Abstract: TH-OR087
Learning Health System Intervention to Increase SGLT2 Inhibitor Use
Session Information
- Late-Breaking Research Orals - 1
November 06, 2025 | Location: Grand Ballroom C, Convention Center
Abstract Time: 05:06 PM - 05:18 PM
Category: CKD (Non-Dialysis)
- 2302 CKD (Non-Dialysis): Clinical, Outcomes, and Trials
Authors
- Ishani, Areef, Minneapolis VA Medical Center, Minneapolis, Minnesota, United States
- Murphy, Daniel P., University of Minnesota Twin Cities, Minneapolis, Minnesota, United States
- Pestka, Deborah, University of Minnesota Twin Cities, Minneapolis, Minnesota, United States
- Kaplan, Adam N, Minneapolis VA Medical Center, Minneapolis, Minnesota, United States
- Huynh, Pearl Huyen, Minneapolis VA Medical Center, Minneapolis, Minnesota, United States
- Rechtzigel, Jessica A, Minneapolis VA Medical Center, Minneapolis, Minnesota, United States
- Kjos, Shari, Minneapolis VA Medical Center, Minneapolis, Minnesota, United States
- Switzer, Lisa Marie, Minneapolis VA Medical Center, Minneapolis, Minnesota, United States
- Taylor, Brent C., Minneapolis VA Medical Center, Minneapolis, Minnesota, United States
- Atwood, Melissa M, Minneapolis VA Medical Center, Minneapolis, Minnesota, United States
- Polsfuss, Beth, Minneapolis VA Medical Center, Minneapolis, Minnesota, United States
- Lee, Joseph Y., Minneapolis VA Medical Center, Minneapolis, Minnesota, United States
Background
We sought to determine if proactive pharmacist outreach to patients with type 2 diabetes (T2D) and chronic kidney disease (CKD) is more effective than usual care at increasing sodium-glucose cotransport-2 inhibitor (SGLT2i) use.
Methods
All patients from 8 Veterans Affairs sites, with T2D and estimated glomerular filtration rate (eGFR) 25-59 mL/min/1.73m2 without SGLT2i were identified and pseudo-randomized by their social security number’s final digit. Odd and even numbers were assigned to early and delayed intervention, respectively. The early arm was invited to meet with a pharmacist by phone to discuss SGLT2i. Patients were then started on empagliflozin if they were both willing and without identified contraindications. The delayed arm received an invitation >1 year later.
All patients were included in intention-to-treat analysis. The primary outcome was SGLT2i prescription filled at 1 year. Secondary time-to clinical outcomes were identified using electronic health records.
Results
8,883 patients were included, with 4,505 in the early intervention arm. In each arm, 97% were male, 86% were white, and mean age, baseline eGFR, and A1c were 79 [standard deviation (SD): 8] years, 46 (SD: 10) mL/min/1.73m2, and 7.0% (SD: 1.2%), respectively. The median length of follow-up was 743 days. Many patients in the early arm declined intervention based on comorbidities (e.g., cancer).
At 1 year, 19.9% vs 10.2% of patients had a SGLT2i fill (P <0.001), favoring the intervention. The early intervention prescribed a SGLT2i to 980 (21.8%) patients after 1,797 (39.9%) patients attended a telephone encounter.
Win-loss ratio was 1.02 (95% CI: 0.97-1.07; P = 0.43), in the direction of fewer adverse outcomes. The hierarchy was mortality, any eGFR <15 mL/min/1.73m2, acute heart failure, heart attack, and lastly stroke.
Conclusion
Using a learning health system approach with pseudo-randomization, SGLT2i use doubled in the early arm compared to usual care. With only limited follow-up and a 9.7% absolute difference in SGLT2i use between arms, we did not observe a statistically significant clinical difference. However, this project shows the value of integrating pharmacists to improve medication use, as well as learning health system methods to pragmatically evaluate system initiatives.
Funding
- Veterans Affairs Support