Mood Swings and Emotional Blunting on a GLP-1

The short version

Two very different complaints get filed under the same heading, and they have different answers.

Irritability and a short fuse are the ones the labels actually speak to — and what they say is unexpected. Feeling flat, muted, or like nothing lands the way it used to is the one nobody has studied. Not "the evidence is mixed": there is essentially none, and that is worth knowing before you read anyone's confident explanation of it.

Underneath both sits a safety question that changed in February 2026. Educational only, not medical advice. If you are having thoughts of harming yourself, that is not a wait-and-see situation — contact your clinician or emergency services now, whatever any study below says.

What the labels list, which is not what you would guess

Across the current prescribing information for Wegovy, Ozempic, Zepbound and Mounjaro, the words "irritability" and "mood changes" each appear exactly once, and in every case in the same place: the Medication Guide's list of the signs and symptoms of low blood sugar^1,2.

Not as psychiatric adverse reactions. As hypoglycemia symptoms, sitting alongside shakiness, sweating, headache, blurred vision and feeling jittery¹.

That is the single most practically useful thing on this page, because it hands you a question you can answer yourself: does the irritability track your eating, or your injection? On a drug engineered to remove your interest in food, skipping meals is effortless and under-eating is the default. A short fuse at 4pm on a day you had coffee and half a yogurt is a different problem from a short fuse that arrives every Tuesday after your shot.

The label change nobody wrote up

Both the Wegovy and Zepbound labels carried a Warnings and Precautions subsection titled "Suicidal Behavior and Ideation". The current versions of both record the same entry in their Recent Major Changes table: removed, 02/2026^1,2. The diabetes labels, Ozempic and Mounjaro, never carried it — the word "suicidal" does not appear in either³.

The removal followed the evidence rather than leading it. A systematic review and meta-analysis in JAMA Psychiatry pooled the 27 randomized placebo-controlled trials that systematically recorded these events — 32,354 people on a GLP-1 receptor agonist against 27,042 on placebo, over roughly 70,000 and 64,000 person-years. Event incidence was very low in both arms (0.047 versus 0.042 per 100 person-years) with no statistically significant difference between them (rate ratio 0.76, 95% CI 0.48–1.21)⁴.

Observational data points the same way and further. A retrospective cohort of 240,618 patients with overweight or obesity found semaglutide associated with a lower risk of recurrent suicidal ideation than non-GLP-1 anti-obesity medications (hazard ratio 0.44, 95% CI 0.32–0.60), replicated in a much larger type 2 diabetes population⁵.

Treat the protective-looking direction with caution rather than enthusiasm. A 2026 meta-analysis of neuropsychiatric outcomes found a similar association in observational studies but with extreme statistical heterogeneity, and no association at all when semaglutide was analyzed alone⁶. The defensible reading is the null one: these drugs do not appear to raise the risk. They are not an antidepressant.

The flat feeling, and why nobody can tell you about it

Search the published literature for emotional blunting alongside GLP-1s, semaglutide or tirzepatide and you get nothing. Not a small or conflicting body of work — no records.

The structural reason is visible in a 2025 systematic review of GLP-1 receptor agonists and psychiatric symptoms: of 26 studies, only five treated psychiatric outcomes as primary endpoints, and those were mostly substance-use trials. The other 21 measured mental health as a secondary outcome, using instruments designed for depression and anxiety⁷. Nothing in that toolkit measures "food used to be the good part of my day and now it isn't."

So what follows is reasoning, offered as reasoning:

• Reward, not mood. These drugs are meant to reduce how compelling food is. For someone whose main daily pleasure was eating, that can feel like a flatness that has nothing to do with depression.
• Habits leave with the appetite. The glass of wine, the takeaway, the biscuit at 3pm — several small rituals often disappear in the same month.
• Fatigue mimics it convincingly. Tiredness reads as flatness from the inside; that one is covered in GLP-1s, fatigue and energy.
• Under-eating flattens everyone. See the previous section.

One line does not belong in the reasoning column. Persistent low mood, loss of interest in things beyond food, or any thoughts of self-harm are not a puzzle to solve from an article. They are a call to a clinician, and the meta-analyses above are not a reason to delay one.

The sorting question

What to actually do

1. 1.Eat on a schedule, not on appetite. Three anchored eating occasions with protein at each, whether or not you want them. This is the intervention with a label behind it, and it costs nothing.
2. 2.Log the week, not the day. Note mood against injection day and against meals for two weeks. Patterns that are invisible day to day are obvious on a page, and it turns "I feel off" into something a clinician can work with.
3. 3.Look at what left with the appetite. If the flatness arrived alongside quitting alcohol, note it — the interaction has its own quirks, covered in GLP-1s and alcohol.
4. 4.Raise it at a review rather than saving it up. The relevant levers — pace of titration, dose, whether this drug suits you — are all prescriber decisions, and none of them work retrospectively. Hold, drop, or push through covers how to ask for the first two, and the best GLP-1 for busy schedules board scores programs on pausing and canceling.
5. 5.Do not stop on your own because of low mood. Stopping abruptly to see what happens removes the only variable you were tracking, costs you the titration progress, and is a decision worth making with someone.