Analysis of the bacterial infection trends in the children population.
Analysis developed with the aim of using the 2.2 4CE consortium data
R/
: contains the files carrying out the analysispublic-data/ICD_bacterial4.csv
: mapping of ICD codes to bacterial infection categoriesoutput
: location of the files saved with by the analysis
First, clone the repository: git clone https://github.com/covidclinical/Phase2.2.bacterial_infections
Then open the folder 'data_extraction' and open the file bacterialInfection.Rmd and go to the section “Variables that need to be checked/modified by each site”:
- change the folder_4ce_files to the directory where your phase 2.2 AllAdm cohort data is located
- determine the obfuscation threshold:
- obfuscation = FALSE if no obfuscation
- the numeric value of the obfuscation threshold if any; e.g. obfuscation = 3 Make sure you comment the existing obfuscation line, set up as FALSE
- change the dateFormat to the one followed in your site (e.g., if your date looks like 03-AUG-20 follows the format "%d-%b-%y"). Further details about how to specificy the format below:
%d day as a number (0-31) 01-31 %a abbreviated weekday (e.g., Mon ) %A unabbreviated weekday (e.g, Monday) %m month (00-12) 00-12 %b abbreviated month (e.g, Jan) %B unabbreviated month (e.g, January) %y 2-digit year (e.g., 07) %Y 4-digit year (e.g., 2007)
After all these changes are done, run the bacterialInfection.Rmd. As an output the next files will be generated in the output folder:
- an html file named: bacterialInfection.html
- two RData files with ICD codes summary for QC