The objectives of the current study were to use the 2006 swine NAHMS samples to determine the prevalence, antimicrobial characterization, and genetic profile of E. coli O157:H7 in the U.S. swine herd and to determine the prevalence of Clostridium difficile in theU.S. swine herd. Fecal samples from swine farms from across the U.S. were sampled for E. coli O157:H7 using the same cultural methods that FSIS uses and for Clostridium difficile using two cultural procedures. A total of 1905 samples from 134 farms were sampled for E. coli O157:H7. Seven samples from one farm tested positive for E. coli O157:H7, but all were found not to have the toxin gene which would make them avirulent. A total of 864 samples from 60 farms were tested for C. difficile. Using the best available method at the time the study was started (single alcohol shock), 5.44% of the samples were found to be positive for C. difficile. However, about half way through the study a more sensitive method (double alcohol shock) was discovered and using this method 15.65% of 345 samples were shown to be C. difficile positive. Ribotyping suggests that no strains of C. difficile from this study are the same as the hypervirulent strain which has been causing problems in human patients in Canada and Europe. Further genetic characterization studies are continuing. The results from this study suggest that little to no E. coli O157:H7 is seen in fecal samples from U.S. swine. These results confirm the results from the 1996 NAHMS swine study where no E. coli O157:H7 were found. Further, C. difficile is found in about 15% of the swine fecal samples, but the strains present do not appear to be the hypervirulent strain that has recently caused so many hospital patient or community acquired problems.