How AI is Changing the Data Management Playbook

In this week’s The Winning Edge webinar, Roopa Foulger of OSF HealthCare and Sarah Pletcher of Houston Methodist discussed the complexities around gathering, assessing and using data – and how those strategies can improve patient care.

Healthcare executives are dealing with more data than ever before, and new technologies like AI and concepts like patient engagement are muddying the data management waters. How can leadership stay on top?

For Roopa Foulger, VP of Digital and Innovation Development at OSF HealthCare, the key is establishing a data governance chain of command and protocols. The Illinois-based health system has a data stewardship group in place to watch not only how data is used within the network, but also how it’s managed with vendors and payers.

“It’s just been crazy, how much we have to work with,” she says.

“Everyone is trying to get their kayaks onto the whitewater river that is data flow,” adds Sarah Pletcher, MD, MHCDS, Chief Digital Health Officer and SVP and Executive Medical Director of Strategic Innovation at Houston Methodist.

Foulger and Pletcher were participants in HealthLeaders’ The Winning Edge panel this week, during which they discussed the complex landscape of data management. In particular, they talked about how AI is changing the data management paradigm and creating a challenge: While the technology promises to better gather, assess and use data, it’s making life more complicated for those in charge.

“There's an irony in that the very thing that is creating a lot more work is also there to help us do that work,” Pletcher noted.

Staying On Top of a Dynamic Landscape

Both agree that AI can manage data much better than humans, but humans have to be on top of governance. And that means understanding how to manage transactional and analytics data, as well as unstructured and structured data, and working with vendors and payers over who has access to what and who owns the data.

“We can't just care about how we, the health system, store and manage the data,” Pletcher says. “We really have to be involved in how our vendor partners store and manage data, how our employees store and manage data as they're communicating in everyday life, doing their jobs, how contractors have access to data, supply chain, insurers and our patients. Nowadays, a patient can click on, agree, agree and integrate, and have data exchange right through the EMR. It has become a very complicated world and no longer can you just focus on your own organization.”

Foulger says healthcare leaders also have to understand that data management means more than just keeping track of data. Thanks to AI, there are “dynamic new landscapes” in healthcare to govern, including metadata, continuous monitoring for drift and decay, and tools that can parse data not only for clinical but also financial value.

“Where the data was acted on or how the data was inputted has a meaning depending on where it is in the clinical workflow,” she says. “We're just scratching the tip of the iceberg in terms of how we're going to leverage that data clinically and financially.”

Pletcher says that while health systems and hospitals are often forming committees to specifically assess and monitor AI programs, data management goes back a lot further. Instead of reinventing the wheel, she says, it’s important to integrate AI protocols into long-standing data rules and standards.

Keep the “philosophy of the original protocol … in place,” she says, and “evolve and refresh the language to better capture the modern landscape.”

Where challenges may arise is in generational AI, for many healthcare organizations a new and evolving landscape. Foulger says continuous data governance is a new concept, and it requires input from all departments as well as new arrangements with vendors.

Using Data to Help Patients

One of the more intriguing possibilities with AI and data management is turning that data around to help patients. With the move toward patient-centered and value-based care, health systems and hospitals are learning how to empower patients to better understand their care journey, ranging from how to improve care management, managing health and wellness and understanding financial responsibilities.

“There's still a long way to go to help patients really understand the insights from their data,” says Pletcher.

That includes using AI to improve healthcare literacy.

“We are looking at how we can translate some of the complex clinical data into plain language or visualization,” Foulger says. Aside from using AI to translate medical jargon, this includes adding experts in human-centered and experience design to make data tools more convenient for patients.

Pletcher says the rapid development of the Internet of Things means that there are countless opportunities to capture data in the healthcare ecosystem. And that also means helping patients to understand when and how they’re not only accessing their data, but creating opportunities for unintended uses or misuse.

“We have to hold ourselves accountable … and be aware of how much data exchange is happening, and how little control we sometimes have over how those permissions are being given away,” she adds.

That will be important as healthcare evolves and more services move from the hospital, clinic or doctor’s office to the home. Both Pletcher and Fouler say they can see  a future where data from outside the enterprise is just as important for care management as data from inside the hospital. Concepts like remote patient monitoring, virtual care and acute care at home will require new protocols and standards to ensure that information being pulled in is entered into the medical record and put in front of the right clinicians at the right time.

Foulger says it will be important to track data for many kinds of value, and remember that clinical data can have financial value as well. The ROI can be as varied as the effect on variations in care, improved billing and reimbursement opportunities, improved treatment and outcomes, patient engagement and satisfaction, and operational efficiency.

Both agree that “archaic” policies around reimbursement and governance also have to be updated. Healthcare organizations are often held back from using data because of the rules around data gathering and use outside the hospital, and innovative programs are held back because of reimbursement barriers.

Pletcher, meanwhile, says she sometimes wonders if healthcare is making data management more complex than it needs to be. Technology should not get in the way of someone accessing the healthcare they need.

“The patient just wants to live their life as well as they can, feel as good as they can, have as much convenience and simplicity as they can,” she says. “And that's kind of the trade-off they're making. Sometimes I'm surprised that we don't get out of our own way to deliver those experiences to patients.”

“I guess I'm surprised that there's still such an abyss between what patients want and the trade-offs they're willing to make for it and what we're delivering to them, in terms of a truly integrated convenient experience because of our worries about their privacy,” she adds. “I think that we may be making things much more complicated than they need to be.”