EMA crippled approach [Software]
Dear John,
let me snap in.
If you aimed for an EMA submission you have to do some sort of that.
To cite the EMA guideline again (we had this already occasionally discussed here, also by yourself!):
"In studies with more than two treatment arms (e.g. a three period study including two references, one from EU and another from USA, or a four period study including test and reference in fed and fasted states), the analysis for each comparison should be conducted excluding the data from the treatments that are not relevant for the comparison in question."
Practically this means to me: Don't change anything in the dataset. Keep periods and sequences as they are. Only forgot 1/3 of your data per comparison under consideration.
I'm not a friend of this crippled approach .
But ... On the other hand the full analysis is crucially depending on the assumption of a common error variance for the three treatments. What happens and what are the implications if this assumption is violated is not really clear to me.
Is the case of heteroscedastic variances in BE studies an issue? Don't ask me . See a quote of Steven Senn here.
let me snap in.
❝ ... How? By collapsing the data into a 2-way study while preserving the order of the T and R treatments are per sequence (i.e., ABC → CB, BAC → AC, CBA → CA etc etc) and run stats? Or remove the 2nd formulation data completely and run stats? So that the width of the 90% CI is attributed to intrasubject CV from both 1st formulation and reference.
If you aimed for an EMA submission you have to do some sort of that.
To cite the EMA guideline again (we had this already occasionally discussed here, also by yourself!):
"In studies with more than two treatment arms (e.g. a three period study including two references, one from EU and another from USA, or a four period study including test and reference in fed and fasted states), the analysis for each comparison should be conducted excluding the data from the treatments that are not relevant for the comparison in question."
Practically this means to me: Don't change anything in the dataset. Keep periods and sequences as they are. Only forgot 1/3 of your data per comparison under consideration.
I'm not a friend of this crippled approach .
But ... On the other hand the full analysis is crucially depending on the assumption of a common error variance for the three treatments. What happens and what are the implications if this assumption is violated is not really clear to me.
Is the case of heteroscedastic variances in BE studies an issue? Don't ask me . See a quote of Steven Senn here.
—
Regards,
Detlew
Regards,
Detlew
Complete thread:
- SAS vs Winnonlin jag009 2013-07-24 17:30 [Software]
- SAS vs Winnonlin Helmut 2013-07-24 17:55
- SAS vs Winnonlin jag009 2013-07-24 22:56
- SAS vs Winnonlin Helmut 2013-07-24 23:53
- SAS vs Winnonlin jag009 2013-07-25 00:56
- EMA crippled approachd_labes 2013-07-25 09:13
- Hypothesis jag009 2013-07-25 15:38
- Two at a Time? Or All at Once? Helmut 2013-08-10 13:52
- Heteroscedasticity Helmut 2013-07-27 18:30
- Hypothesis jag009 2013-07-25 15:38
- EMA crippled approachd_labes 2013-07-25 09:13
- SAS vs Winnonlin jag009 2013-07-25 00:56
- SAS vs Winnonlin Helmut 2013-07-24 23:53
- Update! jag009 2013-07-27 05:59
- Rounding limbo? Helmut 2013-07-27 14:53
- Rounding limbo? jag009 2013-07-27 22:02
- Rounding limbo? Helmut 2013-07-27 14:53
- Winnonlin: exclude volunteers mittyri 2014-01-10 08:05
- Winnonlin: exclude incomplete data! Helmut 2014-01-10 13:41
- SAS vs Winnonlin jag009 2013-07-24 22:56
- SAS vs Winnonlin ElMaestro 2013-07-24 18:06
- SAS vs Winnonlin jag009 2013-07-24 22:59
- SAS vs Winnonlin Helmut 2013-07-24 17:55