
Methods: We conducted a retrospective cohort study of all patients treated for CDI within the US Department of Veterans Affairs (VA) health care system between January 1, 2005 and December 31, 2012. A diagnosis of CDI was based on a positive laboratory result. Initial treatment was defined as treatment with metronidazole or oral vancomycin during the two days prior to or after the positive CDI test. Changes in rates over time were evaluated using Poisson regression. Vancomycin use was modeled as a function of patient and episode characteristics using logistic regression within a generalized estimating equations framework to account for clustering at the patient and facility levels. Episode severity was measured using the SHEA/IDSA criteria
Results: During the study period, 65,296 patients met the criteria for inclusion in the study. Of these, 9,197 (14.1%) received vancomycin either alone or in combination with metronidazole. The use of vancomycin increased significantly over time from 7.4% of treated episodes in 2005 to 21.9% of treated episodes in 2012 (p<.0001). Using SHEA/IDSA criteria, 52.3%, 37.2%, and 10.5% of episodes were classified as mild to moderate, severe, and unknown severity, respectively. Vancomycin was slightly but significantly more common among patients with severe CDI (15.4% vs. 12.7%, p<.0001). Predictors of vancomycin use included age at diagnosis, prior CDI episodes, hospital onset, severity, recurrent episode, and year
Conclusion: Vancomycin is used as the initial treatment in 15-20% of CDI cases. The use of vancomycin has increased since 2005, but it remains underutilized among patients with severe CDI. Targeted treatment of severe patients with vancomycin could be considered as a means of improving outcomes for patients with CDI

V. W. Stevens,
None
R. E. Nelson, None
D. Toth, None
M. Leecaster, None
M. Samore, None
M. Rubin, None