Log in
Register|
krishnamurthy ca ☆ 2007-10-22 10:33 (6820 d 12:35 ago) Posting: # 1231 Views: 4,298 |
|
|
dear all i would like your opinion on deciding as to how we define that a subject has completed a study in case of replicate design? Example 1: In case of a 4 period replicate design study, if the subject has completed only 2 periods (one test and one reference) could this subject be considered as having completed the study? In continuation of the above, if the protocol indicates that all subjects who have completed the study will be included for PK and Stats, should the above subject be included or excluded from PK and Stats? regards krishnamurthy |
|
Helmut ★★★   Vienna, Austria, 2007-10-22 15:17 (6820 d 07:51 ago) @ krishnamurthy ca Posting: # 1235 Views: 3,570 |
|
|
Dear Krishnamurthy! ❝ Example 1: In case of a 4 period replicate design study, if the subject has completed only 2 periods (one test and one reference) could this subject be considered as having completed the study? No, since the design called for 4 periods, the subject did not complete the entire study... ❝ In continuation of the above, if the protocol indicates that all subjects who have completed the study will be included for PK and Stats, should the above subject be included or excluded from PK and Stats? What was laid down in the protocol? If completion meant 4 periods, the subject should have been excluded from the BE assessment - but PK data of the 2 completed periods must still be presented and accounted for (why did he drop out: personal reasons, AEs, were the concentrations higher than 'normal', etc.) If completion meant at least 2 periods - where like in your example test and reference are covered - it's possible to recover some information and include it in the BE assement by means of linear mixed-effects modeling (LMEM). See this post. Remark: whereas the former is the 'conventional' approach, the latter is also possible, but the gain in statistical power (i.e., the width of the confidence interval) generally is minuscule. — 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 |
|
shankar ★ India, 2007-11-22 06:04 (6789 d 16:04 ago) @ Helmut Posting: # 1318 Views: 3,384 |
|
|
Dear HS, As you mentioned in this post "were the concentrations higher than 'normal'" should be excluded, subjected with the reason. I think you mean to say the subject is an outlier. Which should be dropped based on pk/statistical analysis. Cheers, Shankaranarayanan |
|
Helmut ★★★ Vienna, Austria, 2007-11-22 10:46 (6789 d 11:22 ago) @ shankar Posting: # 1322 Views: 3,482 |
|
|
Dear Shankaranarayanan! ❝ As you mentioned in this post "were the concentrations higher than 'normal'" should be excluded, subjected with the reason. I think you mean to say the subject is an outlier. Which should be dropped based on pk/statistical analysis. I did not mean that. At least according to European guidelines biosamples of a subject who dropped out from a study (for any reason), must be analysed and accounted for. Just imagine a situation where a subject withdraw consent continuing the study because of adverse events (e.g., heavy headaches). If you administered a modified release formulation which might be subjected to a sudden release of drug, it's quite interesting to see such a relationship. This has nothing to do with exclusion of a subject. — 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