Re: $PK ADVAN6

Robert I do not think you can do it directly. I've done something like this by introducing a parallel system (exactly the same as your model), adding a positive amount (that correspond to your subtraction) and then combining two solutions to get the prediction. However, you need linearity (of absorption in this case) to use superposition. Sample code: DADT(1)= -KA*A(1); depot; put all real doses here DADT(4)= -KA*A(4) ; depot for negative doses; put all doses that need to be subtracted here DADT(2)= KA*(A(1)-A(4)) + ...; Central You need to check that you subtract less than you have in the depot. If this is a concern, it is more difficult to implement. I would try something like F4=1 IF(NDOSE.GT.A(1)) F4=A(1)/NDOSE ; here NDOSE is the dose do subtract You need to be careful since A(1) in the PK block is the previous-event amount. You may need two records for each subtracted dose: one with EVID=2 to get A-value, and another with EVID=1 to get dose to compartment 4 Leonid -------------------------------------- Leonid Gibiansky, Ph.D. President, QuantPharm LLC web: www.quantpharm.com e-mail: LGibiansky at quantpharm.com tel: (301) 767 5566