Asking the Wrong Questions About the Electronic Health Record

The wrong question always produces an irrelevant answer, no matter how well-crafted that answer might be.  Unfortunately the debate on health information technology seems to be increasingly focused on the wrong question.  An Op-Ed in the Wall Street Journal argues that we have had a “Major Glitch” in the use of electronic health records (EHRs).  This follows on a series of recent studies that have asked the question “do EHRs save money?” Or “do EHRs improve quality?” with mixed results.

While the detractors point to the systematic review from McMaster, boosters point to the comprehensive review published in Health Affairs that found that 92% of Health IT studies showed some clinical or financial benefit. The debate, and the lack of a clear answer, have led some to argue that the federal investment of nearly $30 billion for health IT isn’t worth it.  The problem is that the WSJ piece, and the studies it points to, are asking the wrong question.  The right question is:  How do we ensure that EHRs help improve quality and reduce healthcare costs?

The fundamental issue is that our healthcare system is broken – our costs are too high and the quality is variable and often inadequate.  Paper-based records are part of the problem, creating a system where prescriptions are illegible, the system offers no guidance or feedback to clinicians, and there is little ability to avoid duplication of tests because the results from prior tests are never available.  Even more importantly, the paper-based world hampers improvement because it makes it hard to create a learning environment.  I have met lots of skeptics of today’s health information technology systems but I have not yet met many physicians who say they prefer practicing using paper-based records.

The problem is that some Health IT boosters over-hyped EHRs.  They argued that simply installing EHRs will transform healthcare, improve quality, save money, solve the national debt crisis, and bring about world peace.  We are shocked to discover it hasn’t happened – and it won’t in the current healthcare system.

Most EHR vendors today sell their products to doctors promising increased “revenue capture” (that is, improved billing resulting in greater payments to physicians and higher costs to the health care system).  In a fee-for-service world, the EHR, which is nothing but a tool, helps you get more “fee” for your “service”.  It’s not surprising that we aren’t seeing huge savings.

To understand how to best leverage the potential of EHRs to help the US improve care and save money, we will have to answer a series of other related questions:  how do we create incentives in the marketplace that reward physicians who are high quality?  How do we allow physicians to capture efficiency gains?  Today, if a physician becomes more efficient, he/she will likely lose revenue to insurance companies or to government payers.  When Kaiser Permanente installed an EHR and gave patients the ability to use the electronic system to message their physicians, they saw their ambulatory care visit rate fall by 20%.  This is a disaster in a fee-for-service world.  Sure, Kaiser was able to see real financial gains from their EHR – but how do we help the thousands of other physicians and hospitals that are not Kaiser gain efficiencies from their EHR?  That’s the question I’d like to see answered.

Now that we have made an important investment in EHRs, we need to figure out how to use this new technology to address the fact that the healthcare system is a mess.  We need to figure out how EHRs can promote coordination of care across sites, seamless flow of good clinical information, and smart analytics, to name a few things.  We simply can’t do that in a paper-based world.  I am sure that the healthcare industry single-handedly keeps the fax machine industry alive.  We need to stop. Period.  Every other part of our lives has become electronic and the benefits are clear.  Our lives are better because we bank online, communicate online, shop online.

The debate over whether we should have EHRs is over.  Can we fix our broken healthcare system without a robust electronic health information infrastructure?  We can’t.  Instead of re-litigating that, we need to spend the next five years figuring out how to use EHRs to help us solve the big problems in healthcare.

Ashish Jha, MD, MPH is the C. Boyden Gray Associate Professor of Health Policy and Management at the Harvard School of Public Health. He blogs at An Ounce of Evidence.

Categories: Uncategorized

Tagged as: , , ,

35 replies »

  1. in fact the tool helps to drive the system into the wall faster and highlights all of the dysfunctions at the systems and even cultural levels.

  2. Hi Sara! Wasn’t sure if “user friendliness” was hinting at the challenge or using most of them or a compliment?

    One unique quality of the original VA Vista System is that it was/is open source and providers and developers were co-located.. It is an interesting story politically – check out the Underground Railroad people at the VA Sadly during the Bush years they centralized it and then attempted to outsource some parts.

    Since nearly all of our residents do a rotation through a VA facility at least they all have worked with one by the time they finish their internships.

  3. Thank you, Sherry, for calling our attention to the VA. I’d add that the level of user-friendliness of the VA EHR is comparable to that of most proprietary systems.

  4. Watching the AAFP fight with NPs for the crumbs from the health care table is better than Saturday night WWW. That they’re reduced to this indicates that they’ve already lost.

