Abstract P146: Diabetes, Prediabetes and Risk of Hospitalization: the Atherosclerosis Risk in Communities (ARIC) Study
Introduction: Diabetes represents a significant proportion of healthcare costs. The magnitude of risk and types of hospitalizations that occur in persons with prediabetes and diabetes are not well characterized.
Hypothesis: We hypothesized that persons with prediabetes and diabetes would have higher rates of hospitalization, especially from cardiovascular causes, than those without diabetes, and that, in addition to hyper-/hypoglycemia, infection would represent a substantial burden of hospitalizations in persons with diabetes.
Methods: Prospective analysis of 13,522 participants in the ARIC Study followed for a median of 20 years (1990-2011) for hospitalizations (primary ICD-9 code defined cause). Self-reported history and clinical cut-points of HbA1c were used to define diabetes status at baseline. Negative binomial models were used to calculate demographic-adjusted rates of hospitalization with tests for interaction by age, sex, and race.
Results: Mean age was 57 years, 56% were female, 24% were black. Persons with diabetes and HbA1c ≥7% had the highest rates of hospitalization (3.1 times higher than those without diabetes; 1.5 times higher than those with diabetes and HbA1c <7%). Persons with prediabetes had 1.3 times higher rates of hospitalization than those without diabetes (Table). Rates of hospitalization by diabetes status were higher for blacks vs. whites (p-interaction=0.01) and men vs. women (p-interaction=0.02); there were trends for higher rates among and older vs. younger (p-interaction=0.14). Among those with diabetes, cardiovascular causes accounted for the highest proportion of hospitalizations (45 vs. 39% among those without diabetes), with hyper-/hypoglycemia and infection causes accounting for 12 and 10% of hospitalizations, respectively.
Conclusions: Persons with diabetes and prediabetes are at a high risk of hospitalization, but a significant proportion of hospitalizations were for hyper-/hypoglycemia and infection, which may be preventable with improved glycemic control.
Author Disclosures: A.L.C. Schneider: None. H. Yeh: None. R.R. Kalyani: None. S.H. Golden: None. S.C. Stearns: None. L. Wruck: None. J. Coresh: None. E. Selvin: None.
- © 2015 by American Heart Association, Inc.