Peripheral arterial disease (PAD) is a frequent complication of diabetes mellitus. In the first phase of the San Luis Valley Diabetes Study, diagnostic criteria for PAD were evaluated in 607 controls and 343 diabetics. Normal ranges, and the lowest 2.5 percentile of the distribution of ankle/arm systolic blood pressure ratios were derived from a non-diabetic subset of the population with a very low probability of PAD. From this subgroup, abnormal ankle/arm ratios were defined as less than: 0.94 at rest, 0.73 after exercise, and 0.78 after reactive hyperemia. Using these criteria, PAD was identified in 130 subjects from the study population of 950 (prevalence of 13.7%). In contrast, a history of intermittent claudication, or an absent pulse in the extremity were uncommon findings in the study population, and thus had a low sensitivity and positive predictive value for PAD diagnosed by vascular laboratory criteria. We conclude that vascular laboratory tests provide a useful, and objective means of determining the prevalence of PAD in a geographically-based population of diabetic and control subjects.