Antimicrobial Resistance (AR) Option: Beginner Analysis

My name is Fergie Phills and I'm a senior public health analyst on the NHSN antimicrobial use and resistance module team. Um I'll be providing a beginners's overview of analysis options available for the antimicrobial resistance or AR option. I'm going to turn my camera off now. So um today we will demonstrate how to run and interpret AR option um analysis reports. By the end of this session you will be able to describe the analysis reports currently available within the NHSN AR option, explain the utility of each analysis report and interpret the output produced in each report. first and we usually start off with uh this slide. Um whenever we're talking about analysis reports, I want you to always remember to generate data sets. Um generating data sets is the first step to performing analysis in NHSN. Newly uploaded data do not appear in the NHSN analysis outputs until you generate new data sets. So first you'll want to click analysis, then generate data sets on the lefth hand navigation menu. Then on the generate data set screen, you can specify the beginning and ending dates if you like or you can leave this blank to include all data ever entered into NHSN. Finally, you'll click the generate reporting data sets button. Once your data sets are finished generating, click analysis, then reports on the lefthand navigation menu. To get to the AR reports, we want to go to the antimicrobial use and resistance module folder, then the antimicrobial resistance data folder. Within the antimicrobial resistance data folder, you can see the different analysis reports that are available and we have quite a few. So there are a number of analysis reports for the AR option. We have three analysis reports that use AR event aka numerator data and that uh is a line list, a bar chart and the antibiogram and percent tested report are available. We have another three reports that use what we call AR organism definitions or phenotype definitions. And so there's a line list, a frequency table, and a rate table that are available. We have an AR summary or aka denominator line list report. We also have six incidents and prevalence reports that use both AR numerator and denominator data. And these are available as rate tables. And then we have two risk adjusted measure reports. The standardized resistant infection ratio or the SRIR and the pathogen specific resistant infection ratio which is the PSIR. So this presentation will not cover the SRIR, the PSIR or the incident and prevalence reports as those reports have their own training presentations. On this slide, I've included links to the recordings and slides for the SRIR and PSIR reports presentation from the 2024 NHSN annual training as well as the incident and prevalence reports presentation from the 2023 NHSN annual training. Um we please note we recently made some changes to the links for our previous annual training presentations. So the slides that were posted to the community do not include the updated links but they are updated on this slide here. However, if you go to the AUR training page that is linked at the end of this presentation, you can access the working links for those recordings and slides. Please note that we have recorded an updated presentation for the antibiogram and percent tested report that will be posted hopefully at a later date. But I will still briefly review that report in this presentation. With any report uh we have the ability to run, modify or export the report. These report options exist for all NHSN analysis reports. If you click run report, this generates the HTML report using the default predefined settings. If you click modify report, that allows you to make changes to the default settings such as filtering to a specific time period like a quarter or filtering to a specific location like the ED. If you click export data set, this allows you to take your AR data out of NHSN and analyze in a different software such as Excel or SAS. So this presentation mostly covers um default output um but there are resources that are available to help you modify the report for your use. There's a reference with uh general tips on how to modify a report. We also have many quick uh reference guides which are available for each report report type we have u for the AR option. These are especially helpful to review for modifying and interpreting the AR option reports. There is also a reference for how to export data um from NHSN that I've linked on this slide as well. So we have our first knowledge check of this session. So the question is what types of analysis reports are available for the AR option and then you are able to select more than one option. Um so we have a line listing all antimicrobial resistant events. B rate table antimicrobial resistant percentages. C frequency table antimicrobial resistant organisms. D bar chart all antimicrobial resistant events. E facilitywide antibiogram and percent tested. F SRR report. G all of the above. And the answer is G. All of the above. All of these are types of analysis reports, line list, rate tables, frequency tables, bar charts, the antibiogram, and our risk adjusted measures like the SRR. These are available for the AR option. So, let's start by talking about our AR event reports. These include a line list, a bar chart, and the antibio and percent tested report. So, the first is our AR event line list. In the analysis reports folder, this is named the line listing all antimicrobial resistance events report. If you are new to the AR option and you just want to see what you submitted at the basic level or you want to review your data for validation purposes, you will want to run this report. The AR event line list provides a detailed list of all AR events uploaded into your NHSN facility. It includes patient, specimen, organism, and drug susceptibility testing variables. It will display one row per drug susceptibility test within the