Ozempic, GLP-1s, and mood

The GLP-1 agonists (Ozempic and Wegovy and Mounjaro and the rest, a class of medications originally developed for type 2 diabetes and now used aggressively for weight loss) are the biggest thing in medicine right now and the second biggest thing in psychiatry that nobody planned for. They were approved for diabetes and obesity. They’re turning out to do interesting things to mood, to impulsivity, to substance use, and the field is trying to figure out what to make of it more or less in real time.

The mechanism, in plain terms: GLP-1 is a gut hormone that signals to your brain that you’re full, slows the rate at which your stomach empties, helps your body use insulin better, and, it turns out, has receptors all over the brain including in the regions involved with reward and impulse control. When you give a synthetic GLP-1 agonist (a drug that mimics the natural hormone, just stronger and longer-lasting), you’re hitting all of those systems at once. The weight loss is the headline because it’s the part the marketing leans on. The other effects are showing up in trials and case reports faster than the field knows what to do with.

What the data is showing

Reduced cravings for alcohol. There are now human studies looking at this directly, including a randomized trial of weekly exenatide in people with alcohol use disorder who were studied specifically for the alcohol question. That trial did not move the main drinking number across the whole group, but it did quiet the brain’s reward response to alcohol cues and it did cut drinking in the heavier patients, which is the kind of partial-but-real signal that gets the field paying attention rather than chalking it up to weight-loss side effects.

Reduced cravings for nicotine, opioids, cocaine, gambling. Most of this is observational data or small trials, not big randomized studies. The signal across categories is consistent enough that the mechanism is starting to make sense as something broader than appetite suppression. If a drug quiets the reward signal across multiple substances, that’s a different conversation than “it makes you less hungry.”

Reduced impulsivity in eating and probably more broadly. The phrase patients keep landing on is “food noise.” The constant low-grade thinking about what to eat next quiets down. Some patients then report similar quieting in other areas… less compulsive shopping, less mindless scrolling, less reactive snapping at family members. Whether that’s the same mechanism extending to other reward loops or whether some of it is just the patient feeling better in general because his sleep and weight and metabolism are all moving the right direction, the field doesn’t really know yet.

The depression signal

This is the messy part. Early case reports suggested some patients on GLP-1 agonists got depressed or had increased suicidal thinking. The FDA looked at it carefully. Large population studies have not confirmed a meaningful association. If anything, the data is leaning toward GLP-1 use being associated with slightly lower rates of depression, not higher, which makes sense if you think about what improving someone’s metabolic picture and pulling sixty pounds off them does for mood over a year.

That said, there are absolutely individual patients who feel low on these medications. Sometimes that’s the calorie restriction doing what severe calorie restriction does to anyone. Sometimes it’s the loss of food as a coping mechanism (which people don’t always know they were using food for until it’s suddenly not available the same way). Sometimes it’s something more direct about the drug. The honest answer is that a small subset of patients tell their prescriber their depression got worse after starting Ozempic, and that should be taken seriously and the medication should be adjusted rather than dismissed.

What the prescribing conversation actually looks like

If a patient comes in already on a GLP-1 for diabetes or weight loss and they’re describing reduced alcohol or food cravings, that’s worth noting. The prescriber might not need to add naltrexone for the drinking if Ozempic is already doing most of the work. The food piece of the depression conversation might not need as much pushing because the patient is already eating differently without it being a daily fight.

If a patient has both depression and heavy alcohol use, and they also have a real weight or metabolic indication, the GLP-1 conversation gets a bigger seat at the table than it would have five years ago. The psychiatrist isn’t usually the prescriber for the obesity indication, that’s primary care or endocrinology, but coordinating across the picture is part of doing the job right rather than running parallel disconnected treatment plans that don’t talk to each other.

If a patient is on a GLP-1 and feeling low, the first move is to look at calorie intake, nutritional adequacy, alcohol use, sleep, exercise, all the basics. A patient who’s eating 800 calories a day because nothing sounds good is going to feel like garbage regardless of mood baseline. If those are okay and the depression is real, treat the depression normally. Most of the time, we don’t need to stop the GLP-1, the underlying mood issue was independent of the medication and predates it.

