Abstract: The microscopic detection of cryptosporidiosis is difficult, time-consuming and relatively insensitive, especially in patients with Iight infections. New techniques based on specific antibodies to Cryptosporidium antigens have become available. One method, an indirect immunofluorescence assay, has increased sensitivity over conventional microscopy. A more recent technique, an antigen detection ELISA, has not been adequately evaluated in a clinical setting. The purpose of this study was to compare the antigen detection ELISA to the quantitative IFA (QIFA) using stool samples from HIV-infected patients with diarrhea. The antigen detection test had a high degree of repeatability with a coefficient of variation of 0.08. This method compared very favorably with the QIFA and yielded a sensitivity of 96.4% and a specificity of 100%. The two tests gave comparable results when high numbers of oocysts were present. In contrast, the antigen detection assay was often positive for QIFA-negative samples from patients intermittently shedding oocysts. Hence, in the lighter infections, the antigen detection ELISA appeared to be more sensitive. The ELISA absorbance value was proportional to the number of oocysts present in the test as defined by QIFA. Antigen detection ELISA appears to be an effective alternative to immunofluorescence for detecting fecal oocyst excretion and may be particularly useful when large numbers of stools must be tested.