panel for each event. Which means that due to the number of AR events that you report, you will likely reach the maximum number of rows that can be displayed when you run the default HTML report within the first few months of reporting. You can work around this by making modifications to narrow your selections to a specific time period or a specific organism, for example. or you can export your data into Excel or SAS which is what we recommend. Um please note that all the data in this presentation are fictitious and are used for illustrative purposes only. So this screenshot shows the HTML output of the AR event line list. I have just one AR event with susceptibility results for the seven drugs that are required to be reported for this organism which is Canada albbeans. In the event ID column um highlighted with the purple box on the left um you can see the event ID is the same for all seven rows. That means this is a single AR event. In the drug description column highlighted in the uh purple on the box on the right you can see that there is one row for each drug susceptibility test in this panel. There are seven drug susceptibility tests that are reported for Canada organisms um in 2025. So I'm going to show you how to modify uh the line list to show data for a specific time period. So after clicking modify report, the modification screen appears. Um click the time period tab on the modification screen. In this example, I am a limiting I am limiting the report to specimen date year beginning with 2025 to show only AR events with specimen collection dates with 2025 and forward. To modify the line list to filter for uh specific AR events, click the filters tab of the modification screen. In this example, I am filtering the report to only review Canada's AR events isolated from blood. If you're trying to add multiple filters, you can click the add rule button to add additional filters. to modify the line list to remove default variables. Um, click the display variables tab of the modification screen. In this example, I remove the default variable patient ID by selecting the variable on the right and clicking the selected button where the arrow is pointing to the left to move the variable to the left to take it out of my selected variables. To modify the line list to show additional variables that are reported with the AR event, you can stay on the same display variables tab on the modification screen. In this example, I added the individual lab test sign value and interpretation for the E test by selecting those three variables and clicking selected where the arrow is pointing to the right. This moves all three variables to the right to be included in the output. So here's the output that is displayed when applying those modifications I just made. We can now see um these three E test variables the sign the value and their interpretation for each drug susceptibility test for this AR event. We can see that for this Canada albbeans isolate the E test was conducted for each drug in the panel except boriconazol. Now switching back and going back to the default output screenshot, the one that we first saw. This is showing me that Canada albbeans was isolated from a blood specimen collected in the medical surgical critical care unit um the Miku highlighted by those two purple boxes. So you can see under specimen group we have blood and then the location is the MCQU. So this is where that Canada Alicans is uh collected. This isolate was resistant to amphetarasin B, penis funen, caspop fungen, fluconazol, mica fungen and posonic posonicol. Um the isolate was not tested against ponicazol that you can see in the last two columns of the report. Um the drug description and the final interpretation description highlighted by the big purple box. So here's our next knowledge check. The question is in the AR event line list there is one row per AR event or event ID. Is that true or false? Okay. And the answer is false. In the AR event line list, there is one row per drug susceptibility test for each AR event. So one event ID will have multiple rows. Okay. Now let's pivot to talk about the AR event bar chart. And the bar chart graphically displays the number of AR events submitted by location over time. It defaults to display monthly AR events by pathogen and location and will show a maximum of 12 time periods such as 12 months. If I wanted to see the number of AR events reported over time, this might be a good chart to run. It could show you the types of AR events being submitted within a given location. So, this is a screenshot of what the bar chart looks like. to interpret the output in this example looking above the bar chart in the header section. It's highlighted in purple. Um we highlighted the um location which was surge ward which is our surgical ward in this facility. Looking at the x-axis or the horizontal axis we have highlighted January 2024 in the purple box. In January 2024, eight AR events were reported from the surgical ward is how you read this. So at the bottom of the chart, you can see a pathogen legend where each pathogen corresponds to a specific color. NHSN uses pathogen codes, and those are like abbreviations for pathogens for display purposes because you can't use the full name. It'll just get too clunky. For example, this light green color uh represents ACS which stands for asintobacctor species. There were five total asinated back specy events reported from the surgical ward location. Four of those four of those were um from January 2024 and one was in February 2024. So, I'm going to spend a few minutes to show you how to determine what the pathogen codes like ACS stands for. One way is by using the modification screen of the AR event spar chart. Um, in the modification screen, go to the filters tab and select the pathogen description variable. Uh, you can use the drop- down menu to determine what each pathogen code refers to. For example, I have highlighted here that asinita um is represented by ACS. Just below that, we can see that Aanita Bmani is represented by ACBA. Another way to determine what the pathogen codes stand for is by downloading the IDM or the information data model that is used by vendors to map codes for pathogens, specimen sources, drug susceptibility tests, etc. for reporting to NHSN. The IDM is located within the ARC CDA toolkit which you can download on the CDA toolkits website linked on this slide and you will need to open the spreadsheet titled IDM vendors. Once you have opened the IDM spreadsheet, go to the pathogen codes 2025 tab. Filter the ARRO pathogen variable which is column S in this screenshot to X to show all the eligible AR option pathogens. The new code variable which is column G contains the list of pathogen codes that are displayed in the NHSN analysis reports. So you can search for the pathogen code in column G and look at the corresponding row in column E which is the description for drop down in app to see what pathogen the code represents. [snorts] Again as backress the first one in that code is ACS as Bmani is ACBA etc. On to our next knowledge check. So the question is what is the maximum number of time periods the AR [snorts] event bar chart can show? A3 B4 C6 D8 E12. Okay. And the answer is E 12. The AR event bar chart can show a maximum of 12 time periods such as 12 months. Now moving on to the AR facilitywide antibiogram and percent tested report. As a reminder, I will be briefly reviewing this report as we have an updated presentation that goes into more detail that will be available for your reference. Um the AR facilitywide antibiogram and percent tested report provides two tables based on AR events reported in NHSN. It will display the calculated percent susceptible which is percent S and the calculated percent tested for each organism antimicrobial combination reported from all locations together. By default, the report will include all of your inpatient and outpatient AR events that you submitted. In addition, it will include all specimen types, blood, CSF, urine, lower respiratory, skin, soft tissue, wound, muscularkeeletal, and all patient ages. Essentially, it's all the isolates regardless of organism category or hospital onset or community onset. The report defaults to show by quarter, but the user can roll up to additional time grouping such as halfyear, year, or cumulative. This report is available to both facilities and groups and users are encouraged to use filters for customization of the report. So our antibiogram will show the percent susceptible which is calculated for each organism antimicrobial pairing using the following formula. Number of isolates tested susceptible divided by the number of isolates tested times 100. Percent S is only calculated if at least 30 isolates are tested for the specific antimicrobial during the time period of your antibio. For example, if your antibio is being generated by quarter and less than 30 isolates of an organism were tested for a specific antimicrobial, you can try rolling up your data to half year or year in order to get more than 30 isolates. in a separate table in an antibiogram report. It would also show the percent tested which is calculated for each organism antimicrobial pairing using the following formula. Number of isolates with a susceptibility result. Those susceptibility results can include S SDDD which is susceptible dose dependent I intermediate R resistant or NS which is non-susceptible divided by the total number of isolates times 100% tested is only calculated if at least one isolate has been reported. So here's an example output of the antibio and percent tested report and these show the two different tables that will be included in the output. Looking at the header section above each table you can determine if you're looking at the percent susceptible which is shown here on the left or the top or the percent tested which is shown on the bottom. Um so you can see there are two different tables that will show you two different information but um very valuable information. On to knowledge check number four. So the question is in the facilitywide antibiogram and percent tested report percented is only calculated when at least 30 isolates were reported. Is this true or false? I can see half are saying true and half are saying false. All right. So the answer is false. Percent tested is calculated if at least one isolate has been reported. Percent susceptible is calculated if at least 30 isolates have been uh tested. So one isolate for percent tested 30 isolates for percent susceptible. So next we will talk about the AR organism reports and those are the lineless frequency table and the rate table. So we have specific phenotype definitions for the AR option. They can be found in appendix I of the NHSN aora module protocol. And I've screenshot the first two phenotype definitions on this slide. For example, any staff orius um that has tested resistant to oxicylin or soxitin will be assigned as a MRSA or MRSA event. Once you upload your AR events and include the drug susceptibility information, NHSN will determine if the isolate meets one or more of our phenotype definitions. Phenotype definitions are also available for the HI reports and they are similar but some are slightly different. So you want to refer to the AUI protocol to ensure you understand how the AR option reports are assigning the different phenotypes. Please note that the phenotype codes are listed in this appendex because some analysis reports only use the phenotype code, not the full name. So if you only see the phenotype code, you can refer to this table in the protocol to determine what the code stands for. We have specific reports for those events that are defined as AR organisms. In other words, you uploaded AR events with drug susceptibility testing information and NHSN has flagged those events that met the AR option phenotype definition such as MRSA, CR, etc. The AR