Healthcare Analytics, Population Health Management, Healthcare Big Data

Quality & Governance News

Epic, Cerner, DirectTrust, and the Quest for EHR Interoperability

By Jennifer Bresnick

- For a concept that is supposed to bring healthcare organizations together, EHR interoperability is certainly one of the most divisive topics in the industry.  With hundreds of different opinions about how to achieve it, where to start, what standards to use, why it matters, who should lead the effort, and how to enforce it, it’s no wonder that Congress has recently stepped in to try to figure out what is going wrong and how to fix it.

EHR interoperability and Epic Systems

Central to these arguments are a few big names, instantly recognizable as friend or foe – depending on your position.  Epic Systems, Cerner Corporation, CommonWell, CareQuality, the Sequoia Project, and DirectTrust have all been making waves in the EHR interoperability space for the past few years, gathering allies and carefully sizing up rivals as they compete for market share, provider loyalty, and the financial support they need to grow to scale.

After several recent Senate hearings prompted a bitter round of finger-pointing and social media sniping, onlookers could be forgiven for thinking that the EHR vendor landscape has coalesced around two major camps: the Epic loyalists who tout the virtues of Judy Faulker’s effective yet expensive monoculture, and the Cerner advocates who applaud the company’s customizable offerings and public commitment to health data sharing.

This shadow battle came to a head in late July, when the Department of Defense snubbed Epic’s contract bid and instead tapped the Leidos Partnership for Defense Health to replace its aging health IT infrastructure, giving Cerner a chance to shine on a $9 billion federal stage.

Epic’s silence on the subject is par for the course, but not every healthcare stakeholder is keeping their opinions to themselves.  EHR interoperability is too important to providers who are trying their best to enter the complex world of population health management and care coordination – and there’s much, much more to the question of health data exchange than watching two companies duke it out in the arena of public opinion.

READ MORE: What Healthcare IT Users Don’t Really Need to Know about FHIR

For one thing, Epic and Cerner aren’t the only players on the field, and they are certainly not the only organizations with skin in the interoperability game.  Epic’s corporate representatives may have had their say in front of the Senate HELP Committee earlier this year, but so did DirectTrust President and CEO David C. Kibbe, MD, MBA, who is also a Senior Advisor for the American Academy of Family Physicians.

Kibbe told the Senate Committee on Health, Education, Labor, and Pensions that EHR interoperability is a much bigger question than EHR vendors can answer.  Healthcare providers, regulators, and collaborative data standardization bodies must push product developers in the right direction, and DirectTrust has already played a major part in doing so.

Conceived in 2010 and brought under the wing of the Office of the National Coordinator in 2013, DirectTrust’s secure messaging protocols play a significant role in the EHR Incentive Programs and the nation’s health information exchange efforts.  With more than 40,000 health care organizations and nearly a million individual end-points capable of exchange, DirectTrust forms one of the most successful interoperability backbones the industry has yet seen, thanks to its neutral, national approach to HIE.

“The ONC was farsighted, at the time, in thinking that this trust framework needed to be a national effort, rather than a state-by-state effort,” Kibbe told in a one-on-one interview. “In 2012 and 2013, there were state entities that were creating their own quasi-accreditation programs for local trust communities. That wasn’t going to work, as health information service providers (HISPs) would have to be charged $5,000 to $25,000 in every state or region that they wanted Direct exchange to work in.  And that was really going to be a mess.”

Approaching interoperability from the state level would have cemented certain problems into the patchwork infrastructure, Kibbe explained.  Not only would the costs be prohibitive for participants, but the requirements would not be standardized across state lines.

READ MORE: Can Application Programming Interfaces Inspire a Better EHR?

“What was developing was a situation in which you’d have a set of rules for Direct exchange in Rhode Island, but they wouldn't be the same rules as was set up in California or Texas or Florida,” he said. “There were certainly people at ONC that thought that was the way to go, but I believe that Dr. Farzad Mostashari, National Coordinator at the time, convinced them that the trust network ought to be national.”

DirectTrust’s standards-based framework for secure data exchange has allowed providers to engage in a certain level of health information exchange with little extra infrastructure investment or training, but it isn’t the only initiative focused on blurring the lines between EHR vendors.  In the same year that DirectTrust received the ONC’s blessing, the CommonWell Health Alliance announced its intentions at the annual HIMSS Conference, and immediately divided an industry that was just learning about the value of cooperation for patient care.

