Failure to Translate: Why Have Evidence-Based EHR Interventions Not Generalized?

The adoption of electronic health records (EHRs) has increased substantially in hospitals and clinician offices in large part due to the “meaningful use” program of the Health Information Technology for Clinical and Economic Health (HITECH) Act. The motivation for increasing EHR use in the HITECH Act was supported by evidence-based interventions for known significant problems in healthcare.

In spite of widespread adoption, EHRs have become a significant burden to physicians in terms of time and dissatisfaction with practice. This raises a question as to why EHR interventions have been difficult to generalize across the health care system, despite evidence that they contribute to addressing major challenges in health care.

Problems Motivating Use

EHR interventions address known problems in health care of patient safety, quality of care, cost, and accessibility of information. These problems were identified a decade or two ago but still persist. Patient safety problems due to medical errors were brought to light with the publication of the Institute of Medicine report, To Err is Human, with recent analyses indicating medical errors are still a problem and may be underestimated. Deficiencies in the quality of medical care delivered was identified about a decade and a half ago and continues to be a problem. The excess cost of care in the US has been a persistent challenge and continues to the present. A final problem motivating the use of EHRs has been access to patient information that is known to exist but is inaccessible, with access stymied more recently by “information blocking”.

Evidence Base

These problems motivated initial research on the value of EHRs. One early study found that display of charges during order entry resulted in a 12.7% decrease in total charges and 0.9 days shorter length of stay. Another study found that computerized provider order entry (CPOE) led to nonintercepted serious medication errors decreasing by 55%, from 10.7 events per 1000 patient-days to 4.86 events, with preventable ADEs reduced by 17%. Additional studies of CPOE showed a reduction in redundant laboratory tests and improved prescribing behavior of equally efficacious but less costly medications. Another analysis found that CPOE increased the use of important “corollary orders” by 25%. Additional studies followed from many institutions that were collated in systematic reviews published first in 2006 and then updated in 2009, 2011, and 2014 that built the evidence-based case for EHRs. There were some caveats about the evidence base, such as publication bias and the benefits mostly emanating from “health IT leader” institutions that made investments both in EHRs and the personnel and leadership to use them successfully.

Failure to Translate

Despite the robust evidence base, why have the benefits of EHR adoption failed to generalize now that we have their widespread adoption? There are several reasons, some of which emanate from well-intentioned circumvention of the EHR for other purposes. For example, both institutions and payers (including the US government) view the EHR as a tool and modify prioritization of functions for cost reduction. There is also a desire to use the EHR to collect data for quality measurement – which should be done – but not in ways that add substantial burden to the clinician. Additionally, there are the meaningful use regulations, which were implemented to insure that the substantive government investment in EHRs led to their use in clinically important ways but are now criticized as being a distraction for clinicians and vendors.

There are also some less nobly intentioned reasons why the value of EHRs has not generalized. One is “volume-based billing,” or the connection of billing to the volume of documentation, which leads to pernicious documentation practices. Another is financial motivation for revenues of EHR vendors, who may be selling systems that are burdensome to use or not ready for widespread adoption. Much of the early evidence for the benefits of EHRs came from “home grown” systems, most of which have been replaced by commercial EHRs. These commercial EHRs do more than just provide clinical functionality; they redesign the delivery of care, sometimes beneficial but other times not. It thus can take a large expenditure on an EHR infrastructure before any marginal benefit from a particular clinical benefit can be achieved, even if the rationale for that function is evidence-based.

Nonetheless, a number of “health IT leader” institutions have sustained successful EHR use and quality of care, such as Kaiser-Permanente, Geisinger, and the Veteran’s Health Administration. These institutions are not only integrated delivery systems but also have substantial expertise in clinical informatics. These qualities enable them to prioritize use of IT in the context of patients and practitioners as well as incorporate known best practices from clinical informatics focused on standards, interoperability, usability, workflow, and user engagement.

