Populating variable based on matching phrases

[smackerz1988]

Hello,

I have this spec for ANRIND 

Here is a sample of the data

data have;
input USUBJID $ LBSTNRC :$40. LBNRIND :$14. AVALC :$200.;
infile datalines dlm = '|';
datalines;
1001|Negative|NORMAL|Negative        
1001|5.0, 6.0, 7.0, 8.0|NORMAL|6.0
1001|5.0, 6.0, 7.0, 8.0|NORMAL|7.0
1001|5.0, 6.0, 7.0, 8.0|NORMAL|6.0
1001|Negative|ABNORMAL|5
1001|None, Occasional, 1-2|NORMAL|Occasional
1001|None, Occasional, 1-5|NORMAL|1-5                     
;

What is the best approach here? I know I could just do INDEX ,FIND, PRXCHANGE for matching but what is the best way to approach the comma separated variables and to make sure LBSTNRC and AVALC match and to account for the condition when they don't? 

[ballardw · Accepted solution]

Is AVALC ever more than that "one word" sort of appearance? If not, I don't understand the $200. input. If so, then you need to provide other examples of values of the AVALC variable.

My first feeling would be to use the FINDW function as you can specify the delimiter.

something like

if findw(lbstrnrc,strip(avalc),', ')>0 then <what you want to do when found>
else <what to do when not found>;

Strip is likely needed as you have defined AVALC as 200 characters in your example and would be looking for all 200 in the FINDW (or Find and Index or indexw functions)

Add in a Not missing(variable) to require the presence of a value where needed.

[smackerz1988]

Perfect thanks!