A long note into muiltiple columns or rows

[X605191]

I have a patient data set with "notes" as a string variable. It is pretty long, and the delimiter used in the notes is "¶". 

Here is an example.

distress ¶XXX... ¶YYY... ¶STOP taking these medications ¶  ¶ XXX .05mg Tab ¶Commonly known as... ¶ YYY 100mg Cap ¶  ¶START taking these medications ¶  ¶ OOO 5mg tab ¶ .... ¶CONTINUE taking these medications  ¶UUU 100mg Cap   ¶   ¶

What I need to do is extracting

1. the lines from "STOP taking these medications" till right before "START taking these medications"
2. the lines from "START taking these medications" till right before "CONTINUE taking these medications"
3. the lines from "CONTINUE taking these medications" till the end of the notes

So I need three columns or rows per patient, but I cannot figure out how to write codes for that. Any advice would be greatly appreciated. Please let me know if anything is unclear. Thank you. 

[ChrisNZ · Accepted solution]

A basic way to do this:

data HAVE;
  STR0='distress ¶XXX... ¶YYY... ¶STOP taking these medications ¶  ¶ XXX .05mg Tab ¶Commonly known as... ¶ YYY 100mg Cap ¶  ¶START taking these medications ¶  ¶ OOO 5mg tab ¶ .... ¶CONTINUE taking these medications  ¶UUU 100mg Cap   ¶   ¶';
  
  POS1=index(STR0, 'STOP taking these medications')+length('STOP taking these medications');
  POS2=index(STR0, 'START taking these medications')-1;
  STR1=substr(STR0, POS1, POS2-POS1);
  
  POS1=index(STR0, 'START taking these medications')+length('START taking these medications');
  POS2=index(STR0, 'CONTINUE taking these medications')-1;
  STR2=substr(STR0, POS1, POS2-POS1);
  
  POS1=index(STR0,'CONTINUE taking these medications')+length('CONTINUE taking these medications');
  STR3=substr(STR0, POS1 );
run;
  

[X605191]

After minor modification, I made it work for my project 🙂 Thank you!