The University of California Health has created a shared data set for COVID-19 research that would accelerate development of treatments and vaccines. The shared data set, dubbed UC CORDS, centralizes data from UC’s 6 health systems used statewide, and provides care for millions of patients yearly.
As such, the data points from the patients’ health records can be aggregated and used to quickly compare treatments of previous patients, map out patterns and draw actionable insights.
“With the scale of the pandemic, we need as many UC researchers as possible to work on treatment options. Having access to this diverse data set that is already integrated may contain insights into COVID-19 that they may not find elsewhere, and can make their work more efficient,” said UC Health’s chief data scientist, Atul Butte. “This type of dataset may provide a window into patterns they might not have otherwise been able to identify.”
Given that the UC’s healthcare systems are spread out throughout the state, they have a much more broad and diverse data pool that includes numerous demographics. In total, there are around 640 million data points in the shared data set for COVID-19 research.
“A significant benefit of UC CORDS is that it gives you insights into clinical practices much closer to real-time and is representative of a broader patient population than any one organization would have on its own, which is critical for research during the pandemic,” said Associate Director at the UC Jonathan Watanabe.
A shared data set for COVID-19 research is the inevitable next step when presented with a global challenge such as this. Now is the best time to develop techniques that would pool knowledge from multiple institutions. Hopefully, a global or multinational version of a centralized data set will be created at some point.