Semaglutide, Depression and HIV

A 2026 study followed depressive-symptom scores in 354 people with HIV before and after they started semaglutide in routine care. The average Patient Health Questionnaire-9 score was essentially unchanged: -0.1 points, with a 95% confidence interval from -0.7 to 0.5^[1]. That is reassuring within this cohort. It does not prove that semaglutide is psychiatrically safe for every patient, because the study had no untreated comparison group and measured symptom scores rather than all possible psychiatric harms.

What the routine-care study measured

The researchers used data from nine Centers for AIDS Research Network of Integrated Clinical Systems sites. They included people with HIV who began semaglutide between April 2018 and October 2024. Depressive symptoms came from PHQ-9 questionnaires collected during ordinary clinical care before and after treatment began^[1]. The PHQ-9 is a nine-item symptom questionnaire. A score can help clinicians assess symptom burden, but it is not by itself a diagnosis or a complete psychiatric evaluation.

The analysis used linear mixed models, a statistical method that can account for repeated measurements from the same person. Results were examined overall and by baseline depression severity, body mass index, diabetes and antidepressant use^[1]. Because the data came from care rather than a scheduled research protocol, the abstract does not report a fixed follow-up interval, the number of questionnaires per participant or how many eligible starters lacked follow-up scores. Those details should not be guessed from the headline result.

Participants had a mean age of 54 years, 77% were male and 38% were non-Hispanic White. Obesity was reported in 78%, while 60% had diabetes. At baseline, 53% had no or minimal depressive symptoms, 28% had mild symptoms, 10% had moderate symptoms and 9% had moderately severe to severe symptoms^[1]. This is a clinically relevant HIV cohort, but its composition and care setting limit how confidently the finding can be applied to other populations.

What changed after semaglutide began

Across all 354 participants, semaglutide initiation was not associated with an overall change in depressive symptoms. The estimated -0.1-point change was close to zero, and its confidence interval included both a small decrease and a small increase^[1]. The publisher's abstract reports the same population, method and estimate^[2]. These are two routes to the same paper, not two independent studies.

The pattern differed when participants were grouped by their starting scores. Scores rose by 1.2 points among those with no or minimal symptoms at baseline, with a 95% confidence interval from 0.5 to 1.8. Scores were stable in the mild and moderate groups. Among people who began with moderately severe to severe symptoms, scores fell by 4.7 points, with a 95% confidence interval from -7.3 to -2.2^[1]. The abstract also reports no worsening across body-mass, diabetes or antidepressant-use subgroups, but it does not provide their numerical estimates.

Those subgroup movements should not be read as evidence that semaglutide worsens mood in people with low starting scores or treats depression in people with high scores. Grouping people by an initial measurement can produce regression toward the mean, where unusually low or high scores move closer to average when measured again. Changes in HIV care, antidepressant treatment, physical health or life circumstances could also affect PHQ-9 scores during follow-up.

Why this is a safety signal, not safety proof

A within-person prepost study compares each participant with their own earlier score. That design is useful for detecting a change after treatment starts, but time itself becomes the comparison. Without a contemporaneous group that did not start semaglutide, the analysis cannot separate the medicine's effect from other events occurring over the same period. It also cannot show what would have happened to these participants without treatment.

The result is about depressive-symptom scores at the group level. It does not establish the absence of new depression diagnoses, suicidal thoughts, self-harm, mania, anxiety or rare individual reactions. The paper's background mentions depression and suicidality concerns, but the abstract reports total PHQ-9 changes rather than a separate analysis of suicidal thinking. A near-zero average can also coexist with improvement in some individuals and worsening in others.

Treatment selection remains another limit. People who receive semaglutide in routine HIV care may differ from those who do not in metabolic health, access to care, clinician judgment or willingness to start treatment. Statistical adjustment and within-person modeling can address some measured differences, but they cannot create random assignment or remove unmeasured changes over time.

How this fits the wider depression evidence

PeptideMethods has separately reviewed the broader and mixed evidence on semaglutide and depression. This HIV-specific paper answers a narrower question. It adds repeated PHQ-9 measurements from a population with a substantial depression burden, but it does not replace randomized trials, adverse-event surveillance or clinical assessment of an individual patient.

The most defensible reading is that researchers did not detect overall worsening after semaglutide initiation in this 354-person cohort. That is different from proving no harm. It is also different from showing a benefit for depression. Anyone who notices persistent low mood, a marked symptom change or thoughts of self-harm needs prompt clinical support, regardless of whether the change appears related to a medicine.

What we don't yet know

The abstract does not expose the timing or completeness of follow-up assessments, treatment duration, semaglutide dose, changes in antidepressants or the distribution of individual score changes. Full article text was not publicly exposed through the publisher route we inspected, so those details are withheld here. Longer controlled studies would be needed to estimate uncommon harms, distinguish temporary from persistent changes and test whether findings differ by sex, HIV treatment or psychiatric history.

This study supports cautious reassurance about average depressive-symptom scores in one routine-care network. It does not support a blanket claim of psychiatric safety, and it should not be used to start, stop or alter semaglutide without a prescriber who can review metabolic and mental-health needs. The site's semaglutide regulation overview provides wider prescribing context, not an individual treatment decision.

Medical disclaimer: This article is for educational and journalistic purposes only and does not constitute medical advice. Peptides discussed may be classified as prescription medicines or research chemicals depending on your jurisdiction. Always consult a qualified healthcare professional before using any peptide product. PeptideMethods.com does not sell, distribute, or facilitate the sale of any peptide product.