Along with athenahealth, Greenway, RelayHealth, McKesson, and Allscripts, Cerner Corporation used CommonWell to declare a new direction in health data interoperability – and Epic, perhaps experiencing a little whiplash as the competition ganged up against its “walled garden” stance, struck the first negative blow.  

The ONC could do nothing but stand by and watch the drama unfold as the private partnership put its shoulder into developing interoperability services for volunteer clients, but Kibbe doesn’t believe that was necessarily detrimental, at least as far as the federal government and DirectTrust are concerned.  “There was some fear on the part of ONC that a group of vendors would be taking this over – which is pretty much what happened,” he acknowledged. 

“But on the other hand, I think it's been a very collaborative effort between private sector health IT firms and the federal agencies CMS and ONC,” he continued.  “There haven’t really been fights over control – the participants are very concerned about listening to collaborative viewpoints and developing what has become a robust public key infrastructure (PKI) that works well. 

READ MORE: EHR, Big Data Analytics Success Hinges on Holistic Change Management

“Of course, it is also the case that ONC still requires EHRs to be capable of doing Direct. So it is really a big success to have a national set of policies supporting Direct exchange in which over 40,000 health care organizations participate, considering the fact that there was such inherent bias against interoperability within the healthcare industry when we started.”

That bias has dissipated quickly, due in part to the cooperation between the growing circle of EHR vendors committed to interoperability and DirectTrust.  “Cerner was very much a leader in the DirectTrust effort to established scalable trust, along with companies like Allscripts, Walgreens, athenahealth, and Greenway,” Kibbe says.

“There was a significantly large group of companies that made the decision that, at least with respect to this sort of simple, push-based kind of interoperability, it was better to do it as an industry collaborative effort and base it on open standards, as opposed to some centralized or proprietary scheme.”

“And I think that that's one of the reasons why Cerner may have gotten the DOD contract,” he added. “They didn't just espouse an open standards-based approach to HIE across vendor lines.  They actually went ahead and did it. So there's now this very large pool of endpoints in the private sector that are reachable via Direct exchange.  That’s valuable because it’s such a large network.”

“That's one of the things that I think all the federal agencies are now aware of. We have the Indian Health Service, the United States Postal Service, and the Veterans Administration all seeking to become HISPs accredited by EHNAC and DirectTrust and to join the community of Direct service providers.  We hope and expect CMS will join that group in early 2016.”

Adding the Department of Defense to the list of large federal organizations integrated into the Direct ecosystem would be a triumph for all involved, but the DOD has taken a step back on the notion during its long search for the right EHR technology to bolster its crumbling infrastructure.

“DOD was very interested until about nine months ago, and then they decided that they had to put their effort into this procurement,” Kibbe said. “But seeing as they chose Cerner, I would imagine that Direct will become part of their toolset going forward. I don't know that for a fact, but I think as things work themselves out, that will probably be the case, because that's the way they appeared to be thinking before they went silent on the issue of Direct.”

“There are around a million DirectTrust addresses out there right now, and Cerner's roughly 1000 hospitals and associated medical practices in the United States account for about a quarter of them. So Cerner's been very aggressive, particularly with its big clients, in making Direct exchange possible for those organizations attesting to Meaningful Use and those who are going beyond that.”

Epic has been aggressive in a slightly different way, mostly by raising its hackles and growling at the competition.  “Epic's approach has been to tolerate Direct as a competitor to their own homegrown proprietary system, which is known as EpicEverywhere,” explained Kibbe.  “Epic's customers do engage in a lot of Direct exchange - it’s not as though Epic is not participating or allowing their customers to communicate via Direct, particularly when those customers aren't in a market where Epic has got everything locked up.  There are actually very few of those markets.”

“But many of the Epic customers I talk to say, ‘Well, we love EpicEverywhere, but it just doesn't get us everywhere we need to go.  We want to connect to that 40 percent or 50 percent of providers using other electronic health records, other than Epic; we want the nursing homes and the facilities that are non-users of EHRs, but still part of a care coordination network, to be able to communicate with us and our care coordinators.’”

Epic’s reluctance to fully embrace bi-directional data exchange has caused consternation among its customers and outside observers alike.  The ONC’s April report on data blocking pointedly declined to mention any names, but it may well have been throwing furtive glances in Epic’s general direction, along with the rest of the industry.

“It’s really hard to talk about information blocking, because you never really know the other party’s intentions,” Kibbe said. “But I think for Epic, their focus on Direct was centered on getting their customers to be able to attest to Stage 2 Meaningful Use via Direct as early as possible, and they’ve done that very well.”

