BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:BSU Virtual Seminar: “Estimating treatment effects from adaptive
  clinical trials” - Prof Ian Marschner\, University of Sydney
DTSTART:20210914T083000Z
DTEND:20210914T093000Z
UID:TALK161632@talks.cam.ac.uk
CONTACT:Alison Quenault
DESCRIPTION:Adaptive clinical trials have design features that adapt to th
 e accumulating data\, meaning that the design itself is informative about 
 the treatment effect. Consequently\, the overall information from an adapt
 ive clinical trial is a combination of information from two sources\, the 
 realised design and the observed outcomes. I will present a general framew
 ork for the analysis of adaptive clinical trials\, based on the decomposit
 ion of overall information into design information and outcome information
 . The framework provides transparent delineation and comparison of uncondi
 tional and conditional approaches to treatment effect estimation. Conditio
 nal estimation involves conditioning on the observed design so that the tr
 eatment effect is interpreted with reference to the particular design that
  actually occurred. Unconditional estimation involves interpreting the tre
 atment effect with reference to all designs that could possibly have occur
 red. Unconditional estimation may be more efficient because it uses more i
 nformation\, but it is potentially subject to bias for some designs. Ident
 ifying such bias in a given clinical trial is a motivation of the proposed
  framework and we show that this is most likely to occur when the outcome 
 information and the design information are inconsistent. Thus\, we can det
 ect the presence of bias by assessing heterogeneity between the informatio
 n provided by the design and the information provided by the outcomes. Whe
 n such heterogeneity is detected\, conditional inference may be more appro
 priate. Various examples will be considered including response-adaptive ra
 ndomization\, multi-stage phase II studies\, multi-arm treatment selection
  and Bayesian adaptive trials. 
LOCATION:Virtual Seminar 
END:VEVENT
END:VCALENDAR