NONMEM Users Network Archive

Hosted by Cognigen

Re: Time to event analysis using NONMEM

From: Anubha_Gupta
Date: Wed, 10 Jun 2009 11:31:06 +0100


Dear all

Thanks a lot for your responses. I managed to run the =
model successfully by
adding TIME=0 in my dataset. I can now work further=
 and try other
suggestions.

Best regards

Anubha


                         =
                         =
                         =
             
  From: Nick Holford <n.holford
> =
          
                         =
                         =
                         =
             
  To: nmusers <nmusers
                         =
               
                         =
                         =
                         =
             
  Date: 09/06/2009 19:17 =
                         =
                         =

                         =
                         =
                         =
             
  Subject: Re: [NMusers] Time to event analysis =
using NONMEM =
     
                         =
                         =
                         =
             
  Sent by: owner-nmusers
                         =
               
                         =
                         =
                         =
             





Leonid,

Your suggestion to use $PRED is fine for this simple e=
xample with a
constant hazard (or easily integrable function). But more =
generally the
hazard may be tough to integrate analytically and thus o=
ne must use $PK
and and $DES.

Serge,

This is an interesting suggestion. Can you be more expli=
cit about when
truncation to zero might be a problem which would be a=
meliorated by
using -2LL?


Nick


Leonid Gibiansky wrote:
> Anubha
> You do not have time=0 records, therefore, A(1)=0
> Try
> HZ=BASE*TIME
> (and use $PRED rather than $PK)
>
> Leonid
Serge Guzy wrote:
> I am not sure it will help but for time to event=
 data I always used
> the log-likelihood instead of the likelihood. This is =
usually done to
> prevent truncation to zero (Like the default when usin=
g the normal
> likelihood). Did you try?
>

> Anubha_Gupta
>> Dear NMUsers
>>
>> I am trying to analyse time to single event data =
using NONMEM VI. I
>> am getting the following termination message. If some =
body could
>> point out what I am doing wrong or may be elabora=
te on the error
>> message that would be very helpful.
>>
>> Thanks in advance
>>
>> Anubha
>>
>>
>> 0PROGRAM TERMINATED BY OBJ
>> ERROR IN CLIK WITH INDIVIDUAL 1 ID=0.40000000E+01
>> CONDITIONAL LIKELIHOOD FOR DATA FROM OBS RECORD 1
>> ESTIMATED TO BE NEGLIGIBLE
>> MESSAGE ISSUED FROM ESTIMATION STEP
>> AT INITIAL OBJ. FUNCTION EVALUATION
>>
>>
>> Data set
>>
>>
>> C STUD ID TIME CS =
 DOSE
>> . 101
>> 4
>>
>> 111
>>
>> 0
>>
>> 0.4
>> . 101
>> 5
>>
>> 51
>>
>> 0
>>
>> 0.4
>> . 101
>> 6
>>
>> 31
>>
>> 0
>>
>> 0.8
>> C Missing DV 101
>> 7
>> . .
>> 0.8
>> . 101
>> 8
>>
>> 59
>>
>> 0
>>
>> 1.6
>> . 101
>> 9
>>
>> 477
>>
>> 1
>>
>> 3.2
>> . 101
>> 11
>>
>> 278
>>
>> 0
>>
>> 3.2
>> . 101
>> 12
>>
>> 58
>>
>> 0
>>
>> 6.4
>>
>> CS=1 is the censored event
>> TIME is in days
>>
>>
>> Model file:
>>
>> $PROB Time To Event data
>>
>> $INPUT C STUD ID TIME=DV CS DOSE
>>
>> $DATA
>>
>> IGNORE=C
>>
>> $SUBR ADVAN=6 TOL=6
>> $MODEL COMP=(RISK)
>>
>> $PK
>>
>> BASE = THETA(1)*EXP(ETA(1)) ;the ETA is a placeholder=
 here
>>
>> $DES
>> DADT(1)=BASE ;hazard
>>
>> $ERROR
>>
>>
>> HZ = A(1) ;hazard
>> SUR = EXP(-HZ) ;survival probability
>> DENS=HZ*SUR
>> Y = (1-CS)*DENS + CS*SUR
>>
>> $THETA (0,.025) ; BASE
>> $OMEGA 0 FIX
>>
>> $ESTIM MAXEVAL=9990 METHOD = COND LAPLACE LIKE PRIN=
T=1 MSFO=msfb01
>> $COV PRINT=E
>>
>>
>>
>> THIS EMAIL AND ANY ATTACHED FILES ARE CONFIDENTIAL.
>> If you are not the intended recipient you are notif=
ied that any
>> disclosure, reproduction,copying, distribution, or action t=
aken in
>> reliance on the contents of this information is stric=
tly prohibited.
>> If you have received this transmission in error pleas=
e notify the
>> sender immediately and then delete this email. Email =
transmission
>> cannot be guaranteed to be secure or error free as =
information could
>> be intercepted, corrupted, lost, destroyed,arrive late or=
 incomplete,
>> or contain viruses. The company/sender accepts no liabi=
lity for any
>> changes made to this email during transmission or any=
 damage caused
>> by any virus transmitted by this email. Any views o=
r opinions
>> expressed in this email are solely those of the aut=
hor and do not
>> necessarily represent those of the company. The company=
/sender
>> accepts no liability for the content of this email, =
or for the
>> consequences of any actions taken on the basis of t=
he information
>> provided, unless that information is subsequently confirm=
ed in writing.
>>

--
Nick Holford, Dept Pharmacology & Clinical Pharmacology
University of Auckland, 85 Park Rd, Private Bag 92019, A=
uckland, New
Zealand
n.holford
mobile: +33 64 271-6369 (Apr 6-Jul 17 2009)
http://www.fmhs.auckland.ac.nz/sms/pharmacology/holford



______________________________________________________________________
This email has been scanned by the MessageLabs Email Sec=
urity System.
For more information please visit http://www.messagelabs.com/ema=
il
______________________________________________________________________


THIS EMAIL AND ANY ATTACHED FILES ARE CONFIDENTIAL.
If you are not the intended recipient you are notified =
that any disclosure, reproduction,copying, distribution, or =
action taken in reliance on the contents of this info=
rmation is strictly prohibited. If you have received =
this transmission in error please notify the sender immedi=
ately and then delete this email. Email transmission ca=
nnot be guaranteed to be secure or error free as infor=
mation could be intercepted, corrupted, lost, destroyed,arrive=
 late or incomplete, or contain viruses. The company/sende=
r accepts no liability for any changes made to this =
email during transmission or any damage caused by any =
virus transmitted by this email. Any views or opinions =
expressed in this email are solely those of the author =
and do not necessarily represent those of the company. T=
he company/sender accepts no liability for the content =
of this email, or for the consequences of any actions =
taken on the basis of the information provided, unless t=
hat information is subsequently confirmed in writing.


graycol.gif
(image/gif attachment: graycol.gif)

ecblank.gif
(image/gif attachment: ecblank.gif)

Received on Wed Jun 10 2009 - 06:31:06 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.