Moving Forward

How, then, do we move forward? We can start by building on the technology foundation, albeit imperfect, that has come about from the HITECH Act. We must focus on translation, aiming to understand how to diversely implement functionality that is highly supported by the evidence while carrying out further research in areas where the evidence is less clear. As with any clinical intervention, we must pay attention to both beneficial and adverse effects, learning from the growing body of knowledge on safe use of EHRs. We must also train and deploy clinician informatics leaders who provide expertise at the intersection of health care and IT.

Finally, we also must reflect on the perspective of the larger value of IT in health care settings. Approaches to cost containment, quality measurement, and billing via documentation must be reformulated to leverage the EHR and reduce burden on clinicians. We should focus on issues such as practice and IT system redesign, best practices for the patient-practitioner-computer triad, and practitioner well-being. We must build on value from other uses of EHRs and IT, including patient engagement and support for clinical research. Leadership for these changes must come from leading health care systems, professional associations, academia, and government.

William Hersh, MD is Professor and Chair of the Department of Medical Informatics & Clinical Epidemiology at Oregon Health & Science University. Dr. Hersh also maintains the Informatics Professor blog.

Categories: Uncategorized

24 replies »

  1. Simplicity would lead one to respond directly to the question asked, but you didn’t respond seemingly preferring your answer to be cloaked in rhetoric. In other words you have unnamed disagreements with the organic growth of EHR’s and prefer them to be imposed from above. That is a problem.

  2. Read my posting. You may not agree with my answers, but I did answer your questions in my original posting above. To wit, medicine has known safety, cost, and information access problems. There is a line of research from the 1990s showing that EHRs (home-grown, developed by informatics experts, and with focused clinical decision support) can improve on these problems. But this research has failed to translate, in part due to the reasons I described. So the answer is not to give up on trying to fix the safety, cost, and information access problems, but instead to figure out how to generalize the successes we have seen in some settings.

  3. Dr. Hersh, what is your objection to organic growth of EHR’s, meaning physicians voluntarily use the EHR’s they find of value and when they find it important in their practice of medicine? When fax machines first arrived physicians didn’t run to buy them until the fax machine proved itself valuable.

  4. I do not receive a penny from EHR vendors. I work for an academic medical center, which is the largest healthcare provider in our state, and my work is mostly in academic clinical informatics. Call that a conflict of interest if you like.

    Again, I am not uncritical of current EHR vendors. Anyone who in any way reads that into my posting is not reading my words. This is getting frustrating.

    But there is plenty of evidence that *some* EHR systems work quite effectively. To denigrate the early groundbreaking work of David Bates and colleagues is an insult to Dr. Bates. The positive findings do occur in some settings, as shown in the most recent systematic review by Jones et al. from 2014. We need to learn from what works and what does not.

    Given my expertise, I should no less advise clinicians and their organizations than an expert radiologist should speak to the optimal use of radiography.

  5. So Big Bill…how many patients do you see a week? This past week? Did you ever see any, for how many years, 1? 3? Do you (yourself) use an EHR to care for your patients currently? Do you get consulting fees from any EHR vendors, HIT companies? Or does you institution get money due to your involvement? Do you have a COI? Your pitiful studies are VERY old, like 1998 to 2000 old, WAY prior to current EHRs. Kaiser cannot seem to replicate itself out of California (IT fairly easy when all providers are on same IT/IDS, they just pulled out of Ohio completely), and the VA as a standard for quality care? Really? How well does VA play with DOD EHR when its just outside their IDS? Terrible obviously. As you can tell we are VERY sick of supposed MDs that do not practice, that are touting the wonders of the current state of EHRs. We do not NEED more nonfront line MDs telling front line MDs to listen to them about HIT, encouraging HITECH, academic MDs to tell us how to use IT or practice. And you SHOULD apologize to mommy doc, Dr Niran. You want to see our claws come out? Then continue to disrespect her.

