Prevalence of Antimicrobial Use in US Hospital Patients, 2011 vs. 2015

Abstract

Background Antimicrobial stewardship (AS) is increasingly recognized as an essential component of patient safety programs. In a US hospital prevalence survey in 2011, 50% of patients received antimicrobial drugs (ADs). The survey was repeated in 2015 to describe changes in inpatient antimicrobial use, approximately one year after CDC published the “Core Elements of Hospital Antibiotic Stewardship Programs.”

Methods Emerging Infections Program (EIP) sites in 10 states recruited up to 25 hospitals each, seeking to re-engage hospitals that participated in the 2011 survey. Hospitals selected survey dates from May to September 2015 and completed AS questionnaires. Patients were randomly sampled from the hospital census on the survey date. EIP staff retrospectively reviewed medical records to collect AD data. Percentages of patients on ADs on the survey date or the day before were compared using chi-square tests (SAS 9.4, OpenEpi 3.01).

Results In 2015, among 148 hospitals participating in both surveys, 29 (19.6%) reported having no AS team (AST); 63 (42.6%) had ASTs for <4 years, and 56 (37.8%) had ASTs for ≥4 years. Antimicrobial use prevalence in 2015 was approximately 50% in hospitals with and without ASTs. Percentages of patients on ADs was not different in 2015 (4,590/9,169, 50.1%) compared with 2011 (4,606/9,283, 49.6%, P = 0.55). Antimicrobial use prevalence in most hospital locations did not change, although the percentage of neonatal intensive and special care unit patients on ADs was lower in 2015 compared with 2011 (22.1% vs. 30.7%, P = 0.005). The percentage of patients on fluoroquinolones was lower in 2015, while percentages of patients on carbapenems or cephalosporins were higher in 2015 than in 2011 (figure).

Conclusion Some observed differences between 2011 and 2015 provide evidence of stewardship impact. The decrease in antimicrobial use in selected neonatal locations may reflect implementation of tools to improve neonatal sepsis prescribing, while decreases in fluoroquinolone use may reflect efforts to prevent Clostridium difficile infections. However, our data also suggest that reductions in some ADs are offset by increases in others, supporting the need for ongoing work to identify the most effective AS strategies.

Disclosures All authors: No reported disclosures.