Glycemic control is fundamental to diabetes care, but aggressive glucose targets and intensive therapy can result in increased episodes of hypoglycemia. We sought to quantify the burden of severe hypoglycemia in an observational cohort study (2005 - 2011) including insured subjects with diabetes from the multi-institutional SUrveillance, PREvention, and ManagEment of Diabetes Mellitus (SUPREME-DM) study. Severe hypoglycemia was defined as any occurrence of a hypoglycemia-related ICD-9 code extracted from an emergency department or inpatient encounter. We examined the rate of severe hypoglycemic events overall and by age, gender, comorbidity status, antecedent A1c, and medication use. A total of 917,440 subjects with diabetes and at least one year of follow-up (mean 4.7 years follow-up) were examined, of which 42,378 (4.6%) experienced at least one severe hypoglycemic event during the 7-year study period. The population had slightly more males than females, and a mean age of 57.9 years. Annual age- and sex-adjusted rates of severe hypoglycemia tended to decrease over time, ranging from 13 - 16 events per 1000 person-years with an overall weighted rate during the study period of 14.7 events per 1000 person-years. Higher rates of severe hypoglycemia were associated with older age, various comorbidities, higher A1c levels, and use of insulin, insulin secretagogues, or beta blockers. The higher rates of hypoglycemia observed here compared to randomized controlled trials suggest that those who are enrolled in trials may be at lower risk of severe hypoglycemia than the broader population of individuals with diabetes either due to trial eligibility or focused clinical attention. Given the higher rate of severe hypoglycemia observed in certain groups, greater attention to management guidelines in populations at high risk for hypoglycemia, including those on insulin therapy and the elderly, may be necessary.