NONMEM Users Network Archive

Hosted by Cognigen

RE: Is it reasonable to add covarite to PD parameters kin or kout?

From: Mats Karlsson <Mats.Karlsson>
Date: Tue, 28 Jul 2015 10:14:50 +0000


There are both pros and cons of using baseline observations as a covariate.=
 We presented four different options for how to handle baseline and investi=
gated estimation properties in the paper below. What we didn't look at, bec=
ause it is a very bad idea, is to use an baseline observation both as a cov=
ariate and dependent variable.

Approaches to handling pharmacodynamic baseline responses.
Dansirikul C, Silber HE, Karlsson MO.
J Pharmacokinet Pharmacodyn. 2008 Jun;35(3):269-83.

Best regards,

Mats Karlsson, PhD
Professor of Pharmacometrics

Dept of Pharmaceutical Biosciences
Faculty of Pharmacy
Uppsala University
Box 591
75124 Uppsala

Phone: +46 18 4714105
Fax + 46 18 4714003<http://www.farmb=>

From: owner-nmusers
 Behalf Of Peiming Ma
Sent: Tuesday, July 28, 2015 3:31 AM
To: Mark Sale; Zhao,Li; nmusers
Subject: [NMusers] RE: Is it reasonable to add covarite to PD parameters ki=
n or kout?

Dear Mark and Li,

I think there is nothing wrong with using baseline as a covariate. Baseline=
 observations contain more information than just some pre-treatment observa=
tions and thus have a lot of explanatory power; this is the reason they app=
ear significant more often than not. Comparisons after treatments should tr=
y to use all information before treatments (thus including baselines).

Statisticians habitually use baselines as covariates in their ANCOVA models=
 for obviously good reasons.


From: owner-nmusers
Sent: Tuesday, July 28, 2015 3:34 AM
To: Zhao,Li; nmusers
Subject: [NMusers] RE: Is it reasonable to add covarite to PD parameters ki=
n or kout?

I'm going to go out on a limb (since I haven't seen your data or understand=
 the biology) and suggest that the baseline value is not a covariate, but i=
s just another observation (presumably with drug concentration = 0). The=
 baseline value is sort of by definition a function of kin. So, you might =
look for predictors of Kin (age? Disease stage?) rather than using the obse=
rved value to predict the parameters (which then predict the observed value=
). Also note that covariates are, by definition, measured without error, a=
nd a baseline value is likely measure with error (the same error as any oth=
er observation?).


From: owner-nmusers
Sent: Monday, July 27, 2015 2:39 PM
To: nmusers
Subject: [NMusers] Is it reasonable to add covarite to PD parameters kin or=

Dear NMusers,

Right now I am doing covarite analysis for an indirect response model.

 I tested a few potential covarites and found it's STATISTICALLY significan=
t if I add observed baseline value to kin.

But I am not sure if it makes sense to add the baseline value as a covarite=
 to kin.

Could you please help me if you have had similar experiences before?

Thank you very much!

Received on Tue Jul 28 2015 - 06:14:50 EDT

The NONMEM Users Network is maintained by ICON plc. Requests to subscribe to the network should be sent to:

Once subscribed, you may contribute to the discussion by emailing: