Disappointing outcomes despite a massive investment in EHRs

The Health Information Technology for Economic and Clinical Health Act of 2009 committed to the expanded adoption of health information technology, expecting electronic health records (EHRs) to transform medical care while promising dramatic improvements in quality, efficiency and safety.  Five years and $25 billion later, the results have fallen short of expectations, and there are multiple reasons for our disappointment.

First, EHRs were designed to document the provision of health care as it was just delivered.  Most EHRs arose from a programming background emphasizing billing and claims processing; software’s priority was data capture of past transactions.  As a consequence, systems were not designed to provide sophisticated guidance to health care practitioners for “what comes next” in the care of a patient. With some important exceptions, the most important part of a patient’s medical care is the ongoing plan, and – unfortunately — EHRs still don’t effectively facilitate planning the future of a patient’s care.

Second, EHRs have been woefully inadequate when used for population health care management.

Software companies are only belatedly realizing their obligation to enable analysis of health care needs and disparities across entire populations of patients. Without a well-designed and implemented patient registry, an EHR cannot identify groups of patients with similar needs, thus impeding a practitioner’s ability to direct limited care resources to patients who would benefit from intensive management.

Third, engaging the patient — presumably an important party in improving health care through IT — has been an afterthought.  Adoption of electronic patient portals has been slow, in part because the design and user interface of portals lack polish, and in part because portals fail to provide patients with actionable information to guide personal health care management.

One of EHRs largest failures is their inability to communicate with one another, a prerequisite to attaining the promised goals. The health care IT industry has been derelict in its responsibility to comply with standards of interoperability, and no funding mechanism has been established to develop the requisite interfaces among software systems in current use.  A patient who transfers from one physician to the next will most likely need to bring a print out of his or her “electronic” health record to be scanned into the “electronic” record in their new practice.  Diagnostic results and hospital records in one system are frequently unavailable within another electronic platform, requiring physicians to access multiple systems for these records, each system requiring a different username and password. If members of the general public were aware that “computers aren’t talking to one another” as patients navigated the multiple providers involved in their care, they would be shocked at the lack of planning and foresight.

Last, we have succumbed to an all-consuming demand for privacy of health care information without considering the implications.  Personal health care information provided to one’s practitioner during a visit is available only to that practitioner, ignoring the potential that the patient may present to another practitioner with a related problem — possibly in another community or even state. The critical allergy or important surgical history reported at one facility is not available when that patient — possibly incapable of providing the relevant history — requires urgent care in another locale.  We should reasonably expect that all relevant information is immediately available to any healthcare practitioner who needs it to provide safe and effective care, regardless of facility or location, and yet we have tolerated the development of laws and IT systems that make it impossible.  This is a dangerous outcome of our reflexive demand for total privacy, especially when the vast majority of patients correctly assume that healthcare information will be responsibly accessed and used.

Until EHRs guide practitioners in the future provision of care rather than simply serve to document care already delivered, facilitate stratification of patient populations by disparity or need, successfully engage patients in their own care, and “talk” to other systems involved in a patient’s care, we will fail to recognize significant benefits from our massive IT investment.  And without a more sophisticated concept of health care privacy, we will continue to hamstring efforts to use information technology to its fullest advantage.

Mark Nunlist is a family physician.  Betsy Nicoletti is president, Medical Practice Consulting and author of Auditing Physician Services. She blogs at Nicoletti Notes.

Comments are moderated before they are published. Please read the comment policy.

  • doc99

    ” the vast majority of patients correctly assume that healthcare information will be responsibly accessed and used”

    Right – responsibly accessed by whom exactly? The IRS? Google? United Healthcare? Employers? Pharma? Anonymous? Oh right – HIPAA is only for the Little People.

    • doc99

      info responsibly accessed? Security fears are far from overblown.

  • guest

    Offhand I would say that to describe current state of patient portals as “lacking polish” is incredibly kind. They are impossibly buggy, and apparently not user-friendly enough for back-office staff to have any idea of how to operate.

  • guest

    As one of the old geezers you’re talking about I feel bound to point out that absolutely everyone I interact with regarding our EHR (Epic) appears to be fresh out of college, if not high school.

    Also, I am certainly no programmer but it seems to me that if you want to design a truly functional EHR, you would have to have an intimate understanding of the workflow of physicians and nurses, which it’s unlikely that a college level programmer would have.

    On a final note, I guess I would say that the only times I ever “struggle” with my email involve problems that turn out to stump my 21 year old tech-savvy son.

    • SteveCaley

      The instrumentalist principles of design are roundly ignored regarding the EMR (See my comments to Ron above) An EMR is supposed to be a tool that allows for the most effective use of trained human manpower, and for this, “you would have to have an intimate understanding of the workflow of physicians and nurses,” absolutely.
      The EMR catastrophe rivals the planning disconnects often seen in the weapons design line from Pentagon to grunt. The ivory tower thinkers come up with snazzy concepts that are pretty but do not work. The grunt has to deal with it.
      Remember, 20% of the firearms on the planet are knockoffs of Kalashnikov’s AR-47. It’s a fairly modest, inaccurate and crude weapon. But it works. 65 years after its rollout, it’s outlived the country that designed it. The only thing it has going for it is its dependability. But it works.
      Nobody thinks like that for EMR’s. But some fool’s going to roll out an EMR with three-D effects and Smell-o-Vision, no doubt, and there ya go.

  • southerndoc1

    Why in the world would you be so naive as to think that EHRs have anything to do with better medical care?

    • ninguem

      Look at the co-author’s book.


      The author’s interest, as usual, is payment. Maybe on the doctor’s side, as opposed to the insurance companies, but either way, patient care is not the concern.

      • betsynicoletti

        It’s true: my interest is in helping physicians code, bill and get paid for their services. When I meet with doctors to review these topics, inevitably the discussion turns to the EHR. I don’t have a clinical background, but as I recall the promise–hype?– of these records it was to reduce medical errors, increase accessibility of information for professionals caring for the patient and provide better care and information. As Dr. Nunlist and I point out, the results are disappointing.

        • southerndoc1

          EHRs are about data collection for the benefit of large corporations. Everything else we heard was, as you say, “hype ,” and the problems you describe are acceptable collateral dam

          • James O’Brien, M.D.

            Nothing confidential should be on a cloud. Ask Jennifer Lawrence.

        • ninguem

          Actually, no insult intended.

          In fact, to your credit, you recognize the problem.

          • betsynicoletti

            Thank you. I didn’t feel insulted.

  • http://onhealthtech.blogspot.com Margalit Gur-Arie

    You are actually taking down the younger generation because there are practically no “old programmers”. The biggest and most common EHR out there has a strict policy of only hiring college graduates with no experience and presumably no “bad habits”, and after a few years they leave.
    Let’s also not forget the middle aged gentleman that designed the iPhone/iPad all the younger generations are swooning over…. Just saying….. :-)
    EHRs are plagued by problems that are very different, and have zero to do with ability to design cool and usable apps.

  • http://www.amerechristian.com/ Ron Smith


    I’m now 32 years in Pediatric practice and I wrote and deployed my own software in 2000 which networked three clinic locations and managed 33,000 patient visits a year. It started as a true EHR though, and was not born out of a practice management system.

    I bought my first Macs in 1985, a year after they first came out. I’ve been database programming since then and wrote about 5 practice management solutions before the EHR project.

    I’m part of the CIN (clinical integration network) arem for The Children’s Healthcare Network in Atlanta (CHOA) where we are in the process of deploying our public health records part of the whole project which will include HIE (health information exchange). We are evaluating software at present which will cost several millions to purchase and deploy and which much integrate with several hundred EMRs.

    I’m 56. I think I’m one of the longest practicing physicians on the board, if not the longest. I’m the only physician with that technical level of software understaning. The base of our Pediatricians are however very competent with their EHRs and were chosen for this particular arm of the whole implementation because of that prowess.

    I think there are still a few things us old guys still can teach you young whippersnappers.

    First you should watch out for the ones like me that would run techincal rings around you who generally have a wrong impression about us. I manage the entire IT and software needs of my office and another that uses my professional software and networking services.

    Second, I take issue that you, as one who has limited professional experience would show such disrespect for fellow physicians, esperically those who struggle to teach you medicine. Maybe I read your comments wrongly, and if so please forgive me, but I’m afraid that the attempts to show such grey medical heads in a bad light, really only reflects dimly on you.


    Ron Smith, MD

    www (adot) ronsmithmd (adot) com

    • James O’Brien, M.D.

      Well done. This whole EHR debate is filled with insane misconceptions. I’ve been using electronic health records for thirty years. The best one is the template you design for your narrative, not a series of checkboxes. EHR on the Web is a nutty idea, it contradicts HIPAA and eventually it will be hacked.

      Most of the critical software running in the world today was originally designed by men who are now 55-65.

    • SteveCaley

      Golly, you’re old – you got two years on me! And I don’t even have store-bought teeth!

      I note the phrase “people who both don’t understand and have no interest in technology” is a straw-man argument – the creation of a fictitious class of troublemakers who simply need to be mowed down for progress to be achieved.

