NONMEM Users Network Archive

Hosted by Cognigen

Re: Time to event analysis using NONMEM

From: Nick Holford <n.holford>
Date: Tue, 09 Jun 2009 20:07:33 +0200

Leonid,

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

Serge,

This is an interesting suggestion. Can you be more explicit about when
truncation to zero might be a problem which would be ameliorated 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 using 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 elaborate 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 PRINT=1 MSFO=msfb01
>> $COV PRINT=E
>>
>>
>>
>> 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 information is strictly prohibited.
>> If you have received this transmission in error please 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 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. The 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 that information is subsequently confirmed in writing.
>>

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

Received on Tue Jun 09 2009 - 14:07:33 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.