NONMEM Users Network Archive

Hosted by Cognigen

[NMusers] RE: Large errors in the estimation of volume of distribution (Vd) for sparse data

From: HUI, Ka Ho <>
Date: Tue, 10 Nov 2015 04:12:59 +0000

Thanks for your responses!

Nitin, I encountered an error when generating VPC by PsN. It says "No DV va=
lues found after filtering original data. At lib/tool/ line 2215=
." What does it mean?

Felix, Past published data suggested similar parameter estimates and models=
 compared to my final model. This is PO and I fixed Ka at a pre-estimated v=
alue (So no estimation of fixed or random effect).

Ahmad, Yes. The CV is even larger.


From: Abu Helwa, Ahmad Yousef Mohammad - abuay010 [mailto:ahmad.abuhelwa_at_my=]
Sent: Tuesday, November 10, 2015 5:34 AM
To: HUI, Ka Ho <>;
Subject: RE: Large errors in the estimation of volume of distribution (Vd) =
for sparse data

Hi Mathew,

Have you tried using an exponential model for vd ? like this: Vd = TEHTA=


From: felix boakye-agyeman []
Sent: Tuesday, November 10, 2015 12:41 AM
To: HUI, Ka Ho <>
Subject: Re: [NMusers] Large errors in the estimation of volume of distribu=
tion (Vd) for sparse data

   Do you have historical data to compare you data to? (Do you know if you =
are hitting a local minimum)
Is this iv or po, if its po how is your Ka?
You may also be over-parameterized due to your data

From: Kaila, Nitin []
Sent: Tuesday, November 10, 2015 12:14 AM
To: HUI, Ka Ho <>
Subject: RE: Large errors in the estimation of volume of distribution (Vd) =
for sparse data


Construct visual predictive check (VPC) plots, using all the estimates of t=
he bootstrap runs, as that will be a more true estimate of overall variabil=
ity in the Cp predictions.

Use the -rawres option in PsN to perform the VPC, and then compare your ori=
ginal final model VPC plot with the VPC plot with all estimates of the boot=


From:<> [ma=] On Behalf Of HUI, Ka Ho
Sent: Monday, November 9, 2015 9:43 AM
Subject: [NMusers] Large errors in the estimation of volume of distribution=
 (Vd) for sparse data

Dear all,

I have some population PK data which are in general very sparse (95% have o=
nly 1 blood sample between 2 successive doses). I developed a population PK=
 model with the one-compartment model with 1st order absorption. The progre=
ss is generally okay except that whenever a random effect, i.e. *(1+ETA(1))=
, is used to describe distribution of Vd, OMEGA would be estimated to be ve=
ry large (around 45% in terms of CV, with 80% Shrinkage), despite statistic=
al significance (dOF approx. -5.5). So I dropped the random effect and expr=
essed Vd in terms of a single fixed effect. When the final model has come o=
ut, I performed bootstrap and found that most estimates are accurate except=
 Vd, which has a very large standard error and bias (mean 232, bias 49, SE =
156), while the estimates for CL and other parameters look normal. I then c=
onstructed the predictive plots for the developed model using both the orig=
inal estimates (i.e. estimates using my original dataset) (#1) and estimate=
s from one of the bootstrap runs which has an extreme estimate of Vd (9xx) =
(#2), and found out that the two plots of plasma profiles are quite differe=
nt in terms of the shape (#1 is "taller", #2 is much flatter) but have simi=
lar average Cp.

These seem to be suggesting that given my sparse data, it is impossible to =
require accurate estimations of both CL and Vd. Apart from fixing Vd to a f=
ixed value, is there any other possible solutions? Or is there anything tha=
t I might have overlooked?

Thanks and regards,

Received on Mon Nov 09 2015 - 23:12:59 EST

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: