combine with GLP-1s.

Disulfiram causes unpleasant reactions if someone drinks while taking it. This is a deterrent approach rather than reducing craving or reward. Some patients find it helpful, but adherence is challenging. Doesn’t directly compete with or complement GLP-1s since the mechanisms are so different.

Topiramate (off-label) affects multiple neurotransmitter systems and has shown benefit for reducing heavy drinking in some studies. Cognitive side effects limit its use. Could potentially combine with GLP-1s though the evidence for combinations doesn’t exist yet.

The registry data suggesting GLP-1s show stronger effects than currently approved medications is intriguing but needs confirmation in head-to-head trials. What we can say is that GLP-1s offer a new mechanism of action that doesn’t exist in our current medication toolkit.

For patients who haven’t responded to existing medications, having a fundamentally different option is valuable.

The Broader Implications: Rethinking Addiction Treatment

If GLP-1 medications prove effective for multiple substance use disorders, it changes how we think about addiction treatment.

A common mechanism across substances. The fact that GLP-1s affect multiple substances through reward pathway modulation suggests shared neurobiological mechanisms in addiction. This isn’t new knowledge, but having a medication that targets this common pathway is new.

Integration of addiction and metabolic treatment. Many people with substance use disorders have comorbid obesity. Many people with obesity have comorbid addiction (sometimes food-related, sometimes substance-related). GLP-1s allow us to address both simultaneously within integrated treatment.

This fits perfectly with the metabolic psychiatry framework I practice. Brain health, metabolic health, and addiction are all interconnected. Treating them together makes sense.

A tool that doesn’t require daily dosing. Most addiction medications require daily or multiple-times-daily dosing. GLP-1s are once weekly. For patients who struggle with adherence, this is a significant advantage.

Potential for prevention. If GLP-1s reduce risk of substance use problems in people taking them for other reasons (as the registry data suggests), could they have a role in prevention for high-risk individuals? This is speculative, but worth considering as evidence accumulates.

Implications for understanding reward and motivation. The GLP-1 system’s role in reward processing isn’t limited to food and substances. It appears to be part of a broader motivation and reward regulation system. Understanding this has implications beyond addiction treatment for understanding human motivation generally.

What Research We Need

The field needs several things to move GLP-1s from promising preliminary evidence to standard addiction treatment.

Large, phase 3 randomized controlled trials specifically in alcohol use disorder with adequate sample sizes, treatment-seeking populations, and clinically meaningful endpoints (abstinence rates, time to relapse, quality of life).

Longer-duration studies showing sustained benefit and safety over months to years, not just weeks.

Studies in people without obesity to understand whether the benefits extend beyond individuals with comorbid obesity.

Trials for other substance use disorders including nicotine, opioids, stimulants, and cannabis use disorders.

Head-to-head comparisons with existing addiction medications and studies of combination approaches.

Research identifying predictors of response to understand who benefits most from GLP-1 treatment for addiction.

Studies examining different dosing strategies to identify optimal doses for addiction treatment, which might differ from diabetes or weight loss dosing.

Several such trials are likely in progress now. Results will emerge over the next few years. The evidence base will strengthen, and we’ll have clearer answers about where GLP-1s fit in addiction treatment.

But while we’re waiting for that research, people are suffering from addiction now. The preliminary evidence is strong enough that many addiction specialists, including me, think careful off-label use is justified for selected patients.

My Take: Genuine Excitement Tempered by Realism

I’m genuinely excited about this, which is rare for me. I tend to be fairly skeptical about new treatments until evidence accumulates.

But the combination of a randomized trial showing medium-to-large effect sizes, massive registry data showing real-world benefit, extensive preclinical research clarifying mechanisms, and my own clinical experience seeing meaningful reductions in alcohol use in patients who’ve struggled for years… it’s compelling.

This could represent a genuinely new approach to addiction treatment. Not a cure, not something that works for everyone, but a novel mechanism that helps some people who haven’t been helped by existing options.

The excitement needs to be tempered by realism. The evidence is preliminary. We need larger, longer trials. We don’t know who responds best or optimal dosing. The medications are expensive and not accessible to everyone. They need to be part of comprehensive treatment, not used alone.

But for the first time in a long time, I feel like we might have something new and valuable to offer patients with substance use disorders. That’s worth being excited about.

