Visitation of hospitalized people by dogs is becoming commonplace, but little is known about the potential health risks of introducing dogs to healthcare settings. This cross-sectional study evaluated the prevalence of zoonotic agents in a group of 102 visitation dogs from a variety of sources across Ontario. Between May and July 2004, owners were interviewed by a standardized questionnaire while dogs underwent a standardized physical examination. One specimen of faeces, hair-coat brushings and one rectal, aural, nasal, oral and pharyngeal swab were collected from each dog and tested for 18 specific pathogens. All dogs were judged to be in good health. Zoonotic agents were isolated from 80 out of 102 (80%) dogs. The primary pathogen was Clostridium difficile, which was isolated from 58 (58%) faecal specimens. Seventy-one percent (41/58) of these isolates were toxigenic. Extended-spectrum beta-lactamase Escherichia coli was isolated from one (1%) dog, extended-spectrum cephalosporinase E. coli was isolated from three (3%) dogs, and organisms of the genus Salmonella were isolated from three (3%) dogs. Pasteurella multocida or Pasteurella canis was isolated from 29 (29%) oral swabs, and Malassezia pachydermatis was isolated from eight (8%) aural swabs. Giardia antigen was present in the faeces of seven (7%) dogs, while Toxocara canis and Ancylostoma caninum were detected in two (2%) dogs and one (1%) dog, respectively. Methicillin-resistant Staphylococcus aureus, vancomycin-resistant enterococci, Campylobacter spp., Microsporum canis, group A streptococci, Pseudomonas aeruginosa and Cryptosporidium spp. were not detected. Further information is needed before the full implications of these findings for infection control can be assessed properly.