Date:         Tue, 4 Mar 2008 14:44:51 -0500
Reply-To:     Mike Rhoads <RHOADSM1@WESTAT.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Mike Rhoads <RHOADSM1@WESTAT.COM>
Subject:      Re: Question abount counting strings
Comments: To: Guylhem Aznar <sasllist@guylhem.net>
In-Reply-To:  <9cc82a4d0803041019v1a8ccddcj237c7b728a86d58d@mail.gmail.com>
Content-Type: text/plain; charset="us-ascii"

Guylhem

So, a "state" is equivalent to a single letter, and a "transition" is
the combination of 2 adjacent states/letters, right?

If my understanding is correct, I wouldn't bother with regular
expressions.  I would just use a DATA step with a DO-loop to create a
restructured data set with one record for each patient and transition.
You could then use PROC FREQ or some other reporting procedure to count
the transitions.

DATA Transitions (KEEP = patient_id transition);
LENGTH patient_id $ 6  states $ 12  transition $ 2;
INPUT patient_id states;
DO I = 1 TO LENGTH(states) - 1;
  transition = SUBSTR(states,I,2);
  OUTPUT;
END;
CARDS;
124883 NH
124884 NHD
124885 NH
124886 N
124890 NPH
124899 NHPHPD
RUN;

Note that this does not create any records for patients like 124886, who
have only one state and therefore no transitions.  You could modify the
code to create a record for these patients if necessary.

Mike Rhoads
Westat
RhoadsM1@Westat.com

-----Original Message-----
From: owner-sas-l@listserv.uga.edu [mailto:owner-sas-l@listserv.uga.edu]
On Behalf Of Guylhem Aznar
Sent: Tuesday, March 04, 2008 1:19 PM
To: SAS-L@LISTSERV.UGA.EDU
Subject: Question abount counting strings


Hello

I have a problem : after various data steps, I have a record of the
list of states a patient has been in, ex:

patient_id states
124883 NH
124884 NHD
124885 NH
124886 N
124890 NPH

I would like to count the transition between the states - in perl I
would simply count the possible strings (NH, NP, PH, HP, HD, PD...)

How could I easily do that with that ?

There's a trick : a patient can return to previous states, or
alternate betweem states like NHPHPD (n=new, d=death)  : here there
should be 2 HP and 1 PH.

I have learn about the prx commands, but I have a hard time
translating from my regexp logic to their logic.

Any help would be welcome.

Thanks
Guylhem