  6. “I am not going to apologize for someone’s baseless attack on my professional activities … But she (and many of the rest you) also are not really reading my posting..”

    You don’t have to apologize, you have choice, something that is being taken away from physicians practicing at the bedside. I read your piece and was only commenting on your response to Niran which was horrible. I think you should be listening a bit more and stop being so defensive.

    EHR’s need to develop more organically. If they do then gradually they will be accepted by the physician community, but there is no way the EHR should obtain the status of the doctor / patient relationship. The EHR is a tool much like an EKG machine.

  7. No one was attacking you. Rather, I put your opinion into context based upon your experience in longitudinal clinical practice. I am making a valid point. This is what physicians are trained to do; assimilate the information and evaluate where bias exists to interpret someone’s assertions.

    Three practicing physicians disagree with your opinion EHR’s improve clinical practice in a small primary care setting over what we already have and we “have an axe to grind?”

    That opinion speaks volumes as to how one has learned to collaborate, compromise, and successfully resolve differences as a professional. We are not computers; we are human beings. There is a huge difference between the two.

  8. I am not going to apologize for someone’s baseless attack on my professional activities. Niran knows nothing of what I do, including how much medicine I have practiced in my career. But she (and many of the rest you) also are not really reading my posting. I am hardly defending the current status quo of the EHRs most physicians are stuck using in their practices. I am trying to make the point that EHRs have been used successfully in many healthcare settings, and have been shown in the studies I cite to improve care delivered. My posting seems to be more like as Rorschach test for those of you who have axes to grind against various aspects of medicine, EHRs, etc.

  9. Somebody got his fee-fees hurt.

    Dr. Niran’s attitude, if you bothered to check her out, is one of complete dedication to her patients, regardless of ability to pay. It speaks volumes that you consider that an embarrassment.

    Shame on you.

  10. I am not certain you read my comment either. My position on EHR (you termed it attitude) is that no one should be able to “force” me to change to a inefficient documentation format in order to practice my profession.

    I did not say your opinion that a “physician-computer-patient” triad does not matter. I disagree with you and do not think you understand the first thing about clinical practice.

    I understood clearly you are not defending the current clunky, archaic systems used now, however there really is not a “usuable” alternative as of yet.

    Your deafness to the plight of practicing physicians on the front lines is the real embarrassment to our profession.

    The point you seem to miss is culling evidence that EHR’s work still requires physicians to document by clicking boxes to “show” that something was done. It is all a game. The data can be just as easily collected (which the insurance company is doing for me as part of a Washington State Transformation Grant) using claims data as it is using internal EHR data.

    Frankly, you do not seem interested in hearing what a practicing physician has to say if they disagree with you. That is a shame. It is your closed-minded attitude that prevents you from developing a system practicing physicians WANT to implement and use.

    If a system was developed that takes me less time to document, is more cost-effective than the dictation system I currently use, and shows better clinical outcomes, I would absolutely use it.

    I have not seen hide nor hair of anything that comes even close. I want the CHOICE to decline until there is scientific proof outcomes are improved over what mine already appear to be. And make no mistake, they are pretty darn good.

  11. As a retired physican I object to the way you mischaracterize Niran who has stood up for patients and the appropriate practice of medicine. Thank goodness we have physicians like Niran that still recognize who we are treating, patients, not numbers, not EHR’s and not political processes.

    You owe her an appology. From your tone it sounds like your beliefs are what has made healthcare a mess. As an M.D. you ought to recognize that an M.D.’s concerns at the bedside should be the patient, not other concerns. You seem to have the desire to place some between the physician and her patient.

  12. Dr. Palmer,

    You are missing my point. In the places where EHRs have been designed with the physician and improving care of patients in mind, they have been highly successful. My whole point is that these best approaches have not been adopted by other institutions, and this is why we have not achieved that translation. We must also not forget the downsides the paper records that I certainly experienced in my career, of records being lost, illegible, or otherwise not available.

