exclude when too few points text missing for aucinf.obs

[kirstenbergmann]

Hi Bill

Some time ago you included a text in exclude when there are two few datapoints to derive the half-life. However, I'm missing this text for aucinf.obs. The value itself (PPORRES) is set to NA but exclude is also NA. See below.

Best regards
Kirsten

library('PKNCA')
PKNCA.options(default=T)

# Last observation at 8 hours
my_conc <- data.frame(conc=c(0, 2.5, 3, 2.7, 2.3),
                      time=c(0:4),
                      subject=1)

conc_obj <- PKNCAconc(my_conc, conc~time|subject)
data_obj <- PKNCAdata(data.conc=conc_obj,
                      intervals=data.frame(start=0,
                                           end=Inf,
                                           aucinf.obs=TRUE))
results_obj <- pk.nca(data_obj)
#> No dose information provided, calculations requiring dose will return NA.
#> Warning in pk.calc.half.life(conc = c(0, 2.5, 3, 2.7, 2.3), time = c(0, :
#> Too few points for half-life calculation (min.hl.points=3 with only 2
#> points)
as.data.frame(results_obj)
#>    start end subject            PPTESTCD PPORRES
#> 1      0 Inf       1                tmax     2.0
#> 2      0 Inf       1               tlast     4.0
#> 3      0 Inf       1           clast.obs     2.3
#> 4      0 Inf       1            lambda.z      NA
#> 5      0 Inf       1           r.squared      NA
#> 6      0 Inf       1       adj.r.squared      NA
#> 7      0 Inf       1 lambda.z.time.first      NA
#> 8      0 Inf       1   lambda.z.n.points      NA
#> 9      0 Inf       1          clast.pred      NA
#> 10     0 Inf       1           half.life      NA
#> 11     0 Inf       1          span.ratio      NA
#> 12     0 Inf       1          aucinf.obs      NA
#>                                                                          exclude
#> 1                                                                           <NA>
#> 2                                                                           <NA>
#> 3                                                                           <NA>
#> 4  Too few points for half-life calculation (min.hl.points=3 with only 2 points)
#> 5  Too few points for half-life calculation (min.hl.points=3 with only 2 points)
#> 6  Too few points for half-life calculation (min.hl.points=3 with only 2 points)
#> 7  Too few points for half-life calculation (min.hl.points=3 with only 2 points)
#> 8  Too few points for half-life calculation (min.hl.points=3 with only 2 points)
#> 9  Too few points for half-life calculation (min.hl.points=3 with only 2 points)
#> 10 Too few points for half-life calculation (min.hl.points=3 with only 2 points)
#> 11 Too few points for half-life calculation (min.hl.points=3 with only 2 points)
#> 12                                                                          <NA>

[billdenney]

I'm about to implement this, but a more complex scenario presented itself. I think that all zeros for concentration yields AUC = 0, including AUCinf even though half-life is not estimable (and therefore PKNCA does, too).

So with the update I'm about to provide, in most cases, the exclusion will flow from the half-life calculation. But, in the case of all zeros, the AUC will not be excluded.

[kirstenbergmann]

Hi Bill

Good point. I didnt think about that scenario. Indeed I agree that all AUC values including AUCinf should be zero in that case.

best regards
Kirsten