BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Use of historical information to supplement a future study: opport
 unity and difficulty  - Nick Galwey\, GlaxoSmithKline
DTSTART:20161121T191500Z
DTEND:20161121T213000Z
UID:TALK62424@talks.cam.ac.uk
CONTACT:Peter Watson
DESCRIPTION:There is currently strong interest in the possibility of utili
 sing historical placebo (or other control) data to supplement the data obt
 ained in a new clinical trial\, thereby increasing its power and/or reduci
 ng its size.   It is generally considered that historical data should be d
 iscounted relative to data from the new trial\, but determination of the a
 ppropriate degree of discounting is a major difficulty.   This presentatio
 n provides methods for assessing the consequences of different choices or 
 decisions.\nThe discount can be expressed in terms of the difference betwe
 en the historical and future means\, represented by an offset variance com
 ponent tau-squared. In a Bayesian analysis\, there must be a strong prior 
 belief that  is small if much value is to be obtained from the historical 
 data\, but a range of such priors can be explored in a sensitivity analysi
 s.   Alternatively\, the same numerical results can be viewed in terms of 
 minimisation of the mean square error when biased historical data are incl
 uded in the estimate\, and a range of values of a bias parameter\, |delta|
 \, can be explored. The choice of value for tau-squared or  |delta| can ha
 ve a substantial effect on the inferences made\, yet a wide range of value
 s may be consistent with the data.   Fortunately\, external information ca
 n sometimes provide additional guidance on the appropriate value. On this 
 basis\, it is likely that historical data can provide a valuable contribut
 ion to exploratory studies in drug development\, where bias can be accepte
 d as one of the research project’s risks.
LOCATION:Statistical Laboratory\, University of Cambridge
END:VEVENT
END:VCALENDAR