Temporal Change of Risk Factors in Hospital-Acquired Clostridioides difficile Infection Using Time-Trend Analysis

Abstract

Background: In the past few decades, the epidemiology of Clostridioides difficile infection (CDI) has evolved. Given recent changes in the incidence of CDI and prevention efforts, we investigated temporal changes over a period of 8 years (2009–2016) in the incidence of and risk factors for CDI. Methods: Both pediatric and adult inpatients discharged from hospitals in metropolitan New York City were included. Individual and environmental (eg, pharmacological) risk factors were identified through a matched case-control by the length of stay at a ratio of 1:4. A Cochran–Armitage test or Mann-Kendall test was used to investigate trends of incidence and risk factors. Results: During the study period, 6,038 of 694,849 (0.87%) patients had a positive test for C. difficile during their hospitalization. Of these, 2,659 of 6,038 (44.04%) were identified as hospital-acquired CDI (HA-CDI) and just over half (3,379 of 6,038, 55.96%) were identified as community-acquired CDI (CA-CDI). There were no trends in total CDI incidence rates; rather, we detected downward trends in HA-CDI and upward trends in CA-CDI (P_trend < .05). Younger patients and patients with lower risk of illness had HA-CDI over time (P_trend < .05). Antibiotics were administered to more patients over time and in longer cumulative days (+3% and +3.1% per year). We detected a reduction in the receipt of high-risk antibiotics in all cohorts (−0.12% per year) and a decrease in cumulative days of high-risk antibiotics in the cohort with HA-CDI (−1.1% per year). When stratified by the type of high-risk antibiotics, the use of carbapenem, cephalosporins, clindamycin, and monobactam increased (+0.53%, +1.8%, +0.5%, and +0.39% per year, respectively), whereas the use of broad-spectrum penicillins and glycylcycline significantly decreased over time in all cohorts (−1.8% and −0.22% per year). Among the cohorts with HA-CDI, only cephalosporins showed a significant upward trend (+ 5.7% per year) and only fluoroquinolones showed a significant downward trend (−2.2% per year). Lastly, a reduction of proton pump inhibitors and an increased use of histamine-2 blockers were detected in all cohorts (−3.8% and +7.3% per year) (all P_trend < .05). Conclusions: Although the incidence of HA-CDI decreased, more effort to decrease all antibiotics use and cumulative days should be emphasized as part of antibiotic stewardship. The downward trends of high-risk antibiotics might have been associated with the decrease in the trend of HA-CDI; however, the impact of the trends of risk factors on the trend of HA-CDI should be further investigated.

Funding: None

Disclosures: None