eral days after completing therapy, she experienced diarrhea, and stool samples demonstrated C difficile toxin by cytotoxicity assay. Her symptoms responded to a 10-day course oforal metronidazole. Since her initial episode, she has had a total of 6 subsequent episodes ofdiarrhea (several ofwhich were docu¬ mented to be related to C difficile). She is currently undergo¬ ing experimental therapies for the treatment of chronic, re¬ lapsing C difficile diarrhea.In each of the cases, a member of the medical house staff self-diagnosed an infection and began therapy with antibiotics that were provided as samples for patient use. In 2 of the patients, complications occurred that resulted in the loss of work time.Although C difficile can be acquired in the community,2 medical personnel are in frequent contact with patients with C difficile, and it is known that the spores of the organism can persist in the hospital environment for prolonged periods.35 Furthermore, many medical personnel have ready access to antibiotics in the form of pharmaceutical samples, and selfdiagnosis and treatment are, in all likelihood, highly prevalent. Our cases of C difficile diarrhea affecting medical personnel illustrate how these circumstances can combine to produce C difficile diarrhea, even in otherwise healthy, nonhospitalized individuals.