    Bill Hersh

  13. Adrian,

    I don’t disagree with you that our health records architecture is archaic and that there are systemic reasons why it is not becoming more modern. But that is somewhat tangential to the real improvements in patient outcomes and healthcare delivery that we see in places where EHRs are used in ways supported by the research evidence.


  14. Niran,

    Your attitude that the facts and opinions of no one who is not a full-time physician matter perfectly exemplifies why medicine is so screwed up in this country, and one of the reasons why we achieve the worst outcomes at the highest costs. I am not sure you even read my piece above, because I am hardly defending the current state of the EHR environment. But your attitude clearly shows that our real problems in medicine in the US will never be solved with physician attitudes like yours that are an embarrassment to our profession.

    I am happy to discuss the real issues here and how we can translate the evidence that EHRs do work to all kinds of physician practices and settings.

    Bill Hersh

  15. This post and most of the comments are about a health records architecture that’s over 50 years old. It’s beyond obsolete. Expensive, insecure, proprietary. Because neither physicians or patients have any purchasing power in this architecture, it’s an obvious market failure. But it’s not regulated as a market failure because a decade of politics and regulatory capture insists that that we’re just about to turn the corner: The regulators have a ten year plan.

    Look around. Machine intelligence is beginning to provide decision support. It will provide this advice to physicians and patients directly, regardless, or in spite of, the fragmented institutional EHRs and it will factor in the social determinants of health (financial, social, environmental) that nobody will trust to large institutional EHRs and unsustainable HIEs. Only a longitudinal patient-centered health record architecture can feed the mobile, computer, and network technologies we already have today.

    Computerized health records will become indistinguishable from medicine itself. The EHR as an institutional resource management system is destined to become an undifferentiated commodity. The patient-centered health record will be open source just as medicine is open source. Interoperable because with open source there’s no incentive to avoid standards. Peer-reviewed for the same reason that today’s security software is open source.

    It’s time to start talking about how we get to open source patient-centered health records and chart a course away from a half-century of institutional EHR.

  16. “physician-patient-computer triad”

    Similar to the driver-traffic-computer triad – aka texting while driving. I’m sure Dr. Hersh highly approves.

  17. Of course I agree with both of you, gentlemen. I do not think computers have really improved the quality of our lives. They simply allow us to be more efficient, so we can do more things. I am not convinced those things are as meaningful or necessary when you really think about it. 🙂

  18. And then of course ….

    there’s the parent-child-computer relationship which turns out to be a big deal these days and may or may not be closely related to the doctor-patient-computer relationship!!!

  19. “You lost me at “physician-patient-computer triad. … Clearly, you like your computer so much you now think it belongs in the physician-patient relationship. ”

    Niran, I’m glad you focused on the triad mentioned in the above blog posting.

    I am now wondering if that triad shouldn’t be invloved in all relationships. Think of the husband-wife-computer triad. I wonder what bedside complaints we would get from that relationship? 🙂

  20. You lost me at “physician-patient-computer triad.” What a ridiculous notion. I am absolutely thrilled you went on to further specialize in informatics and suspect you know a lot about computers. Clearly, you like your computer so much you now think it belongs in the physician-patient relationship. I disagree. According to your bibliography, despite getting an MD in 1984 (back when I was 10 and only dreaming of becoming a pediatrician), you have put in exactly 3 years at an urgent care center.

    I have 16 years of clinical practice under my belt, and I have had 100’s of patients for the entire 16 years. I already have 35 second generation patients. That means I know them backward, forward, inside out, and upside down. I do NOT require a computer to practice medicine and you have no business telling me about how great it can be when you have no concept of what I do every day.

    Dr. Nelson, who commented below, was in practice for 41 years, I bet he still remembers many patients and their medication list off the top of his head. I suspect Dr. Palmer is the same, but would have to ask his specs to know details and such. They were physicians in the REAL sense, as am I.

