Congress Can’t Solve the EHR Interoperability Problem

Niam YaraghiRep. Mike Burgess (R-Texas) has released a draft bill entitled “ensuring interoperability of qualified electronic health records” in which interoperable (Electronic Health Records) EHRs are defined as those that do not block sending and receiving data to and from other EHRs and provide users with complete access to the captured medical data. The draft bill proposes that detailed methods to assess interoperability be defined by a “Charter Organization.” According to the draft bill, this Charter Organization shall consist of one member from each of the standard development organizations accredited by the American National Standards Institute and representatives that include healthcare providers, EHR vendors, and health insurers. To keep its certification after January 2018, an EHR vendor should comply with the definitions of the Charter Organization, publish API’s to enable data exchange with other EHRs and attest and demonstrate that it has not willfully interrupted data exchange with other EHRs. The draft bill suggests that the Inspector General of HHS shall have the authority to investigate both EHR vendors and medical providers with regards to claims that they have interrupted interoperability.

The proposed Charter Organization will not be successful.

The language in the draft bill abandons the traditional approach of focusing on standards and instead is pushing for establishing methods and measures for assessing the level of interoperability in EHRs. However, it is not clear if the proposed charter organization would be able to come-up with such measures. As I have discussed before, some EHR vendors, along with many other medical providers are reluctant to become interoperable and exchange medical information.

It is unclear how vendors and medical providers with diverse interests can work together and come-up with actionable measures of interoperability.

Decertification is not good policy

It is exciting that Congress has finally acknowledged that some medical providers do not want to share their patients records even if there are no technical barriers to interoperability. According to the draft bill, if it is determined that a medical provider has willingly refused to exchange health information with other medical providers, it should be subject to penalties under provisions of sections 11281128A, and 1128B of the Social Security Act. While these penalties will probably prohibit the medical providers to intentionally withhold data, they may not apply to the EHR vendors. The threat of decertification is a bluff. ONC cannot decertify an EHR vendor that has more than 50 percent of the market share. In the best case scenario, after Congressional pressure, such vendors may enable data exchange, but will demand very high fees to overcome a plethora of technical barriers, especially if the EHR vendor has a monopoly in the market.

This is a complicated situation which I believe cannot be resolved through regulation. A better solution is to stop funding HITECH’s meaningful use incentives and avoid negotiating with the EHR vendors over the exchange fees. Without HITECH incentives, the market will force health IT vendors to develop sustainable revenue stream through reasonable exchange fees negotiated with the medical providers.

Niam Yarhagi is a research fellow at the Brookings Institute’s Center for Technology Innovation.

31 replies »

  1. Greetings,

    We have created a platform which solves your problem. The records of the patient gets automatically uploaded by the partnered Health Care Service Providers under one log in ID.
    So, you don’t have to make phone calls or you don’t have to ask for faxes from the patients. You can view those documents if the patient wants you to view the documents since all the records will be by default locked and can only be visible if the patent unlocks it.

    The charges for this is also very reasonable which is approximately $2 p.a.

    As an introductory offer, we are not charging anything as of now.

    Please register at http://www.medicalui.com and review it.

  2. Hello all. I’m a 4th year medical student wanting to help solve the interoperability problem. Although it’s not a perfect fix, I’ve developed a software called FreeMyEHR that helps bridge the gap between different EHR brands. Please check out my website for more info. FreeMyEHR.com

  3. David,

    What if your patient arrived with all his/her records , and the application to manage them, on a flash drive which enabled you with two or three clicks to electronically search for and access specific records including complete progress notes, lab reports, etc.? And what if it also enabled you, with your patient’s permission, to download specific records from your patient’s other providers to your EMR system? Finally, what if it paid you to upload copies of your notes, etc., so your patient’s device was always up to date?

    Sounds to me like this would save you all kinds of time getting records and consents — and it would increase your income to boot!

  4. I presume most people engaged in this discussion talk about interoperability in the context of sharing data between institutions to better coordinate care and avoid repeating tests. This is primarily a non-technical problem. If I request medical records for a patient, this costs me two phone calls, two faxes, and probably two hours. This makes it impractical given the patient volume, and it’s much easier to order a repeat test.

    The paper records I get in the end of the arduous process are a nuisance but are sufficient to take care of the patient. Having all the outside lab values and notes import to the correct view of my EMR would be nice, but is really the cherry on top. I can imagine a scenario in which all EMRs have beautiful APIs using standardized vocabularies, and I still can’t get approval for the outside records. The true solution would be to incentivize outside hospital to get me the records more easily.

  5. Bethany – you can lead a horse to water….you cannot regulate behavior. Until the price of being a data island exceeds the cost of connecting, barriers will remain.

  6. Without interoperability, we can’t achieve the entire purpose of going digital with health care data. It’s not just between EHR vendors, either. Physicians, practices, and hospitals are using a multitude of available technology to not only help care for the patient, but to try to increase their productivity/organization/efficiency for the administrative side. The programs used to submit claims need to be able to link to EHRs without causing headache, mobile charge capture applications should be able to link, and there’s talk of whether or not wearables prescribed by physicians should upload latest data so the patient doesn’t keep having to call in or go in. I appreciate the sentiment of trying to regulate and force EHRs to comply, but since when does forceable regulation work as planned?

  7. Agree John – we’ve hit that point where “natural market forces” may be the last resort. Continuing to throw money at a problem the market participants do not uniformly want to solve is a zero sum game.

    Data sharing will occur when the “business” is at risk because it is disconnected or a data island. This will occur not as consumers embrace “interoperability”, or support providers that use interoperable software but as the management of their health becomes “convenient”. This will take the form of mobile apps, alternate points of care, financial incentives – simple easy to engage apps and services that are consumer friendly and integrate their health data on the fly.

    Pick your analogy – credit cards started at a retailer level. An opportunity to attract more customers to a business based on credit. The retailer with more locations had a distinct benefit – until national cards arrived on the scene.

    More recently – Uber. While all healthcare is local, the taxi industry can also be considered “local”. How could a service that transports its customers by car where the range of transport is optimized at 20 or 30 miles, possibly be transformed by an international service?

    As you accurately point out, healthcare is a business. Only when that business is threatened because their competitor is connected to a larger eco-system and begins to provide a better customer service/is easier to do business with, will all of the “natural forces” come cascading down and the scramble will be to connect.

  8. Paul,

    Bravo! Your diagnosis of what ails Interoperability is spot on.

    Our primary, most important priority should be to give doctors access to their patients’ complete records at the point of care so they can deliver better, lower-cost care. Period. We can accomplish this today if we want to. All it will take is for ONC and HHS to abandon their Meaningful Use requirements which, though well intentioned, have made it impossible to achieve interoperability.

    This simple act would permit us to look for and adopt simpler solutions to achieving interoperability and would free hospital CIOs and execs to try innovative solutions. If you have recently spoken with a hospital CIO or administrator you understand how focused they are on meeting MU requirements. That is their job and that is what they are trying to do — even when they know it won’t work!

  9. Yes, the healthcare market has some dysfunctional characteristics to it that will hinder its ability to be truly market driven. But then again, do we really want healthcare to be a market-driven sector? That will raise a whole host of other issues, some not too pleasant to think about.

    My main point is: Prescriptive guidelines won’t work. What has proven to work are market-based incentives – this is what truly drives innovation. If we can move to a PHM model of care with associated incentives, there will be a need for interop, deeper clinician and patient engagement, etc. Healthcare orgs will start demanding such functionality from their vendors or simply take their business elsewhere.

  10. “Economics is not physics. The “natural” forces that drive it are forces we (“marketplace actors” – ugh, the terminology) bring to bear, not merely with dollar votes but with insistence on what behavioral standards will “well order” the market under consideration.”

    Thank you.

  11. The proper customers of a medical product like electronic health records should be doctors and patients! In the current system, Epic and its competitors serve two customer-masters:

    1. The technocrats and social engineers who devised Hitech and ACA provisions. Their objective is a utopian system that allows monitoring and control of doctors and patients for population management and cost control. Many see this as a threatening dystopian world of surveillance and coercion (some combination of Orwell’s Big Brother and Kurbrick’s HAL).

    2. Hospital system administrators who are fearful of falling out of compliance with federal regulations and want to take advantage of per doctor incentives for buying systems and avoiding penalties…..and while they are at it try to lock in their market share by rolling up independent doctor practices to ensure intra system revenue by capturing referrals.

    I like Bill McCann’s introduction of the concept of market structure…but I think he has drawn the wrong conclusion. He thinks if deregulation were to occur that there would be consolidation. The problem is that right now we have the huge dominance of Epic….and the morass of Hitech and CMS and ACA rules and regulations act as a barrier to entry for new innovators in health care IT. In addition, the hospital system administrators who make decisions are understandably risk averse and prefer to stay out of trouble by going along with the pack…..where Epic dominates. Removal of mandates would reduce the investment in health care IT in its current form…..and would enable firms to shift their focus to health care IT that serves the needs of doctors and patients. Investment capital would drain from the Epics and flow to the real innovators who figure out how to help doctors and patients.

