Dear all,
I am modeling some PK data from one rabbit. And I have some doubts how to
codify the dataset. The rabbit received different single doses of a drug at
different times.
E.g,
First of May it received a 100 mg dose i.v. ,
Then, 2 months latter (enough washout period), the first of July, it received
and oral dose of 100mg.
Finally, after other two months the rabbit received the last oral dose 200mg.
I could codify in the TIME column the real times, including the jumps of two
months, but I think that there is an option in Nonmem to “restart” the dose
through the dataset.
Could somebody clarify me this point? and perhaps include a simple example of
dataset.
Thank you in advance
Mark
dataset codification
Dear Mark,
You may want to read about "EVID" in the html help. EVID = 4 is what you mean. It is a dosing event but implies that the system is reset before the dose is introduced into the system.
Sebastien
Mark Gutierrez a écrit :
> Dear all,
>
> I am modeling some PK data from one rabbit. And I have some doubts how to codify the dataset. The rabbit received different single doses of a drug at different times.
>
> E.g,
>
> First of May it received a 100 mg dose i.v. ,
>
> Then, 2 months latter (enough washout period), the first of July, it received and oral dose of 100mg..
>
> Finally, after other two months the rabbit received the last oral dose 200mg.
>
> I could codify in the TIME column the real times, including the jumps of two months, but I think that there is an option in Nonmem to “restart” the dose through the dataset. Could somebody clarify me this point? and perhaps include a simple example of dataset.
>
> Thank you in advance
>
> Mark
I have seen EVID=4 used, but don't remember the details. How does the time
variable look? I always convert Date and time to a time counting variable that
starts at 0 time (time of first dose) and always increases within each subject.
I recall that when EVID= 4 is used, the time variable does NOT need to restart
with the second dose. Do I remember correctly?
Note I do not include DATE and clock TIME in my NONMEM input, since I have seen
NONMEM do nefarious things with these.
Susan
Quoted reply history
________________________________
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Sébastien Bihorel
Sent: Wednesday, September 24, 2008 5:52 AM
To: Mark Gutierrez
Cc: [email protected]
Subject: Re: [NMusers] dataset codification
Dear Mark,
You may want to read about "EVID" in the html help. EVID = 4 is what you mean.
It is a dosing event but implies that the system is reset before the dose is
introduced into the system.
Sebastien
Mark Gutierrez a écrit :
Dear all,
I am modeling some PK data from one rabbit. And I have some doubts how to
codify the dataset. The rabbit received different single doses of a drug at
different times.
E.g,
First of May it received a 100 mg dose i.v. ,
Then, 2 months latter (enough washout period), the first of July, it received
and oral dose of 100mg..
Finally, after other two months the rabbit received the last oral dose 200mg.
I could codify in the TIME column the real times, including the jumps of two
months, but I think that there is an option in Nonmem to "restart" the dose
through the dataset.
Could somebody clarify me this point? and perhaps include a simple example of
dataset.
Thank you in advance
Mark
Susan,
NM-TRAN's translation of DATE and TIME is a critical part of the system. If you have found a problem then I encourage you to report the details to nmconsult <[EMAIL PROTECTED]> so that it can be investigated.
Willavize, Susan A wrote:
> Note I do not include DATE and clock TIME in my NONMEM input, since I have seen NONMEM do nefarious things with these.
Sebastian,
Sébastien Bihorel wrote:
> You are right Susan: the Time values have to increase within a subject. Using EVID=4 allows you to use a more convenient time scale
In general I try to avoid using EVID=4 because it requires the assumption of complete washout from previous doses. The convenience of every dose starting at time zero may be offset by the inconvenience of getting the wrong parameter estimates.
