Geographic Variation of Lower-Extremity Major Amputation in Individuals With and Without Diabetes in the Medicare Population

Abstract

OBJECTIVE—To describe geographic variation in rates of lower-limb major amputation in Medicare patients with and without diabetes. RESEARCH DESIGN AND METHODS—This cross-sectional population-based study used national fee-for-service Medicare claims from 1996 through 1997. The unit of analysis was 306 hospital referral regions (HRRs) representing health care markets for their respective tertiary medical centers. Numerators were calculated using nontraumatic major amputations and the diabetes code (250.x) for individuals with diabetes. Denominators for individuals with diabetes were created by multiplying the regional prevalence of diabetes (as determined using a 5% sample of Medicare Part B data identifying at least two visits with a diabetes code for 1995–1996) by the regional Medicare population. Denominators for individuals without diabetes were the remaining Medicare beneficiaries. Rates of major amputations were adjusted for age, sex, and race. RESULTS—Rates of major amputations per year were 3.83 per 1,000 (95% CI 3.60–4.06) individuals with diabetes compared with 0.38 per 1,000 (95% CI 0.35–0.41) individuals without diabetes. Marked geographic variation was observed for individuals with and without diabetes; however, patterns were distinct between the two populations. Rates were high in the Southern and Atlantic states for individuals without diabetes. In contrast, rates for individuals with diabetes were widely varied. Variation across HRRs for individuals with diabetes was 8.6-fold compared with 6.7-fold in individuals without diabetes for major amputations. CONCLUSIONS—Diabetes-related amputation rates exhibit high regional variation, even after age, sex, and race adjustment. Future work should be directed to exploring sources of this variation.