BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20251122T033318EST-2435eD8ped@132.216.98.100
DTSTAMP:20251122T083318Z
DESCRIPTION:Title: Calculation of power and type I error rate of methods fo
 r borrowing data from historical controls: Should we sample the historical
  data?\n\nAbstract: Joost van Rosmalen is an assistant professor at the De
 partment of Biostatistics of Erasmus MC in Rotterdam\, the Netherlands. Hi
 s research mainly focuses on methods for including historical and real-wor
 ld data in the analysis of clinical trials and the analysis of repeated me
 asurements. Link to website: https://www.erasmusmc.nl/en/research/research
 ers/rosmalen-joost-van\n\n\nMany methods for borrowing information from hi
 storical controls in the analysis of a randomized trial have been proposed
 \, such as the (modified) power prior and the meta-analytic predictive pri
 or. These methods aim to increase the power and reduce sample size require
 ments while providing a safeguard against prior-data conflict (i.e. substa
 ntial differences between data sources). However some methods can increase
  the type I error rate\, which may not be acceptable to regulatory authori
 ties. Power and type I error rate are calculated using simulation\, which 
 can be done using a conditional (taking the historical data as fixed) or a
 n unconditional approach (sampling also the historical data in the simulat
 ion).\n	\n	In this talk we give a brief introduction to these borrowing meth
 ods and we discuss the ideas behind the unconditional and the conditional 
 approach. We propose a framework for the unconditional approach that can i
 ncorporate a systematic bias of the historical data as well as between-stu
 dy heterogeneity based on an assumption of exchangeability. In this framew
 ork\, prior information on the similarity of the current trial and the his
 torical data can be included. The framework is illustrated using a simulat
 ion study. The results of the unconditional approach show that if we assum
 e exchangeability and no systematic bias of historical and current control
 s\, a gain in power while keeping the type I error rate fixed is possible.
  Using the conditional approach it is difficult to incorporate knowledge o
 n the similarity of the historical controls.\n\nVia Zoom: Please visit our
  website for the Zoom Link: /epi-biostat-occh/seminar
 s-events/seminars/biostati...\n\n \n\nweb : /epi-bios
 tat-occh/seminars-events/seminars/biostatistics\n\n \n
DTSTART:20221123T203000Z
DTEND:20221123T213000Z
SUMMARY:Joost van Rosmalen\, PhD\, Erasmus University Medical Center\; Rott
 erdam\, Netherland
URL:/mathstat/channels/event/joost-van-rosmalen-phd-er
 asmus-university-medical-center-rotterdam-netherland-343760
END:VEVENT
END:VCALENDAR