Beginning in 2018, antimicrobial stewardship pharmacists from the four hospital systems in Des Moines, Iowa (Broadlawns Medical Center, MercyOne Des Moines, UnityPoint Health Des Moines, and Veterans Affairs Central Iowa Health Care System) were solicited for their current antibiogram (representative of information from 2017) and were asked to respond to a 24-item survey regarding antibiogram data collection, current breakpoints, testing equipment, time frames for collection, reporting of resistant organisms, and adherence to the Clinical Laboratory and Standards Institute (CLSI) M39 guidelines.[1]           

After confirming breakpoint similarities at each of the hospitals and 1 year exclusion of same patient isolates of organisms, a pilot antibiogram was built using Tableau data visualization software. In order to combine the four antibiograms, the percent susceptible for each organism in each antibiotic cell was multiplied by the number of isolates tested at the given hospital to provide a surrogate count of susceptible organisms. These surrogate counts were summed for all hospitals and then divided by the sum of the isolated organisms from all hospitals reporting that organism versus that antibiotic. This is consistent with methodology published previously.[2-3]  Any cell that contained a sum of less than 30 isolates after combining antibiograms was removed from the final version. This impacted one organism (Acinetobacter lwoffi).           

Following combination of antibiograms, the antimicrobial stewardship pharmacist from each institution participated in a debriefing session in order to determine the best approach for addressing inconsistencies in the final combined antibiogram.    

• One institution presented staphylococcus aureus in total instead of separating into methicillin-sensitive and methicillin-resistant isolates. Because this represented only a small amount of the total staphylococcus aureus in the antibiogram, this data was removed.

• One hospital of the four created an 18 month antibiogram which extended 6 months prior to the 12 month antibiograms of the other three facilities. It was determined that these 6 months of data would be included, but the team would work with that institution’s microbiology department in the future to obtain a 12 month antibiogram consistent with the time frame of the other institutions.

• One institution did not classify certain organisms into species based on their microbiology department’s cascading policies. There were five organisms that were not have their species specified. Because of the quantity of these isolates they were reported in the antibiogram with other members of that genus as “unspecified” (example: enterococcus faecalis, enterococcus faecium, and enterococcus unspecified).

• Two hospitals had separate antibiograms to specifically identify outpatient cultures only. Because of the over-reliance on a single institution as this data source, a separation of an outpatient only antibiogram was not pursued.

• Because the full antibiogram contained 36 antibiotics and 32 unique organism, the group felt that an executive summary including the top 10 isolated organisms and most common outpatient antibiotics should be created. While available cultures from each antibiogram report from all cultured sites, other types of executive summaries based on type of infection could be created based on likely etiology of the site of infection (respiratory, urinary, gastrointestinal, skin).

Once the pilot antibiogram was completed, this process was repeated using antibiograms from the four hospitals from 2018 to 2020 with some notable differences:

• For 2018 data, one hospital did not create an antibiogram due to planned changes in microbiology software platforms. Instead, microbiology reports were provided to the antimicrobial stewardship pharmacist group and an ad hoc antibiogram was created for the sole purpose of inclusion into the combined community antibiogram. Because this institution’s reporting cascade policies for cephalosporins with Enterobacteriaceae were complex, their cephalosporin sensitivities were censored for Enterobacteriaceae only. Reporting for the other three institutions was not impacted.

• For 2019 data, only three hospital systems were represented. The smallest institution (in terms of isolate counts) did not submit an antibiogram. Another institution had antibiogram data available only for the second half of 2019 (July through December) because of major changes to microbiology software platforms. 

• For 2020 data, only three hospital systems were represented. The smallest institution (in terms of isolate counts) did not submit an antibiogram. Another institution reported data that did not separate staphylococcus aureus as methicillin-sensitive and methicillin-resistant isolates. All staphylococcus aureus isolates from that institution were excluded.

• For 2021 and 2022 data, four hospital systems were represented. Two institutions reported data that did not separate staphylococcus aureus as methicillin-sensitive and methicillin-resistant isolates. All staphylococcus aureus isolates from those institutions were excluded. One institution separated cefazolin susceptibilities for E. coli, K. pneumoniae, and P. mirabilis total organisms based on systemic CLSI breakpoints; however, a majority of these isolates were actually cultured from urine. A subgroup of susceptibility of these three organisms from the urine was made available. These urine isolate subgroup numbers were all assumed to be uncomplicated urine infection and removed from the total isolates; the remainder was considered the systemic infection subgroup. The susceptible urine and susceptible systemic infection subgroups were then combined for a calculated total susceptibility. The total susceptibility from the other three institutions was accepted as reported on their published antibiogram.

We were able to publish our methodology and findings of our educational session in the journal Antimicrobial Stewardship & Healthcare Epidemiology at the end of 2023.[4] You can read our article here.

1. Clinical and Laboratory Standards Institute. M39-A4: Analysis and Presentation of Cumulative Antimicrobial Susceptibility Test Data; Approved Guideline (4th ed.), 2014.

2. Halstead DC, Gomez, N, and McCarter YS. Reality of developing a community-wide antibiogram. J Clin Microbiol. 2004;42:1–6.

3. Guarascio AJ, Brickett LM, Porter TJ, et al. Development of a statewide antibiogram to assess regional trends in antibiotic-resistant ESKAPE organisms. J Pharm Pract. 2019;32:19-27.

4. Miesner AR, Williamson B, Bushman AM. AntibiogramDSM: a combined local antibiogram and educational intervention. Antimicrobial Stewardship & Healthcare Epidemiology. 2023;3(1):e179. doi:10.1017/ash.2023.450