Why this doctor loves electronic medical records

A recent post in the Wall Street Journal’s Health Blog noted that a study found electronic medical records don’t improve outpatient quality.  The authors of the Archives of Internal Medicine article, Electronic Health Records and Clinical Decision Support Systems, correctly points out that we should be skeptical and “doubt [the] argument that the use of EHRs is a “magic bullet” for health care quality improvement, as some advocates imply.”

This should surprise no one.  Were we that naive to think that simply installing health information technology (HIT) in the medical field would generate significant improvement in outcomes?  Does simply installing computers in our classrooms improve educational test scores?

Of course not.

The excellent commentary after the article makes some plausible reasons why the clinical decision support (CDS) didn’t seem to improve outcomes on 20 quality indicators.  First, it isn’t clear that the CDS implemented across the various doctors’ offices and emergency rooms actually addressed the indicators studied.  Second, the data studied is already dated (from the 2005 to 2007 National Ambulatory Medical Care Survey), a long time in technology terms (iPhone first debuted in 2007).  The authors of the original article also point out that there is some evidence that institution specific use of CDS actually improves quality.  Whether this can be scaled to the national level is the question.

In other words, it isn’t just that perhaps CDS failed, but rather the robustness of the system was inadequate, that doctors failed to use them, or just as importantly patients were unswayed by the doctors reminded by the CDS to do the right thing. One of the 20 quality indicators studied was in fact the appropriate antibiotic use in viral upper respiratory infections.  As most people know already, there is not typically an appropriate antibiotic to use for a virus.  It’s a virus.  It does, however, take good bedside manner to inform and educate an ill patient!

As someone who has had the benefit of a robust electronic health record since the spring of 2006, I know I’m incredibly lucky.  In an April 2009 New England Journal of Medicine article, only 4 percent of doctors nationally have a fully comprehensive EHR that I take for granted daily. Only 1.5 percent of hospitals have a comprehensive EHR, which I also have access to.  CDS is also an incredibly helpful tool and an excellent reminder to provide the right care every time.  It is a safety net.  Understandably some EHRs aren’t that good, the CDS is clunky, and certainly the one I use is good, but not perfect.

The real issue isn’t finding a perfect EHR, but rather how do we address the culture of the medical profession. There is something still heroic and mystical about a lone doctor, independent, smart, and getting the job done.  Indeed, to get into medical school, one has to be self-motivated, persistent, and determined.  Why on earth would we need a computer to help us?

Frankly, because it makes us better doctors.  CDS frees up time and mental energy.  I don’t have to remember the latest guidelines on immunizations, repeating blood work, or treatment of illness like coronary artery disease, congestive heart failure, and hyperlipidemia.  Most of these diseases are well understood and often under a protocol, something known as precision medicine, a term used by Harvard Business School professor Clayton Christensen.

Now I can focus on if the patient in front of me is an exception to the protocol as well as thoughtfully diagnose and treat their ailments which don’t fit any protocol (cognitive medicine) because science hasn’t evolved to that level of understanding.  I’m a big believer in the history and physical exam and how the use of HIT can make care more personal.  Having real-time access quickly and reliably to medical information and data 24/7 is important to make this happen.  Instead of hunting for lab work in a paper chart or trying to find a specialist’s consultation, I can access the information I need rapidly and focus on the patient in front of me.

Sadly, however, many doctors don’t feel the same way. Perhaps it is a generational thing. Perhaps it is because their EHR is inadequate.  It might also be, however, our training and tradition which limits us from improving.  If anything, the medical profession needs to emulate ourselves after the aviation industry where technology is used to support decision making and make pilots and flying even safer and better.  We are where our aviation colleagues were in 1935 as noted in Dr. Atul Gawande’s New Yorker piece, the Checklist.  Because, really, CDS is essentially a checklist.

In the situation where a patient doesn’t fit CDS, then we get to do what we do best and that is use all of our training to get a patient better.   HIT, EHR, and CDS are things the next generation of doctors must accept that will make the care we provide more personal than ever before.  In the end, that is what patients really want.

This is why I love my EHR so much.

