EMRs are ugly, and what the next generation of doctors can do

I was in our family practice clinic today and couldn’t help but noticing once again that the electronic medical record (EMR) system they use looks like it was designed in the 1980s.

Gray boxes with tiny free text fields and little check boxes. Unfortunately, the aesthetics are the least infuriating design element.

Why is nothing connected in current EMRs? The single most important thing that has led to the proliferation and usability of information on the internet is linking. I can be reading a story in the New York Times and by simply clicking links within the story I’m taken to a dozen different sources giving me more in-depth information on unfamiliar topics.

In an EMR (I have used close to half a dozen of the most popular) nothing is connected. You have to have a mental picture of where information is located within the hidden folders of the EMR to find the information you need. How is that useful?

And the thing that truly baffles me –- why are hardware and software companies churning out things like Google Chrome, the iPhone, Wikipedia, YouTube, Android phones, the iPad, Gmail, Facebook, and Twitter but we still have EMR systems that look and act like Windows 95?

Never mind that no patient records are connected or even electronically exchangeable (at least not on any scale to make it useful). I can access my e-mail, bank accounts, travel information, even my research library, from any computer with an internet connection. But my patients still have to hand me paper copies of their records from other physicians, which they often forget or are illegible (and remember that it takes time, money, and paper to copy those records).

The patient I saw today — it would have been useful to know she had a prescription written last week by a physician at an urgent care clinic, then another one at the ER.

It also would have been useful to have notes, treatment plans, and medication orders from the endocrinologist and cardiologist who are treating my patient with poorly controlled diabetes. I might have been able to review these before I saw him, instead of wasting 14 minutes of our 15 minute visit trying to help him remember all his medications and instructions.

The state of medical records in our country is shameful. It’s not a particularly sexy problem or one that is going to be easily solved. Conflicting evidence exits regarding implementation of EMR systems and decreased morbidity and mortality.

However, the next generation of physicians have grown up in the digital age. They have learned to leverage technology to work more efficiently and effectively. Demand for well-designed EMRs will grow as these young physicians move onto the wards and further into administrative positions. Maybe that will be the catalyst for change.

I hope not, I don’t think I can wait that long.

Josh Herigon is a medical student who blogs a Number Needed to Treat.

Submit a guest post and be heard.

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

  • rezmed09

    Excellent points. In 20 years the EHR’s that we use today are going to seem like the big old TV’s with tiny black and white screens in the middle.

    EHR’s seem to have no standards. No regulation. No conventions. Any interface is an extra cost . Your comparisons to browsers is excellent, but not as dangerous. Better yet, Imagine having to electronically drive a fleet of new cars all with electronic controls and data sets hidden in different locations. Oh and seat belts and airbags are an extra cost.

  • ninguem

    “….The state of medical records in our country is shameful…..”

    Wow. Such sharp criticism from someone who hasn’t even graduated medical school yet. Is there some Nirvana of a country where the the medical records system is so wonderful?

  • Peter, MD

    I concur completely…the legacy – as well as the vast majority of so-called ‘next gen’ EMRs – still feel and act as if they were developed 30 years ago.

    I know of a solution that is attempting to address all of these important UI and work-flow issues (developed by doctors, for doctors/providers), and it also introduces a completely novel approach to data collection as well as interoperability, but it won’t be ready for another 6 months. Stay tuned…and please keep posting your frustrations and concerns about EMRs here at KevinMD.

  • me

    Funny to read this. We just spent thousands of dollars to “upgrade” our EMR. I’m reading this and I thought someone read my mind. EMR’s appear to be doing less to improve office efficiency and more to document erroneous information. EMR software programmers are at least 10 years behind. CLICK CLICK CLICK CLICK CLICK CLICK CLICK CLICK CLICK CLICK CLICK “here’s your refill of lisinopril”.

  • KP Internist

    KP’s system is pretty complete and we are very happy with it. It is not the EMR’s fault that most of the providers of care are on separate systems and choose to stay that way. I would argue that most of the EMRs are 5 years behind in terms of style and usabilty. But, the medical system (outside of KP) is at least 30 years behind. It doesn’t matter what the GUI looks like, does the EMR have the power to integrate all the data that is generated?

  • stargirl65

    EMR’s need to be able to talk to one another easily. You should not require an IT degree to send things from one office to another.

    EMR’s need to talk the same language. Currently they all use proprietary information.

    EMR’s are designed to document information for HEDIS, insurance companies, lawyers, PQRI, etc. They are not designed to improve health care delivery and efficiency.

    Many entries in EMR’s are duplicative. They need to be more automated and intuitive.

