HealthApril 26, 2024

Performance improvement in your antimicrobial stewardship program

Even in the best-performing healthcare organization, there is always room for improving the quality of patient care and health outcomes.

Committing to continuous improvement is a key element to delivering quality patient care, but it’s also an expectation from accreditation and regulatory bodies, such as the Joint Commission which explicitly states in its Antimicrobial Stewardship standard MM.09.01.01, element of performance 21:

The hospital takes action on improvement opportunities identified by the antibiotic stewardship program.

Some common models, such as Plan-Do-Study-Act (PDSA) cycle, emphasize continuous improvement with iterative cycles, whereas LEAN and Six Sigma focus on reducing waste and standardizing processes to minimize defects. For antimicrobial stewardship (AMS), the CDC introduced a Target-Assess-Steward (TAS) model to provide a framework for improvement in antimicrobial use. In this article, we will discuss how to adopt the TAS framework through some concrete examples to improve antimicrobial utilization.

The TAS framework and metrics for improvement in antimicrobial use

The TAS framework comprises of three different components to help AMS leaders work methodically to improve antimicrobial use in their organizations. 


In the Target step, facilities that submit antimicrobial use and resistance (AUR) data to NHSN can obtain metrics such as standardized antimicrobial administration ratio (SAAR) and antimicrobial use cumulative attributable difference (AU-CAD) to help them identify opportunities for improvement.

SAAR is a ratio that compares a facility’s antimicrobial use to a standard referent population that has been statistically adjusted to account for the hospital’s unique characteristics in order to provide as close to a benchmark as possible with the available data. It is obtained by dividing a hospital’s actual antimicrobial utilization by the predicted antimicrobial utilization. As such a SAAR of 1 means that the hospital’s usage of antimicrobial is exactly the same as the predicted utilization, whereas a SAAR greater than 1 indicates the hospital’s usage is greater than expected. SAARs are available in adult, pediatric, and neonatal populations and are further separated by drug classes and patient care locations. Note that SAARs are not available in all NHSN-defined patient care locations, where insufficient national data exist to provide a denominator in the ratio calculation.

The CDC introduced a new metric called AU-CAD in 2022. Mathematically, AU-CAD is the difference in the number of days of therapy (DOT) between the hospital’s current state of antimicrobial use and the desired state of antimicrobial use based on its selected SAAR target. The calculation can be represented by the following equation:

AU-CAD = Observed antimicrobial days – (predicted antimicrobial days x SAAR target)

AU-CAD can be calculated for each SAAR antimicrobial group and patient care location in the NHSN AU module. Based on the equation presented above, a positive AU-CAD value indicates that your current SAAR value for the antimicrobial group selected is greater than your target SAAR. The AU-CAD number will then represent the number of DOTs you would have to reduce to reach the SAAR target. AU-CAD provides you with numeric target DOTs that you should reduce to reach your goal.

AU-CAD should be interpreted based on a defined period of time. For example, the observed antimicrobial days and predicted antimicrobial days may represent usage over 12 months, and then the AU-CAD will represent the number of DOTs that have to be decreased or increased over 12 months to reach the target SAAR. There is no national goal for SAAR and each facility may select a target SAAR based on historical trends, organizational and clinical priorities, resistance patterns, and other local factors.


As a reminder, metrics such as SAAR and AU-CAD simply compare your actual antimicrobial utilization to predicted antimicrobial utilization based on statistical adjustments to account for your hospital’s attributes and types of patients served. However, these metrics do NOT measure the appropriateness of antimicrobial prescribing. Once AMS leaders use SAAR and AU-CAD to identify locations and/or drug groups to target their efforts, they will need to assess the prescribing of antimicrobials to determine if there are gaps and opportunities to improve. Such assessment would typically involve methods such as Medication Use Evaluation. 


Once AMS leaders identify the opportunities and gaps, we recommend convening relevant stakeholders such as prescribers, nurses, and microbiology personnel to design systemic interventions to improve antimicrobial use. Leaders will need to periodically monitor process and outcome metrics to ensure that the interventions are implemented and executed consistently and that the antimicrobial usage is trending in the right direction.

Putting TAS framework into practice

While there are only 3 steps in the TAS improvement model, the volume of available data can appear overwhelming at first. Let’s walk through a hypothetical example of how an AMS leader can take advantage of the data provided by NHSN to drive improvement.


In this facility, as we can observe in the SAAR by Antimicrobial Category graph, most of the SAARs are around or below 1. The group with the highest SAAR is broad-spectrum agents typically used for hospital-onset infections. Over the past year, this group has a SAAR of 1.25 and was consistently above 1 every month. Thus the AMS leader should focus on this drug group based on the TAS model.

The next step is to assess which antimicrobial agents and which patient care locations are driving this elevated SAAR for the hospital. As shown in the Antimicrobial Use by Agent pie chart, piperacillin-tazobactam, and meropenem contributed 50% and 33% of DOTs to this category, respectively. 

