Kaspar Rufibach

Family & sports person. Biostatistician for @roche in @BaselStadt. @[email protected]

[Kaspar Rufibach @numbersman77]

Thread to save my longer threads for (my own) future reference.

[Kaspar Rufibach @numbersman77]

@Theholisticpsyc @threadreaderapp unroll

[Kaspar Rufibach @numbersman77]

@Theholisticpsyc @threadreaderapp unroll

[Kaspar Rufibach @numbersman77]

@kareem_carr @threadreaderapp unroll

[Kaspar Rufibach @numbersman77]

@Theholisticpsyc @threadreaderapp unroll

[Kaspar Rufibach @numbersman77]

@Conqueror_Mind @MMNotes_28 @threadreaderapp unroll

[Kaspar Rufibach @numbersman77]

@Theholisticpsyc @threadreaderapp unroll

[Kaspar Rufibach @numbersman77]

@Theholisticpsyc @threadreaderapp unroll

[Kaspar Rufibach @numbersman77]

@ildiazm @f2harrell @threadreaderapp unroll

[Kaspar Rufibach @numbersman77]

@ADAlthousePhD @yudapearl @robertwplatt @threadreaderapp unroll

[Kaspar Rufibach @numbersman77]

@SahilBloom @threadreaderapp unroll

[Kaspar Rufibach @numbersman77]

@angelosgeo @biobrainbox @threadreaderapp unroll

[Kaspar Rufibach @numbersman77]

@LeilaJTF @tmorris_mrc @statsepi Reference based mean imputation: https://t.co/3InovUiHr4

https://t.co/WtZcjnlbAQ

[Kaspar Rufibach @numbersman77]

@Theholisticpsyc @selfhealerscirc @threadreaderapp unroll

[Kaspar Rufibach @numbersman77]

@Theholisticpsyc @threadreaderapp unroll

[Kaspar Rufibach @numbersman77]

@threadreaderapp unroll

[Kaspar Rufibach @numbersman77]

1) Why do we run group-sequential trials in drug development?

2) How does the effect we power at relate to the effect(s) needed to stop?

3) Is stopping such a trial early (like the one for the Pfizer vaccine) "cheating"?

[Kaspar Rufibach @numbersman77]

So group-sequential designs reduce expected number of events needed and provide valid inference in reasonable cases.

This all within the framework of hypothesis testing, as required by Health Authority guidelines.

I hope this thread is useful. Comments welcome!

The end.

[Kaspar Rufibach @numbersman77]

@LGHemkens @amancayork Provide more clarity around what we are estimating. My personal opinion and answer to the question asked would be "yes".

[Kaspar Rufibach @numbersman77]

@Codie_Sanchez @threadreaderapp unroll

[Kaspar Rufibach @numbersman77]

@Codie_Sanchez @threadreaderapp unroll

[Kaspar Rufibach @numbersman77]

@PhDemetri Reference-based mean imputation:

https://t.co/WtZcjnlbAQ

https://t.co/qWnELcnlK1

https://t.co/j9idaIKY8t

https://t.co/VxFVBcSofQ

[Kaspar Rufibach @numbersman77]

@Theholisticpsyc @threadreaderapp unroll