Saturday, March 31, 2018

EHR - Smart Regulate, Don't Deregulate

Who's to blame for the Electronic Health Record debacle? And a debacle it certainly is, at least for clinicians. The latest bite of criticism has hit the oped pages of the Wall Street Journal. Drs. Mass (pediatrician) and Fisher (nephrologist) from the Massachusetts General Hospital do a good job of chronicling its ills. MGH uses Epic, the near monopoly of enterprise-level EHRs. I call Epic “by engineers, for engineers,” so unintuitive that the pathways rival those of the old city in a European capital where only the natives or the guides can find their way, who use their special knowledge for fun and profit. With this and other EHRs, doctors become unproductive data entry clerks typing and clicking away to fill up the boxes and navigate the menus, often at home late at night finishing up charting the day's patient notes. The indictment goes on: the software design often causes errors rather than preventing them. The content of the notes is frequently filled with garbage designed to gather more money according to coding rules and to protect the practice legally. For instance, in pediatrics we are used to receiving ER notes that assure us that our three year old patient has attested that he does not smoke, and the pages and pages of verbiage makes finding what actually happened at the visit a needle and haystack adventure. And of course we often receive these notes not by computer but by fax, because the EHR programs are most often not interoperable as they were envisioned to be. The promise of all patient information available everywhere all the time? No way. Everyone looks at the same record within the system, generally hosted by a large medical complex, but you can't see the record if you are outside the system, as smaller independent practices and other units are. So why did we go through all this investment and work for a more troublesome and less productive state of affairs?

It's a familiar catalogue of frustration and vitiated hopes that we all share. What makes the article particularly congenial to the Journal, however, is its attribution of blame and the proposed solution. Blame goes to the government, for funding, encouraging, and requiring that the work of doctors be computerized before the programs were ready for mass consumption. Subsequent blame comes to government also for requiring far too much certification by vendors which, they claim, inhibits innovation. If only this vendor protection were removed, the authors aver, Amazon and Apple and other consumer electronics companies would invade the medical space and bring it up to speed and down in price. In other words, deregulate. WSJ red meat.

But I have to say to these doctors and WSJ, not so fast, my friends. It's not so easy and it's not so simple. For one thing, consider if the medical field and the field of consumer electronics are really so compatible. Medicine is far more complex than the stereotypic tasks of ordering a household item, and the market is far more constricted. The big tech software companies make their money on volume, and there are a lot more ordinary people in the world than there are doctors. How much money would it take to really attract the A team? Is the profit possibility really there?

But that's not the major objection I have; maybe they would come in, maybe not. My major objection is that I think the source of the problem is deeper than it appears. The physician authors might want to think of their medical training. Sometimes a rash is pretty simple and can be cured with a simple cream. But sometimes that rash is the harbinger of a deeper disorder that needs far deeper intervention. I'm afraid that this is one of those more serious situations, where the symptom of the poorly functioning EHR emerges not only from clunky governmental functioning and their less than acute ministrations to the health care system, but also from the organizational structure and political-economic interests of those we call “stakeholders.” Thus, I would not be drinking the Journal deregulation Kool-Aid just yet.

Should the federal health officials and the other pezzonovante who make up the health care establishment have pressed the Obama Administration to include EHR funding in ARRA, taking advantage of a unique time when big time money would be available? Were EHRs shovel-ready? Probably yes, they should have, and no, the EHRs really weren't completely ready. If they hadn't taken the cash opportunity when it presented itself, how could the medical world be computerized? It would have taken a lot longer time, and the money to buy and install and maintain the EHR systems could only have been raised by the institutions that were already predominant. So, the government's getting into the game seems well founded to me.

But did the government, under the Office of the National Coordinator of Health IT (known as ONC) screw the pooch in their administration of the program? To me, unequivocally, yes. They went for micromanagement of what was “meaningful use” (MU) of the EHRs that the practices and institutions bought – making sure that they weren't ripped off, that the government got what it paid for, it seems. He who pays the piper calls the tune, and the ONC didn't want to be accused of a giveaway to industry, understandably. Understandable.

But what the ONC screwed up was in hitting the wrong notes. What they should have concentrated on was interoperability, not every little use modality that took so much effort to document, prone to such error and inconsistency. What they didn't understand was that interoperability is the key. They should have mandated that all EHRs be absolutely interoperable, and then let the smaller details take care of themselves. This was their cardinal sin. Was it a sin of ignorance or one of influence? Were they not sophisticated enough about the usual pathway of progress in free market systems, or were they influenced by the most powerful pezzonovante in the EHR world, and the world of medical institutions?

I can't answer that question because I don't know how the process went. But we do know that interoperability is technically quite feasible, and that lack thereof is a political rather than technical issue. And it's quite clear that the ONC decision to go easy on interoperability only reinforced the controlling forces in our health care system. I detailed in a blog post last November why this is so important, as I described how large organizations have essentially weaponized the EHR.

