Heterogeneous Cardiometabolic Effects of GLP-1/GIP Agonists in Real-World CKM Care
Description: Cardio-kidney-metabolic (CKM) syndrome affects nearly 9 in 10 U.S. adults and is a major driver of cardiovascular disease (CVD), kidney failure, and premature mortality. CKM stages 2–3 include individuals with metabolic risk factors, chronic kidney disease, or subclinical/high-risk CVD and represent a large, clinically important population in whom disease progression remains modifiable. Glucagon-like peptide-1 and dual GLP-1/GIP receptor agonists (GLP-1/GIP RAs) have emerged as promising therapies, demonstrating improvements in weight, glycemic control, blood pressure, lipid levels, kidney outcomes, and cardiovascular risk in clinical trials. However, evidence guiding their use in real-world CKM populations remains limited, as trial populations are highly selected and do not capture the complexity of routine care, where treatment initiation, dose titration, interruptions, and discontinuation occur dynamically and outcomes are measured irregularly. This project will leverage large-scale electronic health record (EHR) data from the national Epic Cosmos network (>300 million patients) and detailed data from Yale New Haven Health System and Penn Medicine to generate robust real-world evidence on the use and heterogeneous effects of GLP-1/GIP RAs in CKM stages 2–3. Aim 1 will characterize real-world treatment patterns, including initiation, dose escalation, persistence, discontinuation, and switching, and identify patient and healthcare factors associated with these patterns across CKM stages and key clinical subgroups. Aim 2 will estimate the effects of GLP-1/GIP RA initiation on cardiometabolic risk markers using systematic error–corrected target trial emulation methods that address measurement error, exposure misclassification, and time-varying confounding, with analyses stratified by CKM stage and clinical subgroups. Aim 3 will quantify heterogeneity in treatment effects across CKM stages and demographic, clinical, and healthcare subgroups, and identify patient profiles associated with greater or lower benefit using both prespecified subgroup analyses and modern causal machine learning approaches. The expected outcome is a comprehensive, population-scale understanding of how GLP-1/GIP RAs are used and how their cardiometabolic benefits vary across CKM stages 2–3. By integrating rigorous causal inference methods with large real-world data, this project will provide actionable evidence to guide equitable and effective implementation of these therapies.
Funder: American Heart Association
Amount: $300,000
Funding type: research
Funder scheme: GP1 Medications Among Individuals with CKM Disease
Start date: 2026-07-01
End date: 2028-06-30
Investigators: Yuan Lu, Yuan Lu
Institution: Yale University
Funded outputs count: 1
Funder award ID: 26GLPCKMD1646434
OpenAlex ID: G5096963147