[Forum SIS] Seminario prof.ssa S.Giovagnoli alla Sapienza

[fulvio de santis]

*SCUOLA DI DOTTORATO IN SCIENZE STATISTICHE*
Dipartimento di Scienze Statistiche
Sapienza Università di Roma

*AVVISO DI SEMINARIO
*
Data: venerdì 15 febbraio 2013 ore 11
Luogo: Sala 34, IV piano Dipartimento di Scienze Statistiche
Sapienza Università di Roma
Piazzale A. Moro 5, Roma


*Adaptive Randomization for Treatment Comparison
*
*Alessandra Giovagnoli
*ex-University of Bologna, e-mail: alessandra.giovagnoli at unibo.it
Co-authors: Alessandro Baldi Antognini, Maroussa Zagoraiou

The past decade has witnessed an outburst of publications in the
specialized medical literature on
the topic of adaptive designs to compare different drugs or therapies in
clinical trials.
Pharmaceutical industries invest very large budgets but recently the
increased spending in
biomedical research has not reflected in a corresponding increase in
benefits. Thus the FDA Critical
Path initiative of 2004 has encouraged innovative approaches in the design
of the trial, in particular
adaptive procedures that use the available information to modify aspects of
the trial as it progresses,
without undermining its validity and integrity.
The topic has aroused the interest of statisticians with an eye to the
methodological implications.
>From Fisher’s times, the data collection process is often assumed to
include some form of
randomization, in particular in clinical trials randomization is regarded
as a must. The problem of
individual versus collective ethics suggests a form of randomization that
uses available information
for skewing allocations towards the best treatment.
This presentation is an overview of designs recently proposed by the
authors in the context of
adaptive randomization. One approach to the ethical problem is via the
optimization of a
compromise criterion, obtained by taking a weighted average of a design
optimality measure and a
measure of the subjects' risk. The relative weights in the compound
criterion are allowed to depend
on the true state of nature, since it is reasonable to suppose that the
more the effects of the
treatments differ, the more important are the chances of receiving the best
treatment. There follows
the implementation of randomization methods that approach the optimal
target, whatever the true
value of the parameters.
Of major interest are sequential allocation procedures adjusted for
covariates. Balancing methods
that go beyond Pocock and Simon’s traditional minimization procedure and a
general covariate-
adjusted response-adaptive (C.A.R.A.) randomization design will be briefly
illustrated.

A. BALDI ANTOGNINI, A. GIOVAGNOLI Compound optimal allocation for
individual and collective ethics in
binary clinical trials, Biometrika, (2010).

A. BALDI ANTOGNINI, M. ZAGORAIOU The Covariate-adaptive biased coin design
for balancing clinical
trials in the presence of prognostic factors, Biometrika, (2011).

A. BALDI ANTOGNINI, M. ZAGORAIOU Multi-objective optimal designs in
comparative clinical trials with
covariates: the reinforced doubly-adaptive biased coin design, Annals of
Statistics, (2012).
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.stat.unipg.it/pipermail/sis/attachments/20130204/cb38b323/attachment.html>