Log in
Register|
K2K ★ India, 2012-01-12 06:15 (5263 d 22:15 ago) Posting: # 7907 Views: 7,746 |
|
|
Dear sir we want to calculate intersubject CV from reference literature in that they given mean, CI we calculated intra CV from CI? then how to calculate Inter CV? how to calculate inter subject CV from arithmetic mean or Standard devation? its possible to calculate? or we can able to calculate inter subject CV from intrasubject cv? pls guide ME thanks in advance Dear Karthik, questions in the BEBA-Forum only, please! Some hints:
Best regards, Helmut Dear Sir Good morning  ❝ why do want to get CVinter? according to WHO BTIF(bioequivalence trial information form) in that section 7.4. WHO are asking to compare our study report with reference literature report in that they are asking to calculate intersubject cv how its possible to calculate with mean and standard deviation so please guide me. thanks and regards karthik Edit: First section of the post was copypasted from the guestbook. BTIF linked. [Helmut] |
|
Helmut ★★★   Vienna, Austria, 2012-01-12 14:55 (5263 d 13:35 ago) @ K2K Posting: # 7915 Views: 6,821 |
|
|
Dear Kartik! ❝ according to WHO BTIF(bioequivalence trial information form) in that section 7.4. Thanks for bringing this document to my attention; didn’t know it. ❝ WHO are asking to compare our study report with reference literature report in that they are asking to calculate intersubject cv how its possible to calculate with mean and standard deviation so please guide me. Looking at this table I think that it is less complicated than you think. Ignore the word ‘Interindividual’ and give CV% as the ratio of the two preceding columns with CV%=100·SD/x. Unless the study was performed in a replicate design (specifically: one where both T and R were repeated) it is not possible to calculate CVinter / CVintra separately for the two treatments. Even then you would have to repeat the analysis on untransformed data (which is against the guideline) to approximately (!) match the values in the first two columns. Although TRS 937, Annex 7 set a milestone back in 2006 (especially for countries lacking their own guidelines) it is not free from errors – like all guidelines, our own reports, and papers published in peer-review journals as well. P.S.: Section 7.3(b) Period and/or sequence effects of the form is meaningless at all. — 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 |
|
K2K ★ India, 2012-01-13 06:51 (5262 d 21:39 ago) @ Helmut Posting: # 7921 Views: 6,584 |
|
|
Dear sir thanks for your reply. study was two way crossover. In reference literature they given geo.mean means its possible to calculate. In query they specifically asked intersubject CV for reference literature. but we calculated inter and intra subject CV for our study report for last time query and we mention intersubject CV is not available for reference literature. Again they send same query and asked to calculate intersubject CV from arth.mean and standard deviation. its possible to compare our study report with reference literature? we try to compare but some times the values are not match means what we want to do? sometimes values may change according to condition of subjects, formulation and any other reason. thanks in advance thanks and regards karthik — thanks and regards Karthik |
|
Helmut ★★★ Vienna, Austria, 2012-01-13 18:05 (5262 d 10:26 ago) @ K2K Posting: # 7932 Views: 6,476 |
|
|
Dear Karthik! ❝ study was two way crossover. OK, so definitely no way the calculate (any!) CV from ANOVA separately for the two treatments. ❝ In reference literature they given geo.mean means its possible to calculate. What do you mean by that? ❝ In query they specifically asked intersubject CV for reference literature. Well, even in the few replicate designs published only the intra-subject CV of the reference is given. CVinter is completely irrelevant in BE (unless it’s a parallel design, of course). ❝ but we calculated inter and intra subject CV for our study report for last time query and we mention intersubject CV is not available for reference literature. Correct. ❝ Again they send same query and asked to calculate intersubject CV from arth.mean and standard deviation. If they refer to this strange form, simply give them what they want. You and I know that 100·SD/xR ≠ CVinter. ❝ its possible to compare our study report with reference literature? Sure, but of limited value. ❝ we try to compare but some times the values are not match means what we want to do? Nothing. Especially if CVinter is high (especially for drugs subjected to polymorphic metabolism) differences between studies of up to 50% don’t really surprise me. ❝ sometimes values may change according to condition of subjects, formulation and any other reason. Exactly. I would add: posture, bionanalytics. — 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 |
