Investigators are building an empirical evidence base for real world data through large-scale replication of randomized controlled trials. The investigators' goal is to understand for what types of clinical questions real world data analyses can be conducted with confidence and how to implement such studies.

This is a non-randomized, non-interventional study that is part of the RCT DUPLICATE initiative (www.rctduplicate.org) of the Brigham and Women's Hospital, Harvard Medical School. It is intended to replicate, as closely as is possible in healthcare insurance claims data, the trial listed below/above. Although many features of the trial cannot be directly replicated in healthcare claims, key design features, including outcomes, exposures, and inclusion/exclusion criteria, were selected to proxy those features from the trial. Randomization is also not replicable in healthcare claims data but was proxied through a statistical balancing of measured covariates according to standard practice. Investigators assume that the RCT provides the reference standard treatment effect estimate and that failure to replicate RCT findings is indicative of the inadequacy of the healthcare claims data for replication for a range of possible reasons and does not provide information on the validity of the original RCT finding.

Please see: https://drive.google.com/drive/folders/1WD618wrywYjEaXzfLTcuK-VCcnb6b-gV for full code and algorithm definitions.

Eligible cohort entry dates:

Market availability of saxagliptin in the U.S. started on July 31, 2009 For Marketscan and Medicare: July 31, 2009-Dec 31, 2016 (end of data availability).

For Optum: July 31, 2009-Sep 30, 2017 (end of data availability).

Inclusion Criteria:

• Diagnosed with T2DM based on the current American Diabetes Association guidelines Age ≥40 years Glycated hemoglobin level of ≥6.5% (based on the last measured and documented laboratory measurement in the previous 6 months)

High risk for a CV event defined as having either established CV disease and/or multiple risk factors:

• History of established cardiovascular disease

  • Ischemic heart disease, and/or
  • Peripheral vascular disease (eg, intermittent claudication), and/or
  • Ischemic stroke

• Multiple risk factors for vascular disease - At least 55 years of age (men) or 60 years of age (women), AND at least one of the following additional risk factors

  • Dyslipidemia (based on the last measured and documented laboratory measurement in the previous 6 months and defined as at least 1 of the following):

    • High level of low-density lipoprotein cholesterol (LDL-C), defined as N130 mg/dL (N 3.36 mmol/L) regardless of lipid-lowering therapy
    • Low level of high-density lipoprotein cholesterol (HDL-C), defined as b40 mg/dL (b1.04 mmol/L) for men or b50 mg/dL (b1.30 mmol/L) for women

  • Hypertension, as confirmed at the enrolment visit

    • BP N140/N90 mm Hg, or
    • BP N130/N80 mm Hg on BP-lowering agent

  • Currently smoking, as confirmed at the enrolment visit Women of childbearing potential must take precautions to avoid pregnancy throughout the study and for 4 weeks after intake of the last dose. Men participating in the study should also take precautions not to father a childwhile participating in the study and for 4 weeks after intake of the last dose.

Provision of informed consent before any study specific procedures

Exclusion Criteria:

Current or previous (within 6 months) treatment with an incretin-based therapy such as DPP-4 inhibitors and/or GLP-1 mimetics Acute vascular (cardiac or stroke) event <2 months before randomization Initiation of chronic dialysis and/or renal transplant and/or a serum creatinine >6.0 mg/dL Pregnant or breastfeeding History of human immunodeficiency virus Patients being treated for severe autoimmune diseases such as lupus Any patient currently receiving long-term (>30 consecutive days) treatment with an oral steroid Patients with

• Body mass index >50 kg/m2
• Last measured HbA1c ≥12%
• Sustained BP >180/100 mm Hg
• LDL-C >250 mg/dL (>6.48 mmol/L) (based on the last measured and documented laboratory measurement in the previous 6 months) regardless of lipid-lowering therapy
• Triglycerides >1,000 mg/dL (N11.3 mmol/L) (based on the last measured and documented laboratory measurement in the previous 6months)
• HDL-C b25 mg/dL (<0.64 mmol/L) (based on the last measured and documented laboratory measurement in the previous 6 months)
• Known liver function tests >3 times upper limit of normal (ULN) (based on the last measured and documented laboratory measurement in the previous 6 months) Involvement in the planning and/or conduct of the study (applies to both AstraZeneca and Bristol-- Myers Squibb or representative staff and/or staff at the study site) Previous randomization in the present study Participation in another clinical study with an investigational product and/or intervention within 30 days before visit 1 Individuals at risk for poor protocol or medication compliance

This study will involve a new user, parallel group, cohort study design comparing saxagliptin to the 2nd generation sulfonylurea (SU) antidiabetic class as a proxy for placebo. SUs are not known to have an impact on the outcome of interest. In addition, SUs were the most frequent background treatment in SAVOR-TIMI53 (after metformin), and DPP4i and SUs are preferentially prescribed to similarly older patients in real world (Patorno et al., 2019). The patients will be required to have continuous enrollment during the baseline period of 180 days before initiation of saxagliptin or a comparator drug (cohort entry date). Follow-up for the outcome (3P-MACE), begins the day after drug initiation. As in the trial, patients are allowed to take other antidiabetic medications during the study.