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 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 through 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.

Inclusion Criteria:

• Diagnosis of Type 2 Diabetes mellitus [0,0] days
• 3 or more recorded HbA1c values [-365, 0] days
• At least two dispensing claims for metformin [-180,0] days
• 1 or more recorded HbA1c values [-90,0] days
• Age between 18-80 [0,0] days

Exclusion Criteria:

• Cancer, malignant neoplasm, or MEN2 [-1,825,0] days
• Use of other anti-diabetic medications [-180,0] days
• Cirrhosis or Hepatic decompensation [-180,0] days
• Viral hepatitis B or C [-180,0] days
• Stage 5 CKD, ESRD, dialysis or renal transplant [-180,0] days
• MI or HF [-180,0] days
• Treatment with Anti-VEGFi or Photocoagulation [-180,0] days
• Hypertensive crisis or uncontrolled hypertension [-180,0] days
• Two diagnoses of hypoglycemia [-180,0] days
• Alcohol/drug abuse/dependence [-180,0] days
• Dementia and brain damages or non-compliance [-180,0] days
• Pregnancy [-180,0] days
• Type 1 DM, DKA, or other metabolic acidosis [-180,0] days
• Systemic corticosteroids [-90,0] days

This study will involve a new user, parallel group, propensity score-matched, retrospective cohort design comparing injectable subcutaneous liraglutide (1.2 or 1.8 mg/day) to oral glimepiride (4 mg once daily). Treatments in both arms are administered in combination with metformin (up to 2,000 mg/day). Patients will be required to have continuous enrollment during a baseline period of 180 days before initiation of liraglutide or glimepiride. The analyses will be restricted to individuals with type 2 diabetes mellitus who have been previously treated with antidiabetic drugs. In the replication, previous anti-diabetic treatment was defined as the presence of at least 2 prescriptions for metformin within 6 months before and including cohort entry.