Are Healthcare and Health IT in a Dysfunctional Relationship?

What a week last week! First the disgraced cyclist confession and later the baffling college-football-player-and-his nonexistent-(dead)-girlfriend story, with the RAND report sandwiched somewhere in between. It’s positively a scandal-palooza.

What’s that? You don’t feel like the recent RAND report, which basically says that a 2005 RAND study financed by GE and Cerner was wildly optimistic in predicting about $81 billion in potential health care cost savings through widespread adoption of electronic health records, qualifies as a genuine hoax, controversy, scandal?

Me neither.

But it does neatly frame what is arguably a unique characteristic of the healthcare industry—a trait that extends to peripheral industries as well. Basically, healthcare is an interconnected environment. Call it the systems theory of healthcare, co-dependency … or just regular dependency. Call it what you want, but there is an interconnectedness in healthcare that we ignore at the expense of national wellness.

Witness key data points provided by the RAND report:

  • Modern health IT systems are not interconnected and interoperable, functioning “less as ‘ATM cards,’ allowing a patient or provider to access needed health information anywhere at any time, than as ‘frequent flier cards’ intended to enforce brand loyalty…”
  • Neither are they widely adopted, with an estimated 27 percent of hospitals utilizing a basic electronic record. Without broad adoption, interoperability is far less relevant.
  • Improvements in quality of care / patient safety and reductions in healthcare costs (which have grown by $800 billion since 2005) are not manifesting with EHR adoption, in part because hospitals and clinics are rushing to adopt mediocre solutions and garner federal funds.
  • The provision of care is the same as it ever was, even though EHRs are frequently promoted as the optimal tool for a different kind of care.

The reasons for these disappointing stats are readily apparent and unalterably interconnected.

  1. We still live in a fee-for-service healthcare system: Doctors and hospitals are still paid based on the procedures and examinations they perform. And because most EHRs evolved from billing systems, it now appears EHRs have made it easier to bill but not easier to provide quality care.
  2. Many doctors are not on board: In a fee-for-service system, doctors understandably don’t want to adopt a technology that will make their jobs harder. They don’t want to see fewer patients in a day and bill less than before they invested millions in health IT. And they don’t want to stay at work until after dark updating patient records. They can hardly be blamed for not embracing this scenario when …
  3. EHR solutions are hard to use: We’ve succeeded in creating technological solutions that would be most impressive to a physician in 1985. Now? Not so much. And the vendor community really doesn’t want to do the interoperability dance to the extent that the RAND report said some industry insiders are convinced many health IT vendors are “opposed to interoperability.”

As I survey the health IT landscape, I see four groups of stakeholders—health IT vendors, hospitals and clinics, government, and patients—with sometimes overlapping and sometimes conflicting goals. If this health IT project is going to make a difference in healthcare, each group may have to take some initiative and make a few sacrifices to keep this whole endeavor on the rails.

  • Health IT Vendors: Are we really making systems doctors want to use? If not, we need to step it up and improve the quality of our offerings, and we need to make our systems communicate. Can the format wars between Blu-Ray and HD DVD, Betamax and VHS, in consumer electronics serve as some kind of industry standards guide? Can HIMSS play a constructive role?
  • Hospitals and Clinics: Admittedly, significant change will be difficult while fee-for-service is predominant, so some initiative will be required. Until EHRs are used to keep patients well instead of billing for itemized treatments after they are already sick, costs will not come down.
  • Government: Personally, I think the use of incentives / penalties by the government is appropriate. I know many people disagree and think the market should be permitted to function. But for how many years did the AMA, AHA and other industry groups fail to act before the government stepped in? Healthcare threatens to bankrupt the nation. We need action. And if healthcare IT can’t establish standard formats ourselves, I would argue the government is correct in doing so.
  • Patients: Even when you visit those hospitals and clinics that offer personal health records, you aren’t using them. As a consumer of healthcare, you need to take some control over the product you receive and insist on quality and prevention. Do your research. Be aware of your own personal health data. Ask your physician why he still uses paper.

Can we all work together to ensure this grand health IT experiment contributes to saving healthcare in America? I don’t know. If it requires putting aside personal and organizational concerns for the greater good, I’m skeptical.

What are your thoughts? Should we slow the Meaningful Use train, as an increasingly loud chorus suggests? Who should step up, and how, to make this all work? Are we all just rearranging the deck chairs?

Edmund Billings, MD, is the chief medical officer for Medsphere Systems Corporation.

26 replies »

  1. Greetings! Quick question that is certainly entirely away from subject. Are you aware learning to make your website portable friendly? My own blog site seems to be unusual any time surfing around coming from my apple iphone4. I am just looking for a theme or extension that might be able to solve this issue. In case you have any suggestions, please share. Regards!

