NEED ASSISTANCE? Click here for our new AMR for R Assistant!

[msberends]

👋 Welcome!

We’re using Discussions as a place to connect with other members of our community. We hope that you:

• Ask questions you’re wondering about.
• Share ideas.
• Engage with other community members.
• Welcome others and are open-minded. Remember that this is a community we
  build together 💪.

AMR for R Assistant

https://chatgpt.com/g/g-M4UNLwFi5-amr-for-r-assistant

[ghost]

Hello, i would first like to thank all the team working on AMR package, this is wonderful.

As I am quite new to coding in R, I would like to have some help. I am currently working with an antimicrobial resistance dataset, focusing on multi drug resistance patterns in methicillin resistant staphylococcus aureus (MRSA).

1. is it possible to differentiate and get the exact count between MDR, XDR and PDR, using the CMI2012 guidelines?

2. concerning the MRSA, termed as such when resistant (R) to cefoxitin (FOX), is it possible to pool the proportion of resistance of MRSA to other antibiotics (in other word this is to specifically answer the question: what proportion of MRSA are drug/multi drug resistant ?)

I am working with a standard dataset with column headers as: patient_id, service, bacteria, and the rest of the columns containing susceptibility/resistance data of antibiotics.

[msberends]

Hi! Thank you so much for using the package and letting know that you find it helpful - that’s the sole reason we developed it 😊

Sure, your use case is probably quite common and indeed supported by the AMR package. We wrote an article on the determination of MDR/XDR/PDR: https://msberends.github.io/AMR/articles/MDR.html. In short, you can use the mdro() function to determine multi drug resistance.

You can just add an MDRO column like this:

# using tidyverse/dplyr 
library(dplyr)
library(AMR)
your_data %>%
  mutate(MDRO = mdro(col_mo = "bacteria"))

# using base R
your_data$MDRO <- mdro(your_data,
                       col_mo = "bacteria")

Now you can retrieve the counts you requested for using either of these methods:

# using tidyverse/dplyr 
library(dplyr)
library(AMR)
your_data %>%
  count(MDRO)

# using base R
table(your_data$MDRO)

To answer both your questions, you have to filter first before you use the count() or table() function. For you first question, you should filter on S. aureus isolates, for your second question you can afterwards filter on oxa or fluclox resistant isolates.

If you find it hard to filter, please let me know! I can add some examples for you.

[ghost]

thank you for your response. And yes, I will find it very helpul if you could provide me with an example on how to filter.

thank you again

[sherubpvet]

Hi, How do we calculate the MDR using "The CLSI guideline? "

[msberends]

Hi! I’m not sure I follow - is there a CLSI specific guideline or definition for MDRs? I might be a bit oblivious here coming from Europe where EUCAST is mostly used. We don’t have CLSI specific definitions in the package at the moment, but could add these if they exist.

You can now determine MDROs in these ways: https://msberends.github.io/AMR/reference/mdro.html

I’m happy to help if any help is needed.