Why IT is the core of the healthcare renaissance

Why IT is the core of the healthcare renaissance

Warning!  I am a practicing doctor who sees real patients using an electronic medical record (EMR).  My sole agenda is to provide the best patient care.  I have no financial stake in information technology (IT).  However, unlike the editorial board at the Wall Street Journal, Mr. Stephen Soumerai of Harvard or Mr. Ross Koppel of the University of Pennsylvania, I have actually used digital patient records for over a decade and I have news for them;  EMRs work.

In a reactionary opinion in the WSJ entitled “A Major Glitch for Digitized Health-Care Records,” the authors expanded to the point of silliness the conclusions of a review of healthcare IT by McMaster University.  The McMaster analysis abstracts data from 36000 studies over five decades of healthcare IT and concludes that computerization has yet to save dollars nor improve health care.  WSJ editorialists proposed that the concept of a common medical database has “already failed” as is “common knowledge.”  While they portend to “fully share the hope” in the success of a computerized healthcare system they express doubt as to “why are we pushing ahead to digitalize.”

The question is so ridiculous as to barely require an answer.  We are pushing ahead to digitalize because the healthcare industry, which is 18% of our GDP, is the last major industry to go electronic.  Despite how critical medicine is to our citizen and nation’s vitality, health care is most often documented with paper and ink.  Can you imagine any other industry where this would be acceptable?  Would you go to a bank where they use a hand-written ledger?  Travel on an airplane without GPS, fly by wire technology or a minimum of three computers?  Do you yearn for rotary phones?  Credit cards left paper money behind decades ago and will soon move on to the next phase, pay by smart-phone. The world is digital and one of the core problems with medical care is its failure to follow.

The average doctor writes his notes on parchment and scribbles orders on contact paper.  He wastes time writing prescriptions by hand that cannot be read and will produce unneeded drug interactions and side effects.  The data on billions of health care events cannot be mined, monitored, analyzed or improved, because it is not digital.   Millions of hours are wasted, billions of dollars vanish and tens of thousands die because of preventable medical complications, the result of massive variation in quality and safety.  We are doomed because we cannot access or evaluate most medical care data; “If you can not measure it, you can not manage it.”

Taking health digital is key to fixing and affording care.  Standard, unified medical records will significantly decrease the risk of providing unneeded or dangerous medical care.  Massive efficiencies will result by reducing duplication, speeding communication and reduction in waste (and fraud).  Critical improvements will follow the use of guidelines to study clinical databases and drive quality. This means that whether one lives in Manhattan, in the mountains of Tennessee or potentially deep in Africa, the finest care will be possible.

How do I know this to be true?  Our practice of seven doctors and three nurse practitioners was an early EMR adapter.  We put in our first basic system in 2000, upgraded three times and have been fully electronic for four years.  This has resulted in marked efficiencies and obvious quality improvement. Encrypted electronic records cannot be lost, are unlikely to be stolen and are always available from anywhere.  Ordering tests is instantaneous, as is reviewing results, organizing treatments, scheduling appointments or communication with outside health providers. On the cost side we reduced non-clinical staff by more than 50% and in an account receivable analysis our billing cycle dropped more than 60% and bad debt fell to low single digits.

Today in the office, I saw four new patients.  Without assistance of clerical staff and without leaving my desk I reviewed their entire surgical, laboratory, pathological, and radiologic records.  By the time, I entered the room to meet each of them a significant part of their medical history was entered into our EMR, based on outside records, so that the care and observations of previous doctors was not forgotten. Half way through each visit documentation of their history and physical exam was complete, leaving more time to talk with each patient.  Tests were ordered, treatments scheduled, disability letters printed, medications e-scripted, instructions created and follow-up appointments setup.  The patients were given codes to electronically access their records from home.  Letters were sent online to referring doctors, as well as any clinicians we were consulting in that patient’s care.  Billing was complete before each patient got to the parking lot.  Such is the power, efficiency and quality of electronic medical records.

