The Data Doldrums

Two generations have passed since the computer age inspired the founding of companies seeking to unlock the promise of health-care information technology. Another generation has passed since the term “Big Data” was introduced to suggest ways that huge volumes of information could help discern population-health trends and create pathways for delivering care that would lower overall costs.

As the restless 4-year-old in the back seat might ask on vacation, “Are we there yet?”

Providers, insurers, health-policy and IT specialists, along with various trade organizations representing the physician and hospital communities, insurers, consumers and others have answer that will rarely stop Junior’s interrogations from the back seat: “We’re getting closer.”

Just last fall, a survey of health-care organizations responding to a survey by Health Catalyst emphasized just how important the topic is to their futures: 83 percent said it was “extremely important” to see a return on electronic-health record investment.  That same group of 1,100 professionals, however, isn’t encouraged: Nearly 20 percent said their return on EHR incentive programs had been “terrible;” 42 percent rated their experience as “poor.” And only 10 percent felt that the $30 billion spent on incentives to implement EHRs was justified.  Worse, that sentiment was measured not long after the American Medical Association reported that the challenge of creating EHR documentation was taking a bite of six hours out of a primary-care physician’s day—with one-fourth of that coming after their shifts were to have ended.  Row those results downstream to employers offering health-care insurance to their employees, and it’s easy to understand why many executives are frustrated with annual increases in premiums for their staffs. Where, they ask, is the ROI for their organizations?

None of this comes as breaking news to Charles Rhoades, a physician and CEO for Kansas City Orthopedic Institute. Fully 25 years ago, he was part of a study group for a project called the Community Health Care Initiative, working to get area hospitals and practices onto the same page with shared records.  “It failed for corporate reasons, it failed for regulatory reasons, and out of the fear that you might be helping your competitor if you provided access to care received at your place,” he said. “It was very frustrating.” And in some ways, still is.

Nearly 40 years after the 1979 founding of both Cerner Corp. in Kansas City and Epic in Madison, Wisc., each has risen to become one of the world’s five biggest health-care IT companies. Their systems have transformed the way data is compiled, aggregated and used, and their products have vastly improved the ability of providers to track and deliver care.  But, critics say, EHR systems providers have largely failed to address concerns about the way their products interact with others, and as a result, health-care silos are thwarting the promise of that technology. The need to safeguard proprietary software that is at the heart of the business model for health-care IT companies, and that sector has resisted calls for collaboration that could truly unleash the power of data, connect consumers to providers across the spectrum, improve efficiencies and cut costs.

With concerns about protecting corporate secrets dragging on efforts to create systems that are truly interoperable, and with regulatory issues like the Patient-Protection Act, “you have identified two of the greatest obstacles to the digital age in medicine,” Rhoades says. “One is simply sharing data and having interconnectivity; the other is having identified outcomes, if you will, which help evaluate quality of care. That is a much more difficult problem, because it is hard to define exactly what needs to be measured.”  For all the excitement about the clues that can be derived from huge tranches of data, there is caution in medical circles—the front line of data aggregation—about what’s being collected, and how. Physicians and nurses note that every specialty, for example, will have its own preferred metrics—the heart patient’s record must by definition differ substantially from that of someone with a construction-related work injury.  EHR systems can chronicle their particulars, but how, then, do those records become part of a larger data set with discernible information for policymakers, providers and payers?

“Data is only as good as the information it produces and the actions people take based on the information it produces,” says Karen Johnson, vice president for health-care insights and partnerships at Blue Cross Blue Shield of Kansas City. “That’s our biggest challenge around translating big health-care data, which is, ‘where are we?’ The bigger challenge is, what to do with what we’re learning, based on aggregated data.”  The business model around health care is so complex, she says, it is ingrained in and sits in so many silos, that coordination among all is enormously complex. But the health-care world is starting to see the opportunities emerge. When you look back at the High-Tech Act of 2009, which authorized billions of federal dollars as incentives for implementation of systems by hospitals and providers, the industry has come a long way.  “Where we are today, yes, do all hospitals have EHR? Yes. Do most physicians? Yes, but the process is relatively young,” Johnson said. “If you look at the digital revolution, the iPhone moved a lot faster. But it’s important to understand that the implementation for providers has meant not just having new tech, but changed their whole workflow, staffing models, skills they had to learn. And that’s just the front end. We’re still trying to have that conversation about what to do with the data in there.”

Even at the point of collection, says former Google and Uber executive Brian McClendon of Lawrence, “there’s a lot of data in medicine that comes from a very numeric form—lab tests, MRIs and X-rays—that can be digitized and figured out. But there’s a lot of data in the medical system that comes from doctors’ notes and nurses notes.”  Those used to be hand-written, he notes, and the structures within them could change from practice to practice, even within a given field like cardiology.  “So one of the challenges that Cerner faces is how to interpret nurse’s and doctor’s notes and instructions so that it’s consistent across all their patients,” a process called data normalization,” he said.

That’s hugely important, because while there may be consistency of data across patients with height or weight, day to day interactions with a patient are not.  But, said Blue KC’s Johnson, EHRs are only one way that providers are upgrading. Referencing a podcast titled “It’s Time to Face the Fax,” she noted that the health-care industry is what keeps the fax machine alive and well. What we have in health-care today is kind of crazy; we have millions of dollars invested in these complex, sophisticated EHR systems, but we’re still trading a lot of information with each other via fax machine.”  The upside, she said, is that “the tolerance of healthcare leaders around that is waning. There is going to be lot of change over the next five years.”

Rhoades, for all his concern about barriers to change across a quarter-century, is also optimistic about what’s on the horizon.  “I’m a huge fan of the digital age, and lecture nationally on its adoption,” he said. “Without question, it solves a lot more problems than it creates. This is a big problem, but don’t misunderstand:  I would never want to go back to the paper chart.  These are things those of us who were early adopters and advocates knew could occur. We knew it was going to be difficult, and it has proven to be.  “I’m passionate about this subject, and have been a big advocate for electronic records within our own organization—but that does not mean I don’t recognize there’s a problem.”