I think the following is correct:
The LAST dose of a steady state series is given AT THE TIME of the
SS record. So, for your q8h example, a level at 16:15 is 8:15h after the
last of the steady-state series.
Your non-regular-interval steady-state dose coding is almost right, but
unfortunately a miss is as good as a mile. Each steady state
as you now have it is REPLACING the last one, so at the time of
your observation, only the last SS will be present - that is, the system
will appear to have been given a dose of 200 q 24 hours, each day
at 2200 hours, the last dose of the series having been given 9:59h
before your observation. It should be coded as follows:
Date Time Dose SS II DV
9/1/94 08:00 200 1 24 0
9/1/94 16:00 200 2 24 0
9/1/94 22:00 200 2 24 0
9/2/94 07:59 0 0 0 23
Note the SS=2 on the SS records after the first - this ADDS IN the
SS solution for the corresponding SS pattern to the previous SS
solution. This then coresponds to SS with doses at 0800, 1600 and
2200 each day, the last dose of this pattern having been given
2200 9/1/94, and hence, again 9:59h before your observation.
Your last set of data records would generate a SS pattern of doses
at 0600, 1400 and 2200 each day, the last one of these given at 2200
on 9/1/94, as previously.
OK?
Intrepretation of Dosing Intervals
I need to clarify how the program interprets dosing intervals.
Suppose I have a dose given at 8:00 and indicate steady state with a dosing
interval of 8 hrs, but my level is obtained at 16:15, 8 1/4 hours later, does
the program assume that a dose was given at 16:00 and my 16:15 level is
actually a 15 minute level?
In most dosing scheems even if it is stated the drug is given Q8 the doses
are ofter 8:00 16:00 and 22:00 (especially in the pediactric population). If
I have the dosing information, I have been coding it like this.
Date Time Dose SS II DV
9/1/94 08:00 200 1 24 0
9/1/94 16:00 200 1 24 0
9/1/94 22:00 200 1 24 0
9/2/94 07:59 0 0 0 23
I don't know how the program would respond if I entered it like this.
9/1/94 22:00 200 1 8 0
9/2/94 07:59 0 0 0 23
because the interval between dose and level is actually 10 hours. I have
population data with the majority of levels drawn at trough concentrations.
I am wondering if my approach is correct.