Grant: Discovery Grants Program - Individual
University of Manitoba
$14,000.00
Statistical design and analysis of response adaptive clinical trials
FY 2008-09
Summary
) The use of randomization is essential in clinical trials because it masks the treatment identity, promotes the similarity of treatment groups and validates the use of statistics. However in desperate medical situations, the gold standard of balanced randomization favors the collective ethics (i.e., the benefits to the patients after the trial) but hinders the individual ethics (i.e., the paramount interests of the patients in the trial), and hence becomes ethically infeasible. This conflict is alleviated with the use of response adaptive randomization, the goal of which is to evaluate and utilize the accumulating information so as to improve the efficiency and ethics of the trial without undermining the validity and integrity of the clinical research. Most patients in the trial are expected to receive the potentially superior treatment without sacrifice to the collective ethics. This is a very important issue because in the 21st century, there is a greater demand of personal care in medical research.
) There are still many unresolved issues and problems in this relatively young but fast-growing field, mostly on developing optimal randomization procedures and methods of statistical analysis. The specific aims of the proposal are to develop a new class of statistical designs and analysis methods that improve the existing ones, to assess the sensitivity of the designs, and to search for robust designs. The methodology is as follows.
) (i) To develop the new class of designs, we extend our current achievements on the variance penalized criterion. The proposed approach is to combine the decision-theoretical design (using the Bayesian method and bandit models) and randomization (using the doubly adaptive biased coin design) and form new designs.
) (ii) The sensitivity analysis examines the stability of the available designs with respect to departures of model assumptions (such as model misspecification or non-normality) and measurement errors. The proposed method is to extend the standard ideas for independent data to dependent data from response adaptive designs.
) Both the proposed designs and sensitivity analysis have not been addressed in the literature. We anticipate achieving potentially significant contributions and impact of opening up new directions of research.
Program
Discovery Grants Program - Individual
Recipient
Location
Manitoba, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data
As published
Application id
217441-2008
Name nom
Wang, Xikui
Department département
Statistics
Organization id
19
Institution établissement
University of Manitoba
Province en
Manitoba
Province fr
Manitoba
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2008
Competition year année de concours
2008
Award amount
14000
Program id
RGPIN
Program nane en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
Discovery grants programs
Group fr
Programmes de subventions à la découverte
Committee code
14
Committee name en
Statistical Sciences
Committee name fr
Statistique
Area of application code
1211
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Medical and health sciences
Area of application fr
Sciences médicales et de la santé
Research subject code
3005
Research subject group en
Statistics and probability
Research subject group fr
Statistique et probabilité
Research subject en
Biostatistics
Research subject fr
Biostatistiques
Partie
1
Nb partie
5
Application title
Statistical design and analysis of response adaptive clinical trials
Keyword
NA
Application summary
) The use of randomization is essential in clinical trials because it masks the treatment identity, promotes the similarity of treatment groups and validates the use of statistics. However in desperate medical situations, the gold standard of balanced randomization favors the collective ethics (i.e., the benefits to the patients after the trial) but hinders the individual ethics (i.e., the paramount interests of the patients in the trial), and hence becomes ethically infeasible. This conflict is alleviated with the use of response adaptive randomization, the goal of which is to evaluate and utilize the accumulating information so as to improve the efficiency and ethics of the trial without undermining the validity and integrity of the clinical research. Most patients in the trial are expected to receive the potentially superior treatment without sacrifice to the collective ethics. This is a very important issue because in the 21st century, there is a greater demand of personal care in medical research.
) There are still many unresolved issues and problems in this relatively young but fast-growing field, mostly on developing optimal randomization procedures and methods of statistical analysis. The specific aims of the proposal are to develop a new class of statistical designs and analysis methods that improve the existing ones, to assess the sensitivity of the designs, and to search for robust designs. The methodology is as follows.
) (i) To develop the new class of designs, we extend our current achievements on the variance penalized criterion. The proposed approach is to combine the decision-theoretical design (using the Bayesian method and bandit models) and randomization (using the doubly adaptive biased coin design) and form new designs.
) (ii) The sensitivity analysis examines the stability of the available designs with respect to departures of model assumptions (such as model misspecification or non-normality) and measurement errors. The proposed method is to extend the standard ideas for independent data to dependent data from response adaptive designs.
) Both the proposed designs and sensitivity analysis have not been addressed in the literature. We anticipate achieving potentially significant contributions and impact of opening up new directions of research.