Interrupted time series analysis of inpatient C. difficile treatment in response to the 2017 and 2021 SHEA/IDSA updates using Oracle EHR Real-World Data

Abstract Purpose To determine the prescribing patterns of metronidazole, vancomycin, and fidaxomicin for inpatients with Clostridioides difficile infection (CDI) in response to the 2021 Society for Healthcare Epidemiology of America (SHEA)/Infectious Diseases Society of America (IDSA) update. Methods We evaluated metronidazole, vancomycin, and fidaxomicin use utilizing Oracle EHR Real-World Data (2014-2024). Utilization data were measured by treatment courses, defined as 10 or more days of use or use on the last day of hospitalization. To determine if there was a statistically significant change in the use of each drug by CDI severity before versus after the SHEA/IDSA guideline updates, interrupted time-series analyses were conducted using multinomial logistic regression to model the probability of being assigned specific treatment and evaluate prescribing patterns over time, with vancomycin designated as the reference. Covariates included a continuous time variable, intervention indicators for both the 2017 and 2021 updates, and time by intervention interactions for both updates. Results A total of 70,200 CDI treatment courses were included (38,357 for nonsevere and 31,843 for severe disease). Metronidazole use dropped precipitously after 2017 for nonsevere and severe CDI. Fidaxomicin use was only statistically significantly increased for nonsevere CDI with the 2021 event (odds ratio [OR], 1.80; 95% CI, 1.30-2.50) although a nonsignificant increase continued post 2021. For severe CDI, fidaxomicin use increased after 2017 (OR, 1.56; 95% CI, 1.12-2.17), with the 2021 event (OR, 2.32; 95% CI, 1.60-3.35), and post 2021 (OR, 1.21; 95% CI, 1.04-1.41). Conclusions The 2021 SHEA/IDSA update was associated with statistically significant increases in fidaxomicin use for inpatients with CDI. However, the increases were small.