  5. Yes, Kaiser has reported some impressive outcomes associated with its EHR, including a sharp drop in their ambulatory care visit rate with no apparent ill effects. But it is also worth noting that it had several failed attempts at EHR installation it got it right, including a write off of more than a billion dollars after one of its earlier attempts ended in a fiasco. The benefits of EHRs get lots of publicity, but the costs and failures tend to be airbrushed out of the literature on EHRs. Which is a shame, because the health care industry could learn a lot from the failures yet there is a conspiracy of silence among software vendors, providers, and (of course) CIOs when it comes to explaining what went wrong and why it went wrong.

  6. Your views are anything but worthless – it’s just that you have never used an EHR. Period. You have played with them, studied them, analysed them, but you have never used one. Ever. But the real point was that Midwest doc’s views are not worthless either, despite him/her having not evaluated the number of EHRs that would impress you.
    So answer this question – which of the 40 systems you are familiar with are able to understand how humans think, work, and solve problems? If there are none, then you owe Midwest doc an apology.

  7. Perhaps world peace or at least peace in the Middle East is easier. This process is paralleling the introduction of electronic billing in the late 1970s. Remember when claim forms were written by hand, bundled, and mailed to BCBS? And there a clerk keyed the data to be crunched by a big IBM computer and then print a check and remittance form back to the doctor?

    It took about 20 years and multiple coding updates for that to be “perfected” to the point checks appeared quickly. It suspect it will take about as long for providers, payers, and patients to deal with and benefit from EMR/EHRs. I know physician offices that still have their fax machines and use them daily!

    We still have massive amounts of data and very little information. I don’t recall anyone ever asking a doctor what information they need to effectively and efficiently care for their patients. When we know and agree on what we want, I’m sure there are smart computer engineer folks who can design a system to produce it. We just need to agree on what it is we want/need and then make up the codes for it.

  8. You read too much into it. We work with about 40 systems here, about 2.5% of the 1,578 currently “Certified complete” EHR systems (ambulatory and inpatient). So, I would never claim to know how well ALL systems work.

    Oh, and, yeah, I know; since I’m not a physician, I have nothing but worthless views in this space. Heard that one many, many times.

    “A culture of denial subverts the health care system from its foundation. The foundation— the basis for deciding what care each patient individually needs— is connecting patient data to medical knowledge. That foundation, and the processes of care resting upon it, are built by the fallible minds of physicians. A new,secure foundation requires two elements external to the mind: electronic information tools and standards of care for managing clinical information.

    Electronic information tools are now widely discussed, but the tools depend on standards of care that are still widely ignored. The necessary standards for managing clinical information are analogous to accounting standards for managing financial information. If businesses were permitted to operate without accounting standards, the entire economy would be crippled. That is the condition in which the $2½ trillion U.S. health care system finds itself—crippled by lack of standards of care for managing clinical information. The system persists in a state of denial about the disorder that our own minds create, and that the missing standards of care would expose.”…

    …this subject matter may seem like just a variation on current policy concerns with using “health information technology” to bring “evidence-based medicine” to “patient-centered” care. Yet, current policy fails to comprehend the needed discipline in medical practice and thus fails to define precisely what is needed from health information technology. A dangerous paradox thus exists: the power of technology to access information without limits magnifies the very problem of information overload that the technology is expected to solve. Solving that problem demands a meticulous, highly organized, explicit process of initial information processing, followed by careful problem definition, planning, execution, feedback, and corrective action over time, all documented under strict medical accounting standards. When this rigor is enforced, a promising paradox occurs: clarity emerges from complexity.”

    Lawrence Weed, MD and Lincoln Weed PhD, “Medicine in Denial”

  9. Although it was only implied by your comment, if the implied statement is true then it is also true that It is the height of arrogance to think that the fact that you may have evaluated dozens of EHRs means anything at all. Unless I missed it somewhere and you are in fact a provider of clinical services. I’ll say it for you – (bleep) me too.

  10. I agree with all you write in your last post, but I still stand by what I said above … there is no empiric evidence for the upsides, but strong “common sense” and anecdotal evidence supporting the advantages above. But even if I was right about the upsides, the question remains whether
    1. it’s worth the effort or not (that in part a hard to impossible to research empirical and in part a judgment question)
    2. whether EMR could be made cheaper and userfriendly (=less costly)
    3. whether we can weed out a lot of the downsides (as discussed). For instance, one can address issues like stupid templates and note bloat. I heard from old doctors that in the very early days of the beta blockers, they felt that patients were killed due to wrong use/overdose. But usage was adjusted, and the betablockers prevailed.

  11. EMRs prevent some errors, and cause new ones. I don’t think we have any good data yet that shows EMRs leading to better quality care.

