Jump to content


Photo

Population Model with Samples from Two Compartments

nlme population

  • Please log in to reply
3 replies to this topic

#1 amh79@georgetown.edu

amh79@georgetown.edu

    Member

  • Members
  • PipPip
  • 11 posts

Posted 31 October 2018 - 02:21 PM

Greetings,

 

I have samples from individuals collected in plasma and then, 30 minutes later, in CSF. 

The drug given has a well established one compartment model but nothing done for the CSF. 

As I have samples from both, I am aiming to build a population model for both plasma and CSF. 

 

Questions:

 

1- Can phoenix model this in one model? That is, can I construct one model to characterize both at the same time or does this require separate modeling attempts?

2- If I can do this with one model, do I need to structure the data differently as there are multiple time points from different compartments with missing values? Can I use MDV for two separate observations?

 

 

 



#2 smouksassi1

smouksassi1

    Advanced Member

  • Members
  • PipPipPip
  • 231 posts
  • LocationMontreal

Posted 31 October 2018 - 02:27 PM

Hi,

 

1- Can phoenix model this in one model? That is, can I construct one model to characterize both at the same time or does this require separate modeling attempts?

 

Yes you can have a "joint" model where you model both at the same time you need to make sure you have two observation types say

Cplasma

Ccsf

( you will have to go to graphical or textual mode)

 

2- If I can do this with one model, do I need to structure the data differently as there are multiple time points from different compartments with missing values? Can I use MDV for two separate observations?

 

yes you can have something like that

 

ID TIME DV1 DV2             MDV

1     10     100     5                0

1       12    300                       0

1        12                                1

( note that in phoenix you can have empty DV so it wil lbe interpreted as missing no need for MDV)

but if you want MDV that applies differently to the different observations you need to split inot multiple lines so the MDV applies only to the line with the type of obs you want to declare as missing.



#3 amh79@georgetown.edu

amh79@georgetown.edu

    Member

  • Members
  • PipPip
  • 11 posts

Posted 31 October 2018 - 02:37 PM

Thank you so much - so I wouldnt need to specify MDV at all, except where both were missing or BLQ?



#4 smouksassi1

smouksassi1

    Advanced Member

  • Members
  • PipPipPip
  • 231 posts
  • LocationMontreal

Posted 31 October 2018 - 02:39 PM

bql you need a bql flag

mdv is more when you have a value in your data that you want to exclude due to a good reason







Also tagged with one or more of these keywords: nlme, population

0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users