Log in
Register|
Chiku ☆ India, 2012-02-15 07:33 (4822 d 07:52 ago) Posting: # 8125 Views: 4,061 |
|
|
Dear all, For calculating sample size of parallel design, I need to know total CV. In the absence of any previous knowledge of a parallel study, how can I estimate this total CV? Is it possible to estimate the total CV from simple arithmetic CV of reference from literature (assuming that test will have similar intra and inter subject variability)? Regards, Chiku:) Edit: Category changed. [Helmut] |
|
Helmut ★★★   Vienna, Austria, 2012-02-15 15:19 (4822 d 00:06 ago) @ Chiku Posting: # 8126 Views: 3,402 |
|
|
Dear Chiku! ❝ For calculating sample size of parallel design, I need to know total CV. In the absence of any previous knowledge of a parallel study, how can I estimate this total CV? If you have the ANOVA from a cross-over it’s possible (see here). ❝ Is it possible to estimate the total CV from simple arithmetic CV of reference from literature (assuming that test will have similar intra and inter subject variability)? I wouldn’t recommend that. Arithmetic mean / CV implies normal distributed data. — Dif-tor heh smusma 🖖🏼 Довге життя Україна! ![[image]](https://static.bebac.at/pics/Blue_and_yellow_ribbon_UA.png) Helmut Schütz ![[image]](https://static.bebac.at/img/CC by.png) The quality of responses received is directly proportional to the quality of the question asked. 🚮 Science Quotes |
|
Chiku ☆ India, 2012-02-16 08:23 (4821 d 07:02 ago) @ Helmut Posting: # 8127 Views: 3,390 |
|
|
Dear HS! Thank you very much. ❝ If you have the ANOVA from a cross-over it’s possible (see here). Yes, I was referring this presentation of yours. Fortunately, we have ANOVA for a Cross over study. This study was for the strength of 10 mg. Can we use total %CV obtained from this study to design a parallel study for 20 mg (there's pharmacokinetics linearity between 10 and 20 mg). I think we can use it because CV is a relative measure of variability, it is not absolute. Kindly correct me if I am wrong. Chiku |
|
Helmut ★★★ Vienna, Austria, 2012-02-16 14:11 (4821 d 01:15 ago) @ Chiku Posting: # 8130 Views: 3,312 |
|
|
Dear Chiku! ❝ […] we have ANOVA for a Cross over study. This study was for the strength of 10 mg. Can we use total %CV obtained from this study to design a parallel study for 20 mg (there's pharmacokinetics linearity between 10 and 20 mg). I think we can use it because CV is a relative measure of variability, it is not absolute. Yes, you can. For one of the APIs I have worked on the CVs were pretty reproducible in 10/20/30/40/60 mg studies (linear PK). Since CVtotal includes between-subject variability note that you have to keep the inclusion/exclusion criteria in the parallel study at least as stringent as in the cross-over. Don’t forget to add a safety margin. — 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 |
Ing. Helmut Schütz