  2. As an IT guy that butts heads with IT security all the time I have to say that I love the idea of patient data being open within the system I am however nervous about it popping out into the mainsteam through small holes in the security systems. Is this an issue that we have to worry about? How is the security costs hampering both the encryption of patient data and the layers of security complexity growing costs?

  3. The key issue is that the EHR is just the first step in a long journey to transform healthcare using information technology to enhance the system. The holy grail is intelligent clinical decision support tools, but to get there you need to first build information repositories upon which to base that capability.
    We don’t really know what we can do with the systems until we have a chance to experiment. I use the stethoscope as an example, at first it was only used to listen to breath sounds, now we use it for many other investigations. Same will apply to the EHR and IT. Put it into the hands of smart clinicians and they will develop new uses for it they we cannot predict.

  4. Until Health IT vendors get their acts together and finally create a comprehensive EHR system that is simple to use and does what physicians want, the simplest solution seems to be good old document management. Although not a replacement, a good HIPAA-compliant document management system can be a great alternative for storing health information. For example, this one is an online solution with scan-to-cloud features for digitizing paper records. Hope this helps! http://www.dynafile.com

  5. Dr. Billings,
    What you wrote above is precisely why the “newer” EMRs don’t work either, not to mention that those “at-risk” supporting EMRs are the same as the FFS ones (see KP).

    EMRs should not support business models of any kind, and they should certainly not drive business models. They should support patient care and nothing else. Leave the business to practice management systems.

    More to come on this from my side as well… 🙂

  6. The EHR is only a tool.

    It’s the business/reimbursement model that determines how it is used. Fee for service will drive services, fees and costs. At-risk will drive toward cost effective quality. The report does not clearly describe this basic driver.

    The EHR’s built primarily for Fee of Service setting will be designed to drive coding and throughput. Good luck, trying to use them to control costs and drive quality, good luck…

    The EHR’s built to support an at-risk model: KP and the VA etc, will support the outcomes needed to control costs and improve quality.

    It’s the nature of the business model, not the tool that determines the result.

    More to come on this…


  7. Brian,
    I am certain that additional customization is needed and will be needed for the foreseeable future. Both HL7 and X12 have been around for a very long time and are used for millions of very useful exchanges every day, and both still require customization when new partners come on board. Sometimes very extensive (and expensive) customization.

    For example, go to the same randomly chosen small physician practices and see if any hospital is willing to provide lab orders/results interfaces to any of them. Almost none would (I tried many times), even though the HL7 lab interface standard and protocols are ubiquitous everywhere else. There is just no benefit to the hospital to justify the effort and the exact opposite may be true.

    Exchanging clinical summaries with CCD is fairly new and although there are challenges, exchange is occurring through private and public HIEs from Alaska to Kansas to the East Coast. Could it be improved upon? Absolutely. This is just the beginning of the road.

    The point I am trying to make is that there is no business interest on the part of technology vendors to prevent such exchange. If anyone is interested in holding information hostage, it is the same entities who are interested in limiting patients’ choices of providers.

  8. Ok so all the cynical responses win! EHR is a freight train most providers are sitting back watching go by at the speed of sound, to nowhere. Think of the investment without true coordinated thought process and real input by the docs in the trenches. Oh by the way I agree an attempt by the government to try “pay for performance” is at least a swing in the right direction ( go ahead slam me on that at least someone is standing up!) In my world loud objections are just the first step to being a “disruptive physicians”. To object means we must just not want what’s best for the patient. In the mean time vendors and hospital administrators cozy up for a comfortable ride. Winston Churchhill said it “the Americans eventually do it right but they try everything wrong first”. I doubt if I will lie that long on this one.

  9. I am so glad that I am enrolled in the Kaiser Permanente system where medical IT works! If I see a doc about a problem, my history is all there. My immunizations and any drugs prescribed are visible on the screen. Moreover, I can see the same records myself on the web.

    Kaiser invested in this technology and — of course — it is not a fee for service system. All medical care should be delivered through similar systems. What’s in the way? — most obviously profit incentives for hospitals, drug companies and individual practitioners. Besides, there are cultural changes that need to come. Docs lose any aura of omniscience where data informs medicine.

    Those changes will come. Sure, Kaiser can make mistakes and be slow to respond, especially if the patient has something rare and unusual. But scientific medicine based on aggregate data is coming and I’m glad.

  10. Margalit,

    This would be true if the CCD were actually a standard and the MU criteria hadn’t been gamed by the vendors through different implementation approaches. But the fact is, they have. See, for example, this: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3243208/. The fact remains that, the MU criteria notwithstanding, the EHR vendors have not agreed to a definitive interoperability standard. As a practical matter, this means that it remains impossible to easily transfer clinical data across EHRs.

