Original articleIncreasing incidence of serious hypoglycemia in insulin users
Introduction
Hypoglycemia occurs frequently in patients with diabetes mellitus treated with insulin therapy [1]. Beyond the milder symptoms, hypoglycemia can result in coma, seizure, injury to the patient and others, and death [2]. Patients' fears of hypoglycemia impact their quality of life and deter the pursuit of lower blood glucose levels 3, 4, which reduce the risk of microvascular complications 5, 6.
Physicians classify the severity of hypoglycemic episodes according to treatment requirements. Mild episodes of hypoglycemia can be treated by the person experiencing the episode provided that he or she is aware of the symptoms and acts promptly [1]. “Severe” or “major” episodes require the assistance of another person to treat the episode 6, 7. Shorr and colleagues used the term “serious hypoglycemia” to refer to episodes that required medical attention (i.e., an emergency department visit or hospitalization), often because of a loss of consciousness [8]. Our epidemiologic investigation concerns serious episodes only.
Most information on the burden of hypoglycemia is based on carefully screened and selected patients who were followed in randomized controlled trials with protocols 6, 7. The protocols and systematic patient monitoring in trials may not reflect disease management in the broader population. To address this gap in knowledge, Shorr and colleagues conducted a cohort study to measure the incidence of serious hypoglycemia in insulin users in the Tennessee Medicaid population who were 65 years of age or older [8]. Our cohort study builds on the work of Shorr and colleagues by reporting the incidence of serious hypoglycemia in patients 20 to 64 years of age who were enrolled in a health maintenance organization (HMO), Harvard Pilgrim Health Care.
Our primary study objectives were to calculate the population-based incidence of serious hypoglycemia (and the risk of recurrence), and to evaluate if the incidence changed over calendar time. Our secondary objective was to describe the positive predictive value of various International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) [9] codes that can identify potential episodes of serious hypoglycemia in healthcare insurance claims databases.
Section snippets
Cohort eligibility
We identified retrospectively a cohort of patients 20 to 64 years of age with diabetes at Harvard Pilgrim Health Care. Harvard Pilgrim Health Care offers a range of health insurance options to residents in the greater Boston area. During the mid-1990s, approximately 300,000 members belonged to Harvard Pilgrim Health Care's staff-model health maintenance organization (HMO), formerly known as Harvard Community Health Plan [10]. We restricted eligibility to the HMO members because we could access
Results
We identified 203 confirmed episodes of serious hypoglycemia in 1,113 patients who met our cohort inclusion criteria. Of the 1,113 patients, 129 (12%) experienced at least one episode of serious hypoglycemia, and 43 patients experienced at least one recurrent episode during a median follow-up of 3.5 years (4,005 total person-years). Every patient contributed at least one person-year of follow-up with a maximum of five years. Patients contributed a maximum of six episodes during follow-up. Table
Discussion
Our primary study objective was to calculate the population-based incidence of serious hypoglycemia and to evaluate if the incidence changed over time. We found an overall incidence of 5 per 100 person-years, which increased by approximately 24% per year. Ninety-four percent of episodes were treated exclusively in the emergency department. In the year following their first episode of serious hypoglycemia, 22% of patients experienced a recurrent episode. The positive predictive value of ICD-9-CM
Acknowledgements
Several people and organizations contributed to this research and we are grateful for their assistance. Harvard Pilgrim Health Care (HPHC) provided the data, and the Channing Laboratory at Brigham and Women's Hospital provided the computer for analyzing the data. Ms. Emily Cain of HPHC supervised the data extraction. Mr. Jim Livingston of the Channing Laboratory provided expert programming to link the cohort records. Dr. Peter Choo of the Channing Laboratory explained the nuances of the
References (23)
- et al.
Hypoglycemia associated with use of inhibitors of angiotensin converting enzyme
Lancet
(1995) - et al.
Analytic strategies for recurrent events in epidemiologic studiesbackground and application to hospitalization risk in the elderly
J Clin Epidemiol
(2000) - et al.
Hypoglycemia
Diabetes Care
(1994) Adverse events and their association with treatment regimens in the Diabetes Control and Complications Trial
Diabetes Care
(1995)The psychological care of patients with insulin-dependent diabetes mellitus
N Engl J Med
(1996)Banting lecture. Hypoglycemiathe limiting factor in the management of IDDM diabetes
Diabetes
(1994)The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus
N Engl J Med
(1993)Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33)
Lancet
(1998)Hypoglycemia in the diabetes control and complications trial
Diabetes
(1997)- et al.
Incidence and risk factors for serious hypoglycemia in older persons using insulin or sulfonylureas
Arch Intern Med
(1997)