LAUSR

OBJECTIVES Clostridioides difficile infection (CDI) represents a challenging issue, with an evolving epidemiology. Main objectives of our study were: - to assess the frequency of diarrhea of overall etiology, including CDI, as a cause of hospital admission or occurring during hospital stay;- to determine the rate of underdiagnosis of community-acquired (CA-), health care associated (HCA)- and hospital onset (HO-) CDI, and explore factors associated with its clinical suspicion by physicians. METHODS A prospective cohort study included all hospitalized patients with diarrhea at two acute-care hospitals. C. difficile (CD) tests were performed on every stool samples, irrespective of the treating physician request. Factors associated with the likelihood of CD test request by physicians were assessed. RESULTS We enrolled 871 (6%) patients with diarrhea. CD test performed on all diarrheic stool samples was positive in 228 cases (26%); 37, 106, 85 cases of CA- (14%), HCA- (42%) and HO- diarrhea (24%), respectively. Treating physicians did not request CD test in 207 (24%) diarrhea cases. The rate of CDI underdiagnosis was 11% (24/228); it was higher in CA-CDI (27%, 10/37). Logistic regression analysis identified age >65 years (RR 1.1; 95 CI 1.06-1.2) and hospitalizations in the previous 3 months (RR 1.2; 95% CI 1.1-1.3) as independent factors associated with the likelihood of requesting the CD test by the physician. These risk factors differed by epidemiological classification of diarrhea and by hospital. CONCLUSIONS Our study confirmed the relevance of CDI underdiagnosis and provided new insights in the factors underlying the lack of CDI clinical suspicion.