BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Principal Stratification for Causal Inference with Extended Partia
 l Compliance - Professor Donald B. Rubin (Harvard)
DTSTART:20071120T170000Z
DTEND:20071120T180000Z
UID:TALK8782@talks.cam.ac.uk
CONTACT:Helen Innes
DESCRIPTION:Many double-blind placebo-controlled randomized experiments wi
 th active drugs suffer from complications beyond simple noncompliance.  Fi
 rst\, the compliance with assigned dose is often partial\, with patients t
 aking only part of the assigned dose\, whether active or placebo.  Second\
 , the blinding may be imperfect in the sense that there may be detectable 
 positive or negative side-effects of the active drug\, and consequently\, 
 simple compliance has to be extended to allow different compliances to act
 ive drug and placebo.  Efron and\nFeldman presented an analysis of such a 
 situation and discussed inference for dose-response from the non-randomize
 d data in the active treatment arm\, which stimulated active discussion\, 
 including concerning the role of the intention-to-treat principle in such 
 studies.  Here\, we formulate the problem within the principal stratificat
 ion framework of Frangakis and Rubin\, which adheres to the intention-to-t
 reat principle\, and we present a new analysis of the Efron-Feldman data w
 ithin this framework.  Moreover\, we describe precise assumptions under wh
 ich dose-response can be inferred from such non-randomized data\, which se
 em debatable in the setting of this example.  Although this article only d
 eals in detail with the specific Efron-Feldman data\, the same framework c
 an be applied to various circumstances in both natural science and social 
 science.\n\n\n
LOCATION:Wolfson Room (MR 2) Centre for Mathematical Sciences\, Wilberforc
 e Road\, Cambridge
END:VEVENT
END:VCALENDAR