    There is a big difference between theory and practice, or policy and practice, especially if those with a financial stake in creating obstacles are in a position to demonstrate compliance while creating impediments.

    If you doubt any of this is true, test it. Go to 10 randomly-chosen physician practices with EHRs and 3 health systems. Try to effect a seamless exchange, and see what happens and how many of the organizations can send and receive the information successfully, without a great deal of additional customization.

  11. Brian,
    Meaningful Use Stage 1, Core Measure #14, requires that all EHRs have the “Capability to exchange key clinical information among providers of care and patient-authorized entities electronically”. The standard is CCD. MU does not prescribe the protocol (e.g. web services, secure email, private VPN) but there are very well established protocols used every day to carry scripts to pharmacies, claims to payers and orders/results to laboratories. Any of those could be used and are used in several places where systems chose to connect to exchanges or each other.

    In order to be certified an EHR must demonstrate the ability to generate, send, receive and consume a standard clinical summary, and in order to receive MU money, every hospital must attest that at least one such test of interoperability was performed by the hospital.

    All the EMRs you mentioned above are certified and should be able to exchange data. If a vendor tells you that they are incapable of interoperability, their MU certification should be revoked by ONC. If a vendor tells you that it will cost you X number of dollars to set up the interfaces, that is their prerogative because it does cost money to operate an exchange.

  12. Margalit:

    I’m sorry, but your comment is factually incorrect. Not only is an Epic implementation incapable of seamlessly exchanging patient data with a different vendor’s EMR, it is also, by design, incapable of exchanging patient data with another Epic client.

    The same principle holds for ambulatory EMRs. Take any Allscripts or NextGen or Practice Fusion EMR, and try to get them to talk to one another. Can’t currently be done. My clinics would like nothing more than to be able to send and receive patient data from our EMR, which was built to be open, to other EMRs. This is a huge barrier to care coordination, and it rests at the feet of the EMR vendors.

    I agree that health systems’ reluctance to share data for proprietary reasons is also a huge problem. But that is a separate issue.

    There are no required Meaningful Use EMR interoperability standards at this point, which is why David Kibbe MD is spearheading DirectTrust.org, which would establish protocols for secure clinical data exchange.

    As to the Chamber nationally, we should have no illusions that they’re not representing anyone’s interests except those of the largest Fortune firms. But what is clear is that, as Florida Governor and former HCA CEO Rick Scott recently noted, nobody is interested in seeing their revenues drop, which why the health care industry won’t fix the health care industry. Only a mobilized, collaborative effort among non-health care business can be a counterweight to the health care industry’s current practices. People in health care may not like hearing that, but so far we haven’t seen meaningful movement by the industry to fix the structural problems that allow them to extract more money than they’re legitimately entitled to.

  13. EHR vendors don’t get to decide whether they adhere to interoperability standards or not. They all do, because Meaningful Use requires such adherence and customers will only but EHRs that are certified to adhere so they can collect taxpayer funded bribes.
    Deploying standards based interoperability in practice is not up to the EHR vendor. It is up to the client, and clients have business needs that run contrary to data sharing (e.g. customer capture and retention, referral retention). By clients, I mean large health systems, because individual and small practices, who are not providing all services for all their patients and are not benefiting from referrals, are not opposed to exchanging information.
    Since we are now morphing all health care delivery from small business to monopolistic large systems, this is just going to get worse, or maybe the walled garden will be big enough to create an illusion of interoperability for those inside it.
    As to driving one nail out by hammering another one in (i.e. letting the Chambers of Commerce be our “champions”), I don’t think this is an optimal solution, but I think we agreed to disagree on this one, Brian 🙂

  14. I think Brian has IDed the naked emperor embedded in the rush to EHR implementation. Policy follows the money, as in “who contributed the most to my campaign fund?” All you have to do is follow that money down K St., and similar streets in 50 state capitals, and you’ll see exactly why EHR is an entirely unfulfilled promise.

    Dr. Billings’ tiered recommendations for the four groups involved in this process (which is, BTW, everyone, since we’re all in the patient group) are aces. However, as long as we’re still playing the game from the stacked deck of the Holy Billing Codes, we’ll be stuck right where we are.


  15. Edmund,

    An excellent post, particularly for the clarity of its explication of the problem. Contrary to what many of the comments, this is not a technical problem but a political one. So long as special interests can bribe our US and state legislators with campaign contributions in exchange for influence over policy, our laws will favor the special over the public interest. A lack of interoperability is a barrier to care coordination, costing us thousands of lives and hundreds of billions of dollars, and taking a toll each year on our global competitiveness as well. Allowing EHR vendors to decide whether they’re going to adhere to standards that would permit seamless information exchange may serve their interests it is directly counter to the national interests and should not be tolerated.