If you’re struggling with alcohol use disorder or other addiction, particularly if you also deal with obesity, talk with an addiction specialist about whether a GLP-1 medication might be appropriate as part of your treatment plan. Make sure comprehensive addiction treatment including therapy is part of the approach. Be realistic about the preliminary nature of the evidence but also open to trying something new if existing treatments haven’t worked well.

This is an area where the field is moving quickly. The article I’m writing today will probably need updating within a year or two as new evidence emerges. Stay informed, work with knowledgeable providers, and remain hopeful that better treatments are coming.

In our next article, we’ll bring together everything we’ve discussed throughout this series and talk about creating a personalized, integrative approach to metabolic and mental health. How do you put all these pieces together into a comprehensive treatment plan that works for you?

References

1. Hendershot CS, Bremmer MP, Paladino MB, et al. Once-Weekly Semaglutide in Adults With Alcohol Use Disorder: A Randomized Clinical Trial. JAMA Psychiatry. 2025;82(4):395-405. doi:10.1001/jamapsychiatry.2024.4789.
   
2. Lähteenvuo M, Tiihonen J, Solismaa A, et al. Repurposing Semaglutide and Liraglutide for Alcohol Use Disorder. JAMA Psychiatry. 2025;82(1):94-98. doi:10.1001/jamapsychiatry.2024.3599.
   
3. Gonzalez-Rellan MJ, Drucker DJ. New Molecules and Indications for GLP-1 Medicines. JAMA. 2025;:2838996. doi:10.1001/jama.2025.14392.
   
4. Moiz A, Filion KB, Tsoukas MA, et al. The Expanding Role of GLP-1 Receptor Agonists: A Narrative Review of Current Evidence and Future Directions. EClinicalMedicine. 2025;86:103363. doi:10.1016/j.eclinm.2025.103363.
   
5. Guo J, Hayes MR, Leggio L, Oru E, Rinaman L. GLP-1 Receptor Agonists and Research to Treat Overeating and Substance Use Disorders. The Journal of Neuroscience. 2025;45(46):e1375252025. doi:10.1523/JNEUROSCI.1375-25.2025.
   
6. Bruns Vi N, Tressler EH, Vendruscolo LF, Leggio L, Farokhnia M. IUPHAR Review – Glucagon-Like Peptide-1 (GLP-1) and Substance Use Disorders: An Emerging Pharmacotherapeutic Target. Pharmacological Research. 2024;207:107312. doi:10.1016/j.phrs.2024.107312.
   
7. Marquez-Meneses JD, Olaya-Bonilla SA, Barrera-Carreño S, et al. GLP-1 Analogues in the Neurobiology of Addiction: Translational Insights and Therapeutic Perspectives. International Journal of Molecular Sciences. 2025;26(11):5338. doi:10.3390/ijms26115338.
   
8. Jerlhag E. GLP-1 Receptor Agonists: Promising Therapeutic Targets for Alcohol Use Disorder. Endocrinology. 2025;166(4):bqaf028. doi:10.1210/endocr/bqaf028.
   
9. Scheen AJ. Glucagon-Like Peptide-1 Receptor Agonists and Alcohol Use Disorders: An Emerging Unexpected Beneficial Effect. Diabetes, Obesity & Metabolism. 2025;27(8):4083-4091. doi:10.1111/dom.16453.
   
10. Petrie GN, Mayo LM. GLP-1 Receptor Agonists for the Treatment of Alcohol Use Disorder. The Journal of Clinical Investigation. 2025;135(9):e192414. doi:10.1172/JCI192414.
    
11. Farokhnia M, Tazare J, Pince CL, et al. Glucagon-Like Peptide-1 Receptor Agonists, but Not Dipeptidyl Peptidase-4 Inhibitors, Reduce Alcohol Intake. The Journal of Clinical Investigation. 2025;135(9):e188314. doi:10.1172/JCI188314.
    
12. Martinelli S, Mazzotta A, Longaroni M, Petrucciani N. Potential Role of Glucagon-Like Peptide-1 (GLP-1) Receptor Agonists in Substance Use Disorder: A Systematic Review of Randomized Trials. Drug and Alcohol Dependence. 2024;264:112424. doi:10.1016/j.drugalcdep.2024.112424.