Risk factors for Clostridium difficile-associated diarrhea on an adult hematology-oncology ward

Abstract

Nosocomial diarrhea caused by Clostridium difficile causes significant morbidity and mortality in an increasing proportion of hospitalized patients annually. This case-control study of patients admitted to the hematology-oncology ward of a tertiary academic medical center over a 2-year period demonstrates that patients with Clostridium difficile-associated diarrhea (CDAD) were 22 times more likely than ward-matched controls with diarrhea to have received any antibiotic either during hospitalization or in the month preceding admission (p < 0.005), and they were nearly three times as likely as controls to have received a cephalosporin during the same period (p < 0.005). Diarrhea among lung cancer patients was approximately three times more likely to be caused by this organism than to be due to other causes (p = 0.04). A trend towards CDAD patients receiving higher numbers of different antibiotics during hospitalization (3.3 vs. 2.6, 95%CI -1.42-0.02, p = 0.06) was noted. Administration of interleukin-2 either during hospitalization or in the 30 days preceding admission was seven times more likely to have occurred in CDAD cases (p = 0.04), raising the question of whether or not this agent increases risk.