Facility-Level Variation in Major Leg Amputation Among Patients With Newly Diagnosed Diabetic Foot Ulcer

The prevalence of diabetes is increasing over time, fueling an epidemic of diabetic foot ulcers (DFUs) and subsequent risk of leg amputation. However, little is known about the variation in outcomes for patients with DFUs according to the health care facilities treating them.

To examine facility-level variation in major leg amputation among veterans with incident DFUs using the Veterans Health Administration (VHA) cohort.

A retrospective cohort study was conducted from January 1, 2016, to December 31, 2021, of all veterans with a new diagnosis of DFU at 140 VHA facilities across the US. Patients were followed up to 1 year from DFU diagnosis. Analyses were conducted between March 22, 2024, and January 13, 2025.

A facility was assigned to each patient corresponding to the health care site where the initial DFU diagnosis was made.

The primary outcome was major leg amputation during the follow-up period. A multivariable mixed-effects regression model with random facility intercepts was applied to assess variation in major leg amputation rates across facilities, adjusting for social drivers of health, comorbidities, and complicated DFU at initial diagnosis. The median odds ratio (MOR) was calculated to quantify facility-level variation in outcomes.

A total of 86 094 veterans (98.3% male; mean [SD] age, 73.0 [8.1] years; age range, 55-102 years) were included. Major leg amputation was performed for 3279 veterans (3.8%) within a year of DFU diagnosis. The MOR for facility-level variation in major leg amputation was 1.85, indicating that the odds of major leg amputation were 1.85 times higher between 2 randomly selected facilities for an average patient (P < .001). In contrast, the MOR for facility-level variation in 1-year mortality was 1.16 (P < .001).

This cohort study of veterans with newly diagnosed DFU found significant facility-level variation in major leg amputation rates within 1 year of DFU diagnosis. Facility-level variation in 1-year mortality rates was much smaller, suggesting variation in leg amputation was likely to stem from variation in DFU-specific care. The VHA should strive to minimize the odds of major leg amputation and interfacility variation.