In this second installment of our series, Donna Fedor, Managing Director for Digital Health at Mavericks Capital, discusses the lingering confusion around HIEs with three industry-leading healthcare data and information experts.
This series intends to shed a bit of light on healthcare information sharing to help both the healthcare novice as well as those that reside inside the healthcare value chain but have largely ignored the decade-long muddled evolution of HIEs.
Three healthcare information experts have generously shared their thoughts about the progress of health data exchange in the United States. Enrique Estrada is Solutions Architect at Sandlot Solutions, a fourth generation health information exchange (HIE); Bob Yencha is the President of RTY, LLC, a consultancy focused on information standards development and data sharing programs; and Bill Beighe holds CIO positions at Santa Cruz Health Information Exchange (HIE) and Physicians Medical Group of Santa Cruz.
The Lingering Confusion
When asked about the lingering confusion around HIEs, one of our experts, Bob Yencha, got right to the heart of it:
“It is tough to be a middleman, and HIEs are the middleman. The thing about middlemen is that if you add no value, trading partners will start to cut you out of the transactions.”
“In other industries like consumer marketing or financial services, middlemen that host and analyze data can provide profitable analytical services. Data is a commodity in other domains; it’s the raw material for creating information in the age of Big Data. But, in the case of healthcare, the content is much more complicated semantically (and from an encoding perspective, syntactically) with a far more intricate weave of information models for core concepts of consent, security, privacy, validity, traceability, and other regulatory constraints. The initial service that HIEs provided was secure collection and access to healthcare data for a community where the value proposition was to share the heavy load of building infrastructure for moving data. As those basic functions become commodities, sustainability will depend on providing higher-level services, i.e., become information brokers vs. data brokers.”
Yencha believes that it is much more difficult to set up a viable business model based on mining and monetizing healthcare data.
“We have a wide mix of technologies to convey the same concept in different ways at each stage of Personal Health Information (PHI) data exchange. If you provide basic library and post office services, there’s less burden to deal with those differences. Providing EHR/EMR services such as clinical decision support, alerts, or any other processing of PHI where the results could engender risk to patients requires a deeper understanding of the semantic variance and accounting for it algorithmically. This is no small feat, and can engender additional costs for certifications, etc. Any promise of lower cost for better outcomes is going to be realized in other businesses, not the HIEs themselves even though they will be enablers for those savings. That’s a tough ecosystem to survive in.”
Enrique Estrada feels that HIEs have served a great need in the early stages as the interoperability broker between silo EMRs, connecting clinical assets across disparate locations. He stated:
“The difficulty with the evolution of HIEs has been that they have meant different things to different people. First starting as Community Health Information Networks (CHIN), they morphed into regional Health Information Organizations (RHIO) and then into the HIEs of today. But many of these organizations have not been able to show sustainable business models. And now with ACOs becoming more prevalent, HIEs will take on a different role of providing the core infrastructure for clinical interoperability. HIEs will act as the ingestion point, and will service the data up the value stack to ACOs, where analytics, reporting and predictive modeling will now take center stage. The new data models will align closer to the shift to value, as well as drive a new method to leverage data to focus on the triple aim of healthcare, which is better health, lower cost, and better outcomes.”
Interestingly, the Santa Cruz HIE was formed in 1995, well before federal funds and grants were available. Bill Beighe explains:
“The team’s aim was to set up a sustainable model from the start with $1.2M in seed funding and ongoing operational revenues from 25 individual and organizational stakeholders including Dominican Hospital (now part of Dignity Health), Unilab (now part of Quest Diagnostics), and Physicians Medical Group of Santa Cruz. Building a model that was sustainable through stakeholders and members has been critical to our success.”
A 2012 report, HIE 2.0 Clinical Network Management, from Chilmark Research revealed the inability for the majority of HIEs to make money and supports the statements by our experts.
Out of 322 organizations in the US identified as facilitating healthcare data exchange, 178 participated in the Chilmark study. Operational HIEs numbered at 119, which increased from 75 in 2010 and 55 in 2009. Many HIEs in 2012 had only been exchanging data for less than 2 years.