Davis Liu is a family physician who blogs at Saving Money and Surviving the Healthcare Crisis and is the author of The Thrifty Patient – Vital Insider Tips for Saving Money and Staying Healthy and Stay Healthy, Live Longer, Spend Wisely.

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

  • Vox Rusticus

    It frees up time? Maybe in some ways, but it seems to eat up much more time in other ways, and it costs enormously in the process.

    The products private offices are being asked to buy are not living up to the hype at all. They are a very long way from being what they were promoted as being. For the most part, they are electronic notekeepers, inconveniently tied to proprietary systems that sit in isolation from thousands of other systems, hostage to the companies that produce the software. This was a boondoggle orchestrated by the IT industry largely for its own benefit but sold as a time-saving, health-improving, risk-reducing product. It hasn’t come close to living up to those promises.

  • http://briarcroft.wordpress.com Emily Gibson

    I completely agree with Dr. Liu. Our clinic is in its seventh year of using an EHR, and not only does it improve our efficiency, our interoffice communication and patient flow, as well as increase consistency in documentation and patient care standards, it is unparalleled in allowing secure communication with the patient. Most of our advice, feedback about lab and xray reports and patient education materials are delivered to the patient via a secure web portal and automatically recorded in the EHR. The patient not only can see the chart notes and lab work we send to them but can comment and provide feedback electronically, all recorded in the chart.

    That ability to have a recorded real time electronic dialogue with a patient makes a huge difference in the quality of documentation (rather than the time consuming redundancy of summarizing a phone conversation) but the patient becomes a partner in the documentation of their history.

    I would go back to paper kicking and screaming…

    • http://www.davisliumd.blogspot.com Davis Liu, MD

      I completely agree. Would not go back to paper either, ever.

      • http://briarcroft.wordpress.com Emily Gibson

        We “rediscovered” the hand-written medical record for a long 2.5 hours yesterday in our clinic when the entire university network went down in a windstorm, taking our server with it. In a blink of an eye we had twenty patients in the waiting room and had no idea who they were or who they were waiting to see, and 15 rooms of patients with no electronic charts to give us their history. We do have a protocol for this rare event, and made do with paper, all of which was scanned into the EHR when it came back up. The outage reminded me how our internal clinic communication system (orders, consults, messaging, and encouragement of the staff on a very busy influenza day) happens with far more efficiency than if we were limited to sticky notes, paper charts and verbal orders.

        • pcp

          “a long 2.5 hours yesterday in our clinic when the entire university network went down in a windstorm, taking our server with it. In a blink of an eye we had twenty patients in the waiting room and had no idea who they were or who they were waiting to see”

          That’s a very convincing argument for going electronic.

          • http://briarcroft.wordpress.com Emily Gibson

            Just like the common occurrence of lost or unavailable paper charts is an argument for staying with hard copy records?

          • http://www.davisliumd.blogspot.com Davis Liu, MD

            Paper system far worse. Missing or misfiled charts, lab work, specialty notes. Illegible handwriting by colleagues and staff. No ability to do drug-drug interactions.

            We took for granted this is how medicine should be because that is how our attendings were trained and practiced. The world is a different place. We communicate differently (note the blog here) and are more connected than ever.

            The medical profession must move towards the 21st century and adapt as well. We can disagree the incentives (or penalties) and who is responsible to make it happen, but first we must have a mindset that despite some of the imperfections of HIT, the benefits are far better.

          • pcp

            20 years in practice, one lost chart (found the next month, it only had five pages). Illegible writing not tolerated (all applicants take writing test). Trained staff gets all outside notes in advance. Been using hand-held drug-interaction monitor for 15 years. It works.

          • pcp

            Addendum to previous post:

            I’ve also been permanently exempted from any type of quality review by the largest insurer in our state, because they could never find any deficeincies in my charts. First practice that they did that for.

            All this is not to boast, but to give some anecdotal evidence that the quality of medical care provided is completely independent of the charting medium chosen.

      • horseshrink

        Never say never.

        Computers and software are less reliable than paper. Fail points are countless.

        Even basic requirements we take for granted, e.g. access to electrical power, are susceptible to myriad, unexpected disruptions. (… and what about the sun? … We’re headed into another solar activity cycle, raising worries of 1859 & 1989 phenomena again.) And when the generators are on during a prolonged power outage … which should get preference … the surgical suite or hospital-wide IT functions?

