Lower 1-yr Mortality in Diabetes Patients with Sleep Apnea Prescribed GLP-1 RAs

Data revealed one-year mortality in patients with type 2 diabetes prescribed GLP-1 RAs was substantially lower than those not prescribed the drugs, and disproportionate benefit for those also diagnosed with OSA.

A real-world study of nearly 1.8 million patients found that individuals with type 2 diabetes (T2DM) and obstructive sleep apnea (OSA) prescribed glucagon-like peptide-1 receptor agonists (GLP-1RAs) experienced greater survival benefits than those without OSA, suggesting that OSA status may enhance the mortality benefit of GLP-1RA therapy.

While GLP-1RAs such as tirzepatide, semaglutide, and dulaglutide are known to reduce cardiovascular and all-cause mortality in patients with T2DM, their effects in patients who also have OSA remain unclear.

Study Details

• Building on the findings from the SURMOUNT-OSA1 trial, which demonstrated tirzepatide’s efficacy in improving OSA in nondiabetic patients, this study investigated whether OSA modifies the mortality benefit associated with GLP-1RA use among individuals with T2DM in a large, real-world population.
• The primary end point was all-cause mortality within one year of metformin ± GLP-1RA prescription. Propensity score matching (PSM) adjusted for demographics and comorbidities, and Cochran–Mantel–Haenszel testing assessed for effect modification by OSA status.
• The study identified 1,799,261 patients with T2DM prescribed metformin, of which
  • 1,083,492 (60.2%) had neither OSA diagnosis nor GLP-1RA prescription;
  • 361,492 (20.1%) were prescribed GLP-1RA without OSA diagnosis;
  • 207,947 (11.6%) were diagnosed with OSA without GLP-1RA prescription; and
  • 146,330 (8.1%) were diagnosed with OSA and prescribed a GLP-1RA.
• The most prescribed GLP-1RA was semaglutide, followed by dulaglutide and tirzepatide.

Study Results

• Following PSM (in which standardized mean differences were found to be <0.1, indicating successful matching), GLP-1RA-prescribed groups continued to show lower 1-year mortality, both without (acute respiratory distress, [ARD] 0.9% vs 1.8%, respiratory rate [RR] 2.04, CI 1.95–2.13) and with OSA (ARD 1.0% vs 2.5%, RR 2.45, CI 2.29–2.61), all p < 0.001.
• The ratio of RRs for mortality between non-OSA and OSA cohorts was 1.2, indicating a 20% greater mortality benefit in the OSA population, and CochranMantel–Haenszel statistic was significant (p < 0.001).

“We observed 1-year mortality in patients with type two diabetes who were prescribed GLP1RAs to be substantially lower than patients not prescribed GLP1-RAs with a disproportionate benefit observed in those also diagnosed with OSA,” said Cosmo Fowler, MD, lead researcher and CHEST 2025 presenter.

“This large-scale analysis suggests that OSA status may act as an effect modifier in the association between GLP-1RA prescription and mortality.” The results of this study may lead to OSA diagnosis identifying T2DM patients more likely to derive mortality benefit from GLP-1RA therapy. The 20% greater relative mortality reduction in patients with OSA supports considering OSA status when making GLP-1RA prescribing decision. Future research should examine prospectively the relationship between OSA and T2DM diagnosis and long-term outcomes of GLP-1RA therapy.

Further results will be shared at the Chest Annual Meeting 2025 as part of the GLP-1 Agonists and Implications on OSA Comorbidities Rapid Fire original investigation presentations titled, “Enhanced Survival Benefit for GLP-1 Receptor Agonist Prescription in Patients With Coexisting Type 2 Diabetes and Sleep Apnea: A Real-World Analysis of 1.8 Million Patients.”

Read the study abstract here. 🔍 Page 7054