Grant: Discovery Grants Program - Individual
University of Toronto
$14,000.00
Expected value of information and optimal trial design
FY 2011-12
Summary
It is becoming increasingly common in randomized controlled trials (RCTs) to collect patient-level cost data along with the measures of effectiveness (health effects). This added dimension to RCTs has motivated the development of new statistical methodology designed to answer questions of health policy in addition to those of purely clinical importance. Initially attention was focussed on estimating the incremental cost-effectiveness ratio. More recently, due to the acknowledged problems associated with ratio statistics, attention has shifted to making inference about incremental net benefit (INB), using test of hypothesis and confidence intervals. An advantage of expressing the results of a cost-effectiveness analysis of an RCT in terms of INB is that it allows the use of decision analysis for interpreting the results. Faced with positive, but uncertain, INB, the clinical decision maker must choose to adopt the new intervention now or delay the decision while another trial is performed. Value of information methods based on decision analysis can be used to determine which choice is optimal, and if another trial is to be performed, the optimal sample size.
) In recent articles my colleagues and I have proposed replacing tests of hypotheses and power arguments for the interpretation and design of RCTs, which we consider arbitrary, with methods based on a Bayesian decision theoretic approach which determine the sample size that maximizes the difference between expected cost of the trial and the expected value of information it provides. The research proposed centres on further developing value of information methods for interpreting RCT results and determining optimal designs for future RCTs. In particular, the following topics will be covered.
1. ) Relaxation of restrictive model assumptions,
2. ) Adaptive multi-stage designs,
3. ) Partial value of information.
Program
Discovery Grants Program - Individual
Recipient
Location
Ontario, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data
As published
Application id
44868-2008
Name nom
Willan, Andrew
Department département
Dalla Lana School of Public Health
Organization id
31
Institution établissement
University of Toronto
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2011
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
4
Nb partie
5
Application title
Expected value of information and optimal trial design
Keyword
NSERC
Application summary
It is becoming increasingly common in randomized controlled trials (RCTs) to collect patient-level cost data along with the measures of effectiveness (health effects). This added dimension to RCTs has motivated the development of new statistical methodology designed to answer questions of health policy in addition to those of purely clinical importance. Initially attention was focussed on estimating the incremental cost-effectiveness ratio. More recently, due to the acknowledged problems associated with ratio statistics, attention has shifted to making inference about incremental net benefit (INB), using test of hypothesis and confidence intervals. An advantage of expressing the results of a cost-effectiveness analysis of an RCT in terms of INB is that it allows the use of decision analysis for interpreting the results. Faced with positive, but uncertain, INB, the clinical decision maker must choose to adopt the new intervention now or delay the decision while another trial is performed. Value of information methods based on decision analysis can be used to determine which choice is optimal, and if another trial is to be performed, the optimal sample size.
) In recent articles my colleagues and I have proposed replacing tests of hypotheses and power arguments for the interpretation and design of RCTs, which we consider arbitrary, with methods based on a Bayesian decision theoretic approach which determine the sample size that maximizes the difference between expected cost of the trial and the expected value of information it provides. The research proposed centres on further developing value of information methods for interpreting RCT results and determining optimal designs for future RCTs. In particular, the following topics will be covered.
1. ) Relaxation of restrictive model assumptions,
2. ) Adaptive multi-stage designs,
3. ) Partial value of information.