The soybean protein isolate used in powdered soybean formula is hydrolyzed more extensively than the isolate which is used in liquid soybean formula in most commercial soybean formulas. Previous in vitro studies have shown differences in human antibody response to these soybean protein isolates. Therefore, a prospective clinical study was undertaken to determine if there were differences in adverse reaction rates to these soybean protein isolates. Forty-three patients with possible milk- and/or soy-protein enterocolitis were enrolled in this study. Patients had 3 separate oral food challenges; using milk formula, soybean powder formula and soybean liquid formula. Ten (23%) patients challenged with milk had positive challenges. Fourteen (33%) patients challenged with powdered soy formula had positive challenges while thirteen (30%) challenged with liquid soy formula had positive challenges. In the 10 patients with positive milk challenges, 6 (60%) had a positive soy challenge. In the group with positive soy challenges, 5 reacted to the powdered soy challenge done first, but not the second challenge with the liquid soy formula, and 4 patients reacted to the liquid soy formula challenge done first, but not the second challenge with the powdered soy formula. These results indicate that a significant number of patients with milk protein enterocolitis have soy protein enterocolitis. In addition, an order effect can be demonstrated in the soy challenges because of the tendency to react to the first soy challenge regardless of the type of isolate. These results suggest that a local immune effect caused by the protein may be present.