What new technologies are changing the way we conduct health research, and how will this impact the use of evidence in decision-making in the future? This was the subject of the latest Sax Forum, an initiative that allows the Sax Institute to share knowledge among Institute Members and help understand what we can do better together.
More than 60 people from the Institute’s member organisations joined the forum online on July 4 to hear three fascinating presentations. First up was Professor Trish Greenhalgh, an internationally recognised expert in primary healthcare at the University of Oxford, who spoke about new developments in evidence-based medicine.
In her talk, Professor Greenhalgh commented on the inadequacies of the commonly-used “pyramid of evidence”, where evidence is placed in a hierarchy, with systematic reviews and randomised controlled trials at the top and expert opinion and case-based reading at the bottom.
In some complex situations, we have to accept that there’s no universally applicable hierarchy of evidence, Professor Greenhalgh said. She pointed to one of the first COVID superspreader events in 2020, where dozens of infections occurred among members of a choir. Airborne transmission of the disease seemed the probable cause and universal mask-wearing in indoor settings a reasonable precaution. And yet as late as 2023, a Cochrane review based on the conventional pyramid of evidence was still able to conclude that there was no firm evidence that masks work.
“Good science needs multiple methods, applied adaptively and pragmatically to build a nuanced narrative that explains all the evidence. Theory and explanation are at least as important as method,” she noted.
An “argument framework” grounded in the sciences of reasoning provides an alternative to medicine’s measurement framework for evaluating and synthesising evidence in healthcare, she concluded.
Professor Julian Elliot, a leading researcher in evidence synthesis at Monash University, spoke about the “living evidence” paradigm in evidence synthesis. Evidence review systems need to change, he said. Producing rigorous summaries of evidence takes an enormous amount of time, and we don’t have the systems that can move at the speed they need to.
The living evidence model meets this challenge by developing useful summaries of evidence that can be kept up to date with the rapid flow of new evidence. Compared with the conventional model of reviews of evidence which are updated every three to five years, living evidence models are updated every one to three months. This provides clarity in complex research fields and improves the engagement of decision-makers, Professor Elliott said.
Professor Bronwyn Hemsley, a leading speech pathology researcher at the University of Technology Sydney, spoke about applied AI in speech pathology. Among the possible uses in this area are using AI to inform the screening and diagnosis of swallowing disorders; visual supports for language learning in autism; monitoring early development milestones; and preschool and school-age speech development.
In fact, almost every area of speech pathology has potential for AI technology, Professor Hemsley said.
Learn more about our Sax Forums here.
Watch now:
PUBLICATION DATE
11 July 2024
