This study compared oral and rectal administration of O157-specific bacteriophages for mitigating the fecal shedding of Escherichia coli O157 by experimentally inoculated steers. Fecal shedding of nalidixic acid-resistant (Nal(R)) E. coli O157:H7 was monitored over 83 days after oral (ORL; 3.3 x 10(11) PFU), rectal (REC; 1.5 x 10(11) PFU), both oral and rectal (O+R; 4.8 x 10(11) PFU), or no (CON; control) treatment with a four-strain O157-specific bacteriophage cocktail in multiple doses. Bacteriophages were enumerated by plaque assay, and NalR E. coli O157:H7 by direct plating on sorbitol MacConkey agar supplemented with cefixime, potassium tellurite, and nalidixic acid. Orally treated steers produced the fewest Nal(R) E. coli O157:H7 culture-positive samples (P < 0.06) compared with REC and O+R steers, but this number was only nominally lower (P = 0.26) than that for the CON steers. The overall mean shedding level (log CFU per gram of feces) was higher for REC steers (P < 0.10) than for steers of the other treatment groups. Despite the shedding of higher mean bacteriophage levels (log PFU per gram of feces) by ORL and O+R than by CON and REC steers, there was no difference (P > 0.05) in the number of E. coli O157-positive samples among treatments. Bacteriophage was isolated from CON steers, indicating that these steers acquired the bacteriophage from the environment and shed the phage at a level similar to that of REC steers (P = 0.39). Continuous bacteriophage therapy may be an efficacious method for mitigating shedding of E. coli O157:H7 in cattle, providing that the host bacterium does not develop resistance. This therapy may be especially advantageous if nontreated cattle can acquire this biocontrol agent from the feedlot environment.