Date:         Tue, 2 Sep 2008 21:06:08 -0400
Reply-To:     Muthia Kachirayan <muthia.kachirayan@GMAIL.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Muthia Kachirayan <muthia.kachirayan@GMAIL.COM>
Subject:      Re: counting comorbidities in a longitudinal dataset
In-Reply-To:  <2fc7f3340809021502k1cbffce1g369b19ddeeea628c@mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1

On Tue, Sep 2, 2008 at 6:02 PM, Muthia Kachirayan <
muthia.kachirayan@gmail.com> wrote:

>
>
> On Tue, Sep 2, 2008 at 5:36 PM, Magda <magdatolea@yahoo.com> wrote:
>
>> I have a dataset that looks like:
>>
>> id   visit   diagnosis
>> 1    1       ICD9 code
>> 1    1       ICD9 code
>> 1    2       ICD9 code
>> 1    2       ICD9 code
>> 1    2       ICD9 code
>> 1    2       ICD9 code
>> 2    1       ICD9 code
>>
>> etc  ....
>>
>> I am only interested in counting the number of diagnoses per visit,
>> assuming there are no duplicates within visits. In other words, I
>> would like to have a dataset that looks like:
>>
>> id   visit   dx_no
>> 1    1       2
>> 1    2       4
>> 2    1       1
>>
>> I am sure there is a very easy way to come up with this, I just don't
>> know enough SAS to do it. Your help will be very much appreciated.
>>
>> Thank you,
>> Magda
>>
>
> Under the stated assumptions, a datastep approach follows:
>
> data need;
> do until(last.visit);
>    set have;
>    by id visit;
>    if first.visit then dx_no = 0;
>    dx_no + 1;
> end;
> drop diag;
> run;
>
>
>
Still simpler. Assumed pre-sorted by id and visit and that diagnosis is
unique within visit.

data need;
retain id visit;
do dx_no = 1 by 1 until(last.visit);
   set have;
   by id visit;
end;
drop diag:;
run;