Project Summary

Health Condition: Type 2 diabetes in patients who have a moderate risk for heart disease, defined as a 2 to 3 percent chance of having a heart attack or stroke, or dying in a given year.

Importance: Type 2 diabetes is one of the most common and serious chronic health conditions in the United States. Keeping blood sugar levels in a safe range is important for preventing long-term complications of diabetes, including heart disease, stroke, heart failure, nerve damage and pain, kidney disease, dialysis, limb loss/amputation, and vision loss, as well as short-term complications like severe hypoglycemia (low blood sugar). Among people with diabetes, these complications are significant causes of death, disability, and poor quality of life. However, there is currently not enough information to know whether particular diabetes medications or medication classes are more effective at preventing these serious complications than are others. This is particularly true for patients at moderate risk for heart disease, who have not been the focus of prior studies examining diabetes medications.

Several classes of medications are used to treat type 2 diabetes. When patients start taking medication for their diabetes, metformin is recommended first because it is effective, well tolerated, and widely available. However, when metformin is not able to control blood sugar levels, or is not tolerated, patients and clinicians require guidance on which medication is best to start next (i.e., a second-line medication). The four classes of second-line medications used most often, and which we will examine in this study, are GLP-1RA, SGLT2i, DPP-4i, and SU.

Study Aims: The goal of our research is to improve diabetes care and prevent short- and long-term diabetes complications. To achieve this goal, we will conduct three parallel studies that answer the following questions that routinely arise when managing diabetes: (1) Which class of second-line diabetes medications is most effective at reducing risks of heart disease, kidney disease, and other complications? (2) Which medication class is best for a specific set of outcomes desired by an individual patient with type 2 diabetes? (3) Of the available medications in each class, is there one that is best?

Study Design: We will use existing, anonymous, information from OptumLabs Data Warehouse (a data set of over 140 million Americans with private and Medicare Advantage insurance) and Medicare to mimic randomized clinical trials.

Who Can Use These Results: This research will help patients and clinicians choose diabetes medications, professional societies develop guidelines for how to manage diabetes, and health systems and insurance companies identify medications to be included in their formularies.

Patient Population and Comparators: In our data set, we will identify adults with type 2 diabetes who are at moderate risk of heart disease and started a GLP-1RA, SGLT2i, DPP-4i, or SU medication between January 1, 2014, and December 31, 2019.

Outcomes and Analyses: In the first study, we will compare the rates of cardiovascular events (i.e., heart attack, stroke, and death) as well as of hospitalizations for heart failure, revascularization (i.e., bypass and stenting) procedures, kidney disease, dialysis initiation, amputations or foot ulcers, blindness or treatments for diabetic eye disease, severe hypoglycemia, and other adverse events between patients taking each of these four medication classes. In the second study, we will ask a group of patients with type 2 diabetes about the health outcomes that are important to them (i.e., which complications are most important for them to avoid), combine these outcomes into one, and compare the four medication classes specifically for that combined outcome. This will, in essence, mimic a clinical trial designed specifically with the goals of each individual patient in mind. The third study will compare the specific medications within each drug class regarding the primary and secondary outcomes of the first study.

Stakeholder Engagement: Three patients with type 2 diabetes and four practicing clinicians (an endocrinologist, an internist, a pharmacist, and a diabetes care and education specialist) helped design these studies with input from two health system leaders and a senior health plan official. These stakeholders, as well as a representative from the American Diabetes Association, will be involved in conducting the study, sharing our findings, and implementing them into practice.

Project Information

Rozalina McCoy, MD, MS
Mayo Clinic

Key Dates

March 2021
August 2025


Award Type
Populations Populations PCORI is interested in research that seeks to better understand how different clinical and health system options work for different people. These populations are frequently studied in our portfolio or identified as being of interest by our stakeholders. View Glossary
State State The state where the project originates, or where the primary institution or organization is located. View Glossary
Last updated: March 4, 2022