For progressive HIE leaders, the future is now
Not everyone embraces the term ‘health data utility,’ but HIE leaders are nevertheless charging ahead with the concept.
Even amid a rather jumbled and uncertain healthcare policy landscape, the future of health information exchanges is becoming clear over time. HIE leaders nationwide are embracing the idea of becoming far more than simply storers and sharers of patient data.
Rather, they're looking to become key players in the use of patient data to improve clinical outcomes and community health regionally and even nationwide.
On the one hand, the current administration’s rejection of the previous administration’s vision for TEFCA — the Trust Exchange and Common Agreement initiative managed by the Office of the National Coordinator for Health Information Technology (ONC) inside the Department of Health and Human Services (HHS) — has meant a rejection of mandatory participation in what was once seen by many as a nationwide network of networks that virtually all patient care organizations and all HIEs would participate in. That shift has led to some degree of uncertainty about how TEFECA will evolve forward over time.
On the other hand, what absolutely is emerging now is a broad consensus around a different concept, that of the HIE as a health data utility (HDU), integrating clinical, behavioral, social and research data for the public good. Embedded in the concept is the idea of public-private partnership, including across public health management and biosurveillance, with collaboration along numerous dimensions.
That suggests that HIEs increasingly are working not only with providers and private payers, but also state governments, public health departments, community organizations and other entities, to optimize patient data use.
The Consortium for State and Regional Interoperability (CSRI) notes that the Health Data Utility Capability Model that it has developed over the past few years “is a practical, stakeholder-driven method for characterizing and assessing the capability and maturity of HDUs that provide comprehensive health data and analytics services to a wide range of public and private sector stakeholders across the U.S.”
What’s more, the organization states that, “Building on CSRI’s HDU Maturity Model first launched in 2023, the HDU Capability model establishes a method for describing what an HDU can consistently deliver at scale that is outcomes-oriented, evidence-verifiable, and stakeholder-specific, mapping capabilities to the distinct needs of providers, public health agencies, Medicaid and other state programs, payers and patients.”
And though some HIE leaders find the term “health data utility” slightly problematic, the concept of the HDU is being universally embraced. It certainly fits into the paths forward that they are forging as they move to chart the future of health data exchange.
Charting capabilities across a variety of dimensions
“To date, we’ve surveyed about 55 HIEs nationwide around their capabilities, services, business models, and projected trajectory,” says Pam Arlotto, president and CEO of the Atlanta-based consulting firm Maestro Strategies.
“It’s very, very rare that any one state government would give all of those elements to one agency or organization, because it’s all operationally very complex. Instead, most states are giving some capabilities to one organization and others to another,” she explains.
Arlotto reports that her consultancy has been able to articulate four stages of development for emergent and evolving health data utilities. “1.0,” she says, involves original HIE capabilities and interoperability and infrastructure; 2.0 gets into service offerings and basic analytics capability. “3.0 means moving into the stage at which the users are not just sending and receiving data, but actually integrating it into their workflows, and data is contributing to outcomes improvements. In 3.0, they’re changing the way they work.”
And, she says, “4.0 is contributing to statewide goals and not just delivery outcomes, but true public health initiatives and outcomes and statewide initiatives. That’s a true HDU.
"Most HIEs are still in the 2.0 or 2.0-plus stage.” Importantly, she adds, “The public perception is that HIEs are still only in the 1.0 stage, that they’re still just moving data. But the innovative ones are already past that stage.”
Engaging with community health in New York state
Leaders at progressive HIEs are already doing the work of which Arlotto speaks. One example is New York City-based Healthix, which operates in New York City, Long Island and large swaths of the southern part of New York state.
“We’ve moved into a more mature phase here,” reports CEO Todd Rogow. “First, you have to establish trust and obtain trust in the community and get people to send you their data, and become both a data aggregator and a data utility. Then, you have to move forward to leverage that data effectively. We already have 50 million unique individuals in our EMPI” (enterprise master person index).
