To longitudinally assess fecal shedding and animal-to-animal transmission of Clostridium difficile among finishing feedlot cattle as a risk for beef carcass contamination, we tested 186 ± 12 steers (mean ± standard deviation; 1,369 samples) in an experimental feedlot facility during the finishing period and at harvest. Clostridium difficile was isolated from 12.9% of steers on arrival (24/186; 0 to 33% among five suppliers). Shedding decreased to undetectable levels a week later (0%; P < 0.001), and remained low (< 3.6%) until immediately prior to shipment for harvest (1.2%). Antimicrobial use did not increase fecal shedding, despite treatment of 53% of animals for signs of respiratory disease. Animals shedding C. difficile on arrival, however, had 4.6 times higher odds of receiving antimicrobials for respiratory signs than nonshedders (95% confidence interval for the odds ratio, 1.4 to 14.8; P = 0.01). Neither the toxin genes nor toxin A or B was detected in most (39/42) isolates based on two complementary multiplex PCRs and enzyme-linked immunosorbent assay testing, respectively. Two linezolid- and clindamycin-resistant PCR ribotype 078 (tcdA+/tcdB+/cdtB+/39-bp-type deletion in tcdC) isolates were identified from two steers (at arrival and week 20), but these ribotypes did not become endemic. The other toxigenic isolate (tcdA+/tcdB+/cdtB+/classic tcdC; PCR ribotype 078-like) was identified in the cecum of one steer at harvest. Spatio-temporal analysis indicated transient shedding with no evidence of animal-to-animal transmission. The association between C. difficile shedding upon arrival and the subsequent need for antimicrobials for respiratory disease might indicate common predisposing factors. The isolation of toxigenic C. difficile from bovine intestines at harvest highlights the potential for food contamination in meat processing plants.