Shiny improvements

[tcoroller]

• Could you introduce "delta Alt" at the time people enter the plausible beneficial effect (or only use the name "plausible beneficial effect" throughout) ?
• It would be good to say "expected deaths" when entering the numbers for interim analyses, since those will be triggered by another endpoint. For the final, you can say "targeted deaths at final OS analysis" to be very clear.
• When setting statistical thresholds, it could be good to keep the layman's language explanation in a consistent way with the rest of the App. For example: "Interpretation: this controls the final-analysis reassurance threshold when the true OS HR is equal to your unacceptable detriment" could also read "Interpretation: this controls the final-analysis threshold for positivity. It is the risk of being wrongly reassured when the true OS HR is in fact equal to your unacceptable detriment"
• For the power, it could also be good to provide an "Interpretation: this controls the interim analysis threshold for positivity. It is the chance of being correctly reassured when the true OS HR is in fact equal to your plausible beneficial effect"; I am not sure about the language "depending on how strict the primary-analysis reassurance should be." given here, since the primary analysis refers more to the primary PFS analysis, and I don't know what "strict" means when talking of a power. How about "depending on the possible impact on the program of wrongly flagging a potential detriment at the interim OS analysis" ?
• Is there a way to reduce the upper band presenting the thresholds and monitoring table ? On the little screen of my laptop, it takes 2/3 of the screen and I only have the bottom narrow part to enter the input parameters.