Question In patients with {type 2}* diabetes, are incretin-based drugs associated with increased risk for heart failure (HF) compared with combinations of other oral antidiabetic drugs? Methods Design Nested case-control study, using administrative databases in Canada, a general practitioner database linked to hospital data in the UK, and insurance claims data in the USA. ClinicalTrials.gov NCT02456428. Setting Canada (Alberta, Manitoba, Ontario, and Saskatchewan), the UK, and the USA. Patients 1840467 patients ≥18 years of age who had a first-ever prescription for a noninsulin antidiabetic drug at any time were included in the base cohort. Exclusion criteria included previous insulin use. From the base cohort, 1499650 patients were prescribed a new antidiabetic drug after incretin drugs (e.g., dipeptidyl peptidase 4 inhibitors and glucagon-like peptide 1 analogues) became available and were included in the study cohort (new drug prescription date = baseline), with stratification for previous HF at baseline. Prescriptions could be for new diabetes or for a new drug class in patients already using antidiabetic drugs. 29741 study cohort patients were hospitalized with HF during the study and classified as cases (admission date = case index date): 6536 had previous HF at baseline (mean age 74 y, 59% men, mean treated diabetes duration 1.8 y), and 23205 had no previous HF (mean age 69 y, 57% men, mean treated diabetes duration 0.7 y). Using risk-set sampling, each case was matched with ≤20 controls (n =100480 with previous HF and n =435777 with no previous HF) who were at risk for HF hospitalization at the case index date. Matching was based on age, sex, duration of treated diabetes, baseline date, and duration of follow-up. Risk factors Current use (prescription ongoing or ending ≤30 d before the case index date) of incretin-based drugs, {alone or with other antidiabetic drugs and used as first-, second-, or third-line treatment}*. Because incretins are intended to be second- or third-line drugs, the primary analysis used a reference group of patients using combinations of other oral antidiabetic drugs. Outcome HF hospitalization. Main results The main results are in the Table. Conclusion In patients with type 2 diabetes, incretin-based drugs were not linked to increased risk for heart failure hospitalization compared with combinations of other oral antidiabetic drugs. [ABSTRACT FROM AUTHOR]