OBJECTIVE To determine the period prevalence of Clostridium difficile disease and asymptomatic carriage in the residents of long-term care facilities (LTCF) and to characterize the risk factors for colonization or associated disease. METHODS Period prevalence survey. METHODS Two long-term care facilities in St. Paul, MN. METHODS Specimens were collected from 225 LTCF residents. METHODS The dependent variable was the culture result for C. difficile, which was isolated and identified using selective culture media and a commercial anaerobe identification kit. Tissue culture assay was used to detect the ability of each C. difficile isolate to produce toxin. Independent variables (including gender, age, race, current medical diagnoses, severity of underlying disease, case mix, current clinical symptoms, current medications, antibiotic use within 4 weeks prior to specimen procurement, and other pertinent history) were obtained from the current medical record of each participant. RESULTS Of 225 stool cultures that were obtained, 16 (7.1%) were positive for C. difficile. None of the residents with a positive culture was symptomatic. History of nosocomial infection and the use of antibiotics in general, cephalosporins, trimethoprim/sulfamethoxazole (TMP/SMX), and histamine-2 blockers were significantly associated with positive C. difficile culture (P < or = 0.05) by univariate analyses. Trends towards significance (0.05 < 0.10) were noted for narcotic use, previous hospitalization, LTCF, and non-insulin-dependent diabetes mellitus. Logistic regression analysis revealed significant, independent predictors of positive culture: antibiotic use in general (P = 0.028; relative risk = 3.31), histamine-2 antagonist use (P = 0.038; relative risk = 3.27), cephalosporin use (P = 0.038; relative risk = 4.66), and TMP/SMX use (P = 0.007; relative risk = 8.45). CONCLUSIONS The use of antibiotics, particularly cephalosporins and TMP/SMX, is a significant risk factor for asymptomatic carriage of C. difficile in long-term care facilities. The use of H-2 blockers was also a significant risk factor for carriage; however, this finding has not been reported previously and should be confirmed by independent studies. These medications should be used judiciously in the LTCF population. When diarrheal diseases are encountered in LTCF residents, a high index of suspicion for C. difficile infection should be maintained and the appropriate diagnostic and therapeutic measures taken.