No abstract

Objective: We sought to determine the value of an audit-and-feedback monitoring method in facilitating meaningful practice changes to improve vancomycin dosing and monitoring. Design: Retrospective, multicenter, before-and-after implementation observational quality assurance initiative. Setting: The study was conducted in 7 not-for-profit, acute-care hospitals within a health system in southern Florida. Methods: The preimplementation period (September 1, 2019, through August 31, 2020) was compared to the postimplementation period (September 1, 2020, through May 31, 2022). All vancomycin serum-level results were screened for inclusion. The primary end point was the rate of fallout, defined as vancomycin serum level ≥25 µg/mL with acute kidney injury (AKI) and off-protocol dosing and monitoring. Secondary end points included the rate of fallout with respect to AKI severity, rate of vancomycin serum levels ≥25 µg/mL, and average number of serum-level evaluations per unique vancomycin patient. Results: In total, 27,611 vancomycin levels were analyzed from 13,910 unique patients. There were 2,209 vancomycin serum levels ≥25 µg/mL (8%) among 1,652 unique patients (11.9%). AKI was identified in 379 unique patients (23%) with a vancomycin levels ≥25 µg/mL. In total, 60 fallouts (35.2%) occurred in the 12-month preimplementation period (∼5 per month) and 41 fallouts (19.6%) occurred in the 21-month postimplementation period (∼2 per month; P = .0006). Failure was the most common AKI severity in both periods (risk: 35% vs 24.3%, P = .25; injury: 28.3% vs 19.5%, P = .30; failure: 36.7% vs 56%, P = .053). Overall, the number of evaluations of vancomycin serum levels per unique patient remained consistent throughout both periods (2 vs 2; P = .53). Conclusions: Implementation of a monthly quality assurance tool for elevated outlier vancomycin levels can improve dosing and monitoring practices resulting in enhanced patient safety.