In 2023 the Icelandic medicines agency passed on a small number of reports — a handful of cases of suicidal thoughts in people taking semaglutide and liraglutide. Under the rules that govern pharmacovigilance, a small number of reports is enough to open a review, and both the European and American regulators did.
What followed was two years of headlines considerably more confident than the underlying evidence, in both directions.
What the investigations found
The EMA’s Pharmacovigilance Risk Assessment Committee reviewed spontaneous reports, trial data and observational studies across the class. It concluded in April 2024 that the available evidence did not support a causal association between GLP-1 receptor agonists and suicidal or self-injurious thoughts, and did not require a labelling change on that basis.
The FDA’s preliminary review, published in January 2024, reported that its evaluation of trial and post-marketing data had not found evidence that these medicines cause suicidal thoughts or actions, while noting the limitations of the data and that it would continue monitoring.
Observational work has generally agreed. A widely cited 2024 analysis in Nature Medicine, using a large electronic health record network, found no increased risk of suicidal ideation associated with semaglutide compared with other weight-loss and diabetes medications — and in several comparisons found a lower rate.
Why people still describe feeling different
Here is where the regulatory answer stops being useful, because a great many people report mood changes and the reviews do not explain them away. Several things are plausibly happening, none of which requires the drug to be acting directly on mood circuitry.
Food was doing something. For a substantial number of people, eating is a functioning emotional regulator — the thing that takes the edge off a bad evening. These drugs remove it, quietly and completely, in the space of about a fortnight. If that was your primary coping mechanism, its removal will be felt, and it will be felt as flatness rather than as loss.
Rapid physical change is destabilising. Even entirely wanted change. People respond to you differently. Attention arrives that you may not have asked for, or attention you were used to disappears. Clothes, photographs and mirrors stop matching your internal model. This is well described after bariatric surgery and there is no reason to expect it to be different here.
The mechanical causes of low mood are all present. Undereating, dehydration, low iron, poor sleep from reflux, and the sheer fatigue of the first two months each depress mood independently. A person eating 900 calories a day and sleeping badly will feel dreadful regardless of what is causing the appetite suppression.
The identity problem. For some people, the weight had become an explanation. Losing it removes the explanation and leaves whatever was underneath, which is not always a welcome introduction.
The drug quiets the noise. It has no opinion about what the noise was covering.
The alcohol observation
One of the more interesting findings in this area is that many people report a sharp drop in interest in alcohol on these drugs, and there is now a serious research effort into GLP-1s for alcohol use disorder.
This cuts both ways for mood. Drinking less is good for depression. But for someone whose drinking was itself a coping mechanism, its removal at the same time as food’s removal takes two regulators away in the same month. That is worth naming in advance, and it is covered further in GLP-1s and alcohol.
April 2024
EMA safety committee concluded no causal association with suicidal thoughts
No increase
Finding of large real-world cohort analyses versus comparator drugs
Nat Med, 2024
Monitor
What the labelling asks of prescribers for people with psychiatric history
What to actually do
Say it out loud to someone clinical. Mood is not on the standard follow-up checklist for these drugs, which are typically reviewed on weight, dose and gut symptoms. If it is not going well, it will not come up unless you raise it.
Rule out the mechanical causes first. Are you eating enough? Drinking enough? Sleeping? Is your iron or B12 low? Those are cheap to check and frequently the answer.
Do not stop abruptly on your own. Discontinuing without a plan produces returning appetite, weight regain and a fresh set of feelings about that, which is rarely the improvement people hope for.
Consider that the drug may be revealing rather than causing. That distinction matters, because the response to it is treatment rather than discontinuation.
If you have a psychiatric history, front-load the conversation. Not to be talked out of the medication — a history of depression is not a contraindication — but so that someone is watching alongside you.
The honest position
The specific fear — that these drugs directly generate suicidal thinking — has been examined seriously by two major regulators and is not supported.
The broader observation, that changing your relationship with food this quickly does something to people, is real, under-studied, and largely absent from the follow-up appointment. Those are different claims, and conflating them has served nobody: it frightened people off a drug that was helping them, and it gave everyone else permission to dismiss mood complaints as debunked.
Both things can be true. The signal was not what it looked like, and you can still be having a difficult time.