Nick
--
Nick Holford, Dept Pharmacology & Clinical Pharmacology
University of Auckland, 85 Park Rd, Private Bag 92019, Auckland, New Zealand
[EMAIL PROTECTED] tel:+64(9)923-6730 fax:+64(9)373-7090
http://www.fmhs.auckland.ac.nz/sms/pharmacology/holford
Mark,
When you use EVID =3, you reset the system. Time should be set to whatever time
appears at the new event.(0 in your case). This option is usually used in
inter occasion study within an individual record.
If you use EVID=4, you reset the system and introduce an new dose record. It is
identical to EVID=3 plus EVID=1.
ID TIME EVID...
1 0 1
1 1 0
... ... ...
1 0 4
is the same as
ID TIME EVID...
1 0 1
1 1 0
... ... ...
1 0 3
1 0 1
Xia Li
Xia Li
Department of Mathematical Science
University of Cincinnati
Quoted reply history
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Mark Gutierrez
Sent: 2008年9月24日 14:11
To: [email protected]
Subject: Re: [NMusers] dataset codification
Dear all,
First of all, thank you for your prompt reply.
Following with my doubts. In my case where the time between first and second
dose is two months, I understand that I could substitute in Time column the
real time (24hours * 2 months) by 0 when the second dose is administered. Is
that correct? Or as you mentioned I should include the real time
ID TIME EVID...
1 0 1
1 1 0
... ... ...
1 0 4
... ... ...
On the other hand,
In the help guide, when the interoccasion variability example is described, the
dataset is arranged with EVID = 3 including an extra line previous to the
second dose (second occasion) and restarting TIME with 0
ID TIME EVID...
1 0 1
1 1 0
... ... ...
1 0 3
1 0 1
1 1 0
... ... ...
This could be another solution to my problem. Isn’t it?
Thank you again.
Mark
----- Mensaje original ----
De: Sébastien Bihorel <[EMAIL PROTECTED]>
Para: "Willavize, Susan A" <[EMAIL PROTECTED]>
CC: Mark Gutierrez <[EMAIL PROTECTED]>; [email protected]
Enviado: miércoles, 24 de septiembre, 2008 19:03:43
Asunto: Re: [NMusers] dataset codification
You are right Susan: the Time values have to increase within a subject. Using
EVID=4 allows you to use a more convenient time scale
ex:
ID TIME EVID...
1 0 1
1 1 0
... ... ...
1 24 4
... ... ...
instead of
ID TIME EVID...
1 0 1
1 1 0
... ... ...
1 26546 1
... ... ...
Assuming you don't have any measurement after 24h during the first period of
your study.
Willavize, Susan A a écrit :
I have seen EVID=4 used, but don’t remember the details. How does the time
variable look? I always convert Date and time to a time counting variable that
starts at 0 time (time of first dose) and always increases within each subject.
I recall that when EVID= 4 is used, the time variable does NOT need to restart
with the second dose. Do I remember correctly?
Note I do not include DATE and clock TIME in my NONMEM input, since I have seen
NONMEM do nefarious things with these.
Susan
_____
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Sébastien Bihorel
Sent: Wednesday, September 24, 2008 5:52 AM
To: Mark Gutierrez
Cc: [email protected]
Subject: Re: [NMusers] dataset codification
Dear Mark,
You may want to read about "EVID" in the html help. EVID = 4 is what you mean.
It is a dosing event but implies that the system is reset before the dose is
introduced into the system.
Sebastien
Mark Gutierrez a écrit :
Dear all,
I am modeling some PK data from one rabbit. And I have some doubts how to
codify the dataset. The rabbit received different single doses of a drug at
different times.
E.g,
First of May it received a 100 mg dose i.v. ,
Then, 2 months latter (enough washout period), the first of July, it received
and oral dose of 100mg..
Finally, after other two months the rabbit received the last oral dose 200mg.
I could codify in the TIME column the real times, including the jumps of two
months, but I think that there is an option in Nonmem to “restart” the dose
through the dataset.
Could somebody clarify me this point? and perhaps include a simple example of
dataset.
Thank you in advance
Mark