The objective of this study was to evaluate the relationship between foot ulceration and short-term mortality in veterans of the American military services with diabetes mellitus. A total of 725 diabetic subjects participated in a prospective study of risk factors for lower extremity complications between 1990 and 1994. Mean follow-up was 691.8 days (+/-SD 339.9, range 28-1436 days). Subjects who died during follow-up (n = 72) had a similar mean duration of diabetes to those who survived (12.6 years vs 11.2), but their mean age was greater (65.9 years vs 63.2, p = 0.026). The relative risk (RR) of death was 2.39 (95% confidence interval (CI) 1.13 to 4.58) in the subjects who developed foot ulcer (n = 88) compared to those who did not. The risk of death for those with foot ulcer was 12.1 per 100 person-years of follow-up compared to 5.1 in those without foot ulcer. Cox regression analysis demonstrated a greater than two-fold increased risk of death in ulcerated subjects after adjustment for age; diabetes type, duration, and treatment; glycosylated hemoglobin level; history of lower extremity amputation; and cumulative pack years smoked. Higher ankle-arm index was significantly related to lower mortality risk, independent of foot ulcer occurrence. We conclude that foot ulcer and lower extremity vascular disease are related to a higher risk of death in diabetic subjects. The reasons for this excess mortality require further investigation.