    My great uncle physician visited families in the Pacific Northwest travelling by horse and carriage. He told my grandfather once that having patients come to an office was a silly notion because a physician cannot properly care for a patient without seeing the environment in which they live.

    Every few decades, we “advance” the physician-patient relationship according to new innovations. If one chooses to track patients by computer that is fine; I don’t and have not found ONE part of EHR to be helpful to my practice yet. Let me use my “homegrown” paper charts and stop forcing your “niche” on me and my work.

  21. Why would we expect something that was not designed for doctors and patients to be particularly useful? After all, its real utility function was to help other stakeholders: cleaning up billing problems, studying costs, making prescriptions easier for the pharmacist to read, helping with inventories, studying public health and population health….essentially helping insurers and health plans and government. The proof is by asking ” were we complaining to one another about our old method of record keeping?”

    The EHR also obliterates the deeply intimate and personal value of physician note-taking. These styles had much meaning for us. There was great non-written and hidden meaning in our scribbles, in the order in which we jotted to ourselves, in the size of our written letters, in the intensity of the ink or pencil marks, where on the page we placed our notes, in our spelling. Anyone who has written a grocery list for his own shopping purposes knows what I mean by this.

    But, anyway, this discussion may be moot if we ascend into a future of constant ransomware, viruses, hacking, and extortion. Is our real future one of tightly protected local area networks, LANS, designed from the ground up, with firewalls in every direction?

    Forget the term “interoperability.”

  22. Increasingly, healthcare reform has focused on the cost and quality issues of the @16 million citizens who represent 70-80% of our nation’s healthcare spending. Meanwhile, we have no means to nationally focus on the equitable availability and ecologic accessibility of each citizen to Primary Healthcare. This process should be community driven to acknowledge the local over-riding social determinants of HEALTH. The actual problem-solving process should be community planned including all of its significant stakeholders and implemented by its institutions already existing. The rate-limited process of the cost and quality problems of healthcare are related to its responsive accessibility.
    Remember now that our nation’s maternal mortality ratio (WHO definition) has worsened 25 years in a row, the ONLY world-wide developed nation with that record. The most significant analysis of this problem appeared in 2000, reported by Amnesty International USA. Their important observation is the lack of our nation’s healthcare traditions that emphasize responsive accessibility.
    see https://www.amnestyusa.org/files/deadlydeliveryoneyear.pdf
    In 2000, our nation’s maternal mortality ratio was ‘12.7’. In 2014, it was 23.8 (OBSTETRICS & GYNECOLOGY October 2016). High quality Primary Healthcare for everyone would be the most efficient strategy for its translation to obstetrical care. There also is good evidence that a state with a lower maternal mortality ratio is correlated with longer longevity after age 65 for its women citizens. It is a social capital issue. In the final analysis, there are at least 500 women who die annually because they live in the wrong nation. The best 10 of the 51 advanced world-wide nations have an average maternal mortality ratio of ‘3.8’. (See WHO 2015 WORLD-WIDE MATERNAL MORTALITY report).
    It is unlikely that the trajectory associated with a time-line of high-level healthcare spending for persons during a 1-2 year period of time could be significantly changed without rationing. In the meantime, we need to renew our efforts to fix the “front-end” of healthcare. Currently, the estimate is that 160 million citizens (50%) currently account for 5% of healthcare spending. To do this, a nationally planned, regionally attenuated and locally driven strategy should be implemented. The Design Principles defined long ago by Elinor Ostrom (Nobel Prize winner 2009) for managing a Common-Pool Resource should be applied, specifically its requirement for decentralized governance.
    After 40+ years as a Primary Physician, accessibility requires a skilled registered nurse TEAM to be the first persons to answer the phone during office hours. This ‘desert island’ assignment required a resilient involvement with the physician team, especially for the patients who had not learned how to participate in a ‘caring relationship.’ My own perception is that over the years, the payers for healthcare have increasingly allocated their premium dollar to Complex Healthcare Needs to the exclusion of the Basic Healthcare Needs. Currently, the ability to effectively and efficiently manage a person’s over-all healthcare has become nearly impossible mainly because it is under-capitalized.
    Virtually, nothing that is part of healthcare reform will solve these issues, community by community.