“The problem is that when they did that, they focused on the messages going out of their system, and didn’t appear to have an equivalent interest in receiving messages from somebody else, the inbound message side of the exchange,” he continued. “In fact, in some reported cases, the Epic system refused to accept inbound messages from trusted parties. And that angered some people who wanted Epic customers to be recipients of their Direct messages in order to meet their own Meaningful Use transitions of care objectives.”

“That's really what raised the question of information blocking, in my experience,” he stated.  “Some non-Epic users perceived that Epic's customers were getting their numerator for their transitions of care requirement by using other providers as endpoints, but then turnaround wasn't fairly executed.”

A less-than-wholly-committed attitude towards EHR interoperability, added to periodic reports of sticker-shocked hospitals sinking into dire financial straits after a failed implementation of Epic software, has left Epic Systems looking a lot like the bad guy. 

Their disinclination to speak to the media on a regular basis has left outsiders wondering what exactly is going on in Verona, Wisconsin.  Most of the industry can only wait until the company makes one of its rare public statements, and those don’t always go quite as well as Epic executives might hope.

“Epic hasn’t necessarily helped themselves when it comes to their public image,” Kibbe observed. “During the first Senate HELP Committee hearing in March, Epic’s spokesperson Peter DeVault had a perfect opportunity to get up in front of the senators and say, ‘You know, there is an interoperability problem in this country. We're part of the problem, but we're also part of the solution. We're making it work.’ But for some reason he didn't do that.  He basically just said, ‘There isn't any problem. We have got everything covered.’  And then he went on to criticize the competition.”

“I don't understand that behavior. But I think they have a culture that's very strong, and if you talk to people at Epic, they believe that they are the only electronic health record in the United States.  It's one of the reasons they've been so successful as a company.”

“Epic has been very good at creating this image of themselves as a giant, because their business has been built on going after large academic medical centers,” he continued.  “They’ve been extraordinarily successful with it, but they still only account for a relatively small percentage of the actual total number of hospitals in this country.  Cerner and MEDITECH both have larger numbers of hospital customers than does Epic.”

“So despite this perception of Epic having a lock on so many marketplaces, nine out of ten communities still operate in a multi-vendor environment, where you have Epic and Cerner and McKesson and MEDITECH and 40 different ambulatory EHRs. That’s the norm for most people and most medical communities.  The market is likely going to continue to consolidate to some degree, but Epic isn’t going to end up with the majority of providers out there. The numbers are against them, and they're too expensive.”

“Epic hasn’t quite gotten the memo that their operating environment is changing.  They’re not going to succeed if they appear to be prohibiting information exchange from occurring with their competitors if that's deleterious to the interests of patients.”

Time and again, rule makers and stakeholders have tried to force the issue of EHR interoperability back to those “patients first” roots, but a half-baked business case for a truly seamless continuum of care has left many healthcare organizations wondering how they can meet the complex needs of patients if they can’t make enough cash to keep the lights on. 

With the majority of organizations still receiving most of their revenue from fee-for-service reimbursement models, developing a robust health IT infrastructure that allows for large-scale data sharing has not always been top priority.  Keeping patients and their spending in-house – and using high-volume, high-expense services to keep the coffers full – is how hospitals and providers have traditionally turned a profit.

“The growth of the healthcare industry has largely occurred because it has not been a requirement to produce value,” Kibbe said. “It has been a requirement to produce quantity. And I think that's almost run to the end of its rope. If you're a provider organization, you're now at risk for the excess spending that you do above the revenues that you collect.  It’s really that simple. It's all about risk. And if you don't manage that risk, you're going to do poorly.”

The EHR Incentive Programs hoped to spark a value-based revolution, and started the slow burn of accountable care by allotting billions of dollars to providers who wanted to get in on the ground floor of the EHR adoption game.  But tight deadlines, immature technologies, and slow certification processes have left many well-meaning organizations with practically unusable EHR products.

“It’s a very complicated political, technical, and cultural decision that these organizations make with respect to the choice of an EHR vendor,” Kibbe said. “At least for the last ten years, it has traditionally been a very, very big expense. There have been strong biases within the health care sector to buy enterprise systems as opposed to cloud-based systems, for example.”

“I think they have been obsessed with controlling the information at the enterprise level. And it's not always because the enterprises don't want the information to get out for any competitive reasons, although that certainly remains a concern. Many of them have been obsessed with HIPAA - with very good reason - because of the consequences that can occur if information is released improperly.  Keeping that information in-house sometimes seems like the safest and wisest plan.”

