NHSN SAAR Assessment Using Visual Analytics: Tools to Improve Population Antimicrobial Use

Background:  The National Healthcare Safety Network (NHSN) recently began reporting Standardized Antibiotic Administration Ratio (SAAR).  The SAAR uses antimicrobial use (AU) data transmitted to NHSN to generate an observed-to-expected ratio of Days of Therapy (DOT) using risk adjustment for facility characteristics. Currently, 49 VA Medical Centers participate.  While the SAAR provides a high-level overview for benchmarking AU, it lacks granularity.  Antimicrobial stewardship (AS) decision-making nodes can be further characterized using a cognitive framework of Choice (initial selection), Change (de-escalation), and Completion (duration of therapy) (CCC). Adding this framework to SAARs can improve granularity for targeting AS opportunities.

 

Methods:  We developed visual analytics tools (i.e., a dashboard) based on data extracted from the VA Corporate Data Warehouse that combines antibiotic days, SAAR, CCC, and Spectrum Score (a measure to quantify antimicrobial spectrum of activity) for three common infectious disease conditions: pneumonia, UTI, and SSTI. Stewards can use the dashboard to identify antimicrobial use patterns and potential areas for improvement.  Data benchmarked to other NHSN-participating VA facilities can be categorized by condition, decision-making node (CCC), antimicrobial class or agent, administration route, and ward.  Results can be expressed visually using graphs, tables, and charts.

 

Results:  The dashboard is currently being pilot tested by stewards in 3 VA facilities with 5 more pending.  The tools have identified areas for improvement in measures for SAARs > 1 and with SAARs not different than 1. Figure 1 shows how the dashboard is able to refine analysis of piperacillin/tazobactam (P-T) in a facility with an MDRO Gram negative SAAR not different than 1 to uncover increased empiric P-T use for pneumonia patients admitted to medical-surgical wards.  This data led the stewards to update pneumonia order menus and algorithms.

Figure 1

 

 

 

 

 

 

 

 

Conclusion:  The visual analytics tools provide additional granularity in evaluation of AU data beyond SAARs, and have the ability to identify AU improvement opportunities in an efficient manner.