Development of risk-standardized metrics for benchmarking hospitals on their inpatient antibiotic use

INTRODUCTION: Benchmarking hospitals on their antibiotic use may be facilitated by metrics that adjust for inter-hospital differences in patient case-mix, such as types of infections, procedures, and comorbidities. Metrics that capture antibiotic spectrum [e.g., days of antibiotic spectrum coverage (DASC)] can be more sensitive to stewardship activities than metrics based on days of therapy (DOT). In this study, we developed risk-standardized metrics for both DOT and DASC. METHODS: We performed a mixed-methods study to build risk-standardized metrics for inpatient antibiotic use, using a modified Delphi process integrating expert- and data-driven strategies to identify nonmodifiable risk factors associated with appropriate inpatient antibiotics. These factors were used to create risk-standardized ratios (RSR) for DOT and DASC. A standardized antimicrobial administration ratio (SAAR)-like metric was also constructed. RESULTS: In 2021, there were 497,061 patient-admissions across 121 Veterans Health Administration (VHA) hospitals. The median hospital RSR was 1.00 (interquartile range (IQR) 0.95-1.05) for DOT and 1.00 (IQR 0.96-1.04) for DASC; the median ratio for the SAAR-like metric was 0.85 (IQR 0.68-1.03). The Kendall's tau for RSR-DOT and the SAAR-like metric was 0.48; RSR-DASC and the SAAR-like metric was 0.33; and RSR-DOT and RSR-DASC were 0.48. Compared to the SAAR-like metric, 60 (49.6%) and 80 (66.1%) hospitals ranked in a different quartile for RSR-DOT and RSR-DASC, respectively. CONCLUSIONS: Hospital performance on the SAAR-like metric was weakly correlated with the RSR-DASC and moderately correlated with the RSR-DOT. Hospitals' performance on the SAAR-like metric differed from that of the RSR metrics, suggesting the RSR metrics may have added value over the SAAR.