Identifying ICD-codes in 30 different variables

[David Bergman]

Hi,

I have a dataset containing one variable identifying study participants and 30 variables containing ICD-codes (one per variable, strings). I need to identify every code indicating diabetes and generate a new variable, coded as 1 or 0 to indicate the presence or absence of diabetes. Of course I can write gen diabetes==1 if var1=="xxx" etc and repeat this up to var30. Alas there are quite a lot of different ICD-codes for diabetes so it would take a lot of time. Does anyone know of a faster way?

Best wishes,

David

[Carlo Lazzaro]

Please see https://www.statalist.org/forums/for...79321-for-loop, #2. Thanks.

[Nick Cox]

Cross-reference should be to https://www.statalist.org/forums/for...79321-for-loop Please use the CONTACT US button (bottom right) to contact the forum administrators to re-register with your full name.


The following can be extended:

gen diabetes = 0 

forval j = 1/30 { 
     replace diabetes = 1 if var`j' == "xxx" 
}

See also documentation on the inlist() function

[William Lisowski]

If you are working with ICD codes, Stata's commands that are designed specifically for working with ICD9 and ICD10 codes may help you. If you are not familiar with them,

help icd

is a place to start.

[David Bergman]

Thank you so much! I´ll request the name change. Best wishes,
David