The AMS leader may also examine the specific patient care locations that the highest SAARs, in this case the top 3 locations include Peds Med Surg, Med 5, and Med 2 units in this fictitious hospital. 

Using the AU-CAD data, the AMS leader may examine the usage data in another way. As explained above, the AU-CAD indicates how many DOTs are above the target SAAR over a period of time. Examining the broad spectrum agents for hospital onset infections data, the number 2,626 under the Target SAAR column of 1.00 indicates that this hospital is using 2,626 excess DOTs in this category over the past 12 months. Using historical data, this hospital would have to reduce antibiotic use in this category by 2,626 DOTs to achieve a SAAR of 1.00.

The AU-CAD metric can also be helpful when AMS leaders formulate their annual strategic plan and establish goals. Considering the AU-CAD of 2,626 DOT over a 12-month period, you would have to reduce antimicrobial use in this category by over 7 DOTs per calendar day (2,626 DOT/365 days = 7.19 DOT/day). Is this an achievable goal? This would depend on hospital size, appropriateness of current prescribing, technology, and resources available, just to name a few factors to consider. Perhaps setting an improving target of 1.10 would be more reasonable by reducing DOTs in this category by 4.39 DOT per calendar day (1,603 DOT/365 days = 4.39 DOT/day).


Once the AMS identifies the targeted drugs for improvement, in this case, piperacillin-tazobactam and meropenem, a detailed assessment should be pursued. Prescribing patterns can be assessed at a high level first to determine if there are any concerning areas to focus on. For example, the use of these broad-spectrum agents may be expected in infectious diseases, critical care, or hematology/oncology specialists, but high usage among cardiologists or rheumatologists may warrant further investigation. A more formalized assessment can be conducted using Medication Use Evaluation (MUE). The American Society of Health-System Pharmacists provides detailed recommendations on how to conduct an MUE. It’s often helpful to convene a panel of stakeholders to define what appropriate usage of these agents looks like prior to conducting the MUE. This will ensure objective, agreed-upon criteria for appropriate use and mitigate the likelihood of disputing findings upon the conclusion of the MUE.


Based on the findings in the Assess step, AMS leaders can now design interventions that would help improve antimicrobial use in the targeted areas. Many tactics and practical interventions have been promoted in the literature and guidelines. A few of the common interventions to target specific areas of opportunity include:

  • High empiric usage: standardize order set based on hospital resistance patterns; introduce rapid diagnostics
  • Inappropriate prescribing based on indication: pre-authorization; prospective audit and feedback; require indications in computerized provider order entry system
  • Inappropriate usage among selected providers: individual academic detailing; prospective audit and feedback; peer education
  • Suboptimal dosing: pharmacy dosing adjustment protocol; precision dosing 
  • Delayed or missed optimization based on microbiology results: clinical decision support alert; pharmacy profile review; antibiotic timeout
  • Prolonged duration: antibiotic timeout; duration reviews

The TAS framework is a robust model for process improvement. By submitting data to NHSN’s AUR module, AMS leaders will have access to useful metrics such as SAAR and AU-CAD to help them identify areas of strengths and opportunities for improvement in their programs. 

The volume of data may appear overwhelming, this is why leveraging a contemporary analytics tool such as those in Sentri7 can help you make sense of the data to arrive at an evidence-based improvement plan. While we only addressed the initial steps of identifying areas to improve in this article, AMS leaders should monitor their progress periodically to ensure the interventions implemented in the Steward phase of the TAS framework are leading to desired outcomes. By deploying an interactive, drillable analytics dashboard, AMS leaders can quickly monitor the relevant metrics throughout the year. Unlocking the buried insights inside static data tables goes a long way to improving the efficiency and impact of your AMS program.

Sentri7® Pharmacy
Sentri7 Pharmacy identifies patients early and accurately to improve patient care, safety, and drive pharmacy cost savings by optimizing medication therapies.
Empower your pharmacy team to impact patient care through evidence-based recommendations, standardizing practices via pharmacy-specific workflows, and using robust analytics to drive improvements in medication therapy, Opioid Stewardship, and Antimicrobial Stewardship.
Manager of Pharmacy Services and Fellowship Director
Dr. Steve Mok has over a decade of experience in the areas of antimicrobial stewardship, infectious diseases and clinical pharmacy management. He has practiced in a variety of settings.
Sentri7® Pharmacy
Sentri7 Pharmacy identifies patients early and accurately to improve patient care, safety, and drive pharmacy cost savings by optimizing medication therapies.
Empower your pharmacy team to impact patient care through evidence-based recommendations, standardizing practices via pharmacy-specific workflows, and using robust analytics to drive improvements in medication therapy, Opioid Stewardship, and Antimicrobial Stewardship.
Back To Top