The Canadian Bayesian Adaptive Trials (CanBAT) Network

Overview

Who we are

The Canadian Bayesian Adaptive Trials (CanBAT) Network brings together (i) Bayesian statisticians with expertise in designing Bayesian adaptive trials and (ii) clinical researchers to advance the theory and application of these trial designs in Canada. CanBAT is funded by the Canadian Statistical Sciences Institute (CANSSI) as a Collaborative Research Team (CRT).

Clinical trials are prospective research studies that aim to measure the effect of medical interventions in humans and are the cornerstone of testing the safety and efficacy of these interventions. Adaptive trials respond to accumulating data during the trial to efficiently reach conclusions, while maintaining valid statistical inference. As such, they make better use of time and money than fixed designs.

The Bayesian approach is advantageous for adaptive trials as; (i) parameter information can be continuously updated based on accumulating data, (ii) information can be synthesized across multiple sources, (iii) statistical efficiency can be improved using hierarchical modeling, and (iv) probabilistic statements about the effect of interest are more easily interpreted by clinicians. However, the design and analysis of Bayesian adaptive trials requires bespoke statistical methodology and statisticians integrated within the study team to ensure the integrity of these trials. Thus, the CanBAT Network aims to develop capacity for statistical innovation and support for Bayesian adaptive trials in Canada to support a growing applied interest in these trials.

We are focusing on four research areas:

-        Platform trials

-        Dose finding trials

-        Computationally efficient design of trials

-        Decision theoretic trials

The CanBAT Network combines 13 researchers at eleven institutions across five provinces; Ontario, Quebec, British Columbia, Alberta and Manitoba and mentorship from University of Texas MD Anderson Cancer Center. Our team has expertise in dose finding trials, efficient trial design, decision theoretic trials, applied trial design, platform trials, paediatrics, critical care, emergency medicine, cardiology and decision modelling.

Learn more

Our Vision

The CanBAT Network aims to develop novel statistical methodology to address issues in the design of Bayesian adaptive trials, including evaluating the characteristics of complex platform trials, proposing new dose finding designs and reducing the computational complexity of Bayesian adaptive trials. We expect to have varied outcomes from this network with national and international impact, improving the landscape of clinical trials in Canada.

 

We will

  • demonstrate the value of conducting Bayesian adaptive trials,

  • create a range of novel methodologies to improve the statistical properties and feasibility of Bayesian adaptive trials in practice and

  • develop a sustained set of statistical expertise and resources for Bayesian adaptive trial design in Canada.