Splitning one variable to multiple variables

[Chris_LK_87]

Hello,

In my dataset each row cointains information about one individual and their diagnosis code. The information on each individual's different diagnos is stored in one variable that can contain multiple diagnosis ( see 'data have'). I would like to split each diagnos code in to separate variables.  ('data want'). 

data have;
input id $ ICD $;
cards;
1 I63 H45 T11
2  C03 D25
3  J11 S54 A10 R45
;
run;

Data want;

id $ ICD1 $ ICD2 $ ICD3 $ ICD4 $;

1 I63 H45 T11 .
2  C03 D25 . .
3  J11 S54 A10 R45
;
run;

Thanks in advance!

[Kurt_Bremser · Accepted solution]

Much better: split it into observations:

data want;
set have (rename=(icd=_icd));
length icd $3;
do i = 1 to countw(_icd);
  icd = scan(_icd,i);
  output;
end;
keep id icd;
run;

[novinosrin]

Hi @Chris_LK_87  Do all those ICD's consistently have a length of 3 bytes?  

[Chris_LK_87]

Yes, they have a length of 3 bytes.

[ScottBass]

1.  Format your SAS code using the "SAS Code" icon.

2.  When I paste your code into SAS, the first data step doesn't give the correct results.  The second data step fails.

BZZZZZZZZZZZZZZZZZZZZZZZ!  Try again.

P.S.: Use the SCAN function and PROC TRANSPOSE (or an array - TRANSPOSE would be more flexible).  Read the doc for more details.