        Time to hardware & software failure/obsolescence is much shorter, requiring frequent shifts of data from platform to platform over time … additional opportunities for data loss/corruption.

        And I don’t know about you … but if I had the power and stature of a Steve Jobs, Bill Gates, Warren Buffet, or Barack Obama, I wouldn’t want my highly sensitive medical information in any computer that had a connection to any network. As a Joe Schmoe, I’m not terribly worried about hackers targeting my information. I’m a nobody whose information is not uniquely valuable. But, as a high profile person, my information becomes the constant target of a new breed of paparazzi and spies … hackers. So … safer even than a computer … paper. No hacker can reach it.

  • pcp

    ” Were we that naive to think that simply installing health information technology (HIT) in the medical field would generate significant improvement in outcomes?”

    But that is EXACTLY what we have been told, by two Presidents and everyone on down from there. Those of us who begged to differ are labelled “Luddites” and worse.

  • Marc Gorayeb, MD

    Electronic CDS, eh? My CDS is a collection of reference texts and well-outlined printed treatment protocols for certain specific problems in the doctor’s work room. I also highly recommend the Sanford guide and some of the Tarascon Pharmacopoeia guides for adults and for pediatrics. They are kept up to date, and provide the latest best practices guidelines. I must admit that I have to use my glasses to read the fine print, but it’s much faster than getting on the computer. What am I missing?

    • http://www.davisliumd.blogspot.com Davis Liu, MD

      I’ve used both Sanford guide and some of the Tarascon Pharmacopoeia guides in medical school and residency. Not sure, however, I agree that it is faster than the computer.

      Ask any student today – encyclopedia / dictionary or internet search?

      I think if you feel that the computer system it too awkward to use or that there is a learning curve, then that is a different issue. Doctors in the 21st century will need both speed and accuracy and that is what HIT can offer. But you are correct, CDS in your work room is one option. I prefer mine because I can provide the same consistent care whether in my office or if working out of a colleague’s office – my information is available 24 / 7 wherever there is an internet connection.

  • http://www.healthscareonline.com Richard Young, MD

    Dr. Liu,

    I know I can’t remember everything so I support memory aids that can exist electronically or on paper.

    As a family physician you should at least question the enormous investment in EHRs pushed down our collective throats by well-meaning but ridiculously optimistic pundits and bureaucrats. If EHRs were such a naturally effective tool, why didn’t the market for them explode the way i-phones have? Why did the feds have to come up with a list of financial carrots and sticks to coerce hospitals and doctors to buy these systems?

    It’s because the bang for the buck of EHRs is awful.

    I don’t need a reminder to tell me that my patient with 4 chronic diseases and 3 new symptoms might also be depressed. I need to be paid for the time it takes to deal with more than one or two issues per visit so I can thoroughly explore the possibilty she is depressed. Unfortunately for family medicine, national healthcare resources that could have gone into meaningful payment reform have been sucked into the great EMR black hole, and family medicine has been hurt in the process.

    • jsmith


    • http://www.davisliumd.blogspot.com Davis Liu, MD

      I agree that primary care, whether internists or family physicians, are not adequately recognized for our cognitive specialty and are at a disadvantage in a reimbursement system that is fee for service.

      Fixing how doctors are compensated for their time however is unrelated to the funding of HIT. I agree that it is a conversation the nation should be having – how to maintain and sustain a robust primary care workforce.

      • pcp

        I disagree. I think the amount of money and effort that has been put into promoting the use of EMRs was definitely done at the expense of rebuilding our primary care physician work force. The cart was put way before the horse, and the AAFP was the worst offender in this regard.

      • horseshrink

        Currently, HIT is more expensive over time than paper records.

        EHR related productivity losses are real.

        If payroll is not met, rent paid, utilities paid, insurances paid, equipment purchased, etc., etc., , there is no practice.

        Remembering how doctors are compensated for their time is highly relevant in the real world of medical practice.

  • http://www.emrandhipaa.com EMR and HIPAA

    “HIT, EHR, and CDS are things the next generation of doctors must accept that will make the care we provide more personal than ever before.”