|
Ohlbe ★★★ France, 2012-01-13 19:41 (5262 d 08:49 ago) @ Helmut Posting: # 7938 Views: 6,529 |
|
|
Dear Helmut and Karthik, ❝ ❝ its possible to compare our study report with reference literature? ❝ ❝ Sure, but of limited value. On a PK and BE point of view, I would say the global CV has no value at all. But I've heard some regulators say that they have seen studies with a global CV which was surprisingly low (10 %, for AUC and Cmax). When inspected, the studies turned out to be totally invented, falsified... So on a regulatory point of view, the global CV and its comparison with other trials (including literature data) may provide some other level of information... Regards Ohlbe — Regards Ohlbe |
|
Helmut ★★★ Vienna, Austria, 2012-01-13 19:54 (5262 d 08:37 ago) @ Ohlbe Posting: # 7939 Views: 6,792 |
|
|
Dear Ohlbe! ❝ ❝ ❝ its possible to compare our study report with reference literature? ❝ ❝ Sure, but of limited value. ❝ On a PK and BE point of view, I would say the global CV has no value at all. I did not mean a comparison of the CV but of the means. But you are right, Karthik likely meant CVs. ❝ But I've heard some regulators say that they have seen studies with a global CV which was surprisingly low (10 %, for AUC and Cmax). When inspected, the studies turned out to be totally invented, falsified... So on a regulatory point of view, the global CV and its comparison with other trials (including literature data) may provide some other level of information... Absolutely. A CVtotal of 10% is not possible. Maybe for a bunch of monozygotic twins. Even then a very narrow BMI range must be observed. Should ring all alarm bells. BTW, twins (or even better clones) violate the assumption of independence of the statistical model. — 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 |
|
K2K ★ India, 2012-01-16 13:51 (5259 d 14:39 ago) @ Helmut Posting: # 7942 Views: 6,574 |
|
|
Dear sir! thanks for your reply ❝ I did not mean a comparison of the CV but of the means. But you are right, Karthik likely meant CVs. no sir i also meant means only some time means of value is also showing more different so only i asked. according to there query 'the intrasubject variability in the reference literature is not consistent with your value' thanks and regards karthik — thanks and regards Karthik |
|
Helmut ★★★ Vienna, Austria, 2012-01-16 14:26 (5259 d 14:05 ago) @ K2K Posting: # 7947 Views: 6,536 |
|
|
Dear Karthik! ❝ according to there query 'the intrasubject variability in the reference literature is not consistent with your value' Are you trying to confuse me? Is WHO asking for inter (between) of intra (within) subject CV? You are quoting their request differently: — 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 |
|
K2K ★ India, 2012-01-16 14:35 (5259 d 13:55 ago) @ Helmut Posting: # 7948 Views: 6,473 |
|
|
❝ Are you trying to confuse me? Is WHO asking for inter (between) of intra (within) subject CV? You are quoting their request differently: sorry sir i am not trying to confuse u they asked to calculate both intrasubject and intersubject CV. for intersubject they asked to calculate from art.mean and std dev. for intrasubject they given the statement like that according to the query 'the intrasubject variability in the reference literature is not consistent with your value' thanks you karthik — thanks and regards Karthik |
|
Helmut ★★★ Vienna, Austria, 2012-01-16 15:50 (5259 d 12:40 ago) @ K2K Posting: # 7950 Views: 6,448 |
|
|
Dear Karthik! ❝ they asked to calculate both intrasubject and intersubject CV. ❝ for intersubject they asked to calculate from art.mean and std dev. OK, we had that before. That’s not correct, but calculate it according to their form anyhow. ❝ for intrasubject they given the statement like that ❝ according to the query 'the intrasubject variability in the reference literature is not consistent with your value' I guess your CVintra is (much?) lower than the one given in literature? If published studies were not using an awful bioanalytical method (high CV) WHO might be suspicious that something went wrong in the study, e.g., conversion of a metabolite to the parent in the ion-source in LC/MS-MS. Since metabolites generally have lower CVs that the parent you will get a lower CV than with a method which separates parent & metabolites. Are your mean concentrations substantially higher than in the literature? If yes, get prepared for an inspection. — 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 |
|
K2K ★ India, 2012-01-18 10:36 (5257 d 17:54 ago) @ Helmut Posting: # 7959 Views: 6,440 |
|
|
Dear sir thanks for your valuable reply thanks and regards karthik — thanks and regards Karthik |
Ing. Helmut Schütz