  23. I am not going to address the conspiracy theory presented in the comment, as I think there are more fundamental theories as to why we have not been able to reap the potential benefit of health IT. The two most prominent being: Mal-alignment of business (financial, etc.) incentives and the focus on local optimization (i.e. it is not just a technology solution).

    Bill touches on the first issue of mal-alignment of business incentives by discussing about volume based payments and its consequence of payment for documentation. While I believe this is the most important mal-alignment, it is not the only one. With the lack of transparency throughout the health care ecosystem, there are limited business pressures to improve other than to improve billing and documentation. Of course there is professionalism and ethics driving individuals and organizations, but there is limited business value (i.e. revenue, market share, brand loyalty, etc.). This is exacerbated by the high fragmentation and complexity of the health care system. Luckily we are seeing shifts in payment and transparency that may be the light at the end of the tunnel. Until there are true, tangible business incentives to improve, we will continue to be pushing health IT up a hill (and fighting the inertia of the status quo).

    My second issue is based on the LEAN principle of local optimization and work systems. I worry about the fidelity of the early (and perhaps current) studies in electronic health records. First, an electronic health record is actually a very complex set of functionalities that are frequently configured and customized with each installation (esp. true in the early days). Second, these EHRs were deployed into health care organization, which are also complex and varied. In my view, this leads to some serious variability and potential dependent variables. Even if the researchers are able to account for all dependent variables, we still only have focused on the technology, not a work system. Industrial and systems engineer has taught us that we cannot think of just the technology but rather must look at a system of work (work system) that is made up of people, the organization, internal environment, external environment, the tasks, and the technology and tools. I would encourage a review of the Systems Engineering Initiative for Patient Safety (SEIPS) model – https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3835697/. If we are locally optimizing the technology (i.e. EHR) then we are missing a huge portion of the work system. We need to work on optimizing the work system not just the technology. Only then do I believe we will reap the benefits of health IT.

    Thanks Bill for pressing the issue with your post.

  24. The Academy of Medicine, previously known as the Institute of Medicine, substantially funded its TO ERROR IS HUMAN study by donations from General Electric and Cerner. It was done by the Rand Corporation. Ironically, it is the errors of the study that now continues to plague the poorly directed strategy for the reform of our nation’s healthcare industry.
    The root cause of root causes for the main problems with our nation’s healthcare reform is that is has no means to fix the turmoil existing between the scientific and the humanitarian realms of knowledge underlying each person’s HEALTH, community by community. The computer epidemic has occurred without an understanding of the humanity that fundamentally drives the outcome of healthcare. And the answer is not a single payer, unless you are willing to accept onerous rationing such as AHCA 2017.
    One answer to energize our nation’s healthcare reform is a nationally sanctioned new institution that is uniquely Chartered by Congress and governed by a semi-autonomous Board of Trustees. The Board of Trustees would be selected by 9 groupings of States, each representing @35 million citizens and Federally funded by $1.00 per citizen annually. Also, it would be Chartered to reduce the portion of our nation’s economy devoted to healthcare by 1/3 within 10 years AND reduce our nation’s maternal mortality ratio by 80% with 10 years. It would be driven by the Design Principles for managing a common-pool resource, community by community. These Design Principles have been defined and validated by Nobel Prize (2009), Winner Elinor Ostrom along with many of her independent colleagues. Chief among these Principles is the importance of decentralized governance.
    Oh yes, I am reminded that the Veterans Administration has decided to eventually dump its EHR and start all over. Great Britain is in the midst of the same process. My home town Hospital Enterprise just upgraded its Epic System. The expense probably represented the cost of about one year of healthcare for about 3,000 representative citizens.
    For a unique proposal to focus our nation’s healthcare reform, see: