Scott Mace, for HealthLeaders Media , October 28, 2014
A massive government procurement will speak volumes about the market share, and clout, of one lucky electronic health records system technology. The rest of the healthcare IT industry will be living with the ramifications of the Department of Defense’s decision for years to come.
In much of tech, market share plays an outsize role in determining which standards get adopted. The dawn of the Internet was never supposed to be based upon TCP/IP, but instead on OSI protocols, which blue-ribbon committees carefully crafted to be the best possible foundation for open networks. But then TCP/IP grabbed all the market share and there was nothing left for the OSI protocols to do but fade into history.
The same thing may be about to play out in electronic health records. While the Office of the National Coordinator reboots its various blue-ribbon committees, and struggles to adapt to losing its deputy, Jacob Reider MD, and the loan of its chief, Karen DeSalvo to the administration’s response to Ebola, a massive government procurement will speak volumes about the market share, and clout, of one lucky EHR technology.
That procurement goes by the acronym DHMSM, commonly pronounced “dim sum” and short for the DoD Healthcare Management System Modernization. Final bids for DHMSM are due this Friday, on Halloween. Nearly 10 million active-duty military personnel and their beneficiaries will be affected by this procurement, which will replace and migrate data out of the incumbent Armed Forces Health Longitudinal Technology Application (AHLTA).
Even though leading EHR vendors such as Epic and Cerner are competing for DHMSM, I did not initially think the DoD procurement would have much impact on the civilian market. But this isn’t the entire story. It turns out that about 60 % of the care provided to DoD beneficiaries is delivered in commercial hospitals, according to Jerry Hogge, deputy group president of Leidos Health Solutions Group, the prime bidder partnering with Cerner.
Enter Epic, Cerner
That kind of statistic is all you need to know about why Epic and Cerner, among others, are grasping for this win. One can easily imagine regions of the country where civilian provider will want to be aligned with the EHR at military facilities nearby. And there can be only one DHMSM winner.
But didn’t large, commercial EHR vendors get the U.S. into its current interoperability mess, requiring ONC to create a fix-it plan that will take 3 to 10 years, if successful?
Well, yes. It’s a funny thing. Talk to each of these vendors and they will boast how many standards they support – HL7 and its Consolidated CDA document format, SNOMED, LOINC, IHE, the list goes on and on. And the DoD procurement calls for support of these standards as well. But as we’ve long since learned, mere support of a set of standards does not assure interoperability.
Perhaps no more ridiculous example of just how ill-equipped today’s EHRs are as communication tools than the fact that Thomas Eric Duncan, the Ebola victim treated by Texas Health Presbyterian Hospital, accumulated a 1,400-page electronic medical record over the course of his treatment. All the standards in the world could not correct deficiencies of EHR technology which often presents too much information. Although the EHR (in this case, Epic, but it could have been any number of others) did its best not to present all 1,400 pages at once. Presenting the right information at the right time is tricky at best.
Committees are not the crucible of the breakthrough technology for which technology is best known. Standards alone did not create the iPhone. Skeptics fear, with some justification, that the ONC 10-year roadmap, or DoD’s procurement, expected to be awarded next year, will produce a camel – a horse designed by committee.
But is that fair? Camels are pretty useful animals, after all.
Memories of Healthcare.gov
Also hanging over the DoD procurement is the fresh memory of last fall’s bungled healthcare.gov rollout. Inside-the-beltway operatives managed to snag the contract while a disjointed team of federal agencies failed to coordinate their efforts, and we all saw the result.
In fact, the DoD tried and failed to jointly procure a new EHR system with the Department of Veterans Affairs, and despite spending $1 billion and five years of effort, it was utterly unsuccessful.
Considering the other troubles the VA’s scheduling system had, some critics can be excused for condemning the old-style DoD procurement and demand outright privatization of both DoD’s healthcare system and the VA.
But technology is just a tool, and so far, private industry hasn’t solved the interoperability problems healthcare faces. Not even close. (For the latest example, look at how Walgreen’s and Rite-Aid snubbed ApplePay so they could float their own proprietary phone payment standard.)
Moreover, amalgams of public and private initiatives bring the kind of creativity that built the Internet. Or have we forgotten the role the National Science Foundation played in doing that?
For that matter, there is a team bidding on a commercialized version of the VA’s VistA EHR as the potential DHMSM winner.
Among the features of the VistA-based entrant are “federal-grade security, military-grade cyber-threat detection and response,” says Dan Garrett, principal, health industries & leader of the HIT practice at PricewaterhouseCoopers, which is leading the VistA-powered bid.
Unbeknownst to many, open-source-based VistA is deployed at numerous civilian hospitals, and has its own industry consortium, OSEHRA, promoting innovation on top of existing software. General Dynamics is supplying IT expertise, and DSS, which has worked with the VA for years on enhancing VistA, has implemented a modern user interface on top of VistA.
Some industry analysts believe PwC’s bid has an edge, because a core requirement of DHMSM is to have seamless interoperability between the DoD EHR and VA’s implementation of VistA, as active-duty personnel retire and enter the VA system.
Is It Truly Open Source?
Then there is the issue of the open-source nature of VistA. Critics point out that open source is not the same as open standards. Open source is a licensing arrangement that permits customers to see and modify the source code of the software. Most software outside of Microsoft (and even some within) contains at its core some open source-derived code – even the iPhone. VistA has a lot more in common with Linux than it does with Apple, though, since companies such as Apple layer millions of lines of proprietary code on top of or aside open source.
I recently learned that every Epic customer has the ability to inspect Epic’s source code. But that is not an open source license, since those customers cannot modify the code in any way that can be shared without Epic’s express permission. They would be in big, big trouble if they tried.
Therefore, VistA proponents are banking on its open source licensing to permit a level of innovation simply not possible on top of Epic, Cerner or other proprietary EHRs.
There is also a problem when the work of agencies such as the VA only makes it partway into the standards world. The VA created an application called Blue Button, which makes it easy for VA patients to download their records. Unfortunately, ONC only championed the idea of patient access, not specific data formats so that the downloaded data could become a standardized platform upon which innovation could flower.
When ONC tried to champion a standardized Blue Button, known as Blue Button Plus, specifying the data formats, EHR vendors didn’t adopt it, probably because it would have helped commoditize their EHRs. DoD, for whatever reason, does not appear to be requesting Blue Button Plus, only the earlier Blue Button.
Therefore, the DoD has minimized the amount of clout it could wield in the interoperability space, and yet it still faces a difficult choice. Try to go with a market-share leader such as Epic or Cerner, and enjoy whatever interoperability that brings, or go with the VistA solution which trails in market share (outside of the VA that is) and unleash the innovation so badly needed to avoid the next 1,400-page Ebola care record fiasco.
Go with the proprietary solution, and face what could be a difficult and costly mission of seamlessly integrating with the VA’s VistA, or go with the open-source solution, and face the costly mission of integrating with the leading commercial EHRs in the private sector. Either way, the rest of the healthcare IT industry will be living with the ramifications of the DoD decision for years to come.