On the Thousand-Mile Journey Into Data Analytics for Value-Based Healthcare, Experts See Moderate Progress

March 26, 2019
Everyone understands that leveraging data analytics will be vital to success in value-based healthcare—but industry leaders are finding out just how hard it is to make it all work together

Where is the U.S. healthcare industry right now, on the “thousand-mile journey” into value-based care delivery and payment, when it comes to leveraging data analytics for success? Well, say industry experts, it depends on how one looks at it. Progress is being made, but things aren’t moving as fast as they might be, and major challenges remain.

Healthcare Innovation spoke with a variety of industry leaders and observers to get a sense of the temperature of the industry around analytics-leveraged progress in value-based healthcare; not surprisingly, they see a mixed picture.

“I think that we are probably further along with the analytics itself than we are in changing the care delivery model,” says Debbie Zimmerman, M.D., corporate chief medical officer of the St. Louis-based Lumeris consulting firm and CMO of Essence Healthcare, the health plan that Lumeris operates. “When we think about what we need to do, we need to change the incentives and the business model; we have a business model that supports fee-for-service,” she emphasizes. Still, there is a “chicken or egg” quality to all of this, she says. “For instance, we know that in order to move to value-based healthcare, I need population-level data analytics. How is my performance at the population level, the sub-population level, and the individual-physician level? How is the performance on all those levels? And what is the variation of care like? We have yet to see a population where there isn’t variation of care. So you need that population level, and you need a range of analytics on population, health risk, care management, and so on.”

In that context, Dr. Zimmerman says, “When I say we’re not very sophisticated at it, the situation is that  we have it, and we probably know where the opportunities are; it’s a little clunky. We’re not as sophisticated at identifying trends and opportunities, but we’ve put people and brains on it. When we talk about variation at the individual-patient level, we talk about who needs outreach. And we all know that the social determinants of health are a big indicator. So how do we incorporate that with other data, like claims data and pharmacy data?” There are vast opportunities at numerous levels, she says, including around embedding key data and analytics into clinician workflow to uncover gaps in care. Meanwhile, she says, “We’re a little bit better now about combining clinical and claims data. But we’ve got some opportunities in being smarter about the data itself, and then increasing usage of data by reducing barriers.”

“Most organizations are just starting to scratch the surface around reaping some level of value from analyzed data,” says Rob Barras, head of the Health Solutions team at the Buffalo-based CTG Consulting. “The maturity curve is starting to accelerate relative to the tools available out there on the market; and it’s happening quickly, but not quickly enough,” says the Philadelphia-based Barras. “So it’s slowing the adoption of risk-based contracts. That’s why you’re seeing hospitals, medical groups, health systems, and ACOs [accountable care organizations], push back against taking on downside risk” right now, in the face of demands by Seema Verma, administrator of the Centers for Medicare and Medicaid Services (CMS), and other senior federal healthcare policy officials, that patient care organizations participating in the Medicare Shared Savings Program (MSSP) take on downside risk as soon as possible. In that context, Barras says, “Hospitals, health systems, medical groups, and ACOs have, especially in the past five years, acquired analytical tools and capabilities, either in the context of integrated platforms, or single tools, or they’ve leveraged tools from their EHR [electronic health record] vendor—everything they can to help them make decisions.”

Scott Weingarten, M.D., who has been working with data analytics around clinical transformation and clinical performance improvement for many years, says, “I think we’re getting better as an industry, but we still have a ways to go. You get a lot of information, a lot of which is not actionable. So you need a way of determining what information is actionable and what is not actionable.” Dr. Weingarten, who is CEO of Stanson, a division of the Charlotte-based Premier Inc. health alliance, and who is also a consultant to the CEO of Cedars-Sinai Medical Center in Los Angeles, says, “My criticism in general is this: whenever you do data analytics and benchmarking, you find that, for example, somebody has lower PMPM [per member per month] costs, somebody has a lower readmission rate or bed days per 1,000. Really, the question is, what is actionable? What is important? That’s the next step for data analytics.”

Meanwhile, Barras says, “What providers have found is that those tools were some combination of not-mature-enough, and they weren’t ready to leverage those tools and capabilities. So they went from 0 to 60, and they skipped a bunch of steps. The good news,” he says, “is that because of those failures, they need to go back and redo some of those things, and they’re doing that now. The second piece is that some of the tools are now maturing. And the other piece of it is that organizations are gaining a lot more visibility into what it is they need to answer. So the components of value-based care are becoming clearer. So those three things are working in conjunction, and will start to create a bit more momentum and readiness for downside risk and value-based care.”

What kinds of problems are leading to “re-do’s”? One of the main challenges is a process issue, Barras says. “There has tended to be an assumption that data is owned by IT and that operations plays no role in managing and using that data, but that’s completely flawed thinking, and organizations outside healthcare are much further along in understanding that. Healthcare is just starting to understand that now.” At CTG, he says, “we work in oil and gas and other industries, and in those industries, they already have data stewards. And they’ve made acquisitions and mergers, and have figured out that you have to standardize data sets and have master data management as a strategy. But healthcare organizations have done it backwards, and so they have seven different definitions of length of stay, for instance. And they’re cutting and pasting and unwinding that stuff in spreadsheets. That’s all well and good, and that’s one specific use case you’ve done; but then you can’t get to that next layer.” Still, he says, there’s clear evidence that there has to be “a programmatic approach” to laying the foundations of data analytics to support value-based healthcare delivery and payment and accountable care.

What it looks like on the front lines

So, how is all this playing out on the front lines, among innovative patient care organizations? Well, the picture remains a mixed one, says Steve Hess, CIO of UCHealth, the 10-hospital, 1,800-bed integrated health system based in Denver. UC Health is a large system, with 5.8 million unique patients, and 7.5 percent of that total population, and growing, in value-based contracts. What are the UCHealth folks learning, as they move forward with their ACO contracts with private payers (UCHealth is not a participant in the MSSP program)? “We’re learning that you need a lot of analytics!” he chuckles. More seriously, he reports that “We have Epic [the core electronic health record system from the Verona, Wis.-based Epic Systems Corp.] everywhere; we use it for care management and for population health. We’re using the Epic Healthy Planet module for registries and population health. We pull together the analytics and feed it back to the point of care.”

Importantly, Hess says, “Our strategy is that we’re not treating value-based patients differently; we’re taking all the information and feeding it back to physicians and clinicians at the point of care. I don’t want the physician at the point of care to be trying to treat a patient differently depending on whether they’re a fee-for-service or value-based care patient. But with patients in value-based contracts, we have care management in order to monitor them and intervene as needed. And we can do outreach to patients and try to prevent expensive ED visits. So the interventions can help prevent patients from going to higher-cost settings. But everything we’re doing around analytics is being brought into the EHR. We’re using cost and utilization data, we’re bringing in the risk scores around the patients at greatest risk depending on their lab values and medications, and bringing data around that to the physicians; we’re also using Healthy Planet, the population health tool that stratifies clinical risk; it’s also a registry tool that helps group patients by chronic disease and other factors. And we’re trying the combine claims and clinical data.”

The biggest challenge in all this? One of them, Hess reports, remains the challenge of marrying clinical and claims data. “We have to partner with the Anthems of the world for attributed patients; and integrating claims data is not trivial, by any stretch of the imagination,” he says. “And every payer is different in terms of their claims data format. So that’s a heavy lift. And we haven’t even gone to some of the home data and the social-determinants data—like, how far are you away from the nearest supermarket, etc.” But, going forward in the next few years, he says, the push will be on to adopt cloud-based analytics. In that, he says, “We actually partner with the University of Colorado School of Medicine; we have a Google Cloud-based platform for the big data analytics. Number two, we all have to figure out how to bring external data in and use it. Claims data, external data from other clinics, other patient care organizations, and then home data—all the non-UCHealth data—bringing that data in and managing it, is the holy grail. So you need a very flexible, scalable platform, like a cloud-based platform. And then you need to bring it all together.”

What should CIOs, CMIOs, clinical informaticists, and all informaticists, be doing right now, in patient care organizations, to move the needle forward? “My sense,” says Zimmerman, “is that some providers have a little too much confidence that their electronic health record is going to meet all their needs; so keeping an open mind that there may be other needs, and also pushing to make sure that the EHRs are prioritizing interoperability,” is really important, she emphasizes. “There’s a lot of assertions that the EHRs will move into the population health area and solving all their problems, but that’s going to take a while. So that’s one priority. The other priority would be to work with the other clinical leaders, who are mentoring practicing physicians, and really partnering on what we need to do, which is essentially to change provider behavior.”

Sponsored Recommendations

How Digital Co-Pilots for patients help navigate care journeys to lower costs, increase profits, and improve patient outcomes

Discover how digital care journey platforms act as 'co-pilots' for patients, improving outcomes and reducing costs, while boosting profitability and patient satisfaction in this...

5 Strategies to Enhance Population Health with the ACG System

Explore five key ACG System features designed to amplify your population health program. Learn how to apply insights for targeted, effective care, improve overall health outcomes...

A 4-step plan for denial prevention

Denial prevention is a top priority in today’s revenue cycle. It’s also one area where most organizations fall behind. The good news? The technology and tactics to prevent denials...

Healthcare Industry Predictions 2024 and Beyond

The next five years are all about mastering generative AI — is the healthcare industry ready?