Log in
Register|
sirajudeen ☆ India, 2013-09-24 09:00 (4250 d 08:31 ago) (edited on 2013-09-24 11:18) Posting: # 11545 Views: 8,837 |
|
|
Dear All, I would like to obtain some clarifications on fully replicate studies (four period with two treatment, two sequence) on some of the followings in terms of PK and statistics.
Edit: Category changed. [Dr_Dan] |
|
Dr_Dan ★★ Germany, 2013-09-24 13:17 (4250 d 04:15 ago) @ sirajudeen Posting: # 11546 Views: 7,758 |
|
|
Dear sirajudeen
Kind regards Dan — Kind regards and have a nice day Dr_Dan |
|
Yvonne ☆ 2013-10-14 11:39 (4230 d 05:52 ago) @ Dr_Dan Posting: # 11660 Views: 7,694 |
|
|
❝ subjects who completed two periods of Test products can be considered for within-subject variability of the test product (this is the idea of the full replicate to evaluate the within-subject variability of the reference as well as of the test product). Is it acceptable to include subjects that completed two periods of reference product (and no periods with test product) in the within subject variability calculation of the reference product used for widening? These subjects are (per definition) different from the subjects ("population") used for the test/ref calculation. |
|
Helmut ★★★   Vienna, Austria, 2013-10-14 19:28 (4229 d 22:03 ago) @ Yvonne Posting: # 11661 Views: 7,479 |
|
|
Hi Yvonne, ❝ Is it acceptable to include subjects that completed two periods of reference product (and no periods with test product) in the within subject variability calculation of the reference product used for widening? I don’t see any reason why this could not be done. More subjects would give a more precise estimate of the CVWR. ❝ These subjects are (per definition) different from the subjects ("population") used for the test/ref calculation. Why? The only difference is that you have no values after the test. These subjects dropped out “at random”. — 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 |
|
d_labes ★★★ Berlin, Germany, 2013-10-14 20:38 (4229 d 20:54 ago) @ Helmut Posting: # 11663 Views: 7,473 |
|
|
Dear Helmut, ❝ ... These subjects dropped out “at random”. Hopefully! On the other hand I cant imagine a random process for a fully replicate design where f.i. in sequence TRTR the subject absolved 2nd and fourth period, but no 1st and third period. Seems very unlikely to me. But statistics makes no statement about the individual case ...  — Regards, Detlew |
|
Helmut ★★★ Vienna, Austria, 2013-10-14 21:21 (4229 d 20:10 ago) @ d_labes Posting: # 11664 Views: 7,493 |
|
|
Dear Detlew, ❝ ❝ ... These subjects dropped out “at random”. ❝ ❝ Hopefully! ❝ ❝ On the other hand I cant imagine a random process for a fully replicate design where f.i. in sequence TRTR the subject absolved 2nd and fourth period, but no 1st and third period. Seems very unlikely to me. True. Haven’t seen that in the wild. Only EMA (in their Q&A document) fabricates such data. @Yvonne: What happened in the study? Was it a fully replicated design (as Detlew assumed)? ❝ But statistics makes no statement about the individual case ... Oh, yes! — 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 |
|
Yvonne ☆ 2013-10-23 17:20 (4221 d 00:11 ago) @ Helmut Posting: # 11737 Views: 7,331 |
|
|
Dear Detlew and Helmut ❝ On the other hand I cant imagine a random process for a fully replicate design where f.i. in sequence TRTR the subject absolved 2nd and fourth period, but no 1st and third period. Seems very unlikely to me. No indeed, but a sequence with RTRT, no 2nd and 4th period seems possible to me. This would generate two R periods and no T periods.... ❝ Haven’t seen that in the wild. Only EMA (in their Q&A document) fabricates such data. I have seen "in the wild" a first and fourth period ❝ @Yvonne: What happened in the study? Was it a fully replicated design (as Detlew assumed)? Currently I am writing a protocol (fully replicate) and thinking about possible theoretic cases.... (I'm sorry, not that exciting unfortunately... ). |
|
Helmut ★★★ Vienna, Austria, 2013-10-23 19:51 (4220 d 21:40 ago) @ Yvonne Posting: # 11740 Views: 7,398 |
|
|
Dear Yvonne and all, ❝ No indeed, but a sequence with RTRT, no 2nd and 4th period seems possible to me. This would generate two R periods and no T periods.... Really? What was written in the protocol? Did the subject go for a short vacation after period 1, came back for the 3rd, and finally decided to be no more interested in finishing the study? — 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 |
|
Yvonne ☆ 2013-10-24 11:01 (4220 d 06:31 ago) @ Helmut Posting: # 11746 Views: 7,367 |
|
|
❝ No indeed, but a sequence with RTRT, no 2nd and 4th period seems possible to me. This would generate two R periods and no T periods.... ❝ Really? What was written in the protocol? Did the subject go for a short vacation after period 1, came back for the 3rd, and finally decided to be no more interested in finishing the study? This scenario did not happen to us, but a subject only fullfilling period 1 and 4 happened..... a subject that did not show up was asked to come again and did  . Maybe he did take a small holiday  . So theoretical cases might happen..... |
Ing. Helmut Schütz