Glucagon-like peptide-1 receptor agonist (GLP-1) medications have been used for glucose control in diabetic patients for several years. Recently, the U.S. Food and Drug Administration (FDA) approved some GLP-1 medications, including semaglutide and liraglutide, for weight management in obese patients.1 Notably, one such medication, Wegovy, which has a main active ingredient of semaglutide, lists depression and suicidal ideation as potential side effects.2 Other GLP-1 medications, including another with semaglutide as a main active ingredient, do not list depression or suicidal ideation as potential side effects.3 An initial evaluation by the FDA suggests no direct link between GLP-1 medications and suicidal ideation.1
To better understand the potential relationship between GLP-1 medications and mental health diagnoses after starting the medications, we studied 3,081,254 diabetic patients and 929,174 non-diabetic patients. We adjusted for patient age, sex, and history of depression or anxiety. For the diabetic population, we also adjusted for HbA1c level and use of other diabetic medications. For the non-diabetic population, we adjusted for BMI classification. For the diabetic population, we compared patients prescribed GLP-1 medications to patients with an HbA1c level documented. For the non-diabetic population, we compared patients prescribed GLP-1 medications to patients prescribed a non-GLP-1 weight management medication.
We first investigated the correlation between GLP-1 medications and depression diagnosis after starting the medication. Among diabetic patients, those prescribed any of the GLP-1 medications studied, except for liraglutide, showed a decreased likelihood of depression compared to those not prescribed the GLP-1 medication. Tirzepatide showed the greatest reduction in likelihood of depression for diabetic patients (65%), as shown in Figure 1. Among non-diabetic patients, those prescribed semaglutide had a lower likelihood of depression compared to those not prescribed the GLP-1 medication. Liraglutide showed no statistically significant difference in depression diagnoses for non-diabetic patients.
Next, we evaluated the relationship between GLP-1 medications and anxiety. Similar to depression, diabetic patients prescribed any of the GLP-1 medications had a lower likelihood of being diagnosed with anxiety after starting the medication compared to those not prescribed a GLP-1 medication. Tirzepatide showed the greatest reduction in likelihood, with a 60% reduction, as seen in Figure 2. Among non-diabetic patients, only those prescribed semaglutide had a lower likelihood of anxiety compared to those not prescribed the GLP-1 medication, while those prescribed liraglutide had no statistically significant difference.