BE and gender [General Statistics]
We have done a BE study in normal volunteers using equal numbers of male and female subjects. I ran the BE data analysis in the usual way in Phoenix WinNonlin 6.3 and obtained the confidence intervals.
Since we suspect a gender difference for this drug I then included gender in the model as recommended by Helmut (see below)
"Have a column denoting gender (0/1, f/m, whateveryoulike...) and map it in the BE setup as "Classification".
Model Specification (Fixed Effects):
sequence+period+treatment+gender+treatment*gender
Example from one of my studies (6×3 Williams’ design):"
I repeated my data analysis for BE and I found that the confidence intervals were slightly different. The outcome for the BE assessment was not changed.
The ANOVA p value was 0.08 so the gender effect was not significant. Under what circumstances would one use the BE data analysis with gender in the model.
Angus
Edit: Category changed. Plese don’t use all capitals in the future. [Helmut]
Since we suspect a gender difference for this drug I then included gender in the model as recommended by Helmut (see below)
"Have a column denoting gender (0/1, f/m, whateveryoulike...) and map it in the BE setup as "Classification".
Model Specification (Fixed Effects):
sequence+period+treatment+gender+treatment*gender
Example from one of my studies (6×3 Williams’ design):"
I repeated my data analysis for BE and I found that the confidence intervals were slightly different. The outcome for the BE assessment was not changed.
The ANOVA p value was 0.08 so the gender effect was not significant. Under what circumstances would one use the BE data analysis with gender in the model.
Angus
Edit: Category changed. Plese don’t use all capitals in the future. [Helmut]
Complete thread:
- BE and genderAngusMcLean 2013-06-11 20:17 [General Statistics]
- BE and gender ElMaestro 2013-06-11 20:47
- BE and gender Dr_Dan 2013-06-12 09:57
- BE and gender (recent lenghty example) Helmut 2013-06-12 13:45
- BE and gender AngusMcLean 2013-06-15 15:10