Related to that, Rogow is leading an initiative around maternal health, noting that “I kicked off our maternal health collaborative for Staten Island specifically. We have the data, we know where the needs are. And what’s also interesting is both the clinical and non-clinical elements” when one attempts to collaboratively address broad sets of issues like maternal health.
“What about the environmental elements? Do they have mold in the home, for example? You can trace so many medical issues back to the home. And we get to see a person’s full clinical journey around maternal health. So whether someone has insurance or not and whether they’re transitioning into the home, or wherever they are, we want to make sure that both the clinical and community sides, and the provider and health plan sides, are involved. The caregiver gives you the baton, the provider does the infant delivery, for example. We have a big 1115 waiver from CMS (the federal Centers for Medicare and Medicaid Services, and its Section 1115 demonstration waivers) in New York state that we’re supporting, around housing, food, and transportation needs, parallel to the clinical needs. So we intermingle those elements into the work we do, anchored in the clinical data we have.”
And it is in initiatives like this one, Rogow underscores, that HIEs can really shine, as they can gather together data from numerous different sources, analyze it and help providers, health plans, community-based organizations and state agencies all participate together in improving the health status of populations.
In Delaware, new horizons around analytics
Jan Lee, M.D., and her colleagues at the Delaware Health Information Network (DHIN) are on a parallel path to that of Rogow and his colleagues at Healthix.
“We’re very focused on analytics,” says. Lee, DHIN’s CEO. “We’re working closely with public health.” DHIN has the advantage that some leading HIEs have in that its state government has designated it as its all-payer claims database. In that context, she says, it’s become clear that “Our biggest data gap here in Delaware is on the ambulatory side.” As a result, she and her colleagues have been focused on collecting, sharing and analyzing data on the outpatient side of the care continuum, while moving forward in some very innovative areas, among them, in post-marketing surveillance.
“When clinical trials are conducted, you’ve got to try to control as many variables as possible, selecting people for your clinical trials,” Lee notes. “And when something passes the FDA and gets into clinical use, the variables are no longer controlled. So there’s a need to survey the population in terms of how a new medication is used. And we’ve got data! So we’re teaming up with an organization now involved in post-trial surveillance. They’ll do post-marketing surveillance for a particular drug or initiative. And does it need to be identifiable data? If so, we need patient consent.”
Another area for innovation is “clinical trials sourcing. Let’s say a pharmaceutical company is ready to do a clinical trial: the most time-consuming and expensive part of a clinical trial is recruiting patients. So, the pharma company says, ‘We want to do a clinical trial on XYZ; here are the inclusion and exclusion criteria.’ And we can do a de-identified search for patients. And if so, are they in certain pockets? And if they’re researching an endocrinology product, do we have enough patients in the data to help providers match their patients with a clinical trial? The patient, the provider, and the pharma company all get wins from that. So we see the life sciences area as the next thing up.”
In Vermont, a push to understand AI-facilitated data optimization
Another voice among innovative HIE leaders is that of Randy Farmer, CEO of the Vermont organization VITL, which runs the Vermont Health Information Exchange. VITL and the Vermont Health Information Exchange were created by state legislation in 2005.
“I do think that the ‘health data utility’ term is somewhat diminishing and limiting,” Farmer says of the term. That said, “Having worked at Delaware’s HIE for 14 years and completing my first year here at VITL, I absolutely see that HIEs need to transition from the movement of data and become health intelligence networks; that really is where the HIEs need to go.
"We connect healthcare systems; EMRs can’t do that. HIEs are really at the heart of connecting ecosystems and connecting care where it’s provided and providing health intelligence.”
Vermont has a strong landscape of stakeholder collaboration, notes Farmer. “The folks in Vermont, and my predecessors at VITL, have built an outstanding relationship with public health, around immunization data, electronic lab reports and disease registry information; there’s been amazing coordination with public health. And full participation in the Vermont Health Information Exchange across all our hospitals.”