    The reason the US lags behind the rest of the industrialized world in quality of care measures is not due to the fact that we lack EMRs. It’s because a significant part of our population has no or limited access to needed care. We’d get far more return on our investment by addressing that elephant in the room, and let HIT develop as needed.

  12. “IMHO, a MR should be a MEDICAL record anyway, not primarily a billing tool.”

    Every practicing doc would agree with that, but no one is interested in what we think.

  13. @midwest and southern docs,
    A reasonable bare bones EMR/PACS should provide availability of legible data, to everyone involved as soon as they are generated.
    That’s a huge achievement: no lost parts of the MR, no illegible scribble in docs’ notes, no mystery patients showing up at the ER and the MR or latest CT scan is not accessible, no lab slips falling behind the file cabinet …
    if we abolish the use of idiotic templates and excessive cutting and pasting resulting in note bloat, the EMR is a blessing, esp. in large institutions (IMO, very little benefit in stand alone private practice). Actually, docs not communicating is much more risky without EMR; without EMR, communication is even more random and difficult.

  14. Not sure about this argument. The number of visits may be lower (the VA where I trained prior to the VA EMR was a teaching hospitaI, so numbers are usually lower), but I heard the VA system is much simpler and user friendly … and I am not sure about reimbursement/billing/coding issue. We are working with Epic here and we still have lots of (wo)manpower doing billing and coding, and I would not be surprised if they use their own additional software. IMHO, a MR should be a MEDICAL record anyway, not primarily a billing tool.

  15. Bobby is rolling over as a REC REP.

    [bleep] you. I shill for nothing. In fact, I have taken substantive shit at work over my REC blog (“exceeding your scope”). I am the Resident Curmudgeon.

    My REC job is done effective end of Q12013 anyway.

    “The problem with the EMRs I have evaluated is that they do not understand how humans think , work and solve problems.”

    “I have evaluated…”

    And, that would be precisely how many? Total, and as a pct of market?

    Remember (per ScienceBasedMedicine.org), “the 3 most dangerous words in Medicine are ‘In My Experience’.”

    apropos, see “Medicine In Denial.” Dispositive. Link on my blog.

  16. But the OP insists on asking the most bass-ackwards wrong question: “How do we ensure that EHRs help improve quality and reduce healthcare costs?”

    Who cares? The question we need to answer, of course, is “How do we improve quality and reduce costs?”

    And we all know what has to be done, and it doesn’t require EHRs: universal access to improve quality, cut RVUs for procedures and diagnostics by 50-75%. But nobody is pushing those.

    EHRs can put a little icing on the cake, but to present them as the necessary and sufficient first step is a tremendous distraction from where the real problems lie.

  17. southern doc you are correct. See 8 to 10 patients a day and do the NY times crossword puzzle. What a great day. I worked at a VA and great gov. job. Do nothing , be home by 5. Get a check.
    “Our lives are better because we bank online, communicate online, shop online”. Very funny. For look at all the potential for fraud, I.D. theft ect..Bobby is rolling over as a REC REP.
    Any of you pups remember going to your local bookshop and greeting a smile. Having a beautiful bound book to hold. The local coffee shop owned by a neighbour (not a corporation).
    I had 6 consult requests this past Sat. and Sun. This hospital has the cream of McKesson EHR. (Next week CPOE goes on line). I take a piece of paper, write notes, and decipher a long , redundant, duplicated database. It is like reading The Iliad and the Odyssey just not as good.. I then dictate the corrected allergy list, PMH, Med. list ect.. When I learned BASIC programming at NYU( yes I am old) the foremost teaching was GIGO. Is GIGO gone? Nope.
    I estimate (anecdote) these patients would have been home 3 days sooner on average if the docs involved had picked up the phone and consulted together. The EHR created a forest of bytes. No one saw the obvious. Assumed communication and sharing of information via the EMR may have kept the patient safe from us but failed to diagnose and treat. This old country fart solved the case. Just saying…………….They are home and doing well, no thanks to HIT and interested investors/vendors/payers.
    The problem with the EMRs I have evaluated is that they do not understand how humans think , work and solve problems. We get so distracted with input that we fail to output. Medical thinking is not banking or day trading. Following a haemoglobin A1 C is fine for the machine. Getting that number down is another story.

    Can’t wait for stage 2 and another couple thousand dollars in the toilet.

  18. Well, i smell wafts of false dilemma — implication of “other overlooked causal factors” — in that analogy.

    MFreeman — “It won’t work.”

    Well, I guess we can all just go home now.

  19. The correct analogy would be that EHRs are as likely to solve problems of expense and quality as ATMs are to prevent another financial meltdown.

