FW: Modeling Hemodialysis clearance
I found the following note from Tom very helpful and want to share with other
users.
"It would be necessary to use the correct HEMO value with the understanding
that the covariate value at TIME=X2 applies to the interval from TIME=X1 up to
TIME=X2, X1<X2."
Quoted reply history
-----Original Message-----
From: Ludden, Thomas [mailto:[email protected]]
Sent: Thursday, July 21, 2011 3:08 PM
To: Xiao Hu
Cc: Bauer, Robert
Subject: FW: [NMusers] Modeling Hemodialysis clearance
Hi Xiao,
If the hemodialysate is being directly analyzed then the ON/OFF feature can be
used as for urine.
However, if the hemodialysate is not being directly analyzed, I am not certain
one needs the ON/OFF feature. Perhaps one just needs to have a
change in clearance.
Without Hemodialysis: CL=CL1
With Hemodialysis: CL=CL1+CL2
I think this could be modeled using a covariate in the data set:
HEMO=0/1
If HEMO=0 then CL=CL1 during periods without hemodialysis
If HEMO=1 then CL=CL1+CL2 during periods with hemodialysis
Where CL1 and CL2 could be modeled with THETA's and perhaps ETA's
It would be necessary to use the correct HEMO value with the understanding that
the covariate value at TIME=X2 applies to the interval from TIME=X1 up to
TIME=X2, X1<X2.
A caveat: The discontinuous change in clearance might be problematic for
derivatives.
Tom