Astoundingly, a whopping 74% of HIEs in operation struggled with developing a sustainable business model, meaning, they would not survive without outside funding. Only 24% were able to balance operating costs with fees and revenues. All of the others were in the red.
According to Chilmark, the make-up and types of HIEs are quite diverse. As discussed, in some cases, states built their own and, in other cases, they fostered existing or new business entities.
Non-profits made up the majority (67%) of HIEs while 11% were government run, and only 9% were for-profit businesses. Funding came from three main areas: 52% from grants/contracts, 28% from participant fees and 12% from other sources. Another interesting note was that 65% of the HIEs received revenues from labs and hospitals but only 40% obtained revenues from payers.
Usage penetration of HIEs in 2012 was still low: only 10% of physician practices, 30% of hospitals (~1400) and 10% of ambulatory orgs (~23K) used HIEs.
So, why were there so few healthcare entities actively participating in HIEs?
Estrada believes that lack of usage has to do with the fact that many Health Care Providers (HCPs) have been investing in Meaningful Use 1 (MU1) implementation to qualify for large government incentives. MU1 has focus on EMR (electronic medical records) and requires minimal interoperability/data share components. But things are changing. Currently, Beighe reports that 70% or more of physicians in Santa Cruz are now part of the local HIE.
Notably, a subsequent 2013 survey of HIE entities from the Robert Wood Johnson Foundation determined that three out of four HIEs were still struggling to develop a sustainable business model for their organization.
Beighe believes that the HIEs needed to “act as a start-up but have been overly burdened by federal and state bureaucracy” as well as the fact that “it takes at least 5-10 years to establish an HIE and many of them just ran out of runway when federal funding ended in 2014.”
Interestingly, one of the reasons sited for the demise of the Nevada state-run HIE in a November 2014 Healthcare Informatics article was this exact reason, starting too late and leaving too little runway to get the HIE operational.
In California, according to Beighe, the California HHS encouraged the formation of a non-profit for governance of HIEs in California. The California Association of Health Information Exchanges (CAHIE), is almost entirely member funded. Another interesting note is that only one of the HIEs funded with HITECH funds in California closed its doors.
Estrada also suggests,
"There has been a general 'wait and see' approach from many healthcare providers on which sustainable HIE business models will succeed. But, now, we are starting to see uptake as ACOs become more prevalent and since Meaningful Use 2 (MU2) incentives will require extensive clinical exchanges and interoperability.”
He believes that the sustainability model is front and center, but it is also about the adoption curve.
“Many HIEs have sold solutions but there is still lack of adoption and implementation of the solutions that they have already sold into the marketplace.”
Yencha expressed his views about the many social, technical, legal, and policy issues that affect and inhibit shared data exchange.
“First, many disparate entities in the healthcare value chain must come together, which is quite difficult. Next is the imposition on doctors to implement more challenging workflow processes. Finally, there exists a layer of complex legal and regulatory policies sitting on top of an already complex proposition that doesn’t adequately address the digital realities of coded data.”
Yencha also shared some insightful comments about unintended consequences when living in a democratic society:
“Our many freedoms also result in many privacy fears. The fear of big brother is ingrained in many US citizens adding to the complexity and costs of our regulations and implementations.”
Yencha describes health data exchange as a “wicked problem” in that the healthcare system is so huge, complex and interconnected that it is hard to stand in a single place and understand the full consequences of an action elsewhere up and down the line.
He compares it to a bump in wall-to-wall carpeting—when you push the bump down in one spot, the bump will later appear somewhere else. There are so many connected moving parts that fixing an issue in one place will often result in a new issue bumping up somewhere else.
Stay tuned for the next segment in our series, where we'll dive into the some of the high level technical aspects of HIEs.
The nuviun industry network is intended to contribute to discussion and stimulate debate on important issues in global digital health. The views are solely those of the author.