The pattern this is changing

The kind of guy this is rearranging the conversation around is somebody in his late forties, drinking ten or twelve beers a night for years, BMI in the obesity range, type 2 diabetes that his primary care has been trying to manage and not really winning. He doesn’t come to a psychiatrist for the drinking, usually, he comes because his wife has had enough. Primary care has him on a GLP-1 for the diabetes. He’s lost twenty pounds without trying that hard. What he hasn’t connected, and his primary care doctor sometimes hasn’t either, is that his drinking has also dropped meaningfully, not because of any conscious cutting-down effort but because he finishes two or three beers a night and doesn’t really want the next one.

The conversation in that scenario isn’t to add naltrexone, it’s to let the GLP-1 keep doing what it’s doing and work on the depression piece separately with whatever antidepressant is going to land for him. A year and a half in, he’s down sixty pounds, drinking two beers a night instead of twelve, off two of his diabetes medications, and the marital conversation got better on its own because half of what was driving the marital conversation was a guy showing up half-drunk every night. He’s not a different person. He’s just less driven by the things that were running him before.

People on these medications describe a shift in what they call food noise. The constant low-grade thinking about what to eat quiets down. Some patients report the same quieting in other areas.

The off-label boundary

Nobody should be prescribing GLP-1 agonists for psychiatric indications alone yet. The data isn’t there for that to be defensible practice. The drugs are expensive, the side effects are real (nausea, occasional pancreatitis, gallbladder issues, GI symptoms that are the dose-limiting factor for most patients), and the long-term effects of being on them for decades are still being figured out.

If a patient has a real metabolic or weight indication and the psychiatric benefits ride along, that’s increasingly part of how the whole picture should be considered. If a patient just wants Ozempic to drink less or curb compulsive behaviors and doesn’t have a metabolic indication, the right move is to send them to a primary care doctor who can have that conversation in the proper context rather than to write it off-label from the psych side.

Where the autonomy stance lands

For patients already on these medications who are weighing whether to stay on them or come off given the psychiatric overlap, the call is theirs. I’m a provider, not a parent. The job is the honest take… here’s what the data shows about the cravings piece, here’s what the depression-risk picture actually looks like at the population level (mostly favorable, not unfavorable), here’s what your specific picture suggests is happening for you. The patient decides what to do with the information.

What’s nice to hear, because the rest of the conversation is about side effects and unknown long-term effects, is that for the patient where the GLP-1 quiets the food noise and the drinking and the compulsive whatever else at the same time, it can pull off a kind of multi-front improvement that no single psychiatric medication has ever managed. That’s a real thing for the people it’s working for. Whether it’ll still be a real thing in ten years on the same medication, nobody knows yet, which is one of several reasons not to overstate the case.

Bottom line

GLP-1 agonists are doing things to mood, impulsivity, and substance use that nobody predicted and nobody fully understands. The signals for reduced cravings across multiple categories are strong enough that the field is taking it seriously rather than chalking it up to incidental weight-loss effects. The depression-risk picture is meaningfully better than the early case reports suggested, mostly favorable at the population level. For patients already on these medications, psychiatrists should be paying attention to the whole picture rather than running a separate parallel plan. For patients without a metabolic indication, this isn’t a primary psychiatric prescription yet. Give it three or four more years and that may change. The conversation is going to be wider in a few years than it is now.

Sources

1. Wang W, Volkow ND, Berger NA, et al. Association of semaglutide with risk of suicidal ideation in a real-world cohort. Nat Med. 2024;30(1):168-176. PMID 38182782.
2. Klausen MK, Jensen ME, Møller M, et al. Exenatide once weekly for alcohol use disorder investigated in a randomized, placebo-controlled clinical trial. JCI Insight. 2022;7(19):e159863. PMID 36066977.
3. Wang W, Volkow ND, Wang Q, et al. Semaglutide and Opioid Overdose Risk in Patients With Type 2 Diabetes and Opioid Use Disorder. JAMA Netw Open. 2024;7(9):e2435247. PMID 39320894.