Why can’t an EHR fit into a mobile device?

It’s been a while since my last rant about electronic health records (EHRs), so let’s remedy that right now. EHRs in their current iteration are — how to put this delicately? — an unmitigated disaster. Nevertheless, much of the criticism of EHRs, including mine, has been in the destructive category. What about some constructive criticism? How could EHR software be made better?

I am not familiar with every EHR system out there. In fact my experience is pretty much limited to one system, which shall remain nameless, though I will give some clues as to which EHR I mean: Its name contains four letters, two consonants and two vowels; the name has no pure rhyme in the English language, though it does have some near rhymes, such as the word septic; and the software is under the delusion that it is running in hyperspace, which may indicate that the programmers possess a sense of humor. There, I hope I have been obscure enough so I don’t get into trouble like I did before. If someone feels there is an outstanding one out there that implements the features I would want, please let me know.

Current EHRs were developed before the mobile revolution, and it shows. Sure there are some mobile clients available, such as the puzzlingly Japanese named mobile app for the above-not-mentioned EHR system, but these mobile apps don’t match the functionality of the parent application, and, at least in my experience, have been virtually useless. It was possible to run the full EHR application  in a virtual machine on a tablet, but, being a Windows-based program, it was necessary to have various non-intuitive gestures in order use it, for example, in order to right-click. This was not a natural interface for an Android or iOS tablet though possibly a Windows-based tablet, such as the Microsoft Surface, might work better. I don’t have experience with the Surface, so I just don’t know how much it would help.

Having your EHR running at full functionality on a mobile device seems to be very important for a number a reasons. First, every doctor already has a mobile device of some sort. Second, the alternatives to mobile devices are immobile devices: desktops, which take up a lot of space, are expensive, are constantly breaking down, and are apt to have security issues, such as the doctor forgetting to log off, thus exposing sensitive patient information to the next person who sits down at the computer. Remarkably, the desktop route seems to be the norm for hospital EHRs, with doctors queued up during busy rounding times waiting to get on a computer. Third, doctors are inherently mobile. In the hospital they go from room to room when they round. It is much more efficient to carry one’s EHR with him or her and just go from room to room than it is to go to a room, return to the nurses station to type into a desktop computer, then go to the next room and repeat the process. Having a truly mobile EHR would avoid the constant trips to the nurses station.  So why can’t an EHR fit into a mobile device?

One reason is that present EHRs try to be all things to everyone.  They are not just for record retrieval and note taking. No, they contain everything and the kitchen sink. The same EHR used by the doctors is used by the admissions office to check in a patient.  You may have no reason to enter anesthesia notes or insurance information but your EHR seems to want to do all that and more. Rather than breaking down EHR functions into different tools for different user roles, all functionality is combined together into one megalithic beast. Such a beast simply can’t fit into the mobile form factor. So we are left with the antiquated desktop computers, taking up precious space in the nurses stations, with quaint, 1990s style user interfaces that would rouse feelings of nostalgia if they weren’t so frustrating to use. And don’t get me started on the do-nothing “click me” buttons that are required for meaningful use.

We used to have mobile record-keeping systems in medicine. They were called charts. Sure they were bulky and unwieldy, and often all the information that we wanted was not in them (most egregiously missing were x-rays). Nevertheless they were relatively portable, could be stacked on a mobile rack, and a doctor could go from patient room to room without having to return to the nurses station (other than to get a cup of coffee). Data input was via a pen, which is actually a very quick, if sometimes illegible, way to enter data. For all the deficiencies of such a primitive record keeping system, it was fast, productive, and allowed more face time with patients — qualities that current EHR systems don’t possess.

So, a well-designed EHR system — something that I don’t believe exists today — would take that old-fashioned model and make it work on a mobile device, such as a tablet. The doctor could go room to room, pull up the patient data, and then record, either by writing, dictating, or typing, a note. The key to making data input work on a tablet is brevity. Get rid of all the garbage that is automatically sucked into a progress note by today’s EHR systems: lab reports, X-ray reports, 12-point review of systems, accumulated cruft from old notes. If you actually look at the notes generated by these EHRs, the amount the doctor actually enters is typically very small. It is contained in the history section which often simply says something like this: “no change” and in the plan, which may be “s/p PTCA, discharge tomorrow.”

All the other debris in the note is added merely to satisfy the coders and billing personnel, who will freak out if there the note isn’t long enough (low complexity of patient care, missing review of systems, etc.). They don’t really care if it is all just cut and pasted from the admission history and physical, as long as all the components are there for them to check off. As I have argued elsewhere, the close coupling of billing and documentation has to change in order to alleviate the current EHR disaster.

A useful EHR system is possible. For it to happen the current desktop-based model has to be thrown out.  We need to start over and develop a truly mobile EHR. One suggestion: Get the input of doctors when designing an EHR. Now there’s a novel idea.

David Mann is a retired cardiac electrophysiologist and blogs at EP Studios.

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

  • Shari Rosenbaum

    Wow. I actually feel lucky while struggling to keep up with Meaningful Use, ACO’s, patient portals, etc. I can run my entire office from my iPad mini. My EHR is eMD’s which has a Citrix based app. This is what I bring into the exam room. I can do just about anything except print from the app. The same goes for my hospital based system.
    I carry my iPad with me on morning rounds and electronically send their discharge meds from my tablet and at the same time schedule their follow up visit. I also carry my iPad while on call and always have everything I need to know at my fingertips.

    • David Mann

      I think your experience shows the potential of a mobile EHR. I tried using the Citrix app on an Android tablet, but found the gestures needed to manipulate the cursor difficult to remember and use. I agree that the tablet was good for looking up data. If they could only make the data input side smoother…

  • http://thematthewreview.com Matthew Durham

    Amen David!!

  • NewMexicoRam

    Makes total sense to me.
    That’s why it will go nowhere.

  • SteveCaley

    An EMR on an iPhone resurrects the old phrase about army chow, “Sh*t on a Shingle.”

    • David Mann

      Agree. iPhone is too small. I was thinking more in terms of a 7 or 10 inch tablet. Changing the shingle though won’t change what’s on it.

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

    Hi, David.

    As a Pediatrician and a software developer who wrote and uses his EHR/practice management solution, it might be helpful to understand how a programmer would look at your suggestion of an EHR on a mobile device.

    The key here is screen real estate. The smaller the screen the more clicks, and the more dissatifying is the user experience. Even to move to an iPad or other tablet, the touch modality is not the be all that people might think. Tablets are good for consuming date, but terribly for productively entering data, at least for the kind of data we have to document.

    User interface design for EMRs is really a challenge and I have to agree with what you are saying; most have poor user interfaces, which satisfy a database prorgrammer but do not help you and me efficiently and logically deal the patient chart.

    I took this principle to heart before deploying my software in 2000. But beware, just because a software developer touts that their software was developed with the aid of physicians or that they even have physician programmers on staff, it means nothing.

    If the physician database programmer does not actively use their creation to deal with their patient documentation, then they will probably not be dealing the nuance annoyances or the ‘why won’t it do this’ criticisms.

    Warmest regards,

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

    • David Mann

      Thanks for your thoughtful comments. I agree that data input is the weak link in a mobile system, but I believe technology will eventually develop to the point that truly reliable voice to text translation will be feasible. I don’t feel we are there yet. In addition the need for duplicate, extraneous, unnecessary, and excessive documentation greatly magnifies the amount of user input, again making mobile systems impractical. But I see this as a technologic problem: doctors should be able to work with mobile electronic charts. EHR companies: Get on it!

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

        Unfortunately, textual data is not the end of the EHR like it is in the paper chart. You will still have to deal with scanned or importat data images. We use images quite often taken from our iphones and inserted into the data record.

        Recently I was involved with the development of an otoscopic attachment for an iphone which would take pictures but then they had to be imported into the medical record.

        And then how do you deal with faxes? The current limitation there is that you still have to provide phone numbers. Not everything is a prescription that can be sent through a prescription fulfilment entity, and even then they use fax machines to get prescriptions to the pharmarcy when no other means is available. They are costly as well, so straight faxing is still the best way to relay certain kinds of information.

        The next big revolution will be sound recording. I have a slight hearing deficit in one ear. I have to listen very carefully to children’s heart. But the day will come when there will be an ausculatory device which will record the heart murmur as part of the medical record. That’s not text of course.

        The smaller the screen real estate the more the clicks or taps. This is a very common complaint of physicians. Intuitive interfaces that make logical sense to you and me that impede our production of medical records that have quality require a certain amount of screen real estate.

        I know this is a dealbreaker for some because I’m a member of the team that is looking at software for our CIN (Clinically Integrated Network) in the Children’s Healthcare of Atlanta physicians network.

        I can see the hope that some physicians hold out for tablet or phone interfaces, but it is often based on the fact that they are unwilling to learn to type.

        Unfortunately, these physicians will probably never be fully integrated with EHRs, nor will they have anything but a hard time trying to move to the digital age of medicine.

        Warmest regards,

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

        • David Mann

          Yes, there are graphical data that need to be imported. I did not mean to imply that the only input into the system would be a mobile device. I can see that for the specific example of an ENT doc, having the ability to import ear pics on the fly would be nice, but most of the graphical data, like X-rays and old medical records, would be scanned into the system by someone else as is the case now. Your comment about docs avoiding typing is true. Maybe in the future typing class will be as much a part of Med School as Human Anatomy, but for now there are plenty of docs who are not good typists and are expected to type a myriad of orders and notes, or use something like Dragon, which I have found to be a good program to generate humorous malapropisms, but not much good for accurate transcription of speech. Designers of EHRs should concentrate more on the physician end users, rather than designing their systems to fulfill various bureaucratic requirements. Thanks again for your thoughtful comments.


  • SteveCaley

    many EMR’s are spin-offs of billing and scheduling systems and mostly created by computer techies under the guidance of center office bean-counting gurus.
    Software functions for the purpose for which it is intended. The “hidden agenda” of EMR’s – not so hidden – is everything but the creation of a brief record relevant to patient illness and treatment.
    The proof of this – since the beginning of the EMR project, there have been millions of dollars made and lost in the video gaming industry. Video game consumers will NOT pay for things which do NOT satisfy their demands. EHR buyers will pay for bad product for physicians as long as it satisfies the hidden agenda. The drivers are payment and regulation, and the future is more of the same. PS: Deployment of ICD-11 is scheduled for 2017. That is not a joke.

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

    What I do in my user interface is have descrete fields for each of the H&P areas examined within which there is free form text. It is not that the governement requires that. Then my solution combines the free-formed text within those fields into a readable H&P format. I use templates for different types of exams which give normals for all the things that I routinely exam on every child. These fill in around my abnormals. The result is a high-quality, readable, non-machine look in a one-page printed exam.

    Ron Smith, MD

Most Popular