Bioequivalence and Bioavailability Forum

avrahan ☆ 2014-09-17 23:40 (3875 d 23:18 ago) Posting: # 13530 Views: 10,073	CV calculation [Power / Sample Size] Post reply
	Hi I have found an old study which the ISCV is not stated in the results. Is it possible to calculate this parameter using IC90 and sample size? The study was performed with 24 subjects - 22 finalized IC90: Cmax: 92.08 - 123.44 AUC0-t 93.08 - 113.87 Can anybody help me on this? Thanks Regards

Helmut
★★★

Vienna, Austria,
2014-09-18 03:21
(3875 d 19:37 ago)

@ avrahan
Posting: # 13531
Views: 8,240

CV calculation

Post reply

Hi avrahan,

❝ I have found an old study which the ISCV is not stated in the results. Is it possible to calculate this parameter using IC90 and sample size?

If you mean CI90, yes. Any of my presentations dealing with sample size estimation will give you the required formulas. Then you can calculate the CV manually. More convenient is R / package PowerTOST / function CVfromCI().

❝ The study was performed with 24 subjects - 22 finalized

❝ Cmax: 92.08 - 123.44

❝ AUC0-t 93.08 - 113.87

library(PowerTOST)

CVfromCI(lower=0.9208, upper=1.2344, n=22, design="2x2", alpha=(1-0.9)/2)

[1] 0.2875014

CVfromCI(lower=0.9308, upper=1.1387, n=22, design="2x2", alpha=(1-0.9)/2)

[1] 0.1956711

—
Dif-tor heh smusma 🖖🏼 Довге життя Україна!
Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes

BRN ☆ Turkey, 2015-08-26 18:52 (3533 d 04:06 ago) @ Helmut Posting: # 15342 Views: 6,806	CV calculation Post reply
	Hi all, I have intra and inter-cv, is there an easy way to calculate the total cv? An R code for example or any other suggestions are appreciated. Thank you... — Regards, BRN

d_labes
★★★

Berlin, Germany,
2015-08-27 10:48
(3532 d 12:10 ago)

@ BRN
Posting: # 15349
Views: 6,884

CV(total) from CV(intra) and CV(inter)

Post reply

Dear BRN!

❝ I have intra and inter-cv, is there an easy way to calculate the total cv? An R code for example or any other suggestions are appreciated.

Have a look into Helmut's lectures, f.i. the Moscow lecture about sample size estimation, slide 18. If your design was 2x2x2 which gave you the intra and inter-CV, and if you obtained the CV's from log-transformed analysis.

Using the world-famous R package PowerTost :cool:

you may use:

library(PowerTOST)

# Methylphenidate on slide 19, Helmut's Moscow lecture

# CV's as ratio

CVintra <- 0.070

CVinter <- 0.191

CVtotal <- mse2CV(CV2mse(CVintra)+CV2mse(CVinter))

print(CVtotal)

will give 0.2038621, i.e. 20.4%.

—
Regards,

Detlew

CV calculation [Power / Sample Size]

CV calculation

CV calculation

CV(total) from CV(intra) and CV(inter)