How Stanford is using health data to predict heart failure, so doctors can help before it happens

SAN FRANCISCO — At the Stanford University Medical Center, electronic medical records (EMR) can do far more than just store information. They function as a starting point for systematically providing better care.

“Using an algorithm that we’ve devised we can determine which patients have heart failure, which patients we need to see,” said Dipanjan Banerjee, a clinical assistant professor of cardiovascular medicine at the medical center, at VentureBeat’s 2014 HealthBeat conference today. “We think we’ve used EMR to allow users to do work they need to do, rather than spin their wheels, trying to find patients [with] heart failure.”

[aditude-amp id="flyingcarpet" targeting='{"env":"staging","page_type":"article","post_id":1589307,"post_type":"story","post_chan":"none","tags":null,"ai":false,"category":"none","all_categories":"big-data,business,","session":"A"}']

And he’s open to new data, too.

That could mean easy-to-use wearable devices, which have commanded plenty of attention in the past couple of years.

“If we had a device that has wireless weight monitoring that can automatically be uploaded to our EMR, that’s golden,” he said.

But other data could well come in handy, too, at least from the perspective of Dr. Pravene Nath, chief information officer for Stanford Hospital and Clinics. He’s open to taking the information that patients feel comfortable offering up to Stanford “and then mining that data as appropriate,” he said.

For instance, he said, information about “what Starbucks you go to” could provide value.

But the more data, the better — as some data experts have argued — isn’t something Nath necessarily agrees with. “I think we have to be very careful,” Nath said.

For instance, working with genetic data, he said, could have “unintended consequences.”

As more data becomes available to health care industry, though, it’s important that it doesn’t sprawl out across many platforms for professionals to deal with.

[aditude-amp id="medium1" targeting='{"env":"staging","page_type":"article","post_id":1589307,"post_type":"story","post_chan":"none","tags":null,"ai":false,"category":"none","all_categories":"big-data,business,","session":"A"}']

Hospitals should be able to “plug them in in a way that, if we do it right probably we shouldn’t have to manage,” Nath said.