In a large real-world study, GLP-1 receptor agonists reduced all-cause mortality by 17% compared to SGLT-2 inhibitors but carried higher odds of gastroparesis and acid reflux.
17% lower all-cause mortalityGLP-1 receptor agonist users had significantly lower odds of death from any cause over 24 months compared to SGLT-2 inhibitor users, in a propensity-matched cohort of nearly 105,000 pairs.
What the researchers found
At 24-month follow-up, GLP-1 receptor agonist users had significantly reduced all-cause mortality compared to SGLT-2 inhibitor users (OR 0.83, 95% CI: 0.80-0.87), representing a 17% relative reduction.
However, GLP-1RA users had higher odds of gastroparesis (aOR 1.24, 95% CI: 1.11-1.38) and gastroesophageal reflux disease (aOR 1.14, 95% CI: 1.11-1.18). Notably, the risk of acute pancreatitis was lower in the GLP-1RA group, alleviating a longstanding safety concern for this drug class.
Why it matters
Both GLP-1 receptor agonists and SGLT-2 inhibitors are first-line diabetes treatments with cardiovascular benefits, so understanding their comparative safety profile matters enormously for prescribing decisions. This study provides the largest real-world head-to-head comparison of their GI safety profiles while also revealing a significant mortality benefit for GLP-1 drugs. The trade-off — lower death risk but more digestive side effects — is critical information for patients and clinicians choosing between these drug classes.
How the study worked
This retrospective cohort study used electronic health records from the TriNetX Multi-Institutional Database covering January 2021 through December 2022. From 3.2 million adults with type 2 diabetes, 104,947 GLP-1RA users were propensity-score matched 1:1 with SGLT-2i users to balance baseline characteristics. After matching, the groups had similar mean age (62±12 years) and HbA1c (8.0±2.0%). About 30% of participants were from underrepresented minority groups. Adjusted odds ratios were calculated for GI adverse events and all-cause mortality at 24 months.
What this study cannot tell us
The retrospective observational design cannot establish causation. Despite propensity score matching, residual confounding may exist — for example, patients prescribed GLP-1RAs vs. SGLT-2is may differ in ways not captured in electronic health records. The study used diagnosis codes for GI outcomes, which may underestimate conditions like mild gastroparesis or GERD that go undiagnosed. The 2021-2022 timeframe means the study primarily captured older GLP-1RA formulations and may not reflect current prescribing patterns.
How to read the evidence
This is a large retrospective cohort study using propensity score matching from a multi-institutional database of 3.2 million patients. The large sample size, diverse population, and rigorous matching strengthen the findings. However, the observational design limits causal inference, and electronic health record data may have coding and completeness issues.
When this study was published
Published in 2025 using 2021-2022 data, this study is current and reflects real-world use patterns during the rapid expansion of GLP-1 receptor agonist prescribing.
The bigger picture
This study enters a heated debate about the gastrointestinal safety of GLP-1 receptor agonists at a time when prescriptions are surging for both diabetes and obesity. While GI side effects like nausea are well-known, the specific risks of gastroparesis and GERD have been less well-quantified in large populations. The pancreatitis finding is particularly noteworthy — early concerns about GLP-1 drugs causing pancreatitis have persisted for years, and this large study suggests the opposite may be true. The mortality benefit adds to growing evidence that GLP-1 receptor agonists have effects that extend well beyond blood sugar control.
Questions still open
- Is the gastroparesis associated with GLP-1 receptor agonists reversible after discontinuation, or does it persist?
- What drives the 17% mortality reduction — cardiovascular protection, weight loss, or another mechanism?
- Do specific GLP-1 receptor agonists (semaglutide vs. liraglutide vs. dulaglutide) differ in their GI adverse effect profiles?
Common questions
Should I be worried about stomach problems if I'm taking a GLP-1 drug for diabetes?
Do GLP-1 drugs cause pancreatitis?
Read the original research
All-Cause Mortality and Gastrointestinal Adverse Effects in Adults With Type 2 Diabetes on Glucagon-Like Peptide-1 Receptor Agonists vs Sodium-Glucose Cotransporter-2 Inhibitors.
Gastro hep advances, 4(10), 100736
Citation
Garg, Samita; Qapaja, Thabet; Hamid, Osama; Kaya, Gizem; Abdel-Razeq, Rashid; Alayan, Dina; Abu-Rumaileh, Mohammed; Lembo, Anthony; Nissen, Steven. (2025). All-Cause Mortality and Gastrointestinal Adverse Effects in Adults With Type 2 Diabetes on Glucagon-Like Peptide-1 Receptor Agonists vs Sodium-Glucose Cotransporter-2 Inhibitors.. Gastro hep advances, 4(10), 100736. https://doi.org/10.1016/j.gastha.2025.100736