organisms line lists all the AR events that meet an AR organism phenotype definition. This report includes patient and specimen information. If you're interested in reviewing all the AR events that meet the MRSA phenotype definition, for example, this will be a good report to run. Just to reiterate, this line list is different from the all antimicrobial resistance events line list which we reviewed first in this presentation. That report includes all AR events submitted to NHSN, while this AR organisms line list only includes those specific AR events that met a phenotype definition. To interpret this report, this screenshot shows an example of the CRA expanded phenotype as highlighted by the purple box in the header section. This report shows me that um this facility reported eight AR events that met the NHSN AR option definition for CR expanded and all were collected in the emergency department. Of those five specimens were E.coli and three were Klepsiala oxytoca. Moving into the AR organisms frequency table report. This table displays the counts of AR events that met the NHSN defined phenotype definitions. The report defaults to display totals by month, but you can modify the report to display by quarter, half year, year, or cumulative. This might be a good report if you just want to take a quick look at how many events you reported that met the phenotype definitions over time. to interpret the output in this screenshot. From January 2024 to January 2025, 51 AR events met the definition of multi-drugresistant sudinamonus aruginosa as defined by the NHSN AR option. 23 events were reported in January 2024. One event was reported in February, May, August, and December 2024 respectively. and then 24 events were reported in January 2025. So our last AR organism report is our rate table that displays AR organism percentages. This report will display the total number isolated, the total number tested, and the total number resistant by month. Using this information, the report will display the percentage of organisms that tested non-susceptible or resistant to certain antimicrobials for each defined phenotype. The percentage is calculated only if at least 30 isolates were tested, similar to the antibiogram report or the percent susceptible. The formula is listed on this slide and it's the number of isolates that tested non-susceptible or resistant to specific antimicrobials in that phenotype definition divided by the total number of isolates tested for susceptibility to those specific antimicrobials in the phenotype definition times 100. So interpreting the AR organisms rate table, the purple box highlights that in January 2025, 46 as species events were reported to NHSN. Of those 44 were tested and 40 tested intermediate or resistant to at least one drug and at least seven three of seven select antimicrobial categories as defined by NHSN. So the phenotype definition is also uh included under the table. This gives me a percentage of 90.9 or 91% because it's 40 resistant over 44 tested. Meaning 91% of asinobacttor specy events reported met the NHSN AR option definition for multi-drugresistant asintobacctor species for January 2025. The last column includes the 95% confidence interval. Um the 95% confidence interval for the percent resistant was calculated to be 79.5 to 97.0. Okay. On to knowledge check number five. The question is in the AR organism percentages report will percent resistant for MRSA be calculated if only 20 staff isolates were tested against oxicylin or soxitin? A. Yes. B. No. The answer is no. Uh, the percent resistant is only calculated if at least 30 isolates were tested against the specific antimicrobials in the phenotype definition. Okay, lastly we will briefly review the AR summary or denominator line list report. So our AR summary line list shows the summary AR data by location and month. For fact in it will show the patient days and admissions. And for the outpatient locations such as the ED or your 24-hour observation areas, it will show your encounters for that specific outpatient location. This report also includes the report no events variable. You can confirm if your facility reported events for that month or and location or if your facility checked that report no events box or if you did neither. This would be good to review for validation to ensure you accurately reported your AR data. to interpret this AR summary line list. Um, we're looking at January 2025. There were a 2,289 total patient days and 486 admissions for Faced in so all inatient locations combined. And AR events were reported um for this month. So if you look at the no AR events column on the right um no AR events equals no which means there were AR events reported for some impatient locations during that month. Please note that the label survey patient days here is incorrect as this patient days is from the AR summary record not the annual hospital survey. Um we have submitted this as a bug to be fixed in a future NHSN release. So in November 2022, for example, there were 375 total encounters in the EMUR or emergency department location. And the no uh AR events is equal to yes. And that means the report no events box that was on the alert screen was checked for that location and month. So that's why the no AR events variable [snorts] is yes here. And then looking at April 2023, the facility has not yet reported AR events for this EMR or emergency department location and has not clicked the report no events box for that month. So the no AR events variable here is blank because NHSN doesn't know um um if because no AR events were submitted for this location this month and there were no um the box wasn't clicked on the alert screen. So then let's get into our next knowledge check. The question is if a facility has not reported AR events or clicked the report no events box for this location and month then what is shown for the no AR events variable in the AR denominator line list. So no AR events were