OBJECTIVE: To measure in-shoe foot pressures in diabetic patients and healthy subjects and compare them with the foot pressures when they walked without their shoes. RESEARCH DESIGN AND METHODS: Forty-four diabetic patients at risk of foot ulceration and 65 healthy subjects were matched for age, sex, race, and weight. Neuropathy was evaluated clinically, and the F-Scan program was used to measure the foot pressures. Foot pressures were measured with the sensors placed in the shoes (S measurements), between the foot and the sock with shoes (H measurements) or with their socks alone (B measurements). RESULTS: In the control group, significant differences were found between S (4.77 ± 1.87 kg/cm2) and H measurements (5.12 ± 1.87 kg/cm2, P < 0.001), between S and B (7.23 ± 2.95 kg/cm2, P < 0.0001), and between H and B (P < 0.0001). In the diabetic group, no difference was found between S and H measurements (5.28 ± 2.22 vs. 5.27 ± 2.39 kg/cm2, NS). In contrast, the B pressure was significantly higher when compared with both (8.77 ± 4.67 kg/cm2, P < 0.02). When compared with the control group, the S and H pressures did not differ significantly, but the B pressure in the diabetic group was significantly higher (P < 0.02). The peak S pressure was above the normal limit in 24 (27%) diabetic and 21 (16%) control feet (P < 0.05), the H pressure in 17 (19%) diabetic feet and 22 (17%) control feet (NS), and the B pressure in 24 (27%) diabetic and 21 (16%) control feet (P < 0.05). CONCLUSIONS: In-shoe foot pressure measurements are significantly lower than the ones measured when walking with the socks only in both diabetic patients and healthy subjects. The shoes of diabetic patients provided a higher pressure reduction than did those of the control group, but the number of feet with abnormally high pressures did not change. The F-Scan system may be particularly helpful in designing footwear suitable for diabetic patients with at-risk feet.