Date:         Mon, 7 Jan 2002 06:40:23 -0700
Reply-To:     "Stanley A. Gorodenski" <vvgsgor@DE.STATE.AZ.US>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         "Stanley A. Gorodenski" <vvgsgor@DE.STATE.AZ.US>
Subject:      Re: Statistical Significance of Change in Virus Levels
Content-Type: text/plain; charset=us-ascii

This looks like it could be a factorial analysis design of patients by drug dose by time.  If so, that would be a better
alternative.
Stan


Melanie Barish wrote:

> Hello all - I am assuming some statistical duties temporarily, and I am
> pretty rusty at it.  I have a sample of data from of study of virus levels
> in blood over time.  There are 5 groups of patients, each having a
> different drug dose, and measurements were taken over 8 time points.  I
> need to find out if the changes in virus levels are significant.  I want
> to do this for each patient overall, looking at overall change from 1st
> measurement to last.  I also want to look at change at each time point.
> And at some point, I need to compare the significance between dosing
> regimens.
>
> My question is this:  Is the t-test the best method for me to do this?  I
> know it tests the significance between means for two populations.  I
> technically have only one population though.  If the t-test is not the
> right test, can you suggest one that I can use?
>
> Thanks in advance,
>
> Melanie
>
> *****************************************
> Melanie J Barish
> Clinical Programmer Analyst
> Progenics Pharmaceuticals, Inc
>
> 914-789-2864
> mbarish@progenics.com
> *****************************************