Date:         Fri, 19 Jun 2009 05:32:25 -0400
Reply-To:     Gerhard Hellriegel <gerhard.hellriegel@T-ONLINE.DE>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Gerhard Hellriegel <gerhard.hellriegel@T-ONLINE.DE>
Subject:      Re: comparision of two datasets with different time stamps

some things are very important for your example:

how should (could) the datasets be joined? How is a patient identified?
Are there more than 1 record per patient in table1? And in table2? If you
don't have a primary key to join (more than one record), which records
should be compared? All, some or one? Which one?

Also what is the target dataset? Eg. should that have 2 variables for the
time-point (actual, previous) together with 2 variables for severity? Or
should the target contain a kind of history with all the previous
timepoints and all the (maybe changed) severities? That would require a
different structure...
Gerhard



On Fri, 19 Jun 2009 02:15:37 -0700, Amar Mundankar
<amarmundankar@GMAIL.COM> wrote:

>On Jun 19, 2:03 am, Al <ali6...@gmail.com> wrote:
>> Hi All:
>> I have an ae dataset recevied at time point 1 and time point 2 .. How
>> can i summarize the differences(if any .. not the updated obs) between
>> these two dataset..if the severity of a patient is 1 in the dataset
>> (time point 1) and was changed to severity = 3 in the dataset(time
>> point 2 ).. How can i programatically spot it..
>>
>> Thanks in advance
>
>Hi,
>It will be better, if you can provide some sample data and table
>structure of the 2 tables.
>
>Rgds,
>Amar.