Introduction: C. diff infection is known to be associated with increased colectomy rate, length of hospital stay, healthcare utilization costs, along with short and long-term mortality in IBD patients. The American College of Gastroenterology (ACG) mandated in 2013 that all patients presenting with a suspected IBD flare should be tested for C. diff. It was unclear to what extent this guideline was being followed at our institution and whether the timing of C. diff testing influenced the length of hospital stay.

Methods: Utilizing the ICD-9 diagnosis codes, we performed a retrospective chart review of patients admitted for IBD flare from January 2013 to December 2014 at our tertiary health center. Patients were noted if C. diff testing was performed. We compared the length of stay (LOS) of patients tested for C. diff with those not tested. We also determined if the LOS was affected by when C. diff testing took place (as when done early during hospitalization compared to after second day of admission).

Results: There were 110 index admissions from 51 patients meeting our criteria. C. diff testing was conducted in 59% (n=65) of the admissions for IBD flare. Incidence of C. diff was 11% (7 out of 65) where it was tested. The median LOS when C. diff testing was performed was 4 days, compared to 3 days when C. diff testing was not performed (p=0.3). When C. diff testing was conducted on day 1 or the same day of admission, the median LOS was 4 days; when it was conducted on ≥ day 2 of admission, the median LOS was 5 days (p= 0.2).

Conclusion: The rate of compliance with C. diff testing in patients admitted for IBD flare was relatively low, 59% at our institution. This could be attributed to lack of awareness of the latest ACG guidelines by ED and admitting physicians. The incidence of C. diff was 11% among those tested. The hospital LOS did not change when timing of C. diff testing was evaluated as a factor (day 1 of admission vs. ≥ day 2). There was also no statistical difference in LOS between those tested and those not tested for C. diff. This study, however, could be limited by small sample size. Further evaluation with enhanced power is warranted.