I feel the 2 ways to evaluate Rett efficacy are both valid and equally difficult (or equally easy with qualitative scales) to meet. Anchoring with CGI is not much different than coprimary endpointing with CGI. The bigger difference is RSBQ AUC vs change in RSBQ from baseline to last visit
The 2 methods
(1) RSBQ-AUC with CGI anchor as Anavex proposes
(2) coprimary endpoints of RSBQ change from baseline to end of study and CGI-I change from baseline to EOS as ACAD has done and used in their NDA that has been accepted by the FDA
With a 7 week duration and only two efficacy timepoints (week 4 and week 7) the AUC method will give you a better result if week 4 RSBQ separates out from placebo more than week 7 and the RSBQ delta method will give a better result if week 7 RSBQ separates out from placebo more than week 4. If week 7 is better than week 4 both will give similar significance. With multiple measurements (i.e week 3, 6, 9 and 12) the AUC is probably superior
I actually already answered your question in February. See posts 349403, 351580, 351655