    The interesting thing is that we don’t have to worry about the next generation of doctors accepting HIT, EHR and CDS. The medical students that I talk to now are so digitally proficient that they can’t imagine practicing medicine by paper. It wouldn’t matter the cost to most of these individuals. They’ll take whatever cost to not have to read a scribbled on paper chart, search for a lab result in a 3 inch paper chart, or do a Where’s Waldo trying to search for the missing paper chart.

    Not to mention as one med student told me, I can type a lot faster than I can write. So, I’m all about documenting electronically.

    The challenge is getting these students out of medical school and into positions of leadership where they can drive the EHR adoption.

  • jsmith

    You might have a good EHR. Great.If EHRs work in your practice, use them by all means. Few have a problem with that. But a lot of us have lousy EHRs that suck time and money and don’t improve pt care. What we have a problem with is EHRs being forced down our throats in the name of illusory quality improvement and cost savings.

  • http://www.electronicmedicalrecords.org Carl from Electronic Medical Records Group

    Fantastic, fantastic article. It’s always great to hear of a doctor benefiting (and being outspoken about) the product we dedicate much of our lives to. In fact, care for a job as a sales writer? ;)

    Seriously, though, I think you make a point we’ve been trying to make for some time: EMR isn’t “automatic patient care improvement”, but does create a system where doctors have the time and wherewithal to focus better on what matters the most: their patient. While productivity and return on investment are a lot of our sales focus, those are just bottom line things.

    Thanks for taking a look at some of those “invisibles” that make EMR software so effective for a lot of MDs.

    • jsmith

      The reality is that EHRs waste time and decrease productivity and take us away from pts.

  • Max

    When they stop selling it as time-saving, cost-cutting, and quality-improving and just tell me they will use it to audit my GUR and my charts in the middle of the night while I’m sleeping or throw me in jail if I don’t go electronic, then I will buy one. Just say that and I will buy it. Sheesh.

  • pcp

    Translation of the original post:

    There is no evidence, either from the U.S. or other countries with longer and more widepsread HIT experience, to support any of the claims made for EMRs, but I think they’re fun to use, and so every doc should be required by law to do so.

    • Fam Med Doc

      OMG that’s funny. Strong work.

  • Joe

    Why is it that problems related to EMRs are considered anecdotal, but positive individual reports are not called anecdotal? There is a clear bias in most EMR supporters to start from the point that HIT improves patient care. This is confirmed by the prevalent characterization of anecdotes.

    The HIT supporters use the same arguments when anyone reveals a study that shows no benefits … old data, incorrect end points to the study, etc etc., HIT believers seem incapable of admitting that any patient is at risk from EMR technology or implementations.

    It is certainly true that HIT has helped patient care in some situations and hurt in other situations. Very little has been done to figure out why.

    Let’s stop with the talking points and start with the work of getting HIT right.

  • horseshrink

    An EHR allows simultaneous chart access by multiple users and is supposed to improve what I can do with the medical record’s information. Analyze, flag, alert, search, support decisions, homogenize care, etc. Nice …,

    SO LONG AS the power is on, the computer is functioning properly, the software is running correctly, and the network is up. Hopefully a hard drive (or something else) hasn’t crashed, or malware hasn’t made it into the system, the last Windows or EHR (or other) update didn’t break something, and backups are occurring properly.

    The EHR is down? So sad, too bad. There are still patients in the waiting room and in the hospital. What now? Losing one paper chart is one thing. Losing access to all the records is paralyzing – especially as can happen at a state level (I’ve experienced this repeatedly.)

    And expense … several years ago my life-long geek bubble was burst by a former hospital administrator from a large hospital chain who enlightened me re: the prodigious cost of creating and maintaining an EHR.

    Previously, for my own clinic’s medical record, I used H.S. educated people (or, during the summer, their kids who were still in H.S.) to assemble and file my paper records. No can do with an EHR. Need more educated (= more expensive) tech savvy people to keep the beast alive. Or me … and I’m even more expensive. And expensive computers don’t last as long as paper records & file cabinets. Gotta keep replacing them and moving the data. Oh … and paper records still work when the power goes out.

