Severe Hypoglycemia Requiring Medical Intervention in a Large Cohort of Adults With Diabetes Receiving Care in U.S. Integrated Health Care Delivery Systems: 2005-2011

Abstract

Objective: Appropriate glycemic control is fundamental to diabetes care, but aggressive glucose targets and intensive therapy may unintentionally increase episodes of hypoglycemia. We quantified the burden of severe hypoglycemia requiring medical intervention in a well-defined population of insured individuals with diabetes receiving care in integrated health care delivery systems across the U.S.

Research design and methods: This observational cohort study included 917,440 adults with diabetes receiving care during 2005 to 2011 at participating SUrveillance, PREvention, and ManagEment of Diabetes Mellitus (SUPREME-DM) network sites. Severe hypoglycemia rates were based on any occurrence of hypoglycemia-related ICD-9 codes from emergency department or inpatient medical encounters and reported overall and by age, sex, comorbidity status, antecedent A1C level, and medication use.

Results: Annual rates of severe hypoglycemia ranged from 1.4 to 1.6 events per 100 person-years. Rates of severe hypoglycemia were higher among those with older age, chronic kidney disease, congestive heart failure, cardiovascular disease, depression, and higher A1C levels, and in users of insulin, insulin secretagogues, or β-blockers (P < 0.001 for all). Changes in severe hypoglycemia occurrence over time were not clinically significant in the cohort as a whole but were observed in subgroups of individuals with chronic kidney disease, congestive heart failure, and cardiovascular disease.

Conclusions: Risk of severe hypoglycemia in clinical settings is considerably higher in identifiable patient subgroups than in randomized controlled trials. Strategies that reduce the risk of hypoglycemia in high-risk patients are needed.

Figure 1

Quarterly rates of severe hypoglycemic events by the level of glycemic control and medication exposure (2005–2011). Quarterly rates of severe hypoglycemic events age- and sex-standardized to the 2010 U.S. Census population for various levels of hemoglobin A1C (A), exposure to glucose-lowering medications (B), and β-blocker exposure (C). For each calendar quarter, A1C and medication exposure were determined in the preceding quarter. Categories labeled as short-acting insulin and long-acting insulin also include basal-bolus and premixed insulin users.