OBJECTIVES: Injectable pen therapy with insulin glargine (GLA-P) or the glucagon-like peptide 1 (GLP-1) analog liraglutide (LIRA) may be initiated by type 2 diabetes mellitus (T2DM) patients failing oral antidiabetic therapy. Real-world treatment patterns among T2DM patients initiating these drugs were assessed using electronic medical record (EMR) data in this retrospective study. METHODS: The study included adult T2DM patients who initiated GLA-P or LIRA at seven US endocrinology practices. EMR data for ≥4 months before treatment initiation (baseline), baseline A1C and weight data, and ≥1 additional A1C measurement were available; patients had ≥2 visits during a minimum of six months follow-up. Baseline characteristics and clinical outcomes were assessed descriptively at the end of 1-year follow-up. Patients were stratified by whether they were injectable-naive (IN) or injectable-experienced (IE) at initiation. RESULTS: The study included 440 patients. At baseline, GLA-P patients (n=117), compared with LIRA patients (n=323), were older, had higher baseline A1C, and lower weight; more were male, fewer IE (60.7% vs 65.3%). Insulin/GLP-1 were the only injectable in 49.3%/31.3% of LIRA IE vs 62.0%/26.8% of GLA-P IE patients at baseline. Through follow-up, 88% of GLA-P and 91.6% of LIRA patients remained on study treatment. A1C was significantly reduced in all four treatment groups (range: −0.42% for LIRA IE to −1.38% for GLA-P IN). Weight loss/maintenance occurred in 74.2% of LIRA vs 41.3% of GLA-P patients. Mean change in weight/BMI was –2.5kg/–0.86kg/m2 for LIRA vs 2.4kg/0.85 kg/m2 for GLA-P patients. No severe hypoglycemic events were reported. CONCLUSIONS: Our data identify a trend in the real-world prescribing of LIRA and GLA-P to T2DM patients with considerably different clinical profiles. This introduces challenges when conducting real-world comparative effectiveness studies. Due to the small sample size and descriptive nature of this study, these data should be confirmed by future large scale studies.