Use of Data Envelopment Analysis to Quantify Opportunities for Antibacterial Targets for Reduction of Health Care–Associated Clostridium difficile Infection

Abstract

Clostridium difficile infection (CDI) is an important health care–associated infection that leads to increased morbidity and mortality. Antibacterial medications used in hospitals serve as targets for antibacterial stewardship programs to reduce C difficile. The objective was to create a benchmark strategy targeting high-risk antibacterials for C difficile. This was a retrospective cross-sectional study using claims data from 58 hospitals. The Data Envelopment Analysis Technique was used to identify best-practice hospitals in terms of less use of 5 classes of antibacterials and fewer CDIs. Of 58 hospitals, 17 (29%) were identified as best-practice hospitals. Antibacterial classes requiring the greatest percentage reduction in use in non-best-practice hospitals versus best-practice hospitals were clindamycin (31%), β-lactam/β-lactamase combinations (30%), and carbapenems (29%). This study suggests that there are areas of improvement in high-risk antibacterial use that could lead to decreased CDIs.