This is just the beginning.  Although EMRs now provide assistance with basic medical care, such as scheduling flu shots, identifying drug interactions, and health screening reminders, future systems will use academic information to assist the doctor in making diagnoses and planning treatments.   Seamless with the EMR will be computer augmentation to create differential diagnoses and recommend treatment alternatives.  In oncology alone there are almost 50,000 articles published each year; Artificial intelligence integration with the clinical EMR will help every doctor penetrate that massive database on a continuous basis as it applies to individual patients.

So, why does the McMaster study not show this obvious benefit? It comes to four factors. The first is the “five decades” of study reviewed by the Canadian authors.  Since most doctors did not start adapting EMRs until 3 years ago, that leaves 57 months of irrelevant data.  Second, we have not reached the critical mass to achieve broad system efficiencies, as hospitals and doctors are still figuring out how to incorporate the technology into their daily practice, and less than 50% of health care providers have converted to EMRs.   Third, we do not yet have a universal common database for medical records. This is a complex technological step, which has been achieved in major industries such as banking, but still must be assembled in medicine.  Finally, as was correctly stated in the WSJ op-ed piece, present EMRs are cumbersome, immature, and several generations short of perfection.  However, these are expected problems when implementing disruptive evolving technology in the complex changing health market. Challenge is a weak argument for giving up and bringing back the fountain pen.

I am not certain what the goal of Sommeri and Koppel was in writing this piece, they offer nothing but “hope.”  The WSJ has been a strong supporter of business technology leading the drive towards quality, and has long recognized the positive contribution of IT to industry.   Those of us in the trenches, putting EMRs in place, ironing out the kinks, know that electronic medical records are now and they are the future.  With IT at the core of the healthcare renaissance we can make medicine cost efficient, producing quality second to none.

James C. Salwitz is an oncologist who blogs at Sunrise Rounds.

Image credit: Shutterstock.com


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

  • southerndoc1

    “Healthcare renaissance”

    I thought it was a transformation? But it’s a renaissance? OK, got it.

  • RhythmDoc

    Yes, but it’s not irrelevant that most if not all EMR’s are “cumbersome, immature, and several generations short of perfection.” I also am a practicing physician who uses electronic health records every single day – in fact, I have personal experience with several EMR environments. And, lest you think I’m a computer-impaired elder statesman, I’m young enough to only barely remember a world before personal computers and something of a technophile to boot. I grew up playing Halo. My iPhone, loaded down with apps both medical and otherwise, is my constant companion. I deal with high tech implantable devices every day. And every EMR I’ve ever worked with *makes me a worse doctor*.

    Let me clarify that – not in a million years would I want to go back to a world where records were on paper. The idea that I couldn’t offer the patient and informed opinion because no one can find the chart strikes me as insane, and the idea of sifting through dozens of pieces of paper to trend a patient’s labwork is no better. Computerized information management is here to stay, and do you know why? Because the products work. They help. They function – and most of them are intuitive enough for me (admittedly of above average computer-literacy) to use with minimal or no instruction. I’d cheerfully spend money for them, because they make my life easier, and they help me take better care of the patient.

    BUT…EMR-generated documentation, in the sense of “generate a consultation or progress note” is an entirely different ball of wax. Oh, I’ve learned to deal with it – largely by evading most of the features that salespeople like to show you. I’m lucky enough to be in a field in which physical exam is largely irrelevant, so I don’t have much problem with my physical exams (since they don’t matter very much anyway). I never, ever use any pregenerated templates; I free-text type the history and my assessment (I’m faster typing than I am with Dragon – I’ve timed myself – and I think better; your mileage may vary). I ignore the computer recommendation on what level of service to bill (since it would result in blatant and probably fraudulent upcoding.) And the fact that the output is *terribly* formatted and difficult to interpret for the human eye isn’t too big of a problem for me, since I have a largely captive referral base, being an employee of a large multispecialty group; as long as the note is intelligible it doesn’t matter if it’s pretty, since I don’t really need to impress anyone. But you know what? I could do all that, better and faster, in my last practice, where we hand-wrote everything and then scanned it in as a PDF. And my handwriting is extremely legible, since I’ve always felt it was unprofessional and dangerous for a doctor to have illegible handwriting. Any time I see one of those seven page level 5 notes generated by mouseclicks rather than text? I ignore it completely, whether it comes from the patient’s primary care provider, from another specialist, or even one of my own partners – since any given piece of data has about a 50% chance of being true. I start over, since I can’t trust the data that was previously gathered.

