350. Validity of Recommended Criteria for Attributable Mortality Associated with Clostridium difficile Infection
Session: Poster Abstract Session: Clostridium difficile - Epidemiology, Diagnosis, Treatment, and Prevention
Friday, October 21, 2011
Room: Poster Hall B1
Background: Previously published studies have used multiple, non-standardized definitions for determining attributable mortality associated with Clostridium difficile infection (CDI).  The purpose of this study was to identify the most common attributable mortality definitions based on these studies and apply these definitions to hospitalized patients at a large university hospital with CDI to determine their prediction of mortality.

Methods: A review of articles reporting mortality associated with CDI identified three definitions for attributable CDI associated mortality.  These 3 definitions were applied to an ongoing observational study enrolling patients with CDI from 2007-2010.  Medical charts and death summary data of the patients were reviewed to assess attributable mortality.  Attributable mortality definitions in the literature were as follows:  1. clinical signs and symptoms of CDI (temperature >100.9, white blood cell count of >15,000 cells/mm3, diarrhea, or radiologic evidence of pseudomembranous colitis), and death occurred within 30 days of CDI diagnosis; 2. clinical signs and symptoms of CDI and death occurred from 48 hours to 14 days after the initial positive stool; 3. judgment by the investigator that the patient would not have died within 30 days with the absence of CDI.

Results: During the three year study period, 47 of 231 (20%) patients died within 90 days, and 9 of 231 (4%) died after 90 days to 444 days after CDI positive toxin assay.  The median time from CDI toxin positive to death was 47.5 days.  Out of the 47 patients that died within 90 days, 21 (45%) were CDI-associated based on one of the three definitions. Using definition 1, 18 of 47 (38%) deaths were defined as CDI-attributable mortality.  Using definition 2, 0 of 47 (0%) were defined as CDI-attributable mortality.  Using definition 3, 3 of 47 (6%) were defined as CDI-attributable mortality.  CDI-attributable mortality was significantly different based upon the definition used (p<0.001).

Conclusion: CDI associated mortality ranged from 0-38% in this university-based hospital setting based upon different published criteria.  Literature review CDI-associated mortality rates range from 7%-48%.  A standardized and validated attributable mortality definition is needed for CDI.

Subject Category: N. Hospital-acquired and surgical infections, infection control, and health outcomes including general public health and health services research

Carolyn Grimes, DrPH, BSMT (ASCP)1, Dhara Shah, PharmD2, Hoonmo Koo, MD3, Herbert DuPont, MD4 and Kevin W. Garey, PharmD, M.S.2, (1)University of Texas School of Public Health, Houston, TX, (2)University of Houston College of Pharmacy, Houston, TX, (3)Baylor College of Medicine, University of Texas-Houston School of Public Health, Houston, TX, (4) University of Texas School of Public Health, Houston, TX


C. Grimes, None

D. Shah, None

H. Koo, None

H. DuPont, None

K. W. Garey, None

Findings in the abstracts are embargoed until 12:01 a.m. EST Thursday, Oct. 20 with the exception of research findings presented at IDSA press conferences.