    Unfortunately, as with the rest of our national health care cost crisis, only one group, non-health care business, has the heft, influence and motivation to be a counterweight to the health industry’s rapaciousness. Unfortunately, to date they’ve been divided and neutralized by the industry through organizations like our Chambers of Commerce.

    So, like Dr. Billings, I am skeptical that we can get the health IT industry to change its practices.

  16. Current HIT is dysfunctional, user unfriendly, and represents an emerging disease that must be mitigated.

    How, you may ask. First off, the devices causing the disease, eg CPOE, must undergo safety and efficacy vetting by the FDA. Then, there needs to be after market surveillance.

    There is not any evidence that these devices improve outcomes or reduce costs. ZERO!

  17. The problem is the order in which things have been done. By stimulating the use of IT on a dysfunctional healthcare system we have automated and embedded a dysfunctional system. What was needed was healthcare reform first and then embedding it with technology.
    It has not been a total waste, but I suspect in hindsight the US will look upon it as a wasted opportunity. Maybe if they had been more willing to give healthcare reform a chance back in Bill Clinton’a day they would now not be in such a fire situation.
    Other countries such as Australia and the UK can’t be too smug either. Failure to first implement the required system change before using technology to embed it is a major failing. It doesn’t need to be centrally planned, but it does need physician involvement and it does need to be courageous enough to change the status quo with a long term view. Payment reform is at the center, but quality and safety need to be intrinsic goals as well.

  18. Love the use of the VA example, Dr. Leng, but some clarification is required. To get to VistA, the VA absolutely did not “get a bunch of doctors in a room with the best minds in IT” and “lock the door.” In fact, there was a sanctioned IT project managed in just that way that was a dismal failure

    The history of VistA development at the VA is much more complicated, but is also rather inspirational and a testament to grass-roots (sometimes clandestine) collaboration between motivated developers and physicians. You can read it from the perspective of some of the participants here: http://www.hardhats.org/history/hardhats.html

    Joe is right, central planning won’t work. It didn’t at the VA, and despite the seemingly centralized nature of Meaningful Use incentives, I think the feds are trying to walk a tightrope on this. They’re going to fall off sometimes.

  19. Nice article that explains fairly well why physicians are approaching EHRs reluctantly, spoiled at the end by the suggestion that patients ask their physician why they don’t have an EHR. How about just have the patient read the article, then they don’t have to ask why.

  20. I worry that health care providers are spending time and energy on learning EHR, fixing EHR, explaining their job to EHR creators (not communicating clearly) while learning about how to heal people is being neglected.

  21. Standardization is a great idea and has been implemented in a lot of areas (especially in the VA, as Shirie stated) yet even where it has been done there’s still a long way to go. Finding the right system that works for everyone is a tricky business, especially in an ever-changing industry that’s constantly developing.

  22. Any EMR designed to work in the our dysfunctional system will be, by objective standards, equally dysfunctional. We’re going at this bass-ackwards: fix the system first and the appropriate IT will arise spontaneously. The current approach is only reinforcing what doesn’t work in American health care. To expect HIT to fix our problems is completely dishonest intellectually.

  23. dr. leng- central planning, even with ‘smart people’ doesn’t work.

    in fact, some of our greatest successes and advances in US history are from people that the experts dismissed as stupid, many are college drop outs, etc.

    the system ‘works’ by allowing many, many to make an effort– the bad ideas FAIL, and good ones can flourish — but when you stick in political influence as a barometer or to increase likelihood of success…

    mediocrity reigns…

    increasingly, that describes more and more facets of our hc system.

    and, doctors are complicit in that they have been slow to absent to push back on claims of ‘bad, inconsistent care’ where ‘if only’ more structure and accountability were in place from the ‘right people’, the system will cost less and help more…

    anyone believe that we are better off today in hc than we were?

  24. Get a bunch of doctors in a room with the best minds in IT, lock the door, and don’t let them out until they have a system that works and is flexible. Then use the incentive money to give that system to every doctor and hospital. The VA does this and it works.

  25. medical research studies 101:

    1. who wrote and who paid for it.

    doesn’t necessarily make the results not true — but every living person knows bias can still be there — with the ‘blockbuster’ effect of this study — ONCHIT and $20 billion in stimulus funds that have benefited most of all — the companies that paid for the study…

    it is a reminder that until we know #1, and are willing to wait to corroborating more independent data, we should not have massive forced changes to any part of the economy…

    it’s called ‘evidence based medicine’.