  20. “How do we ensure that EHRs help improve quality and reduce healthcare costs”

    A perfect example of the fallacy referred to as “begging the question”

    The VA EMR works because it’s a closed system where the docs see 8-10 patients a day and don’t have to deal with multiple insurers, not because the EMR is all that superior to what else is on the market.

  21. OT, but related. I am reading Goodmans book. Anyone familiar with the eHealth EHR? Do they have a usable EHR that could be adapted into clinical practice?

    Back on topic, bring me an EMR that works and I will use it.


  22. “When Kaiser Permanente installed an EHR and gave patients the ability to use the electronic system to message their physicians, they saw their ambulatory care visit rate fall by 20%. ”

    There must be a reference documenting this impressive achievement – could you please provide it?

    ” How do we ensure that EHRs help improve quality and reduce healthcare costs?”
    Uhhm, not sure whether this is the appropriate question. Steven Colbert used to ask: “George W Bush – a great or the greatest president?” I actually do favor EMR and have been using them since my residency more than a decade ago (I think they do increase quality of care at least for complex patients at larger institutions). That does not mean we should not question whether they are worth the immense investment. Dr. Jha, have you heard about the VA EMR? I have never used it, but the system seems to be very easy, robust and affordable … maybe we should move more that direction.

    Yes, the US HC system sucks … but that does not imply that EMR are integral part of the best approach to fix that.

  23. Yes. Similarly, it’d be MUCH better if banking returned to the days of 9-3 M-F hours and double-entry paper bookkeeping.

  24. Look at your premise of HOW DO WE ENSURE THAT EHT WORKS? That’s the point. It won’t work. We are distracting ourselves from the real problem which is cost. Computers will NOT help this.

  25. “The parts of our system that are broken have very little to do with paper charts versus EHRs”

    Exactly. That’s what I was trying to say.

  26. The other point that needs to be made is that I can focus on any number of limited aspects of the system and make them better. I can do this with paper charts or with EHRs. It is the FOCUS that makes the improvement, not the particular tool choosen to do the job. Thus we have a multitude of studies purported to show the benefits of EHRs that actually only show the benefit of focusing on a particular aspect of the system.

  27. You are still asking the wrong question. Without realising it, you are in essence asking, “How can we continue to live with our broken system but make it a little more tolerable through the use of technology.”
    The parts of our system that are broken have very little to do with paper charts versus EHRs, and thus the reasons that even the more muted hype surrounding EHRs continues to lead to disappointment when the reality of their limitations sinks in.

  28. Great post. Your point about fee-for-service specifically, and misaligned incentives generally, as the central problem is right on.

    Nonetheless, there still are systematic issues with our EHR infrastructure which we must address to optimize the potential efficiencies as healthcare migrates from fee-for-service to fee-for-value, namely that of cross-compatibility across the closed-standards of the major EHR players. While RHIOs and HIEs demonstrate that we recognize compatibility as a key problem, today HIEs and RHIOs are not adequately addressing the problem, and adding additional cost to the system. While I am not generally in favor of additional regulation or mandates, it seems we would be very well served by moving to much more open standards.

    Until health records can move fluidly and with ease across any facility, I do not think we’ll ever be able to recognize the value of our EHR infrastructure.

  29. “I don’t believe it is about the tools but the system itself.”

    Agree. Where I disagree with the OP is that major changes in the system have to be made first. The IT we have now is only serving to maintain and strengthen what is wrong in the system. Change the system first, and the appropriate and necessary tools will arise spontaneously.

  30. Great post – and an argument many of us have been making for eyars.

    In many cases an EHR as a tool amplifies the flaws in the existing systems and will drive you into the wall that much faster.. Without associated changes in workflows, financing and dare I say the organizational culture of healthcare itself we will end up blaming the tool instead of realizing that it is the system that is broken..

    The VA has had the longest experience with a less than user friendly (but amazingly customizable) EHR was able to provide some of the highest quality care for the lowest cost (during a time of budget cuts they saw more patients). The biggest complaint that most vets have about the VA (not to be confused with the DOD) is that they can’t get in..

    At places like Group Health when we implemented our EHR we actually gave the patients access first and within a year or so also fundamentally changed primary care and shifted our “members” (not patients) to the medical home model. In many practices 40% of all encounters are happening online via email or telephone encounters and visits are now 30 to 45 minutes long for those with chronic conditions. New clinics are being built, young doctors are applying to work there and they are holding costs down for the most vulnerable – no increase for Medicare last year.

    Again two key differences? The payment system and the ownership of their patients over time. It is time to move away from conversations about the tool and instead focus on how to use the tool to achieve all six pillars of the IOM (including patient centered care) and the triple AIM.

    I am open to feedback as I might be wrong but I don’t believe it is about the tools but the system itself.