sas >> Efficacy Tables

by Al » Fri, 10 Apr 2009 03:13:40 GMT

Hi All

I dont know if my understanding is right , i have two datasets A,
B and they are as follows
A B
sub dsdt sub res resdt
1 1/2/07 1 PD 2/2/07
2 1/5/07 2 PD 2/5/07
3 1/9/07 3 SD 2/9/07


I am asked to calculate time to progression and duration of
response ..Is it not a simple merge A and B and find the find the
difference like TTP = resdt-dsdt
Resdt-- Date of last visit ,Dsdt ---Date of first Dose ,Res-Response


Please let me know i am not right .. any suggestions
are welcome!


Thank in advance

sas >> Efficacy Tables

by mlhoward » Fri, 10 Apr 2009 03:29:40 GMT


Al,

Your data looks a little odd for what you are asking. Usually
if you have a duration of response you have not just one record
for each ID, but multiple records. So I would expect data that
looks like this:

subject, visit date, treatment, response
1, 1/1/09, drug1, responding
1, 2/1/09, drug1, responding
1, 3/1/09, drug1, not responding

Then a duration of response would be to take the date of the
first dose to the date of the non-response (or perhaps the
date of the previous visit of the non-response).

While you say it is not a merge, if your B data is really
like the above, it might be helpful to merge the data first,
so that you are working with:

subject, visit date, treatment, date of first treatment, response,
and then once you get that the duration of the treatment would
be to pull the record before the non-response and then take the
difference in dates between the date on that record and the
date of the first dose (which if you've previously merged, is
right there).


-Mary






From: Al < XXXX@XXXXX.COM >
To: XXXX@XXXXX.COM
Subject: Efficacy Tables
Date: Thu, 9 Apr 2009 12:13:40 -0700

Hi All

I dont know if my understanding is right , i have two datasets A,
B and they are as follows
A B
sub dsdt sub res resdt
1 1/2/07 1 PD 2/2/07
2 1/5/07 2 PD 2/5/07
3 1/9/07 3 SD 2/9/07


I am asked to calculate time to progression and duration of
response ..Is it not a simple merge A and B and find the find the
difference like TTP = resdt-dsdt
Resdt-- Date of last visit ,Dsdt ---Date of first Dose ,Res-Response


Please let me know i am not right .. any suggestions
are welcome!


Thank in advance

sas >> Efficacy Tables

by Brian » Fri, 10 Apr 2009 03:32:54 GMT


> sub dsdt gt;ub res esdt
> 1 /2/07>1 PD /2/07
> 2 > /5/07 2 PD /5>07> > 3 /9/07 3 SD /9/07
>
> I am asked to al>ulate ime to progression and duration of
> response ..Is it not a>simple merge A and B and find the f>nd the
> difference like TTP = resdt-dsdt
> Resdt-- Date of last visi> ,>sdt ---Date of first Dose ,Res-Response
>
> Please let me know i am not ri>ht ..


You probably are not right.


> any suggestions are welcome!

You should 1) read the Protocol, 2) Read the stat plan, and/or 3)
discuss this with your statistician. Your task requires knowledge
fro> t>ese sources that SAS-L will not have.

Similar Threads

1. The model of efficacy table

I have a general question about efficacy table. At pharmaceutical industry,
what's model used for efficacy table, especially in oncology and anti-
infection? What kind of statistical analysis is conducted for efficacy?

2. Safty and efficacy table

3. What are Safety variables and Efficacy variables..?

Hi All,
=A0
=A0=A0=A0=A0=A0=A0=A0 What are Safety variables and Efficacy variables..? I=
=A0appreciate your help.
=A0
Regards,
DSTP=0A=0A=0A

4. Efficacy Data

5. TABLES TABLES TABLES

How would you best describe to a retiscent SAS student that the concept of
TABLES isn't limited to SQL?  This came up during a discussion on table
lookup methods - and lookup tables in particular (I mean, lookup tables pre-
date SQL).  Even SAS data sets are referred to as tables.  This made the
student apoplectic.

Words of wisdom most welcome!!!

Thanks,
Howard

6. Summary: Table combining tables? PROC FREQ or maybe

7. CREATE TABLE recursively references the target table

.. The warning is easy to reproduce, e.g. with

proc sql;
create table test as select *
from test;
quit;

WARNING: This CREATE TABLE statement recursively references the target
table. A consequence of this is a possible data integrity problem.

Normally, I would take a SAS warning seriously - ignoring this one, however,
I've never really run into data integrity problems of any obvious,
easy-to-spot kind.

Can anyone offer insight, or a counter-example?

Thank you.

8. SAS table and EXCEL table