Objectives: The objectives of this 5-year retrospective investigation were threefold. (1) Among patients with dermatitis of the feet consistent with allergic contact dermatitis (ACD), what were the final diagnoses of those with dermatitis only on the feet and those whose foot dermatitis was accompanied by other cutaneous involvement? (2) Among those patients determined to have ACD, what were the relevant allergens? (3) Have the allergens in shoes in the United States changed as a consequence of modifications in footwear manufacture and style design? Methods: Of 704 patients patch-tested at the University of Kansas Medical Center with the North American Contact Dermatitis Group's standard allergen tray and/or the University of Kansas “shoe and rubber” tray, 70 patients presented with a clinical pattern suggestive of ACD of the foot. Results: Compared with those without foot dermatitis, these patients were more likely to be atopic and male with bimodal age distribution: <19 and 41 to 60 years. Despite clinical evidence suggesting allergy, only 23 (32.9%) patients had ACD to components of shoes, whereas 30 (42.9%) had psoriasis. Among the remaining patients, 4 (5.7%) had a non-shoe allergy, 3 (4.3%) had dyshidrosis, 2 (2.9%) had nummular dermatitis, 2 (2.9%) had tinea pedis, and 1 (1.4%) each had pityriasis rubra pilaris, juvenile plantar dermatosis, atopy, id reaction, and traumatic/frictional dermatitis as their primary diagnosis. One patient had not received a diagnosis when last seen. Conclusions: When evaluated by sites of involvement, psoriasis was diagnosed as often as ACD to shoes when the dermatitis was confined to the foot but was more frequent than ACD when both the hands and feet were involved. With respect to shoe dermatitis, rubber components were the principal allergens, followed by chromated leather and adhesives. Iatrogenic ACD was also common, and among all allergens, bacitracin caused the most frequently observed relevant reaction. (J Am Acad Dermatol 2002;47:715-21.)