Global consortium pieces together the COVID-19 data jigsaw

National Kidney and Transplant Institute (NKTI) is the consortium’s lone Philippine partner 

By Patricia B. Mirasol

“It’s like driving a car blindfold—at breakneck speed—with your parents at the back. We don’t have a manual for this,” said Professor John Fraser, Director of University of Queensland’s Critical Care Research Group, as he explained that there is no textbook answer for treating patients with severe COVID-19 symptoms.

The is a global alliance of doctors and scientists who are piecing together the different parts of the novel coronavirus’ data jigsaw. More than 360 hospitals and affiliated research centers from 52 countries have joined the Consortium to build a COVID-19 global database for critically ill patients. The Consortium has a dashboard created by IBM that uses artificial intelligence (AI) and data modeling to generate clinical insights.

The National Kidney and Transplant Institute (NKTI) is the consortium’s lone Philippine partner. 

LISTENING TO THE DATA
The data gathering started in mid-January—before the virus was even named and before anyone knew that it would spread outside Asia. Mr. Fraser, who is also president of (Extracorporeal Life Support Organization), recalled that people in his network began asking him what they should do with the information they had gathered on the novel coronavirus. “We knew we had to do something,” he said. “We decided to create a database.”

Initially called , its goal is to characterize clinical features about COVID-19, specifically: the epidemiology, clinical course of infection, the severity of respiratory failure, , ECMO—a form of life support for those who can’t provide oxygen for their own blood—details, and intensive care unit (ICU) and hospital survival. 

The Consortium will develop strategies to guide the international ICU community on how to best treat those with severe COVID-19 symptoms through the collation and analysis of data from its partner hospitals.

The global alliance ; doctors can glean patterns from the amassed database, and can better understand which demographic groups are most at risk or what treatments work better for which population segment. It gives them decision support and creates change and outcomes at the bedside. The more data is fed into the machine, the more efficient it becomes. 

“Our job is to listen to the data. You need to have enough data to help, but not too much to overwhelm. We trimmed it to key points… We did a best size fits all,” said Mr. Fraser.

The data is managed through a custom-made dashboard for clinicians created by IBM. IBM and another multinational technology company contacted Mr. Fraser right off the bat to offer their assistance at the start of the study. IBM, with its expertise in AI and machine learning, consulted clinicians to answer the question, “What’s the best way for a dashboard to work?” The multinational technology company, meanwhile, is being eyed to help the Consortium consume data in various formats (whether image, audio, or video files).

THE WAY FORWARD
A number of sub-studies and subgroups are coming out of the main study. At least one of these will tackle the long-term consequences of the disease: Can survivors still work after 6 months? Can COVID-19 cause neurological damage?

“COVID-19 hasn’t been here one year,” Mr. Fraser said. “If you lose a part of your workforce that can’t earn taxes and would eventually need nursing care, that’s a huge course of study with massive implications.”

The Consortium is a collaboration among professionals and institutions that have banded together gratis to discuss, listen, and learn. It is seeking grants so it can employ data collectors to enter data and relieve overworked healthcare staff of this task.