Excess mortality related to diabetes mellitus in elderly medicare beneficiaries

PURPOSE: To determine whether diabetes remains an important contributor to mortality among the elderly. METHODS: A 5% national sample of Medicare claims was utilized to create a retrospective cohort study. Participants were 148,562 persons with at least one hospitalization or two outpatient visits for diabetes in 1994, who were alive on January 1, 1995, were greater than or equal to65 years old, and were not in managed care in 1994; and 148,562 age, gender, and race matched controls without diabetes. Deaths were ascertained through 1999. Survival and proportional hazards analysis were used to calculate rates and relative risks. RESULTS: The mortality rate was 100.2/1000 person years (PY) among seniors with diabetes compared with 60.6/1000 PY without diabetes (age adjusted relative risk (RR) = 1.83,95% CI: 1.81-1.86). Non-white minorities had a similar RR associated with diabetes (Hispanic 2.37, Asian 1.95, Native American 2.38, blacks 1.64) as whites (1.86). The mortality risk associated with diabetes decreased with increasing age, but remained significantly elevated even among those aged 85 years and older. In contrast, the absolute excess mortality attributed to diabetes increased with age. CONCLUSION: The excess mortality associated with diabetes amongst all older ages suggests that greater attention to optimal diabetes treatment and prevention is needed. (C) 2004 Elsevier Inc. All rights reserved.