reported for this location of month and the facility did not click the report no events box. So what is shown in the output for the no AR events variable? A blank B Y or B yes C no. The answer is a blank. If the facility has not yet reported AR events for the location and month and has not clicked the report no events box for that month, then no AR events is blank. If AR events were reported, then no AR events would equal no. If the blocks was clicked, then no AR events equals yes. It can get a little confusing. So now taking all that we've learned today about these different AR option reports, we are now going to run through different example scenarios to have you all determine the best report to run for the situation. Some of these reports have are like long titles, but if you can like try to summarize in the chat, that would be great. Um, so the first question is of the AR option reports that we covered today, which report can I run if I wanted to graphically display how many asinita species isolates were reported in a location such as the ED over the last 12 months. seeing different answers in the chat, but the answer is bar chart all antimicrobial resistant events and that's because I said graphically display over the last 12 months. All right. So the next knowledge check, knowledge check eight. Which report can I run if I wanted to see a count of multi-drugresistant asinative bacter species isolates that were reported in a location such as the ED over the last 12 months. So I want to count and a bactive specy isolates that meta phenotype definition. So multi-drug resistant over the last 12 months for this one. The answer is frequency tablet antimicrobial resistant organisms report. The scenario wanted to see a count of MDR or multi-drugresistant asinative isolates reported. So we would turn to one of our AR organisms reports which contain events that have met an AR option phenotype definition and the frequency table will display the number of resistant pathogens reported over time. So which report can I run if I wanted to see the percentage of E.coli coli isolates that were susceptible to gentyosin in 2024. So I want to see some type of percentage and I'm looking at E.coli coli isolates that were susceptible to one drug genty. The answer is the facilitywide antibiogram percent susceptible and percent tested report. This scenario mentioned percentage susceptible to one drug. The antibiogram displays a table for percent susceptible for AR option eligible pathogens and the antimicrobial susceptibility test included in their panel. All right. Knowledge check 10. Which report can I run if I wanted to review all the AR events that I submitted for January 2025 to validate that the data are correct? So I want to do some data validation on my AR events that I submitted for this month. Which report would I run? The answer is the line listing all antimicrobial resistant events report. If you're new to the AR option and you just want to see what you submitted at the basic level or you want to review your data for validation purposes, you will want to run this report. So this is the report that provides a detailed list of all your AR events or your numerator data uploaded into your inerson facility. Knowledge check 11. So which report can I run if I wanted to validate that the data for fact in patient days and admissions and my outpatient encounters are accurate for each month. So I want to look at essentially denominator data. Which report would I run? This would be the line listing a all AR summary data report. And this report can be run to review the AR summary or denominator data uploaded into your NHSN facility. You can see how many sacred inpatient days and admissions as well as any specific outpatient location encounters were reported for each month. I believe this is my final knowledge check of the session. So which report can I run if I wanted to see the percentage I've reported staff orus isolates that met the phenotype definition for MRSA for December 2024. Okay, the answer is rate tablet antimicrobial resistant percentages report. This report displays the total number isolated, total number tested, total number resistant by month. Using this information, the report will display the percentage of organisms that tested non-susceptible or resistant to certain antimicrobials for each defined phenotype, including MRSA. So, here I included what we'll call a cheat sheet for the AR option reports that were covered in today's presentation. I list out each report and provided a high-level overview of the report that you can reference whenever you're interested in reviewing your um AR option data. And we're also going to have um another cheat sheet that will a reference document that will be posted to um the website that will include like pathogen codes and things like that that we've been working on that will also be helpful. So in the next few slides I have listed some reporting resources for the AR option. Um we have our AU all module web page has lots of information and is a great um link to bookmark. It's a one-stop shop for the protocol, our validation materials, um link to our training resources, our analysis quick reference guides, as well as our AU and our AR FAQs, and a link to both uh CDA toolkits. I've also listed direct links to some of our analysis resources including our AR module protocol where you can find our phenotype definitions, our quick reference guides which have detailed information on how to run and interpret each AR option analysis report and other AUR training videos and slides including recordings and quick learns. That is the end of my presentation. Um as mentioned um please post your questions in the annual training community and we can get those answered today and tomorrow. Um but after um the annual training um if you have any questions about AR option reports um you can always uh submit questions to the NHSN help desk via service now or email nhsncdc.gov. Go.