Clinical Summary: GLP-1 Receptor Agonists and All-Cause Overdose Risk in Veterans With Type 2 Diabetes and Opioid Use Disorder
Veterans with type 2 diabetes and opioid use disorder face substantial overdose risk, and standard medications for opioid use disorder remain underused because of side effects, discontinuation, access barriers, and misuse concerns. This study asks whether semaglutide or tirzepatide, already used in diabetes care, are associated with fewer overdoses in a high-risk VA population.
Key Findings
- Using the Benjamini-Hochberg procedure, we find significant differences favoring semaglutide/tirzepatide for the insulin and SGLT2i comparisons, and nonsignificant differences for the DPP4i, metformin, and sulfonylurea comparisons.
- For the insulin and SGLT2i comparisons, the E values suggest that an unmeasured confounder would have to be associated with both exposure and outcome by a hazard ratio of 5.70 and 7.46, respectively, over and above the measured confounders, to negate the observed association.
- Prior to propensity score matching, the median SMDs for the insulin, SGLT2i, DPP4i, metformin, and sulfonylurea comparisons were 0.12, 0.11, 0.09, 0.10, and 0.09, respectively; after matching, the median SMDs decreased to 0.03, 0.05, 0.06, 0.03, and 0.03, respectively.
- Prior to propensity score matching, the maximum SMD of the insulin, SGLT2i, DPP4i, metformin, and sulfonylurea comparisons were 0.93, 0.62, 0.62, 0.64, and 0.68, respectively; after matching, the maximum SMDs decreased to 0.10, 0.15, 0.27, 0.10, and 0.15, respectively.
- To control the false discovery rate across the 5 comparisons to 0.05, the Benjamini-Hochberg step-up procedure tested the hypotheses at alpha levels of 0.01, 0.02, 0.03, 0.04 and 0.05.
In Veterans with T2DM and OUD starting a new diabetes medication, semaglutide or tirzepatide was linked to lower 12-month all-cause overdose risk than insulin and SGLT2 inhibitors, but not than metformin, sulfonylureas, or DPP4 inhibitors. These findings support GLP-1–based agents as a clinically relevant overdose-risk signal, not yet as proof of efficacy for OUD treatment.
Practice Implications
- If a patient with T2DM and OUD is a candidate for semaglutide or tirzepatide, overdose-risk findings may be part of the treatment discussion, especially when the alternative is insulin or an SGLT2 inhibitor.
- Do not interpret these data as evidence that semaglutide or tirzepatide specifically prevent opioid overdose; the outcome was all-cause overdose because opioid-only events were too few for adequate power.
- Apply these results most cautiously to patients who would resemble the analyzed cohort, since patients with ≤30-day cumulative prescription coverage or <2 fills were excluded.
- Continue standard OUD care and overdose monitoring, because the authors emphasize that randomized controlled trials are still needed to establish causal efficacy in opioid use disorder.
