Through integration of the Collaboration to Harmonize Antimicrobial Registry Measures (CHARM) program, healthcare providers and clinicians may better quantify and assess local antimicrobial prescribing practices in outpatient settings, according to information shared at the 2023 MAD-ID (Making a Difference in Infectious Diseases) meeting, in Orlando, Fla.
Michael Klepser, PharmD, a professor at the College of Pharmacy at Ferris State University, in Big Rapids, Mich., sat down with Infectious Disease Special Edition to discuss his work with CHARM. “When we looked at [available data], we realized very quickly that there was not anything or any standard that was useful to us at a clinical level [to quantify and assess appropriateness for antibiotic use].
“Unfortunately, a lot of these data sets lagged years behind, and so they really didn’t provide us with actionable data. So, what we decided to do was look at antibiotic use that was recorded in the electronic medical record in the clinic.”
For more information, read the story here.
—Landon Gray
Video Transcript:
Hi, I’m Michael Klepser. I’m a professor at Ferris State University College of Pharmacy. And I’m the senior director for the Collaboration to Harmonize Antimicrobial Registry Measures, or CHARM.
CHARM is a project that had its inception back in 2015, as I was working with a student in my Ambulatory Care Clinic, and what we’re trying to do was find a way to quantify and assess appropriateness for antibiotic use in the clinic.
When we looked at what was available, we realized very quickly that there was not anything or any standard that was useful to us at a clinical level. Most of the ways that were being developed, we’re looking at large data sets of antibiotic use, and looking at utilization rates at population levels.
Unfortunately, a lot of these data sets lagged years behind, so they really didn’t provide us with actionable data. What we decided to do was look at antibiotic use that was recorded in the electronic medical record in the clinic. And we would identify episodes of antibiotic use, and we would anchor all of our data collection around that site.
We would collect a variety of data, and we would look at patient demographics, we will look at insurance, we would look at antibiotics that were used, ICD codes, so on and so forth.
And what we were able to do is take that data and create a good way to assess antibiotic utilization patterns in the clinic.
What we decided is we started to branch out to make sure that we were having a way to normalize data among large clinics and small clinics, clinics with high patient utilization rates, and so forth.
Essentially what we decided to do was look at normalizing data or antibiotic usage data based on 1,000 clinic visits and 1,000 clinic patients, so normalized for the different sides and utilization patterns.
Once we were able to describe antibiotic use, we had to try to figure out, “How do we assess the appropriateness of antibiotic use?”
We did that by looking at published treatment guidelines and FDA-approved indications for various antibiotics. Once we identified the indication, if there were published guidelines, we’d look to see if the agent that was selected was recommended by those guidelines or approvals, and so forth.
If the agents were approved, or recommended, rather for those indications, then we wanted to try to figure out how do we determine not only was this selection appropriate, but were they being dosed appropriately. That was something that we thought long and hard about.
We decided we were going to create a parameter to give an idea of a drug exposure, and we call that the prescribed therapeutic regimen, or PTR.
The PTR is essentially the dose of the antibiotic times the frequency times the duration. And that gave us an overall idea of antibiotic exposure for that course.
We would compare that then with ranges that were calculated in a similar way from recommendations in the treatment guidelines and FDA-approved indications. We would create what we call a recommended therapeutic regimen range, or RTR.
If the prescribed therapeutic regimen fell in that recommended therapeutic regimen range, we would consider dosing to be concordant. If it fell outside of that range, we decided to call that non-concordant. We would report those frequencies back to the clinic.
Once we determine if there was a non-concordance with dosing, we tried to attempt to describe why: “Was the dose too high? Was the duration too long?” And we provide that as well.
As the project evolved, we actually developed an interactive dashboard, and so we’re able to visually present the data to our users.
Everything on these interactive dashboards is a filter so you can get really granular with your cuts and data. Since [Joint Commission on Accreditation of Healthcare Organizations] put out these performance measures in 2020, there’s been an increased interest in tracking outpatient antibiotic use, and we’ve had a lot of clinics reach out to us about the term project.
The data that we provide is highly visual, as well as interactive, and it’s very conducive to educating clinicians. They can easily see their prescribing patterns compared to their peers.
If people are interested in CHARM, they can reach out to me at Ferris State University, and my email is MichaelKlepser@FerrisState.edu.