Atrium Health, an integrated nonprofit health system, operating 42 hospitals, over 30 urgent care centers and more than 1,500 care locations in North Carolina and Georgia, needed a solution to measure hospital antibiotic use (AU) consistently across its hospitals and to provide quality metrics for its facility scorecard comparisons.  They were generating more than 2 million MAR and ADT records per month.  The necessary clinical data was spread across different Epic and Cerner electronic health system implementations with each requiring a different interface format.  The Atrium IT Department had tried several tools including the AU and Infection Control modules that came with their systems, but they were faced with interoperability challenges, incorrect calculations, or functional limitations.  Standardized enterprise-wide reporting was not possible.  Confronted by the technical obstacles, Atrium turned to Asolva’s MEDICI Antimicrobial Use (AU) software solution.

Challenge #1: EHR Unable to Benchmark Facilities

The antimicrobial stewardship module built into the EHR was designed to calculate antimicrobial use but only with its own native data.  It could not extract and consolidate data or understand the data structure of other EHR systems.  Because Atrium Health is a large hospital system and each facility potentially on a different EHR platform, Atrium was unable to effectively compare and benchmark antimicrobial use across all of its facilities.  They needed a solution that could tie all their EHR data together.

MEDICI Solution

MEDICI AU works with any EHR system.  Popular healthcare data exchange standards such as HL7 and FHIR are very useful, but they could also be unnecessarily complex, expensive, or difficult to implement.  It would be like using a bulldozer to do the job of a shovel.  We needed an appropriate tool to match the job.  In this case, the right tool was a simple comma-delimited text file.  Since Atrium could easily run reports to generate data in comma-delimited file format, we simply loaded those files into MEDICI AU.  No costly data interfaces were needed, and exporting data to a comma-delimited file was something that almost all IT departments could already do.  MEDICI AU was able to easily consolidate the data from each of Atrium’s disparate, EHR-systems into a single data repository for antimicrobial use analysis.

Challenge #2: Existing Infection Control Software Unable to Calculate Days of Therapy (DOT) from MAR

Before using MEDICI AU, Atrium had evaluated another software vendor’s infection control solution to help integrate the data comparison across different EHRs.  Unfortunately, this software vendor’s product was only able to accept and calculate DOT based on medication orders instead of administration data.  Measuring antimicrobial use based on medication orders is known to be inaccurate because it does not reflect what was actually given.  Complicating matters further, Atrium wanted to be able to submit their antimicrobial use data to National Healthcare Safety Network (NHSN), which only accepts DOT results calculated from MAR data.

MEDICI Solution

MEDICI AU was designed with the foundation to calculate DOT based on MAR data.

Challenge #3: Handling Data Differences

No two hospitals are exactly alike.  That truth becomes more evident when comparing the data across Atrium’s facilities.  For example, in addition to having to discern between “amoxicillin” and “amoxicillin/clavulanate”, there were at least 20 different variations of those names despite valiant data cleansing effort by the IT programmers.   How do you compare drug usage across facilities if the drug names don’t match and RxNorm values are not always available?  Similarly, locations might have the same name between facilities but different functions.  For instance, “3W” could be an ICU at one facility but Labor and Delivery at another facility.  How do you compare the same location types across facilities for apples-to-apples benchmarking?

MEDICI Solution

MEDICI AU provides deep data cleansing capabilities utilizing pattern matching algorithms that can be easily re-configured on-demand.  When Atrium did not always provide the correct data the first time, MEDICI AU’s ability to easily back-out data and re-import corrected data proved invaluable.  “Because we understand that real-life data always has inconsistencies and that the data cleaning process is iterative, we’ve built something unique and powerful – a very intelligent tool that can make sense out of varying representations of data,” said Joe Chuong, Product Manager at Asolva.

Challenge #4: Ensuring Data Quality

With millions of records to analyze, Atrium needed to ensure that the calculations were correct and consistent.  These metrics would derive important facility scorecards and drive quality decisions.

MEDICI Solution

The calculations produced by MEDICI AU were thoroughly tested by successfully matching the calculations generated by one of the EHRs–Cerner.  MEDICI AU also became the first software product to officially pass NHSN’s Antimicrobial Use Synthetic Data Set (AU SDS) Validation, further ensuring data quality that meets CDC’s standards.


Atrium was able to start viewing and validating the antimicrobial use data within days of extracting its MAR data.  MEDICI AU proved that it could meet all of the functional requirements of a large, complex hospital system without the expensive data interfaces or lengthy project plans of an enterprise-wide deployment.  With the help of MEDICI AU, Atrium Health was able to support its antimicrobial stewardship program with a system-wide reporting of antimicrobial consumption using the CDC-established metrics.