Effect has a specific meaning in stats and is used to compare two populations not a pre and post of a single population. Effect size (i.e. Cohen's D) is an attempt to qualify the effect in a quantitative way (for Cohen's D small is defined as 0.2 to 0.5, medium as 0.5 to 0.8 and large as > 0.8).
For the Rett trial the effect is 4.62 RSBQ points.
The effect size (Cohen's D) = 0.29 is small
-12.93 (A273)and -8.32 (Plac) refer to mean change which I believe is the term that best describes your intent.
So, compared to the trofinetide trial, both A273 and placebo had larger mean changes in the RSBQ than trofinetide. A273 also had a larger effect of 4.6 vs 3.2. However, the effect size was higher for trofinetide vs placebo at 0.37 vs A273/placebo at 0.29. It is probable that a higher n would have reduced the p value and increased the effect size, assuming that the additional patients mirrored the 92. However, it is also probable that, unlike trofinetide, n about 180 would not have made CGI difference in mean change between A273 and placebo significant. Trofinetide also had a highly significant effect on CGI-I and a near medium effect size(0.47). If Missling had sprung an additional 3 million or so for more patients, there would be a chance that the FDA would have considered a single met endpoint (RSBQ) as good enough but with both endpoints with p above 0.05 the study can only be looked at as a failed trial...regardless of mean change seemingly high at 12.83
News 
Market Data 
Discover 