      This is stale Modernist dogma that is insanely obsolete. Trying to push aside the anti-modernist crowd’s old news. At the turn of the 20th century, there were all sorts of shows and expositions showing novel technologies – the forerunner of the conventions. The Pan-American Exposition in Buffalo, New York was one that the President attended – and he was assassinated there. We’re talking McKinley.
      Anyone who’s read Peabody’s exhortation on the Care of the Patient in JAMA has noted his scolding-down of the technocratic, Modernist wing of medicine. His criticism still stands valid; anyone who hasn’t read it, better do so for their depth in medical education.
      Technology centers on the making and using of tools by humans. By assembly of the practical knowledge needed to construct tools, and by understanding the needs they are to satisfy, one implements technology in a successful manner. “Technology” ought not be an end in itself, except for those experts in tool manufacture. For everyone else, they are instruments needed to perform a certain thing.
      People who flutter about Technology with a warm admiration that it is a path to the Great Future are not technologists, but Romantics, and in the most tacky sense of the term. They are caught up in the thrill, with little sober analysis of the structure of things.
      The history of technology is littered with bad toolmaking – junk that works with no intended purpose in mind.
      RSMD pointed out the embarrassing truth – that nobody’s building an EMR that is to effectively record patient information. It’s all about this or that, and those who build it are not dense, but misinformed. RSMD has built his own EMR, that works, and I’m sure he’s a canny programmer; but the task is not really as vast as people pretend it is. At the sum of $25 Bn dollars, it should have been completed and brought to maturity within a few years, and perhaps $20 Bn returned to the Government {like that’s gonna happen.}
      Throughout history, the younger generations have tried busting all the grey heads, in order that all the knowledge spills out and the newbies can get on with building a new world. Didn’t work in China; didn’t work in countless other attempts. Just history, that.
      The computing power to run a distributed network of USEFUL EMR’s could probably go just fine on the Windows 95 system; the rest of the capacity of modern machines is really for the filigrees and bells and whistles that carry no value.
      I truly believe that the entirety of this EMR effort will be bundled up and heaved overboard within the next 10 years or so. It is warts-on-warts, and cannot be deployed for effect, no matter how much the techno-romantics gush about it.

    • betsynicoletti

      So, Dr. Smith, do you and your partners find that using your EHRs allows you to document and plan for the future care of the patient, and not just document the day’s encounter? Are you able to easily access information from the hospital and lab and integrate them into your EHR?

      • http://www.amerechristian.com/ Ron Smith

        Hi, Betsy.

        Yes, the EMR was designed for smooth patient flow. I’ve encorporated several cutting edge features that are not found in any other EMRs.

        We create HeathPlans for children with that are coming in for checkups. The nurses know not only what vaccines they are to give them, but also what things to ask regarding typical health maintenance based on age.

        All vaccines are barcoded and scanned into the patient chart using the barcode to ensure accuracy as well.

        Charts are usually finished before the paitents get to their cars because we use a combination of descrete fields in the examination (i.e., one field for eyes, left ear, right ear, etc) with popdown textual choices, and templates which fill in negatives around the positives that we mark. I created a portable medical record dongle called a Pedikey which is a small waterproof aluminum capsule, that the parent can update on the kiosk computer in the foyer which contains all of their childrends medical records.

        We fax prescriptions, and scan paper records into the chart as needed and these are part of the record.

        The work I’m doing with the CIN and the HIE at CHOA/TCHN is laying the groundwork for practice integration with all other practices and that will be accessible within the patient’s chart and in real time.

        We already do real-time GRITS (Georgia Immunization registry) both download and upload (have done so since 2003 or so) and am the only practice in Gergia that has that capability. (The technique is a trade secret.)

        This and more make the EMR valuable and extremely streamlined. Even more the practice management is fully integrated and seamlessly. The EMR charges credit cards, and tracks payments and EOBs. My practice manager can do a complex EOB that would take hours and hours in about 15 to 30 seconds. This is for a practice that is seeing almost 10,000 patient visits a year.

        Now that’s the shameless plug. I do very much depend on my EMR for smooth practice flow. It does make a difference.

        The problem with many EMRs is that they are written by programmers and not end users. Programmers don’t generally have any idea what coding they do that will make things bumpy or blazing fast for us.

        There is a lot more that I could say, but good design is the key to a rewarding EMR experience.

        Warmest regards,

        Ron Smith, MD
        www (adot) ronsmithmd (adot) com

        • betsynicoletti

          Thank you.

  • DeceasedMD

    I think the point is the EHR companies are laughing all the way to the bank. there was no incentive to have these systems communicate with each other. here was a rush to produce with what was it 9 billion coming to them? so the end products were often crap. and then got replaced multiple times. One of the many facets of MIC that are adding to exorbitant waste.

  • James O’Brien, M.D.

    I wouldn’t call the outcome disappointing because I predicted it.

  • Eric Strong

    In the VA health care system, since billing insurance companies and Medicare is not a significant focus in general, CPRS (the VA’s free EHR) is much more about patient care. A recent national survey found that CPRS is rated the most highly by users among all widely used EHR’s in the country. This is not a coincidence. Also, since the VA is the largest medical system in the country, the benefits are much more readily apparent of an interconnected system of records where providers can see anything from the medical chart from anywhere in the country. There are obviously legitimate privacy concerns about that extent of access, but from a provider’s standpoint, it is an enormous benefit in taking care of patients.

    Ironically, CPRS is public domain. So implementing it from scratch in a hospital costs a tiny fraction of what it costs to license and implement any alternative such as EPIC.

    • southerndoc1

      Why would a hospital use a free EHR if it’s not focused on billing?

  • buzzkillerjsmith

    Indeed, EHRs don’t work too well.

    Not breakin’ any ground with that, at least not at a blog frequented by clinicians.

    Might be a better idea to send this to a mainstream media outlet, if you have no already done so.

  • buzzkillerjsmith

    A vastly more meaningful system. Well, bless your heart, I. Your pure and trusting little heart.

    I also think the world would be a better place if we could get some of those programmers into Congress. Break the deadlock. Maybe in the Presidency. We sure could all use some world peace. And no more global warming, too.

  • Ellan

    I don’t think anyone who has been an end user since the beginning ever thought EHR’s would transform healthcare. Hospitals bought EHR’s because it made it easy to collect data for billing purposes and to document compliance with regulatory agencies.

    • Chiked

      Well you couldn’t tell from the number of doctors screaming the EHR tagline a few years ago.

  • azmd

    “old people who both don’t understand and have no interest in technology”

    I suppose that somewhere there are some “old people” who don’t understand or have an interest in technology. I have to say I don’t know too many of them.

    A mistake that I see the younger generation frequently falling into is an observation bias: you are more likely to notice an “older” person fumbling a bit with Powerpoint icons, than when a younger person does the same thing, and you are more likely to immediately assume that that person “doesn’t understand and has no interest” in technology.

    It’s easy then to assume a dismissive attitude towards that person’s ideas about technology; being as how they are someone who “doesn’t understand and has no interest.”

    But I think the “older generation” has some pretty valuable ideas about technology, and it’s a mistake to ignore them because they are not quick on their feet with Powerpoint, or whatever.

    In my opinion, here’s the greatest divide between the younger “digital natives” and the older “digital immigrants.” The young whippersnappers see technology as an end in itself, a toy, something to play with. Not too surprising that they don’t see a problem with an EHR that you have to “play with” to operate.

    Us antediluvian types are really too busy for another toy. Unlike the typical college student, we’ve got kids to get into schools, mortgages to refinance, kitchens to re-do, laundry to wash and meals to cook, soccer games and dance recitals to go to in the evenings. Not to mention using our “free time” to keep up with a rapidly expanding universe of medical knowledge.

    We don’t want an EHR that we have to play with at work. We want technology to be a useful tool, not a toy. And if the technology is non-intuitive and doesn’t operate seamlessly, it’s not useful. I think that this is where the younger generation “doesn’t get it,” and so we keep getting products that are all about “gee whiz! That’s so cool that it can do that!” but people need to “play with” the tool to figure out what it can do. It’s like the difference between Google Maps, and the map interface on Mapquest. Or the difference between gmail and Outlook. One is “cool,” but the other is useful. When you have a busy grown-up life, it’s actually usefulness that you’re looking for in technology. Coolness, not so much.

  • Ava Marie Wensko George

    The one thing that was missed here was that the physician workflow and needs were not taken into account with the development of the EHR. EHR companies are kind of backhandedly giving physicians opportunities to contribute now, but even with that EHRs are not a tool that physicians naturally want to use.

  • James O’Brien, M.D.

    It was doomed to fail because the whole concept of health information on the Internet is idiotic.

    As some celebrities recently discovered, put nothing personal on the cloud.


  • James O’Brien, M.D.

    Looks like Tim Cook has my back on this argument. Apple just came out and said that health apps should not use the cloud. Therefore the whole foundation of web based EMR is flawed and dangerous:


    Do we do the right thing and shut down cloud based storage of health information now or do we just stay the course?

Most Popular