2425. Correlating Use of High-Risk Antimicrobials and the Incidence of Hospital-Onset Clostridium difficile Infection: Targeting Prescribing Trends for Antimicrobial Stewardship

BACKGROUND: Hospital-onset Clostridium difficile infection (HO-CDI) has a significant morbidity and mortality risk. It also poses increasing financial strain on the healthcare system. Certain antibiotics have been associated with increased HO-CDI incidence and novel strategies are needed to determine what modifiable risk factors exist. Choices of antibiotic have changed overtime time to overcome potential side effects, leading to a possibility that changed prescribing trends could be linked to significant differences in the rate of HO-CDI. METHODS: This study took place at a 971-bed community hospital from January 2016 to January 2018. Monthly utilization (grams) of 11 antimicrobials considered high risk of HO-CDI was collected, along with monthly HO-CDI rate. Antimicrobials included cephalosporins, carbapenems, fluoroquinolones and clindamycin. Correlational (Pearson’s) and logistic regression analyses were completed to identify association with HO-CDI. A P-value of < 0.05 was considered statistically significant. RESULTS: 215 cases of HO-CDI were identified during the study period with 30 being classified as severe. The average HO-CDI rate was 4.3 cases/1000 patient-days. There were no significant correlations identified for any antimicrobials and HO-CDI rate (p> 0.05 for all interactions). Pearson’s correlation coefficients were not significant for any antimicrobial. The multivariable logistic regression model including all antimicrobials, indicated that only ceftazidime had a statistically significant positive effect on the HO-CDI rate. Bearing in mind that only a small number of ceftazidime was prescribed, additional univariate analysis was performed indicating that there was no significant linear association between the HO-CDI rate and ceftazidime utilization (P = 0.3527). CONCLUSION: Our study shows that there is no significant correlation between specific antimicrobial use and HO-CDI rates, even though there has been a general increase in HO-CDI rates. Additional analysis involving control groups of antibiotic use in patients without HO-CDI as well as incidence of HO-CDI in patients without antibiotic use at all is required to further assess possible modifiable risk factors in the inpatient population. [Image: see text] [Image: see text] DISCLOSURES: All authors: No reported disclosures.