Background: Nosocomial diarrhea is a common problem in hospitals, but the epidemiology, causes, and consequences of infectious and iatrogenic nosocomial diarrhea are rarely documented in follow-up or surveillance studies of nosocomial infections. The objective of this study was to describe the epidemiology of nosocomial diarrhea in a cohort study of hospitalized patients.
Methods: The setting was a general medicine ward with a total of 30 beds in private or semiprivate rooms in a county hospital. A prospective cohort study of all consenting patients in the study ward was performed during an 11-month period. Patients were interviewed and cultures were obtained to check for stool pathogens within 48 hours of admission, and patients were followed up during their stay in the study ward. Data from medical notes, laboratory tests, antibiotic logs, and patient interviews were collected daily.
Results: Among the 382 eligible patients, nosocomial diarrhea developed in 126 (32.9%): 37 cases were of infectious origin, 57 were iatrogenic, and 32 were of unknown origin. The most common causes of nosocomial diarrhea were Clostridium difficile and antibiotic therapy. Multivariate analysis revealed four risk factors for infectious nosocomial diarrhea: age, length of stay, number of antibiotics, and nasogastric tube feedings. Multivariate analysis also revealed four risk factors for iatrogenic nosocomial diarrhea: number of antibiotics, current gastrointestinal conditions, current renal conditions, and recent surgery. Comorbidity associated with nosocomial diarrhea included increased risk of acquiring another type of nosocomial infection, increased risk of death, and increased length of stay.
Conclusions: Nosocomial diarrhea was found to be common and associated with an additional burden of comorbidity. Analyses indicated different risk factors, depending on the etiology of nosocomial diarrhea (infectious or iatrogenic). Further studies are recommended to document the epidemiology and cost analysis of nosocomial diarrhea in other hospital populations.