Over the next few years, as Stage 3 Meaningful Use turns the screws on interoperability and CMS gets serious about its value-based reimbursement programs, hospitals will no longer be able to bank on volume as a way to make up its numbers and please their board of investors.

“It isn’t as easy to shift charges away from the unprofitable insurance entities to the more profitable ones, and they can't play the same games with Medicare that they could play five years ago, because Medicare is really looking for signs of abuse and fraud on a whole new level,” Kibbe said.

“And the public is much more highly aware of what’s going on with healthcare costs,” he pointed out.
“Patients are bearing much more of the costs of their care, even when they’re well insured.  It’s increasingly common to have a $3000 or $6000 deductible.  I don't know anybody today who doesn't think about the cost to them when they seek healthcare.”

“All these things mean that healthcare has to figure out how to coordinate care more effectively.  When the sources of profit become sources of cost, they have to be controlled.  That requires collaboration.  Hospitals have to collaborate with primary care doctors, long-term care facilities, and post-acute care providers. They have to collaborate, in some cases, with the patients themselves in order to manage high-risk populations.”

“When everything is electronic, you can’t collaborate without health data interoperability.  If you don't have ways of communicating, everything gets really expensive really fast.”

DirectTrust has tried to tamp down the potential for ballooning data exchange costs by providing a quick and easy way to send and receive basic health information, but EHR vendors have been not been holding up their end of the bargain, according to Kibbe.

“The large enterprise EHR vendors have done a very bad job of helping their customers to collaborate across this idea of a medical home or a medical home neighborhood,” he says. “And the reason they've not done very well is that they’ve had opposing demands. Providers want to keep their information at the enterprise level - they don’t want it to get out.  But oh, now they need to be able to talk to their partners, right?  You can’t do both.”

So that conflict and tension is at the root of where Cerner and athenahealth and Greenway and many leading EHR companies have split with Epic. I haven’t done any formal surveys or anything, but it seems to me that most vendors now operate under the assumption that they're going to be serving customers in multi-vendor environmentsforever, whereas Epic seems gives the impression that they’re increasingly going to be the only system anyone has to worry about.”

In the meantime, as Epic attempts to achieve its goals of EHR dominance, patients are being left in the lurch.  Despite the industry’s lip service to population health management and care coordination, they can rarely provide sufficient help to individual patients navigating the complicated and confusing network of healthcare providers, each with their own technology, their own processes, their own limitations, and their own notions of health information management.

Administrative staff still wait patiently by fax machines; patients are still told that snail-mailed documents are not available in the provider’s computer system.  Physicians are scheduled for patients they may not be able to help, and payers scold everyone involved for wasting time and money.

Primary care providers may be increasingly embracing the patient-centered medical home and other care coordination frameworks that use dedicated navigators to guide bewildered clients through their maze of options, but much of the health data management that supports coordinated care must be conducted by the providers themselves.  To do that, they must put aside their vendor-based differences and keep the patient at the center of the problem at hand.

“If the situation gets to the point where it has become that the patient's responsible entirely for their own coordination –  they see thirteen different doctors; each doctor is prescribing something different, or maybe the same thing, but not coordinating with other doctors – then it becomes very inefficient and everybody suffers,” said Kibbe. “So much goes wrong. Everybody's got a story about poor care coordination, don’t they? We all know that this is not rocket science, but the ball's being fumbled almost every time we get transferred from one care setting to another.”

“Part of the reason why it seems so crazy is that now we've got these electronic health records, but they don’t seem to make any of this better. In some cases it makes everything worse, because you can transfer the wrong information faster than you can get the information right.”

“But the reason that I believe Direct is going to be useful for solving these problems is that it’s simple,” he stated.  “Everybody knows how to use e-mail.  Everybody understands that there's some other person at the end of the address. It's bi-directional.  It’s not like a fax, where you send off this piece of paper and you don’t know who is going to read it or when it’s going to get there.”

“Many of these problems that we have are solvable with just barely minimal, but sufficient, communication. So I think this dual demand for collaboration and interoperability are going to continue to grow within every healthcare system that sees itself at increasing risk of not taking care of patients well and coordinating their care well.”


Join 25,000 of your peers

Register for free to get access to all our articles, webcasts, white papers and exclusive interviews.

Our privacy policy

no, thanks

Continue to site...