GLP-1 drugs linked to lower suicide risk in diabetes, but the benefit vanished in cannabis use disorder patients

GLP-1 RA use was associated with lower suicide attempt risk (aHR 0.63). Cannabis use disorder was associated with dramatically higher risk (aHR 5.50). Among patients with both CUD and GLP-1 RA use, suicide risk remained elevated (aHR 5.75) with no significant benefit from GLP-1 RAs (aHR 1.00 comparing GLP-1 users vs non-users among CUD patients).

Retrospective study using the TriNetX Research Network covering adults aged 30-85 with type 2 diabetes from over 20 countries (2003-2023). 6,424,228 individuals were categorized by GLP-1 RA exposure and cannabis use disorder status. Cox models estimated hazard ratios adjusted for age, sex, depression, BMI, and HbA1c. Those with prior suicidal ideation were excluded.

GLP-1 receptor agonists (like semaglutide/Ozempic) have generated interest for potential mental health benefits beyond metabolic effects. This study suggests their apparent protective effect against suicide does not extend to patients with cannabis use disorder, a vulnerable population with dramatically elevated baseline risk.

The 5.5-fold suicide risk associated with cannabis use disorder in diabetes is striking and largely independent of GLP-1 treatment. This suggests that CUD represents a powerful risk factor that may overwhelm whatever neuroprotective mechanisms GLP-1 drugs provide, highlighting the need for integrated psychiatric and substance use screening in diabetes care.

Retrospective design with potential confounding despite adjustment. Cannabis use disorder codes may capture only the most severe cases. GLP-1 RA users may differ systematically from non-users. The CUD + GLP-1 subgroup may be too small for adequate power. Claims data cannot capture suicidal behavior that does not result in healthcare contact.