The objective of this research was to evaluate the effect of daily dietary inclusion of specific strains of Lactobacillus acidophilus on prevalence and concentration of Escherichia coli O157 in harvest-ready feedlot cattle. Five hundred yearling steers were housed in pens of 10 animals each. At arrival, steers were randomly allocated to one of five cohorts. Four of the cohorts were fed various strains and dosages of Lactobacillus-based direct-fed microbials throughout the feeding period. Fecal samples were collected from the rectum of each animal immediately prior to shipment to the abattoir. E. coli O157 was detected using selective enrichment and immunomagnetic separation techniques. For positive samples, E. coli O157 concentration was estimated using a most-probable-number (MPN) technique that included immunomagnetic separation. Prevalence varied among the cohorts (P < 0.01). The prevalence in the controls (26.3%) was greater (P < 0.05) than that in cattle supplemented with L. acidophilus strains NP51, NP28, or NP51-NP35 (13.0, 11.0, and 11.0%, respectively). The greatest E. coli O157 concentration was also observed in the controls (3.2 log MPN/g of feces); this concentration was greater (P < 0.05) than that observed in positive animals receiving NP51, NP28, or NP51-NP35 (0.9, 1.1, 1.7 log MPN/g of feces, respectively). Specific strains of Lactobacillus-based direct-fed microbials effectively reduced the prevalence and concentration of E. coli O157 in harvest-ready cattle, whereas others did not. When using direct-fed microbials to reduce carriage of E. coli O157 in cattle, it is important to select appropriately validated products.