Seeking business dominance by patient and clinician capture, the large medical centers and enterprise level software manufacturers have essentially weaponized the EHR by keeping it private and unsharable. When patient information is available only within an EHR network, the patient is “nudged” to access only in-network providers and facilities. Likewise, the externally impenetrable EHR pressures clinicians to renounce their independence and join the network not only to defray EHR costs, but also to achieve “featured” status for referrals on the EHR as the networks “nudge” referrals inward, and to utilize data in treating patients that they would have only laborious access to otherwise.

Maintaining strong EHR boundaries for network commercial advantage is regrettable. If large networks are to achieve dominance, they should do so by lowering costs and raising quality, which has been difficult for them, rather than using the EHR as a cudgel. Closed networks and closed EHRs provide diminished incentives to improve efficiency and quality, as services need to be just “good enough” rather than truly excellent to attract captured patients. A closed system even presents an ethical problem, since the primary care provider, who is ethically bound as a medical fiduciary to seek the best and most efficient referral resource for the patient, is nudged by the system to respect instead the financial needs of the network.

The search for root cause leads us inexorably to the organizational structure of health care. Although you wouldn't know it from the density of the propaganda cloud emanating from the large corporate networks, there is a good argument that smaller, decentralized units strung together by modern communication capabilities would deliver better and cheaper care than the large networks. But fighting to remain dominant is typical of economic behavior in any society. That's what is going on now. For a 2,000 word explanation of this argument, see:

So, to return to the start and Mass and Fisher's capitulation to WSJ ideology, what would be the effect of deregulating EHRs? Unfortunately, deregulation would not lead to interoperability. Since interoperability would simply give ammunition to the competitors of the large integrated enterprises, it is likely that they will be content to keep their systems closed.

What is needed is not deregulation, but smarter regulation. A legitimate role of government in our mixed system is to regulate the marketplace so that competition occurs on a level playing field and benefits accrue to the public. Smart regulation would recognize Epic as a dominant platform and regulate it as such, much as the government regulated Microsoft, another dominant platform. If the government made interoperability mandatory, and if they were to require the EHR to display referral opportunities equally, the playing field for clinicians inside and outside the system would be more level. If they also mandated that Epic and other platforms be open to module substitution, EHR competition would be improved. For instance: in pediatrics we have some EHR programs that work fairly well for us because they are specifically designed for us. Other specialties have something similar. If these practices join a big network to help them gain access to referrals, they must give up their more functional module and accept the more generic and inferior Epic module. If plug-in capability were required of the platform, all the EHRs would be subject to competitive pressure and would improve. It is even possible that Amazon, Apple and other A-list companies would enter the field.

That's it in a nutshell. For a more discursive treatment, check this out also, my best effort to describe the organizational structure dilemma facing our system:

But for the smallest nutshell at all, here is the letter I wrote into the WSJ and, mirabile dictu, they published it.

Budd Shenkin


  1. I read this with one giant sigh.

    First, I should be clear: you got it right, much more so than many who complain about EHRs and how poorly they contribute to patient health and physician satisfaction.

    The fact is, EHRs do _not_ exist for the physician (as they should). They exist to defend/support your coding and to "prove" that you had a certain clinical performance. In other words, your EHR is to get you paid and keep you out of jail. The notes are for the lawyers and insurance companies. You have been turned into a monkey for the people who pay and employ you - that USED to be the patients, but it isn't any longer.

    Of course, the doctors all complain that it's the EHRs' fault ("by engineers, for engineers"), but that misses the point. A well designed EHR that exists for the purpose of improving patient care and physician experience *is not financially viable*. You can't sell an EHR, reasonably, that isn't fully MU certified or supports your E&M chart note. Which means your EHR is dissatisfying by default. What the doctors want, they literally won't pay for.

    One issue that you brushed up against, but didn't fully embrace, is the lack of organization among clinicians to define best practices. Look at SOAPM list: ask a simple clinical question, you have 20 different responses with very different documentation needs - how can a single UI possibly manage that?

    I routinely meet individual doctors telling me, "THIS is how you should display data to me, THIS is how a visit should work, THIS is how I want to enter clinical information." If I share that vision with your peer in another town, I'm laughed at. We can't even get doctors in the same group to behave in a mildly similar fashion. I can name a dozen or more "vanity" EHRs written by doctors whose personal vision worked perfectly...for them. And now they are out of business, every one. The engineers aren't the problem, it is the lack of consistency on even the simplest clinical issues that will forever be a challenge.

    Ask ANYONE who develops EHRs for a living. We'd LOVE to sit down with doctors and design not only exactly what they want, but to collaborate on developing new and more efficient ways to consider improving care. But we can't - we're too busy keeping up with regulations and the 1000s of individual needs of every clinician.

  2. I really impress the author on this blog for sharing on this useful information with us for sure I learn more from here. Interoperability