NONMEM Users Network Archive

Hosted by Cognigen

Re: Priors and covariate model building

From: Ribbing, Jakob <Jakob.Ribbing>
Date: Sat, 23 Jun 2012 11:25:39 +0000

Dear Palang and Martin,

For the published analysis; do you have any information on the covariates t=
hat you would like to investigate? (mean and sd or range). Another factor w=
eighting in the approach you take may be what functional form(s) you consid=
er for continuous covariates (e.g. Linear vs. power).

If you have the means for the previous analysis then one simple solution ma=
y be to centre any investigated covariates around these (prior) covariate m=
eans. If you find any highly important covariates, you may additionally con=
sider a lower omega on that parameter since the prior did not take this cov=
ariate into account. (with a linear cov model and in the simplest case: bas=
ed on covariate sd in the previous study and the estimated covariate coeff=
icient - this correction could be implemented on the fly, but is only impor=
tant if you study pop has any very important cov effects beyond the allomet=
ry correction).

Best regards


Skickat frn min iPhone

22 jun 2012 kl. 19:39 skrev "Palang Chotsiri" <Palang

> Dear NMusers,
> I am trying to model a sparse dataset by using the benefit of previously =
published parameter estimates (based on rich data sampling). When applying =
the $PRIOR subroutine, the THETAs and ETAs estimates of the new dataset are=
 reasonable and the model fit satisfactory.
> My question now relates to covariate modeling when a prior is applied. No=
 significant covariate relationships are included in my prior model (apart =
from allometric scaling). The prior was derived based on rich PK sampling b=
ut a fairly small sample size. The later sparse sampling study is conducted=
 in a larger group compare to the previous study. This might render us a gr=
eater power to detect covariate relationships based on this dataset.
> Or problem lies in that we do not know how we can correctly conduct a cov=
ariate model search with this model? The parameter estimates of the prior a=
re conditioned on the covariate distribution in the dataset on which it was=
 derived and are not necessarily relevant when a covariate relationship is =
> Perhaps there is no ideal solution but we would be grateful for any ideas=
 on how to best conduct covariate model building when a prior is used.
> Best regards,
> Palang Chotsiri & Martin Bergstrand
> Mahidol-Oxford Tropical Medicine Research Unit,
> Bangkok 10400, THAILAND
> Ps. Ideal is of course to model both datasets together but that might not=
 always be possible for practical reasons.
Received on Sat Jun 23 2012 - 07:25:39 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: