PSI VisSIG Wonderful Wednesday Webinar Series
This is a good opportunity to develop your knowledge by thinking through a practical example, practicing how to apply graphics principles, and developing your coding skills.
The keynote speakers for the 2022 conference were Jonas Haggstrom and Thomas Jaki.
Jonas Haggstrom, Vice President of Global Health at Cytel and Chair of the Inernational COVID-19 Data Alliance Statistical Expert Group
The COVID-19 pandemic is the greatest health challenge for a generation and has highlighted the critical importance of generating rapid and rigorous evidence for decision-making for the treatment of COVID-19. As a response to this challenge the International COVID-19 Data Alliance (ICODA) was formed, an open and inclusive global coordinated, health data-led research initiative bringing together for- and non-profit companies and organisations working collaboratively to enable and empower researchers to access RCT and RWD health data in a responsible way, making use of innovative data science, contemporary tools and technology to accelerate knowledge of the prevention and treatment of COVID-19.
This talk was in two parts; the first was less technical in its nature and focus on the ICODA journey from its inauguration in July 2020 to the current state with 12 driver projects, including more than 130 researchers, working on data from over 40 different countries – the ins and outs of how to successfully bringing different companies and non-profit organisations to collaborate towards a common goal, and specifically how statistician played an essential role to make it happen. The second part presented results from one of the ICODA projects evaluating the efficacy and safety of COVID-19 of existing medical interventions to support drug repurposing efforts.
Thomas Jaki, Professor of Statistics and Programme Leader at the University of Cambridge and Professor of Statistics at Lancaster University
Master protocols are a new class of clinical trial designs that allow treatment arms to enter and leave the trial and/or (sub)-populations to be added over time. They include platform, Basket and Umbrella trials and have proven to be particularly popular during the COVID-19 pandemic where no fewer than 58 trials have been registered as a platform trial. A common feature of these designs is the desire to answer several research questions within a single protocol. One of the questions arising from this feature that has generated a lot of discussion in the literature is the need for (or lack of need to) control error rates as well as the most appropriate type of error control. Another core feature of Master protocols is the desire to maximise the utility of information in the study by using the same information to answer multiple research questions (e.g. shared control group in platform trials or information borrowing in Basket trials).
In this talk I began by reflecting on the discussion around error rates and introduce different possible testing strategies that could be considered for Master protocols and platform trials in particular. I then introduced possible strategies to borrow information across different research questions and discuss implications of such strategies on the sample size requirements of the study. Throughout this presentation I highlighted areas where further research is necessary to enable Master protocols to unleash their full potential.