Background.Comparisons of rapid antigen tests for the diagnosis ofStreptococcus pyogenespharyngitis are often made between published studies but may not be reliable given differences in study design, sampling and reference standard definitions. Tests are rarely compared against each other in a single study.Methods.The sensitivity and specificity of twoS. pyogenesrapid diagnostic tests were compared against a multiplate culture standard defined as the identification ofS. pyogeneson any of four culture plates. Paired swabs were tested forS. pyogenesantigen using Genzyme’s OSOM Ultra Strep A Test and compared with Biostar’s Strep A OIA Max Test.Results.Ninety-four (31.1%) of 302 matched samples were identified withS. pyogenes.The sensitivity of Genzyme’s OSOM Ultra Strep A Test against the multiplate culture standard was 92.6%, significantly higher (P= 0.001) than that (75.5%) of Biostar’s Strep A OIA Max Test. Specificities were 92.8 and 97.1%, respectively. Data analysis of culture results and statistical modeling showed that cultures of two or more samples are necessary for a sensitivity of 95% or greater for a comparison standard compared with true disease status.Conclusions.When comparing the performance of rapid antigen tests forS. pyogenespharyngitis, a rigorous culture standard should be used consisting of at least two separate samples (swabs and/or pledgets), ultimately plated on selective agar. Genzyme’s OSOM Ultra Strep A Test was significantly more sensitive than Biostar’s Strep A OIA Max Test in comparison with a multiplate culture standard and a same swab, single plate culture standard.