Discussion

In this study, we found that clinicians considered a novel metric for comparing antibiotic prescribing at hospital discharge to be clear and clinically meaningful, suggesting that it may help hospitals identify opportunities for improved antibiotic stewardship at discharge. However, clinicians emphasized the need to be provided with granular data (eg, prescribing service) for the metric to be utilized. Clinicians also raised concerns about fairness and potential punitive use.

Discharge prescribing remains understudied compared to inpatient stewardship, despite evidence of frequent antibiotic overuse at discharge. ^{Reference Feller, Lund and Perencevich1–Reference Daniels and Weber4} Standard metrics for hospital-based ASP programs do not capture data on postdischarge antibiotic use, though prior work highlights the potential usefulness of a postdischarge antibiotic use metric. For example, an evaluation of a national inpatient stewardship initiative found that benchmarking was identified as an effective strategy because it provided performance context, motivated staff to improve, and guided targeted interventions; importantly, ongoing communication and education were needed to sustain engagement. ^{Reference Cosgrove, Ahn, Dullabh and Gordon7} Previous qualitative work within the VHA found that ASP leaders valued standardized, risk-adjusted benchmarking metrics for inpatient antimicrobial consumption but had concerns about the ability of risk adjustments to account for facility differences. ^{Reference Jones, Marra and Livorsi8} Consistent with prior work, ^{Reference Pinkerton, Bongu and James9} clinicians in our project emphasized that benchmarking would be most effective when framed as a quality improvement strategy rather than punitive or invasive. Addressing these concerns and clearly framing the metric as a tool for quality improvement will be essential for its successful implementation.

In practice, the ASP team could use their facility’s metric performance to determine whether further assessments of discharge antibiotics are warranted and whether additional stewardship strategies focused on discharge are needed. In a prior study across 7 hospitals, less antibiotic overuse at discharge was associated with high-quality pharmacist-physician interactions, infrastructure and knowledge to support stewardship, and highly engaged Infectious Disease physicians. ^{Reference Vaughn, Krein and Hersh10}

Our study has limitations. Findings may not be generalizable beyond the VHA context. Although we interviewed personnel from only 9 sites, the 91 interviews included perspectives from both those leading stewardship efforts and those expected to adhere to them. A strength of the project is that there is little data on how hospital-level postdischarge stewardship metrics are perceived by end-users, making our findings an important contribution for future metric development and implementation efforts.