Grant: Doctoral Award: Doctoral Foreign Study Award (DFSA)
Direct Payment
$105,000.00
Bayesian alternatives to conventional randomized trial designs
EF 2023-24
Résumé
Background:
In order to determine whether new drugs, medical devices, or other tools used in clinical settings are effective, researchers conduct studies intended to objectively measure the potential impact of these tools. These are often in the form of clinical trials, where a new intervention is compared against the existing standard or a placebo. These trials are usually randomized, where participants are randomly divided into groups. Advances in computing in the past two decades have allowed researchers to employ more advanced statistics called Bayesian statistics, which take advantage of computing power to perform more sophisticated analyses. These new analyses have the potential to allow trials to make use of historical evidence, quantify uncertainty, and be more flexible in their structure. However, our understanding of Bayesian statistics is still limited, and may lead to false positive results. As a result, there is a need to examine the quality of trials that currently use Bayesian statistics, and see how they measure against the classical way of conducting trials.
Objectives:
1. Identify what kinds of Bayesian statistical approaches are being used in trials
2. Identify how well the details of the statistical approaches are reported in the published papers
3. Determine how well the Bayesian approach performs when compared to classical methods
4. Identify whether previous classical trials may have differed if a Bayesian approach had been used
Implications:
Through this work, we will be able to identify the advantages and disadvantages of using these new Bayesian approaches to conducting clinical trials. We will also be able to see whether Bayesian approaches are able to reduce the number of patients needed to complete a trial without affecting the reliability of the results. Finally, we will also aim to provide an overview of what kinds of Bayesian approaches have been used in the past, and identify ways in which these approaches can be improved.
Programme
Doctoral Award: Doctoral Foreign Study Award (DFSA)
Bénéficiaire
Emplacement
United Kingdom
Type
Grant
Source
CIHR Funding Decisions Database
Tel que publié
Id
495526
Name
Bansal Rishi
Country
United Kingdom
Orgtype
Research Institution
Instname2
Population and Public Health
Theme2
Clinical
Prcname2
Doctoral Research Awards - B
Projecttitle
Bayesian alternatives to conventional randomized trial designs
Programtype2
Studentship Programs
Orgnameinp2
Direct Payment
Cihramount2
$105,000
Cihrequipment2
$0
Programname2
Doctoral Award: Doctoral Foreign Study Award (DFSA)
Keyworddelim
Bayesian Statistics; Clinical Trial Methodology; Clinical Trials
Abstract
Background:
In order to determine whether new drugs, medical devices, or other tools used in clinical settings are effective, researchers conduct studies intended to objectively measure the potential impact of these tools. These are often in the form of clinical trials, where a new intervention is compared against the existing standard or a placebo. These trials are usually randomized, where participants are randomly divided into groups. Advances in computing in the past two decades have allowed researchers to employ more advanced statistics called Bayesian statistics, which take advantage of computing power to perform more sophisticated analyses. These new analyses have the potential to allow trials to make use of historical evidence, quantify uncertainty, and be more flexible in their structure. However, our understanding of Bayesian statistics is still limited, and may lead to false positive results. As a result, there is a need to examine the quality of trials that currently use Bayesian statistics, and see how they measure against the classical way of conducting trials.
Objectives:
1. Identify what kinds of Bayesian statistical approaches are being used in trials
2. Identify how well the details of the statistical approaches are reported in the published papers
3. Determine how well the Bayesian approach performs when compared to classical methods
4. Identify whether previous classical trials may have differed if a Bayesian approach had been used
Implications:
Through this work, we will be able to identify the advantages and disadvantages of using these new Bayesian approaches to conducting clinical trials. We will also be able to see whether Bayesian approaches are able to reduce the number of patients needed to complete a trial without affecting the reliability of the results. Finally, we will also aim to provide an overview of what kinds of Bayesian approaches have been used in the past, and identify ways in which these approaches can be improved.
Competitiondate
202310
Approvedterm2
3 yrs 0 mth
Pinamesdelim
Bansal, Rishi
Approvedterm
3 ans 0 mois
Cihrcontribution2
~ $105,000~ $0~