  12. The problem, John, is you are staring straight at market dysfunction, and deciding that “natural” market forces are the best solution for it.

    Economics is not physics. The “natural” forces that drive it are forces we (“marketplace actors” – ugh, the terminology) bring to bear, not merely with dollar votes but with insistence on what behavioral standards will “well order” the market under consideration.

  13. It always strikes me as a bit odd that everyone throws stones at the vendors, but few providers look closely at their own practices of sharing data with fellow providers.

    Both are to blame.

    Providers want to retain their customers/patients and the data therein as there is competitive value in it. Vendors have a vested interest in locking-in customers and nothing does a better job than that in software than locking up data.

    And I disagree with Yarhagi. stopping HITECH payments is like trying to saddle a horse after it left its stall an he ago. The train has already left the station.

    The path to true interoperability is when customers demand that their vendors provide it, or they will take their business elsewhere. Accelerating the migration from volume to value based models of reimbursement will force the issue, not another set of regulations, another standards review process, another certifying body, or canceling the HITECH act.

    The only thing that does work, and has worked in other industries are natural market forces. For let us not forget, at the end of the day, healthcare is a business.

  14. The way to achieve interoperability is to a avoid the problem and ignore EMR vendors. Give the patient control of her records and the ability to deliver them to any provider she sees anytime, anywhere. Couldn’t be simpler!

  15. I am not sure what you mean, unless you are defining “transparency” in some other way than I am.

    Your bank account, for instance, is private. When you do a bank transaction (say, an interbank transfer of funds), though, the information shows up at the other bank. Imagine what our banking system would be like if the banking software vendors had convinced each bank that it would be best if their software couldn’t talk to the software of other banks?

  16. There are always going to be unintended consequences. If you eliminate the HITECH incentives, you’ll likely trigger a wave of consolidation among EHR vendors which could impose not the best solution, but the most-profitable.

    While we may agree that less government interference can be a good thing, history has shown–from telecommunications to energy to even beverage manufacturers–that unregulated industries go through a Consolidation Curve that results in lower competition. I want to share a terrific article in the Harvard Business Review about this phenomenon (https://hbr.org/2002/12/the-consolidation-curve).

    So, while many industries benefit from consolidation, we have to wonder whether this makes sense for healthcare.

  17. Let me help you. If a patient had an echo or a cath (anything really) at our competitor network, I don’t have access to those records. Well, I do if I can wait for quite a while. In an emergency setting, even an urgent setting, I effectively have zero access.


  18. William, I’m not talking about security. Almost 100% of security software is already open source. That ship sailed years ago.

    I’m talking about the clinical functions of EHRs: decision support, quality measures, summarization. Those things are Medicine and I can’t think of any part of Medicine that used to be secret before EHRs came on the scene.

  19. Excellent argument, Adrian. I have a fear, however, that the privacy/security needs are going to outweigh the interoperability dictum and that hackers would tell us that universal EHR formats and software standards are going to make it easier for them to breech our PHI data. In other words, maybe it IS better to have a lot of software that doesn’t mix very well.

    Also, open software sounds nice, but take a look at distrowatch.com a website discussing all the linux versions. I have tried many of these and, although free, they are not often compatible. There are hundreds of programmers going off in many different directions, all trying to be a next Ubuntu or Mint.

    So, even if you finally get all the EHR software into your “open” category, this leaves a lot of room for non-interoperability.

    The only place where this has been forced to work by the government, I believe, is in air transportation, maybe sea too, where common standards mean life and death. And the simplicity here–I think they are mainly trying to get radio frequencies agreed upon–is far less than trying to agree on how to display a bunch of serum chemistries or agreeing on what a creatinine of 1.3 means in different ages and different genders and weights. I mean, one could have a committee looking at this for a year and still not agreeing.

  20. I agree, Paul. But, as Joe says above, physicians should be able to advertise that they do care about interoperability and that they don’t use secret software. I think patients will understand that.

  21. I think the vast majority of patients don’t care about interoperability and don’t care about electronic access to their records. The EHR incentives and penalties have little to do with most patients’ concerns…..and in aggregate to date have detracted from medical care quality….not enhanced it.

  22. Everybody in healthcare claims that they are all about the patient.

    Interoperability is one of those fault lines that puts the lie to that claim. You are either truly building for real patient data transparency across all platforms, or you are not, building systems that only communicate with each other, and even that in a limited way, and in even more limited ways with other platforms — and then blaming the other vendors for the lack of real transparency. It’s one or the other. One of the two positions is clearly better for the patients and the actual clinicians working with patients, and the other is clearly worse, more expensive, and dangerous, actually injuring and killing patients.

