topic Re: How to make single column from different variables in SAS Programming

How to make single column from different variables

ambadi007 — Thu, 29 Jun 2017 17:59:54 GMT

Hi Experts ,

I havea a data set that the values are showed like different raws and I want to make that raws in a single column .No problem if the claim and suffixs shows multiple times

the data Have

Claim Suffix ICD10A ICD10B ICD10C ICD10D ICD10E
12589 1 vvg ijk ghy ilk opl
12589 2 rty ilk rtr erd owe

data want

Claim Suffix ICD10
12589 1 vvg
12589 1 ijk
12589 1 ghy
12589 1 ilk
12589 1 opl
12589 2 rty
12589 2 ilk
12589 2 rtr
12589 2 erd
12589 2 owe

How to proceed with this scenario Please advice

Re: How to make single column from different variables

Reeza — Thu, 29 Jun 2017 18:06:55 GMT

PROC TRANSPOSE - wide to long.

https://stats.idre.ucla.edu/sas/modules/how-to-reshape-data-wide-to-long-using-proc-transpose/

Transpose via a DATA step

https://stats.idre.ucla.edu/sas/modules/reshaping-data-wide-to-long-using-a-data-step/

Re: How to make single column from different variables

jdwaterman91 — Fri, 30 Jun 2017 13:09:18 GMT

Would something like this work for your situation?

data Some;
input Claim Suffix A $ B $ C $ D $ E $;
datalines;
12589 1 vvg ijk ghy ilk opl
12589 2 rty ilk rtr erd owe
;
run;

Data Some2;
Set Some;
Array New A--E;
Do _N_ = 1 to dim(New);
ICD10 = New(_N_);
Output;
End;
Keep Claim Suffix ICD10;
Run;

Re: How to make single column from different variables

ballardw — Thu, 29 Jun 2017 21:10:13 GMT

@jdwaterman91 minor suggestion:

Data Some2;
   Set Some;
   Array New A--E;
   Do _N_ = 1 to 5;
      ICD10 = New(_N_);
      if not missing(ICD10) then Output;
   End;
   Keep Claim Suffix ICD10;
Run;

@ambadi007 did not indicate whether or not the data may sometimes only have fewer than 5 codes, or possibly no codes at all, and what the output for those cases might be. I am assuming that a row with a missing ICD10 is pretty useless and excluding them.