GLP-1 Drugs Linked to 24% Fewer Substance Use Hospitalizations in Older Adults with Diabetes | Research

The takeaway

Older adults with diabetes and substance use disorders who started GLP-1 drugs had 24% fewer SUD hospitalizations and 36% fewer opioid-related hospitalizations compared to DPP-4 inhibitor users in a Medicare analysis.

36% fewer opioid hospitalizations
GLP-1RA users had significantly fewer opioid use disorder hospitalizations (HR 0.64) compared to DPP-4i users among older adults with diabetes and substance use disorders

What the researchers found

In the GLP-1RA vs DPP4i comparison (n=4,920 matched pairs), GLP-1RA users had significantly lower risk of SUD hospitalization (HR 0.76; 95% CI: 0.67-0.86) and OUD hospitalization (HR 0.64; 95% CI: 0.43-0.96). AUD hospitalization showed a non-significant trend (HR 0.76; 95% CI: 0.53-1.08). No significant differences were found between GLP-1RA and SGLT2i users (n=4,620 matched pairs).

Why it matters

Substance use disorders in elderly diabetic patients are underrecognized and undertreated. The possibility that GLP-1 drugs — already prescribed for diabetes — could simultaneously reduce addiction-related hospitalizations would be a remarkable dual benefit. The 36% reduction in opioid-related hospitalizations is particularly significant given the ongoing opioid crisis and the vulnerability of older adults to opioid complications.

How the study worked

Retrospective cohort study using 2016-2020 Medicare claims data in a target trial emulation framework. Adults ≥65 with T2D and SUD were compared: GLP-1RA new users versus DPP4i or SGLT2i new users. Propensity score 1:1 matching controlled for confounders. Cox proportional hazards models with intention-to-treat approach assessed hospitalization outcomes.

What this study cannot tell us

Retrospective observational study using claims data, which cannot establish causation. The comparison with SGLT2i showed no difference, weakening the case for a unique GLP-1R-mediated reward pathway effect. Medicare claims may undercount substance use episodes. The older adult population (≥65) may not reflect younger SUD populations. SUD severity and treatment engagement were not captured. The opioid result confidence interval was wide (0.43-0.96).

How to read the evidence

This is a well-designed retrospective cohort study using a target trial emulation framework with propensity score matching. The Medicare database provides a large, real-world population. However, the observational design and null result versus SGLT2i limit the ability to attribute effects specifically to GLP-1R-mediated reward pathway modulation.

When this study was published

Published in 2025 in Obesity, this study adds to the rapidly growing evidence base for GLP-1RA effects on addiction and substance use — one of the most intriguing potential new applications of these drugs.

The bigger picture

GLP-1 receptors are expressed in brain reward circuits, and preclinical evidence suggests GLP-1RAs reduce the rewarding effects of alcohol, opioids, and other substances. This Medicare study adds to the growing human observational evidence supporting GLP-1 drugs as potential addiction treatments. The fact that the effect was not seen versus SGLT2 inhibitors (which also improve metabolic health but don't target brain reward pathways) complicates the pure 'reward pathway' hypothesis — suggesting the mechanism may involve both direct brain effects and indirect metabolic improvements.

Questions still open

Why was there no difference between GLP-1RA and SGLT2i users for SUD outcomes — do metabolic improvements alone reduce addiction-related complications?
Would randomized trials of GLP-1RAs specifically for substance use disorders confirm these observational findings?
Are the effects stronger for specific substances (opioids, alcohol, stimulants) or applicable broadly?

Common questions

Can Ozempic help with addiction?

There's growing evidence that GLP-1 drugs like semaglutide may reduce substance use behaviors — possibly by modulating brain reward pathways. This Medicare study found 24% fewer substance use hospitalizations and 36% fewer opioid-related hospitalizations in GLP-1 drug users. However, this is observational evidence and GLP-1 drugs are not approved for addiction treatment. Clinical trials are underway to test this directly.

How might GLP-1 drugs affect brain reward pathways?

GLP-1 receptors are found in brain areas that process reward and pleasure, including regions involved in addiction. Animal studies show GLP-1 drugs reduce the rewarding effects of alcohol, cocaine, and opioids. In humans, this could translate to reduced cravings and substance-seeking behavior. However, this study's finding that SGLT2 inhibitors (which don't target brain GLP-1 receptors) showed similar outcomes suggests the mechanism may be more complex than direct reward modulation alone.

Read the original research

GLP-1 Receptor Agonists and Substance Use Disorders in Older Adults With Type 2 Diabetes: A Target Trial Emulation.

Obesity (Silver Spring, Md.)

View on PubMed ↗Publisher’s paper ↗

Citation

Dai, Hao; Radwan, Rotana M; Scheiffele, Grant D; Tang, Huilin; Sheer, Amy; Lin, Hsin-Yueh; Zhang, Pengyue; Adirika, Darlene; Luong, Kate; Bian, Jiang; Guo, Jingchuan. (2025). GLP-1 Receptor Agonists and Substance Use Disorders in Older Adults With Type 2 Diabetes: A Target Trial Emulation.. Obesity (Silver Spring, Md.). https://doi.org/10.1002/oby.70024