How technology will drive the future of population health management

By Karen Handmaker, MPP, Global Leader, Population Health Strategy, IBM Watson Health

By Anil Jain, M.D., FACP, Vice President & Chief Health Informatics Officer, IBM Watson Health, Member IBM Industry Academy

Organizations responsible for the care of large groups of people—health plans, hospitals and health systems, large employers, and physicians’ groups—currently have more breakthrough technology at their fingertips designed specifically to improve health outcomes than at any point in history. More importantly, the real-world, practical applications of these technologies have potential to truly improve the health of individuals and drive better outcomes of the population as a whole.

To get a clearer sense of how these technologies are currently being used and which specific breakthroughs may have the greatest impact on population health going forward, IBM conducted a research study that revealed a progression from population health management to precision health and wellness for individuals. Part of that research included qualitative interviews with a group of leading U.S. population health experts representing academic, provider, and payer perspectives.

While each expert approached managing population health from a slightly different perspective, all agreed that technology in many forms will play a key role.

Accountability promotes a quality focus

Utilizing “big data,” which includes the capability to process and analyze vast amounts and types of clinical and administrative information, has made it possible for healthcare payers and providers to move away from the fee-for-service model of healthcare and toward a world where payment is based on outcomes.  According to David B. Nash, M.D., MBA, dean of the Jefferson College of Population Health at Thomas Jefferson University, that’s the key to unlocking the possibilities of population health.

“As the healthcare industry moves from volume to value, organizations will learn to drive waste from their systems by leveraging various tools of pay-for-performance including payment models such as bundled payments and Accountable Care Organizations (ACOs),” Nash explained.

He added that, as these models become more widely adopted, the analytics developed to support them will enable nationwide physician performance benchmarking with “real, cross-system national data sharing that allows people to benchmark and adjust.”

It’s a sentiment that was echoed by Mary Jo Cagle, executive vice president and chief clinical officer at Cone Health in Greensboro, NC. Cone Health is part of an ACO called Triad HealthCare Network, which has been named a standout ACO based on criteria including quality scores. Cone Health’s success, she says, has everything to do with its physicians “taking responsibility and accountability” for the care of their patients. “A big leap we’ve made over the past year is being able to get all of our claims data from our Medicare patients” in order to see and understand total cost of care for populations and individuals. Visibility like that “is huge,” Cagle says, because it lets her team identify the areas that need work. “Where is there waste? Where could we bring better value to our patients? Where could we make our community healthier?”

Targeting therapies with precision

The power of technology in population health extends beyond big data applications for quality measurement and benchmarking capabilities. Today’s data science breakthroughs are creating a revolution in targeted interventions and individualized therapy. Additionally, technology makes it possible to quickly aggregate individual patient data to identify important patterns in large populations, which healthcare professionals can use to intervene sooner and more effectively, ultimately spending wisely and driving better outcomes.

This has been a major priority for health insurer Highmark, where Phil Majewski, M.D., is senior medical director. Majewski explained that Highmark uses sophisticated risk identification and stratification to identify group members who would benefit from targeted disease and complex case management. “We have trained nurse-coaches who work with those members to make sure that their health is as ideal as possible,” he explained. “It’s an involved process, and it certainly is costly, but we see the value in that cost in lower claims” overall.

For individuals, new analytic techniques are being used to target therapies more effectively at the molecular level. Mark Cullen, M.D., director of the Stanford Center for Population Health Sciences and senior associate dean for research at the Stanford University School of Medicine, says genomic data’s influence on population health management “is going to be huge” in the very near future. “Researchers are already using genomic information to identify personalized treatment strategies for cancer,” Cullen noted. “The question now is how to use aggregated genomic and other datasets to make population health management programs more effective while also maintaining data security and patient privacy.”

It is also becoming increasingly possible to marry these two concepts of risk stratification and personalized medicine to measure the value of a particular therapy down to the individual patient level. Humana Senior Vice President and Chief Medical Officer Roy Beveridge, M.D., explained.

“In precision medicine, you need an integrated approach that incorporates machine learning, big data, and analysis to really understand whether an individual is going to get meaningful benefit. You may be paying more for an individual to get a particular benefit. But if precision medicine is done correctly, you’re also are finding a lot of folks who are not going to benefit from a particular treatment, and should not, then, get futile therapy. That is how savings will be realized across populations through precision medicine,” he said.

The transition to a world where powerful data can be fully exploited to improve population health is still very much in the building phase. As Stanford’s Cullen pointed out, there are still some steep hurdles that need to be overcome: “The first big challenge is getting access” to such datasets. “It’s not enough to just say, ‘here’s a portal with 300 million electronic medical records, and here’s another portal’” with administrative data, and yet another containing even more information, like claims or environmental data. Those datasets, Cullen says, need to be in “a place where people can really use them.”

Everyone we spoke with agreed that making notable strides forward to improve population health is no easy task. But they also concurred that recent developments have shown what’s possible when the healthcare industry rallies around a common goal of reducing cost while improving access, outcomes, and satisfaction.

Authors’ Note

Interviews were conducted by Oxford Economics from September-November 2016 for the IBM Institute for Business Value executive report, “Precision health and wellness: The next step for population health management.” See the full report here.