    I want to be able to access any record from anywhere, even on my phone.

  • http://www.healthleap.com Jesse

    For how expensive those things are, they should be much more ‘useable’

  • http://Www.burtchiropractic.com PDF and Email

    Stop fighting each other over EMR
    There many doctors out there that have no idea how this works. When I request records from a different office I aske them email it to me. In respond I get: “Oh, we don’t use email for communication.”
    The solution is simple – all The info that comes into the office on a paper must be scanned right away and saved under that patients name.
    All old fashion faxes must be converted to efaxes where info is converted to PDF format right away.
    Regular handwritten notes can be done on Tablet PC and converted to PDF within seconds.
    It is easy people. Just do it and share patients files via email or secure server.

    • KP Internist

      E-mail is not HIPPA compliant.

  • apurvab

    Working with an excellent integrated EMR daily (a modified version of EPIC), I really can’t get behind the “EMR’s are teh devil!” sentiment that seems to prevail here. Done correctly, as KP-Ncal seems to have done, a robust EMR is HUGELY beneficial. Great for patient safety, helps aovid redundant treatment/workup, avoids treatment delay while records are transferred back and forth, expedites interspecialty communication, reduces medical errors and facilities in patient communication. Even the older sceptics withon our group have been converted. That sufficient time and resources were devoted to adequate training for the doctors and staff (with ongoing voluntary classes every few months) was also integral to getting buy-in and utilization of its full abilities. I can’t imagine going back to a paper-chart system again.

  • http://Meyersmedmal.com Jimeyers

    I hope you are right but from my experience, vendors of emr systems care nothing about patients or the quality of medical care. They care even less about users. Large health systems care about billing more than the efficient access to important medical information.
    As long as health systems are undescrimating in their selections, user friendly emrs will never be developed. A further ominous developement is health systems becoming vendors of their software through wholly owned subsidiaries and claiming falsely the success of their emr system to facillitate sales.

  • KP ob/gyn

    The EMR at KP So Cal by Epic have put almost all of the blogger’s concerns to rest. It is here, it is modern, it communicates between laboratories, pharmacies, individual offices, hospitals, and (gasp!) the member too. The patients have access to their information, secure email communication to their providers and their support staff. Plus they have access to almost all of the other parts of their records as well. Email encounters provide written two way documentation, no answering machines, busy phone lines, or call centers, and can be done when the member wants to flick on their home browser. The database handles millions of users across the country and is disaster backed up across several data centers around the U.S.

    • http://healthrecordcorp.com Merle Bushkin

      That being said, why can’t a KP provider in Northern CA access a Southern CA KP member’s records when the Southern CA member seeks care in Oakland?

      Also, where are a patient’s records for care received before they joined KP or for out-of-network care while a member of KP? Aren’t they relevant to treating the patient?

  • Max

    There are too many systems, too many clicks, and they take up alot of time to learn, train, and money to purchase and maintain. It’s a scam. You either buy ‘support’ monthly or they can come and take your whole system away whenever they want to. Forgot the payment? Fine, we shut off your system. Pay up or we’ll hold it hostage. Huh? How can they do that? I need to see patients and they’re shutting my system down. So now someone else holds your practice by the short hairs, in addition to all the third-party reimbursers. Oh and yeah, you’re still paid for that level 3 visit the exact same whether you had that system in there or not. Oh unless your system told you how to game it into a level 4 which you should probably do to make the support payment and feel better about why you purchased that expensive fancy system in the first place. Oh yeah, it was patient safety, that’s why you did it. Uh huh. You keep thinking that.

    • family practitioner

      Start coding too many 99214′s and you wind up audited.

  • http://silvercensus.com Taylor

    Can’t they just get things right the first time?

  • BladeDoc

    Had a discussion with my hospital IT guy about this stuff, here’s his take. The market is not big enough to justify huge expenditures on the software companies’ side to make these things user friendly. Think about it — there are only about 7500 hospitals in the US so that is the absolutely biggest market there is for physician order entry software — what company is going to spend the millions that it takes to write a really engaging useful user interface (I’m using video games as an example). Similarly there is about 550,000 +/- docs in the US. This is not a huge market and that’s why you get a lot of relatively small companies that sell Beta-test not-quite-vaporware and expect you to “customize” it (i.e. be their worse than free beta testers).

    • csmith

      An answer is for the government to solicit bids for a basic universal system and provide it to physicians for free. The selection could be made by an independent panel of physicians/hospitals (AMA, ACP,AAFP etc.). Applications could be added to the system by free enterprise developers similar to IPhone apps.

    • http://healthrecordcorp.com Merle Bushkin

      Your IT guy isn’t much of a businessman. At hundreds of millions of dollars per major hospital system, many thousands of dollars for systems for thousand of smaller hospitals and group practices, and hundreds of thousand small offices, the market is big enough to warrant large investments in user-friendly IT systems.

      The problem isn’t market size. It’s that legacy vendors are looking through the wrong end of the telescope. They are providing bloated systems laden with features and screens that they as developers like, not what docs want and need. The result is outrageously expensive systems that don’t meet the needs of their potential customers. That’s why, to date, fewer than 30% of hospitals and 15% of physician practices have bought what they have been selling.

  • KP Internist

    The code for such software is not that complicated to write. It is getting useful information from every system and user and sharing it in a HIPPA compliant way is the problem. That is why an integrated system like KP can do it and it doesn’t seem like anyone is able to get it right. In it’s simplest form, EMR is just a notewriter. This is just not much a improvement over pen and paper.

  • LynnB

    I can’t help but think that Josh uses the system that begins with “c” that it advertised on TV–minus pictures of docs in our system leaving practice (and not just old docs) because of it.

    I had such hope for EMR, I was eager to adopt it. I am reasonably technophillic and way into finding the 50% of patients who DON:T make followups without reminders . Really making it work requires an robust IT department. Since the IT department is a “cost” our system enlarged the procedure palace instead. Kaiser and the VA are a different economic model .

  • ljpmt

    I went to my doctor’s office this afternoon with an ear infection. Pretty sure that’s what it was because I had one last year. My doctor said I didn’t have an ear infection last year because it wasn’t in the computer. I told her the name of the doctor I saw and the prescription he gave me. She kept saying it must have been long before last year because it wasn’t in the computer. I repeatedly told her the specifics — I even remembered that I had come in on a Saturday. Nope. It wasn’t in the computer. After 3 or 4 of these exchanges and some fast clicking by my doctor on the computer (she’s a young doctor and appears computer savvy), she finally found the record and said that they had changed the way the computer screen was laid out, so she couldn’t find it. I’m glad it was just my ear that was the problem. This isn’t the first time I’ve had an experience like this where I find myself in the position where I’m defending my knowledge of my health history in my doctor’s office with a computer. As a patient, it makes me fearful, in addition to being frustrated and unhappy with my doctor when I’m not feeling well in the first place.

  • http://www.mmfemr.com Leo

    I agree with the comments in this article. EMR’s need to perform like the cloud systems we have become accustomed to (facebook, twitter, kevinmd ;-). The initiative of the Open EMR is addressing these issues, especially the interlinking within patient records.

  • LynnB

    All this will come someday, I am struggling to care for patients today. Documentation AFTER this visit takes longer than the visit time (which includes documentation that I do real-time)

  • MikeG

    Why are EHR systems immature?
    I am a health care IT veteran (20+ years) and consultant now focusing on designing solutions for interconnecting EHR systems using relatively new technology called Health Information Exchange. The answer to this question is far more complex than the article suggests and most technology people working outside health care really do not understand the complexities. I came into health care IT from other industries and was appalled at the technology when I first started. It took me a while to understand why things as they are. Let me start by making some comments:

    1. The current status of the health care marketplace and how medical services are paid for is the biggest reason for EHR systems being immature.
    2. A medical student is not mature enough in their understanding of the dynamics of the health care to be a valid opinion.
    3. The complexity of health care data is far greater than any other industry I have been involved in and may be more complex than any other IT area.
    3. The argument that the market is too small is completely invalid. The example of there only be 7500 hospitals breaks down because the cost of EHR systems runs from about $5M – $50M depending on the size of the hospital. Assuming the average is $10M that makes the market for hospital-based EHR systems about $75B. Whereas a video game only costs about $50. So even if you sell your game to 5% of all Americans you would only make $625M. And that does not include the number of EHR systems you might sell to physician offices which number approximately 500,000 in the US alone. EHRs for physician offices only cost about $50K-75K but you get the point. The market is NOT small.

    So why are EHR systems so far behind? The answer really is that the payment system is messed up. Vendors go to where the money is because they need to make money to be profitable. The payment system has always paid doctors and hospitals for what they do (services they perform). They are not paid to share information with other providers, as a matter of fact they are disincentivized to share data. When they perform another test they get paid, when they use one from another doctor they are not.

    Secondly departments within a hospital (such as radiology and laboratory) have never had an reason or incentive to share data. Each purchased its own system and implemented it for its own reasons. Sharing data cost more, and you got paid less! Oops.

    Specialty systems are complex. Designing a system for medical images is a big deal. Designing the database of this system so it would allow you to share the data was often an afterthought because hospitals did not want to pay extra for something that would cost them money.

    Think of an EHR as a big database with a user interface of all the data generated by all other systems which are effectively incompatible with one another. Going to your gaming analogy this would be like creating a new “game” that would be a combination of an XBox, Wii and PlayStation games that would look and work well. The three gaming vendors would have no incentive to do this nor the standards to combine these together. The designer would have to use the lowest common standards or perhaps create some custom code to make this work. This is somewhat similar to what EHRs are up against except that the data is very complex, critical to peoples health and has strong privacy and security requirements.

    So while I am also not yet excited by the current state of EHRs I tend to give the EHR vendors a break.

  • John, hospital I.T. geek

    A big issue is health records are “provider centric” (be that physician, hospital, pharmacy, etc) and not “patient centric”. Business software products, whether healthcare or manufacturing, are written to automate an individual business. Does software used at Ford talk seamlessly with GM or Toyota – no.

    Changing a provider’s software to be patient-centric and not provider-centric is a huge task. Plus, don’t forget healthcare generates ENORMOUS amounts of data compared to other industries.

  • Justin

    Let me sell your medical data the way google and facebook sell your other data and I can build you an amazing system.

  • http://ducknetweb.blogspot.com MedicalQuack

    Ok how about 2 cents from someone who wrote an EMR a few years back, me:) Oh granted we have blazed many trails since I was writing and everything integrating with the web now has added it’s share of programming updates, namely security as it’s not all stored on the office server any longer.

    There is a project that has been around for almost 3 years now from CodePlex, the open source arm of Microsoft that is called the Common User Interface, in other words the screens and front end of the program that we use. Is that not a grand idea to have all system similar/same and all vendors could still maintain their working functions at the back end and other departments? Some developers have worked with it as well as the NHS and it’s dynamic and uses Silverlight, lots of dragging and dropping to keep the “click wars” down to a minimum as well as integrate PACS imaging too, pretty intuitive. I remember working with my beta office all about working around to keep the clicking down to a minimum, a valid issue by all means and did my best. Even back in the early days, you didn’t have to make everything gray either, I used images and screens that were user friendly, but most do not, but you can have both if you work at it.

    I chat with an MD/Developer in New Zealand and he did some work with it and I have a post that shows a bit how he developed it.


    Back in 2008 I saw this getting very complicated and look where we are today. Sometimes competition creates some of this too. The post below has a video about how the interface would work.


    I agree with some of the comments above here too though that Epic (system at Kaiser) is about the best out there, they put a lot of work into it for sure. I interviewed a doctor from Long Beach Memorial on their transition and he states that they were very open for suggestions and idea and even contributed back to Epic on ideas they could use too, so a good EMR is a 2 way street for sure.


    Technology is giving everyone fits too as just as soon as we digest one solution there’s 10 more knocking at the door too, and that challenges everyone, so I completely understand the argument of just wanting a system that works and helps you get your record keeping done and I think more collaboration, not competition here will be a real key to get us there.

  • http://www.SurgRad.com/ Allen Bates

    There is an interesting article on EMRs, and a rather fresh perspective on them, at http://www.SurgRad.com. Some of the other EMR providers (i.e. Practice Fusion) have also commented on the article. It is published in the peer-reviewed Journal of Surgical Radiology.

  • lhathorn1

    He may be “just a medical student” but he knows a sorry, tedious program when he sees one.
    Our EMR , which shall remain nameless, is exactly as he describes the ones he used. Gray, boring, no good GUI, no windows, you just have to know all the little places where things are hidden to make any progress at all. It takes 10-15 clicks to fill one rx. The SOAP note screen is ridiculously slow and tedious.

  • pommehomme

    I have been interested in this problem for several years, and became much more focused when I moved my office recently and had over 200 boxes of paper records (from my predecessor) shredded (they dated to 1966.)
    In his book “The Healing of America,” T.R. Reid describes health care systems in other countries. It is worth reading the description of the French universal EMR. This is encoded on everyone’s health insurance card. Progress notes are entered electronically and the system is also used for electronic billing. It seems to be used by all MD’s in the country.
    I think this would never fly here because it is too centralized. Our free-market approach encourages small vendors to build proprietary systems that are DESIGNED not to communicate with each other. Price tags, maintenance and service contract costs are ridiculously high.
    Open source EMR’s may be the answer but they are still clunky compared to other computer based systems.
    Until there is a consensus that single payer healthcare is the only reasonable answer (eliminating the for-profit administrators that skin off 10-30% of each health care dollar, without adding anything to the provision of health care) we are doomed to this EMR hell.

Most Popular