OBJECTIVES:  A lower rate of hypoglycemia occurred in a Phase 3 trial in T2DM patients receiving saxagliptin compared to glipizide (sulfonylurea (SU)) added to metformin. The clinical trial included patient-reported and physician-reported hypoglycemia events.  The objective of this study was to compare the rates of hypoglycemia events that required medical attention in actual clinical practice.  METHODS: MarketScan health care claims data was used to identify patients who initiated saxagliptin or SU. During the 6 months prior to initiation, patients were required to have a prescription (Rx) for metformin and no saxagliptin, SU, or other diabetes Rx. Patients were followed for 4 months to assess the rates of hypoglycemia.  During follow-up, patients were required to have at least one more metformin Rx and one additional saxagliptin/SU Rx and no other diabetes Rx.  A hypoglycemia event was defined as a diagnosis of hypoglycemia on an outpatient or emergency room claim or principal diagnosis on a hospital claim or an outpatient glucagon injection.  SU patients were matched to saxagliptin patients (5 to 1) using propensity scores.  The rate ratio was further adjusted for covariates that were not balanced between the matched cohorts using a Poisson regression model. RESULTS:  There were 1,567 saxagliptin, 21,025 SU, and 7,835 propensity-matched SU patients. The rate of hypoglycemia in the saxagliptin cohort was 1.74 per 100 PY versus 6.68 in the SU cohort (rate ratio 0.31, 95% CI: 0.14 – 0.60) and 4.45 per 100 PY in the matched SU sub-group (rate ratio 0.39, 95% CI: 0.17 – 0.77). After multivariate adjustment, the rate ratio was 0.37 (95% CI: 0.19 – 0.74).  CONCLUSIONS: In real world practice, as was demonstrated in a randomized controlled trial, saxagliptin patients had a lower risk of hypoglycemia than SU when added to metformin.