Baseline correction GeoMean and Median [Regulatives / Guidelines]

posted by mittyri – Russia, 2024-11-11 12:20 (20 d 05:51 ago) – Posting: # 24271
Views: 1,607

Hi BEQool,

❝ why is geomteric mean not 0 if one of the values is 0 (or BLQ)? Are only positive values included in the calculation of geometric mean? I know that log(0) cannot be calculated but on the other hand value of 0 could be included in the calculation of geomean (equation: n-th root of the product of n values) and the geomean would be 0 regardless of other values.


While intuitively, you might think that the geometric mean of a set containing 0 should be 0 (since any product involving 0 is 0), the formal definition and intended purpose of the geometric mean don't align well with this interpretation.
Imagine 2 datasets:
c(1,1,1,1,0)
c(0,0,0,0,1)

Then both will have the same geometric mean value. It loses information about the other values and doesn't accurately reflect the "typical value" or "average rate" that the geometric mean aims to represent.

❝ We are currently studying an endogenous compound for which the endogenous baseline samples mostly below BLQ or slightly higher are expected. What is preferred for the time-averaged method in this case, to take the geometric mean or the median? In case the median is used, are BLQ values replaced with 0 not taken into account when calculating median as shown above (similar to geomean calculation)?


Regarding geometric mean: kind of tricky taking into account the information above. The median is better, but why don't you want to follow the method mentioned in BE GL at the first place (arithmetic mean)? Yes, it is bad, but at least not worse than others :-D

❝ We were also wondering, if the pre dose sample at 0 h timepoint can be simultaneously considered as for detection of baseline concentration and also as the first PK sample? I would say that it can be used and the pro of this is that you dont have to take one more sample as well (in connection with the total volume of blood taken). On the other hand, one could argue that the same value will be taken into account twice; e.g. if we take the median for time-averaged method and this timepoint at 0h is the median, then baseline correction for the first pharmacokinetic sample would be 0 as the same value would be subtracted.


"one could argue that the same value will be taken into account twice" - employing the arithmetic mean instead would effectively mitigate this potential issue, even if it were to arise (though I believe it's unlikely given that all prerequisites are met, and the extra sample appears to be an unnecessary use of resources).

Kind regards,
Mittyri

Complete thread:

UA Flag
Activity
 Admin contact
23,328 posts in 4,898 threads, 1,662 registered users;
85 visitors (0 registered, 85 guests [including 16 identified bots]).
Forum time: 18:12 CET (Europe/Vienna)

Satisfaction of one’s curiosity is one of the greatest sources
of happiness in life.    Linus Pauling

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz
HTML5