Is your model a good fit to the data (e.g. as judged by diagnostic plots)? If
so, I would guess that your units may be the problem and you may need to
include a scaling factor for V in S2. eg. S2= V/1000. So check that your units
are compatible for dose, V and your DV. I assume that your fixed value for Ka
is ok (you have a good basis for the value you are using).
If your model is not a good fit to your data, you may need better initial
estimates.
It's not usually a good idea to impose un-natural limits on parameters, like a
fixed upper bound for PK parameters. (but setting a lower bound on pk
parameters does make sense.)
It's also generally a good idea to give a subject for your posts so others can
judge whether they want to contribute (or not) and there also will be a thread
to follow.
Sent: Saturday, March 5, 2011 12:02:54 PM
Hi
I am trying to build a structural model for an oral drug. I am getting
extremely large and unreasonable values for vd, eg a billion L, putting an
upper limit on theta will give a near boundary warning.
Does anyone have an idea why this might occur and how to solve it.
This is part of the input file:
$SUBROUTINE ADVAN2 TRANS2
$PK
TVCL = THETA(1)
CL = TVCL*EXP(ETA(1))
V = THETA(2)
KA = THETA(3)
S2 = V
$THETA (0,10)
$THETA (0,20)
$THETA 2 FIXED
Best regards
Susan