And, related to that, he says, “We’re in the process of onboarding a new supplier for our clinical data repository, to better leverage FHIR resources. We’ll be easier to work with. And we’re building a better, more elegant profile for our users to interact with. And we’re getting certified in ITIL, the best-practice framework for data services. I have nine certifications in ITIL; when I was in Delaware, we put a lot of emphasis on ITIL, to become a better and more strategic partner for hospitals, ACOs, and labs, etc.
"And 100 percent of our leadership team will be certified in ITIL Version 5 certification. That’s the first version encompassing artificial intelligence and AI governance. So at least 90 percent of our staff will be certified in ITIL Foundation. That’s going to be a very big deal for our company and for our stakeholders. And we are also reorganizing our IT team to create a specialization emphasis on data security, supplier management, and solution development and deployment, and day-to-day operations.”
Heading towards a new patient data capability
With all the various HIEs focused on different aspects of capability-building and collaboration with stakeholder organizations, what are some of the throughlines that connect all that activity? If anyone should know the answer to that question, it is David Kendrick, MD, founder-chair of the Department of Medical Informatics at the University of Oklahoma and the founder and former CEO of the MyHealth Access Network in Oklahoma, as well as a still-practicing internal medicine physician and pediatrician. Kendrick is a member of the board of directors of NCQA and worked as a senior counsel for interoperability at ONC from 2013 to 2016.
And ultimately, one of the key elements of enablement, he says, is that of the patient-centered data home, a data-world parallel to the concept of the patient-centered medical home. “Back in 2015,” Kendrick says, “HIE leaders met to try to figure out how to work together; I was working at ONC at the time, and I proposed a patient-centered data home, and the idea of repatriating data.
"So, for example, if one of my patients is on vacation in California and ends up in the ER, the network says, ‘Oh, this is an Oklahoma patient, and they send the data back to us.’ And it turns out that’s better than any other model out there. It’s a trust agreement” among all the entities involved nationwide, he says.
And Kendrick agrees when asked whether the idea of sending entire CCDs has become passé. “Under the patient-centered data home model, sending those big CCDs is not effective at the point of care," he says. "I don’t have time, and it’s dangerous because it shares the entire record — the Epic/Guard Dog breach involved CCD files, patients’ entire records. So those CCDs are useful in some limited cases, but not useful in general.”
Related to that, Kendrick reports that, “We have a FHIR API enablement in Oklahoma, and we went live with that in 2019. So there’s one FHIR API for every patient in Oklahoma. So that ER doctor in Little Rock can specifically say, 'I just want the latest blood test, cholesterol test, medications list, latest scan, without any oversharing taking place, and with immediacy.' And we’ve already resolved the misidentification problem in Oklahoma and Arkansas, for example.”
In all that, he says, “While policies and presidents were changing and payment models were changing, HIEs have continued to connect hospitals and doctors. That’s why I think they’ve earned the moniker of utility: transparency, place-based — in a place, serving a place.” Meanwhile, he notes, on a technical, level, “Every HIE is marching towards a single FHIR API for their services.”
And whether they like the term “health data utility,” many HIE leaders agree that HIEs individually and as a whole are moving forward with alacrity to make that concept a reality. They also are in firm agreement that the overall trajectory for HIEs — becoming deeply trusted stewards of data on behalf of entire communities and beyond — is their inevitable future.
As VITL’s Farmer puts it, “I just don’t know how we solve the big problems that healthcare has throughout our country without the assets that HIE represents. We’re not just about moving the data; we’re about making it actionable so that the cause of medicine can be advanced. That’s where we become health intelligence networks.
"We give folks a more seamless perspective on patients, across the continuum of care. All of those care entities work inside a very complex ecosystem. And HIEs — or health intelligence networks — bring intelligence to bear on the business of serving patients. There’s no one solution that will solve all the problems in healthcare; but HIEs are a very important piece of the picture.”
Mark Hagland is a Fellow of the American College of Health Data Management and a long-time writer and editor covering the health information technology industry.