Healthcare can strive for actionable data, especially regarding social determinants of health, a Lumeris executive says.
Credit: U.S. Air Force photo by Steve Pivnick
Michael Cousins, PhD, isn’t impressed with talk of “tweaking” electronic health record (EHR) systems. After all, he says, it’s 2017. Companies like Capital One, Target, and Amazon have been gathering and capitalizing on consumer data for the better part of 20 years.
As for medicine? “We in healthcare are only talking about tweaking—and I think the time is well past for us to move beyond tweaking,” Cousins, the chief analytics officer at the population health solutions company Lumeris, tells Healthcare Analytics News™. “It’s really unfortunate because patients are literally dying because of it.”
That’s not hyperbole, he adds. Healthcare’s slow crawl forward results in harm due to what’s not happening. If hospital systems and providers were to more aggressively build out their data infrastructure, they could reap better insights. That could help spur precision medicine for the individual, not the less-targeted demographics of their neighborhood, Cousins says.
He and Lumeris have studied the benefits. If a predictive model looks at 100 people, it may accurately pinpoint 30 who are likely to soon be readmitted to the hospital. So health providers or plans contact those individuals, encouraging them to come in for a visit or book an appointment with a home nurse, Cousins says.
But when the model takes into account social determinants of health (SDH) data, the precision of that forecast spikes, increasing the projected number of people facing readmission to 50, Cousins notes. “So now half the patients are getting that proactive or better care,” he says. “Fewer of them are coming in for emergencies.” That will help more people to avoid heart attacks, strokes, and similar ailments, he notes.
EHR systems are key to unlocking this potential. They already harbor troves of SDH data, touching upon how and where people live, what they do, their habits, and more. Providers, meanwhile, might know that a patient is recently divorced or has a pet cat, which represent other useful SDH data points, Cousins says.
The trouble is, as many in healthcare know, the information tucked inside EHRs is often difficult to access or organize, he says. Further, SDH data gathered in conversation might not ever make it to the EHR. “I suppose I should be happy that [EHR] systems are tweaking what they have,” Cousins goes on, “but, in reality, it doesn’t go nearly far enough.”
After about 20 years in the field, Cousins arrived at Lumeris 6 months ago. Since then, he has seen sales rise, a sign that the industry might be catching up. But progress isn’t occurring quickly or widely enough, he says.
Some bright spots exist. Cousins points to the Mayo Clinic, University of Pittsburgh Medical Center, Carolinas Healthcare System, Kaiser Permanente, and Lumeris’s client list, which includes, to name 1, Wake Forest Baptist Health. These premier institutions are at the forefront of capturing and taking advantage of SDH data, he says.
On the other hand, most healthcare organizations—and big-name EHR vendors—take a “superficial” approach to gathering SDH data, Cousins says. Much like car insurers, they examine US Census data for where a person lives. Is it a poor neighborhood? High crime? And their investigation ends there—without digging into the individual and their family history or integrating the insights into care plans, he says.
Here’s the kicker: It doesn’t necessarily take a ton of money to revamp these systems and protocols, Cousins says. He works with smaller, “blue-collar” hospitals that actively choose to leap into the future.
It’s important to not consider data and analytics a “gee-whiz, Star Wars capability,” he says. Instead, hospitals must think of them as “tools to enhance value-based care” and patient outcomes.
For SDH data to have maximum impact on the delivery of care, hospital systems must also think of the information in a more comprehensive way, Cousins notes. It’s not just about precision medicine, analytics integration, and improving accuracy by 25%. They must be aligned with physicians and, ultimately, patients, he says.