NONMEM Users Network Archive

Hosted by Cognigen

Re: [NMusers] RE: Link different thetas to same omega using mu referencing

From: Emmanuel Chigutsa <echigutsa_at_yahoo.com>
Date: Wed, 13 Apr 2016 20:47:09 +0000 (UTC)

Hi Jacob,
Yes, THETA(4) would represent the typical value (before exponentiation). In=
 summary, writing your covariate code after the definition of MU should cir=
cumvent the time-varying problem. If you want everything on a log scale you=
 could try:OCL=MU_1+ETA(1)CL=EXP(OCL+CLCOV)
The issue with trying to use different etas for the different thetas as you=
 had was that only the variance (omega) was constrained to be same, but not=
 necessarily the same eta for each ID.
Mannie
 
      From: "Leander, Jacob" <Jacob.Leander_at_astrazeneca.com>
 To: Emmanuel Chigutsa <echigutsa_at_yahoo.com>; "rupert.austin_at_btconnect.com"=
 <rupert.austin_at_btconnect.com>; "'Sadler, Brian'" <Brian.Sadler_at_iconplc.com=
>; "nmusers_at_globomaxnm.com" <nmusers_at_globomaxnm.com>
 Sent: Wednesday, April 13, 2016 2:00 PM
 Subject: RE: [NMusers] RE: Link different thetas to same omega using mu re=
ferencing
   
#yiv8370814822 #yiv8370814822 -- _filtered #yiv8370814822 {font-family:Helv=
etica;panose-1:2 11 6 4 2 2 2 2 2 4;} _filtered #yiv8370814822 {panose-1:2 =
4 5 3 5 4 6 3 2 4;} _filtered #yiv8370814822 {font-family:Calibri;panose-1:=
2 15 5 2 2 2 4 3 2 4;}#yiv8370814822 #yiv8370814822 p.yiv8370814822MsoNorma=
l, #yiv8370814822 li.yiv8370814822MsoNormal, #yiv8370814822 div.yiv83708148=
22MsoNormal {margin:0cm;margin-bottom:.0001pt;font-size:12.0pt;}#yiv8370814=
822 a:link, #yiv8370814822 span.yiv8370814822MsoHyperlink {color:blue;text-=
decoration:underline;}#yiv8370814822 a:visited, #yiv8370814822 span.yiv8370=
814822MsoHyperlinkFollowed {color:purple;text-decoration:underline;}#yiv837=
0814822 p.yiv8370814822msoacetate, #yiv8370814822 li.yiv8370814822msoacetat=
e, #yiv8370814822 div.yiv8370814822msoacetate {margin-right:0cm;margin-left=
:0cm;font-size:12.0pt;}#yiv8370814822 p.yiv8370814822msonormal, #yiv8370814=
822 li.yiv8370814822msonormal, #yiv8370814822 div.yiv8370814822msonormal {m=
argin-right:0cm;margin-left:0cm;font-size:12.0pt;}#yiv8370814822 p.yiv83708=
14822msochpdefault, #yiv8370814822 li.yiv8370814822msochpdefault, #yiv83708=
14822 div.yiv8370814822msochpdefault {margin-right:0cm;margin-left:0cm;font=
-size:12.0pt;}#yiv8370814822 span.yiv8370814822msohyperlink {}#yiv837081482=
2 span.yiv8370814822msohyperlinkfollowed {}#yiv8370814822 span.yiv837081482=
2emailstyle19 {}#yiv8370814822 span.yiv8370814822emailstyle20 {}#yiv8370814=
822 span.yiv8370814822emailstyle21 {}#yiv8370814822 span.yiv8370814822email=
style23 {}#yiv8370814822 p.yiv8370814822msonormal1, #yiv8370814822 li.yiv83=
70814822msonormal1, #yiv8370814822 div.yiv8370814822msonormal1 {margin:0cm;=
margin-bottom:.0001pt;font-size:11.0pt;}#yiv8370814822 span.yiv8370814822ms=
ohyperlink1 {color:#0563C1;text-decoration:underline;}#yiv8370814822 span.y=
iv8370814822msohyperlinkfollowed1 {color:#954F72;text-decoration:underline;=
}#yiv8370814822 p.yiv8370814822msoacetate1, #yiv8370814822 li.yiv8370814822=
msoacetate1, #yiv8370814822 div.yiv8370814822msoacetate1 {margin:0cm;margin=
-bottom:.0001pt;font-size:8.0pt;}#yiv8370814822 span.yiv8370814822emailstyl=
e191 {color:windowtext;}#yiv8370814822 span.yiv8370814822emailstyle201 {col=
or:#1F497D;}#yiv8370814822 span.yiv8370814822emailstyle211 {color:#1F497D;}=
#yiv8370814822 span.yiv8370814822emailstyle231 {color:#1F497D;}#yiv83708148=
22 p.yiv8370814822msochpdefault1, #yiv8370814822 li.yiv8370814822msochpdefa=
ult1, #yiv8370814822 div.yiv8370814822msochpdefault1 {margin-right:0cm;marg=
in-left:0cm;font-size:10.0pt;}#yiv8370814822 span.yiv8370814822EmailStyle36=
 {color:#1F497D;}#yiv8370814822 .yiv8370814822MsoChpDefault {font-size:10.0=
pt;} _filtered #yiv8370814822 {margin:70.85pt 70.85pt 70.85pt 70.85pt;}#yiv=
8370814822 div.yiv8370814822WordSection1 {}#yiv8370814822 Thanks for the an=
swers. Just as Rupert pointed out it is not possible to model time-depende=
nt covariates in the MU expression. I am not sure what is happening if you =
do but since the NONMEM manual explicitly tell you to avoid it you should n=
ot do it.   It seems that Mannies code instead uses a THETA(4) to mod=
el the “mean” value and then use CLCOV to change the value =
for different covariates. Or what does THETA(4) mean in your code Mannie? =
 Remember that the problems was to impose same omega element for diffe=
rent thetas depending on time-varying covariate and also estimate all the p=
arameters on a log scale.   //Jacob From: owner-nmusers_at_globomaxnm.co=
m [mailto:owner-nmusers_at_globomaxnm.com]On Behalf Of Emmanuel Chigutsa
Sent: den 13 april 2016 19:01
To: rupert.austin_at_btconnect.com; 'Sadler, Brian' <Brian.Sadler_at_iconplc.com>=
; nmusers_at_globomaxnm.com
Subject: Re: [NMusers] RE: Link different thetas to same omega using mu ref=
erencing   I think the following code should work:   IF(CAT.EQ.=
1) CLCOV = THETA(1) IF(CAT.EQ.2) CLCOV = THETA(2) IF(CAT.EQ.3=
) CLCOV = THETA(3) MU_1 = THETA(4) CL = EXP(MU_1+ETA(1))*(1=
+CLCOV)   $OMEGA BLOCK(1) 0.1   Mannie  


   

Received on Wed Apr 13 2016 - 16:47:09 EDT

The NONMEM Users Network is maintained by ICON plc. Requests to subscribe to the network should be sent to: nmusers-request_at_iconplc.com. Once subscribed, you may contribute to the discussion by emailing: nmusers@globomaxnm.com.