    Many major providers and vendors, especially Epic, are choosing Door #2, in a morally obtuse business strategy decision that stinks up the joint something fierce. They often claim technical difficulty, but the claims are false. It’s very hard building data empires of such complexity, but transparency is not the hard part. Providers are doing it in hopes of so monopolizing their area that they can avoid any real competition and keep their prices high.

    Yet the incentives of the vendors and that providers are drifting apart. To the extent that we are really moving “from volume to value,” providers are increasingly finding themselves in the business of population health management. In that business, it’s increasingly hard to separate out “our customers” from “somebody else’s customers.” To actually run a successful business model based on improving population health, you either need to completely monopolize the region, or you need to be able to transparently communicate patient data across the region. Many of the organizations building accountable care organizations and other networks across regions are finding that they are data blind; they cannot coordinate care across the network, cannot eliminate duplication, cannot drill down and find help for the super-users, because the datasets will not talk to each other. The market inefficiencies are built into the software.

    How to unravel this now? Yarhagi is correct that threats of decertifying a market dominator ring hollow, and threats of docking CMS reimbursements by a small percentage years in the future are too small an incentive. It’s way past time for the federal government to set standard definitions of transparency, and say that no federal money incentive goes to build or upgrade any system that does not meet those standards. The industry has managed to game the standards, so that they don’t do the job. This is the job that Congress could do.

  23. PS: Instead of more regulations, Congress could also fix the problem by eliminating the HIPAA Treatment, Payment, and Operations exclusion. That would force hospitals and payers to seek authorization from the patient in order to communicate and it would go a long way to making patients and physicians part of the value chain for interoperability.

  24. EHR interoperability is the latest holy grail of heath information technology. Once we achieve true universal EHR interoperability, the cost of health care will plummet and all US patients will receive the highest quality medical care possible.

    Since the introduction of the electronic medical records in the early 1990s, health information technology proponents of always said that health information technology will lead to better quality care and lower the cost of medical care. At each step along the way, when this did not happen, I they moved the goalpost and said “oh, we only need to do one more thing” to make the US healthcare system the best/lowest cost in the world. And when that “one more thing” did not achieve the desired and, they created a new goal post. First it was EMR’s, then it was CCHIT certification, then it was Meaningful Use 1, then it was Meaningful Use 2, then Meaningful Use 3, then ICD10 and now it is universal interoperability

    As a result of this process, the EMR vendors control a large segment of all medical information in the United States and they are in control of the future direction of health information technology which ensures their own longevity and guarantees their long term fiduciary interests.

    Whereas the military-industrial-establishment controlled Congress and Federal spending in their realm, we now have the EMR-HIT-insurance company-pharmaceutical industry in control of our healthcare system, which represents nearly 18% of the GDP

    The changing economics of health care has forced physicians to become employees of large healthcare systems. As a result, physicians have been relegated to vendors who are required to use HIT tools which hinder their ability to take care of their patients and patients are nowhere to be seen in the control of the healthcare system.

    Everybody involved in perpetuating this sham should be ashamed of themselves. Clearly, health IT is not the solution to the cost/quality problem. To continue down this road, in the mistaken belief that more HIT will fix our healthcare system, will only ensure a delay in actually fixing our healthcare system and solidify the longterm control of our health care system by the EMR-HIT-insurance company-pharmaceutical industr.

  25. The interoperability problem is due to a market failure because neither patients or even physicians are the customers of the EHR vendors. The customers of the EHR vendors are the hospitals or integrated delivery networks. Because neither hospitals or IDNs can prescribe, their strategy is to lock-in licensed clinicians who can prescribe. The patients are just along as feedstock to the entire process. The payers seem willing go along with this hospital / IDN lock-in strategy because they have not yet figured out how to get to the patients and doctors directly. This lock-in reality is responsible for the interoperability problem.

    The market can only work if patients can choose providers on the basis of respect for patient access over our own data. That law is already on the books as the HIPAA Patient Right of Access. It’s just not enforced.

    If our regulators would simply enforce the patient right of access and the accounting for disclosures to patients, things we expect in finance and telecom already, then more physicians would start working for the patient instead of for the hospital. Imagine that. A relatively small percentage of physicians and payers would need to change over to a low cost concierge model to make the vendor lock-in business unattractive for hospitals.

    Maybe we do need Congress to make new laws to enforce the patient’s right of access. Maybe we need some payers to simply give away open source EHRs to physicians that are willing to work for the patient. Open source EHRs would also steer the market away from vendor lock-in. Medicine based on secret software is not a sustainable national strategy anyway. We should probably plan on doing both.