NONMEM Users Network Archive

Hosted by Cognigen

Re: Mixture model with logistic regression

From: Mark Sale <msale>
Date: Sat, 20 Feb 2016 20:01:50 +0000

Matts,

     Thanks for your insights. But, the issue isn't the post hoc values. W=
ith the mixture model the OMEGA on the intercept is huge (680), and the ent=
ire population is in the low intercept value group (Intercept = -11). Th=
en to accommodate the patients with frequent AEs, it assigns a (post hoc) E=
TA of +15, giving an individual value for intercept of 6 (and a probability=
 of the AE of ~1, as it should. My question is whey does it refuse to simp=
ly put those 8% of the patients in a sub population with intercept = 6, E=
TA=0. rather than saying the expected value is -11, with ETA = +15. Eve=
n when I fix the fractions in the subpopulations for the observed values, a=
nd fix OMEGA to a small, reasonable value, and fix the intercept values f=
or the 3 populations to reasonable values it will still do this. The only =
thing that has worked is to assign each subject to the apparent population =
in the data set.


Mark


Mark Sale M.D.
Vice President, Modeling and Simulation
Nuventra, Inc.
2525 Meridian Parkway, Suite 280
Research Triangle Park, NC 27713
Office (919)-973-0383
msale
www.nuventra.com<http://www.nuventra.com>



Empower your Pipeline

CONFIDENTIALITY NOTICE The information in this transmittal (including attac=
hments, if any) may be privileged and confidential and is intended only for=
 the recipient(s) listed above. Any review, use, disclosure, distribution o=
r copying of this transmittal, in any form, is prohibited except by or on b=
ehalf of the intended recipient(s). If you have received this transmittal i=
n error, please notify me immediately by reply email and destroy all copies=
 of the transmittal.



________________________________
From: Matts Kgedal <mattskagedal
Sent: Saturday, February 20, 2016 2:44 PM
To: Mark Sale
Cc: nmusers
Subject: Re: [NMusers] Mixture model with logistic regression

Hi Mark,
The pattern you see in the posthocs could possibly be a shrinkage phenomeno=
n. I.e. patients with AE most of the time will have the same ETA, while pat=
ients with no AE will have the same ETA and there will be a third group in =
 between. If shrinkage is causing this, you should not expect any improveme=
nt with a mixture model. Before you reject your original model I would ther=
efore also evaluate it by simulation and re-estimation. I think it is quite=
 possible that you will retreive a similar pattern in the posthocs even whe=
n you simulate based on a normal distribution.
Best,
Matts Kgedal
Pharmacometrics, Genentech.

On Fri, Feb 19, 2016 at 2:30 PM, Mark Sale <msale
  

Has anyone every tried to use a mixture model with logistic regression? I h=
ave data on a AE in several hundred patients, measured multiple times (10-2=
0 times per patient). Examining the data it is clear that, independent of =
drug concentration, there is very wide distribution of this AE, 68% of the =
patients never have the AE, 25% have it about 20% of the time and the rest =
have it pretty much continuously, regardless of drug concentration. (in or=
dinary logistic regression, just glm in R, there is also a nice concentrati=
on effect on the AE in addition). Running the usual logistic model, not s=
urprisingly, I get a really big ETA on the intercept, with 68% of the peopl=
e having ETA small negative, 25% ETA ~ 1 and 7% ETA ~ 10. No covariates see=
m particularly predictive of the post hoc ETA. I thought I could use a mix=
ture model, with 3 modes, but it refused to do that, giving me essentially =
0% in the 2nd and 3rd distribution, still with the really large OMEGA for t=
he intercept. Even when I FIX the OMEGA to a reasonable number, I still ge=
t essentially no one in the 2nd and 3rd distribution. I tried fixing the f=
raction in the 2nd and 3rd distribution (and OMEGA), and it still gave me a=
 very small difference in the intercept for the 2nd and 3rd populations.

Is there an issue with using mixture models with logistic regression? I'm j=
ust using FOCE, Laplacian, without interaction, and LIKE.




Any ideas?


Mark


Mark Sale M.D.
Vice President, Modeling and Simulation
Nuventra, Inc.
2525 Meridian Parkway, Suite 280
Research Triangle Park, NC 27713
Office (919)-973-0383<tel:%28919%29-973-0383>
msale
www.nuventra.com<http://www.nuventra.com>




Received on Sat Feb 20 2016 - 15:01:50 EST

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.