    And then … there’s productivity. Oops. Absent the expense of a scribe, most docs take a tangible hit on the number of patients they can see daily once they transition to an EHR. EHR’s simply aren’t written by clinicians for clinicians. They are written by back-office coders for sale to administrative personnel. Human data input/interface issues are not considered sufficiently. Yet, that’s a critical choking bottleneck for EHR adoption from the clinical end-user’s perspective! Like it or not, docs can see more patients by handwriting their notes during patient encounters. If this isn’t solved, docs will dump their new EHRs to paper and use the equipment as door stops.

    My skepticism is especially fueled by current daily experience with a bad product deployed at the state level. This product is an icon of all that should be avoided in an EHR product … clunky, non-user friendly work flow, blind loculation of all information, poor data input and data review interfaces, no data analysis (at all), no dashboarding, cobbling of unfriendly, competing products, sloooooow network, periodic crashes, legions of forced response, bureaucratic radio buttons ….

    And the state persistently pours good money after bad to maintain the current atrocity, because change would be too expensive. Why?

    Data migration cost + new product cost = prohibitive cost.


    I believe EHR consumers would benefit if the data migration expense was removed. This is possible if data constructs are standardized.

    Example = the WWW. Web page data constructs are sufficiently standardized to allow use of a slew of different browsers … IE, Firefox, Konqueror, Chrome, Lynx, Safari … I want to check out a different browser? No problem. Install and use. The whole internet doesn’t have to change its configuration because I decided to use a different browser.

    The same should apply to health record data.

    Then, vendors won’t be able to lock clients into a product via an idiosyncratic database (unless that’s what a client actually wants.)

    When it’s as easy to change EHR products as it is to change browsers, EHR companies will shift their entire development strategy to discover and target what docs ACTUALLY want to use while seeing patients. And lubricated competition will also drive down prices.

    • pcp

      “I believe EHR consumers would benefit if the data migration expense was removed. This is possible if data constructs are standardized.”

      I agree with all you’ve said, but, now that docs are basically being forced to buy the existing products, do the vendors have ANY motivation to remove the restrictive features from their products?

      • horseshrink

        One hopeful step by the ONC is a possible standardizing of data meta-tags for Health Information Exchange. (I submitted a comment re: this to the ONC during their comment period re: this.)

        If data constructs are standardized, an industry for data migration may evolve to move legacy data to standardized format.

        • pcp

          “an industry for data migration may evolve to move legacy data to standardized format.”

          Just another way for EMRs to “save” money!

  • Sue Wolver, MD

    I loved this post and agree wholeheartedly. As the MD co-lead of our ambulatory EHR implementation to a large academic hospital’s clinic system, our mantra has always been: The success of this implementation is 20% about the tool and 80% about the operational side. The success of using the EHR “meaningfully” follows the same formula. Putting in the greatest EHR with wonderful CDS will be less than fully successful without the proper leadership support and operational foundation.

    • horseshrink

      Try convincing a surgeon that what she uses for her work is only 20% about the tool. That’s like saying “It doesn’t matter how sharp the scalpel is … it’s all in how I operationalize it.”

      EHR tool design is premature when it comes to the preservation or enhancement of work flow – where our professional raison d’etre exists. If this isn’t fixed, EHR products will be tried and abandoned by their disgruntled subjects.

      If the EHR tool was already “sharp” enough, spontaneous adoption would render federal policy useless. The phenomenon of nearly ubiquitous smart phone adoption is powerful evidence of this – the feds did not have to wheedle and whip smart phones into our hands. That the feds feel a need to herd us into EHRs with carrots and sticks is not an endorsement of EHR quality. It’s an indictment against it.

      I resonate with Zenfire’s post below.

  • Zenfire

    If they were that great, you wouldn’t have to legislate people to use them. If they really saved time, lowered costs and improved the outcomes of my patients I would jump at them. The fact that legal and social pressures are brought to bear is evidence that they currently suck.

    We are used to high tech in the rest of our lives, but the EMRs currently available are horribly designed and coded. One day I expect there to be a decent EMR, but as of today, nothing that I’ve shopped for was even barely functional.

    I type faster than I write, I’m young, everything else in my life is electronic. I’m not a technophobe. I’m just a savvy shopper.

Most Popular