# Re: estimating Ka from dataset combining rich sample study and sparse sampling study

From: Ethan Wu <ethan.wu75>
Date: Wed, 17 Jun 2009 14:05:19 -0700 (PDT)

Hi Jakob,    sparse data came from MD study. and IIV on CL increase=
d from 0.14 to 0.25, on V from 0.185 to 0.196 after inclusion of sparse dat=
a    both in the same population.   I think what you suggest =
making sense to me. I would keep Eta on Ka first, start exploring IOV on =
CL and V, then explore covariates on CL and V, to see if decreasing IIV on =
CL and V would leads to more reasonable estimate of IIV on Ka.    =
but, overall, I think that it is the stress of shrinkage on Ka leads to =
"dumping" IIV to CL and V, not something wrong with the model itself. =
________________________________ From: "Ribbing, Jakob=
" <Jakob.Ribbing
Bulitta <jbulitta
elliffe <jelliffe
sday, June 17, 2009 4:43:28 PM Subject: RE: [NMusers] estimating Ka from =
dataset combining rich sample study and sparse sampling study Hi Et=
han,   If OMEGA(?) for KA is drastically reduced when including the s=
parse data, then something is wrong with your model and in this case it is =
not the estimation method or assumption on distribution of individual param=
eter). Eta-shrinkage would not drastically reduce the estimate of OMEGA, si=
nce this estimate is driven by the subjects/studies which contain informati=
on on the parameter.   If the sparse data is multiple dosing it may b=
e that KA is variable between occasions, rather than between subjects (assu=
ming the sparse data contain some information on KA). Or if the sparse data=
is from a less well-controlled study or a different population, it may be =
that increased IIV in other parts of the model (e.g. OMEGA on V) is making =
IIV in KA appear low for the rich study, when fitting the two studies toget=
her. If you get the covariate model in place this problem will be solved. F=
or the simple model you have it should be quick to start out assuming that =
most parameters (THETAs and OMEGAs) are different between the two studies a=
nd then reduce down to a model which is stable and parsimonious. Obviously,=
if you eventually can explain the differences using more mechanistic covar=
iates than study number that is of more use.   Cheers   Jakob=
Received on Wed Jun 17 2009 - 17:05:19 EDT

The NONMEM Users Network is maintained by ICON plc. Requests to subscribe to the network should be sent to: nmusers-request@iconplc.com.

Once subscribed, you may contribute to the discussion by emailing: nmusers@globomaxnm.com.