This cohort study aims to assess the comparative effectiveness of tirzepatide versus semaglutide with respect to cardiovascular events in patients with type 2 diabetes and heart failure with preserved ejection fraction.

Both semaglutide, a glucagon-like peptide-1 receptor agonist (GLP-1RA), and tirzepatide, a glucose-dependent insulinotropic polypeptide (GIP)/GLP-1RA, are currently approved treatments to control blood glucose in patients with type 2 diabetes (T2D) and to lower weight in patients with obesity, with or without T2D. Obesity is a leading risk factor for the onset and the progression of heart failure (HF), especially HF with preserved ejection fraction (HFpEF). Due to their potent weight-lowering action, semaglutide and tirzepatide have been recently evaluated in placebo-controlled trials conducted among persons with obesity and HFpEF, with or without T2D. These trials showed that semaglutide and tirzepatide reduced HF symptoms and HF events, compared to placebo.

In this new user, active comparator cohort study, the investigators will identify commercially insured adults (aged > 18 years) with T2D and HFpEF who initiated treatment with tirzepatide or semaglutide between June 1, 2022, and December 31, 2024. The investigators will retrieve information for each study participants using deidentified, longitudinal insurance claims data, including demographic characteristics, health plan enrollment status, inpatient and outpatient diagnoses, procedures, health care visits, hospitalizations, and pharmacy dispensing records.

Cohort entry is the day of initiation of tirzepatide or semaglutide. The follow-up starts on the day after cohort entry and continue until the occurrence of an outcome, discontinuation, switching, death, end of continuous health plan enrollment or end of the study period. Our primary outcome is a composite of hospitalization for HF or all-cause mortality.

The investigators will use propensity score adjustment to mitigate the risk of confounding. The investigators will calculate the number of events, incidence rates, and rate differences per 1000 person-years. Cox proportional hazard models will be used to estimate hazard ratios with their 95% confidence intervals.

Inclusion Criteria:

• Patients who are new users of tirzepatide or new users of semaglutide
• Patients with diagnosis of T2D and HFpEF, i.e., EF >= 45%.
• Age >= 18 years old
• Patients with continuous health plan enrollment before and including the treatment initiation date

Exclusion Criteria:

• Patients with missing age or sex information
• Patients with type 1 diabetes mellitus, secondary or gestational diabetes
• History of diabetes related complications
• Patients with related chronic diseases.
• History of gastrointestinal conditions.
• Previous exposure to other GLP-1RA and pramlintide
• Patients with prescription dispensing for both tirzepatide and semaglutide on cohort entry date

Patients > 18 years old with T2D and HFpEF.