Methods: In this quasi-experimental study, retrospective chart review was conducted on inpatients diagnosed with CDI, defined using a tiered testing algorithm. Those diagnosed with CDI in 2013 served as controls before BPA implementation and patients from 2016 served as cases. Antibiotic prescribing was assessed and guideline compliance evaluated based on institutional guidelines for CDI treatment. Secondary endpoints were resolution of diarrhea, length of stay, in-hospital mortality, 30-day recurrence and readmission rate. Continuous variables were analyzed using a two-tailed Student’s T-Test, or for non-normally distributed data, Mann-Whitney U. Categorical variables were analyzed using Chi-square.
Results: Based on power analysis, 131 CDI cases were randomly selected, 66 in 2013 and 65 in 2016, which accounts for 23.7% (66/278) and 15.9% (65/409) of total inpatient CDI cases, respectively. Mean age was 55.0 ± 19.3 years pre-BPA/Order set and 58.5 ± 14.1 in the post- group. Immuncompormise was present in 53% of the pre-group as compared to 32.3% in the post-group. The majority of patients in both the pre-BPA/Order set group and post- group received metronidazole as initial therapy with 69.7% and 75.4%, respectively. The BPA was opened in 54% (28/57) of triggered encounters and led to signed orders in 82% (23/28) of those patients. Guideline-based prescribing increased from 39.4% in 2013 to 67.7% overall in 2016 (p=0.014). Secondary endpoints were not significantly different between groups.
Conclusion: After implementation of the BPA linked to a severity-based treatment order set there was an increase in guideline-compliant prescribing for CDI. Developing a better understanding of how to optimize guideline adherence using BPAs will aid ASPs in determining future improvement efforts.
C. Reha, None
S. Bergman, Merck: Grant Investigator , Grant recipient .
See more of: Poster Abstract Session