    And those “advanced” features? Don’t make me laugh. CPOE, as implemented in my hospital, is a disaster. I can say with stone cold certainty that I’ve personally created more near-miss medical errors (order on wrong patient, medication not started when I want it, forget to stop something that I intended to stop, etc., etc.) since the implementation of CPOE in our environment about six months ago than in the previous ten years. By, I’d guess, about a factor of five – although, thankfully, as far as I know no one’s actually been hurt by any of them, due in part to the vigilance of our nursing and pharmacy staff, who are under standing orders to call if anything looks wrong – because it probably is. And clinical decision support? Not once has any CDR or automated alarm that came up helped. Not even a little. Not *once*. (On the other hand, the various apps on my iPhone are great, as are some of the online tools like UpToDate.) And the tools down in the pharmacy are actually worse in terms of drug interactions than their previous system.

    Now, obviously some of this is the fault of a particular product, designed badly, implemented clumsily, and then not optimized for our particular environment (or for this decade, for that matter) because of a ponderous hospital bureaucracy. But that’s the point – these are fundamentally immature pieces of software, that are largely not ready for prime time, particularly the enterprise-level ones. They’re not polished pieces of consumer software that function transparently and do the job without having to think about it. The objection I have to the EMR boosterism at the federal level isn’t that I don’t agree the medical world needs to digitize. Of course the medical world needs to digitize. But the HITECH Act of 2009, by pushing immature pieces of software via coercive financial pressure (rather than waiting until the EHR vendors develop a product that physicians might actually *want*), has done an incredible amount of harm to that process, by forcing implementation of systems that are obsolete long before they get to the clinician, and that actively impede clinical care. And it’s wrong.

    • kjindal

      geez sounds just like the crappy system I have to use (some garbage called “ADL”) – so many many near-misses, esp when (frequently) CPOE orders do not match the EMAR! or when the computer says I changed an order on a weekend when I wasn’t even in the facility!

      Yup it’s a revolution alright, kinda like the french revolution in terms of fatalities (kidding, but you get the point).

    • Tom Mendelboim

      Why do you think EMR software are so complicated and hard to use? Why is it so difficult to implement and support not to mention migrate from one system to another… Most vendors have dozens if not hundreds of doctors working hand in hand with designers. I come from manufacturing and while it’s not perfect, the Enterprise Resource Planning (ERP) is a similar concept from software perspective. It took years to get it right but most applications are actually helping companies save, increase revenue and overall be more productive. I’ve been reading lately a lot about EMR and I must say I’m surprise to learn all of these nightmares… To me it seems like none of the EMR vendors are ready to serve the right doctor community and it will take another 5 years before something right comes out…

  • buzzkillersmith

    There’s so much nonsense in this post that it’s hard to know where to start. Just a sampling:

    First, you have an N of one. Your personal experience is of zero relevance. What matters for society at large is the aggregate results of a large number of practices.

    Second, in your fifth paragraph, you state that IT will fix care and make it more affordable. Hell, I could say that cancer is going to be cured tomorrow. You’re guessing, you’re making things up. There is a significant possibility that EHRs will fail to do these things. Fortunately your blog post will have been forgotten by then.

    Third, you attack a straw man. Most of us will not go back to paper. But we are wise enough to know that EHRs might be helpful in some limited ways but also come with a large number of downsides. Whether they will revolutionize medicine is an open question. I don’t have the answer to it, and neither do you.

    • http://twitter.com/Alv_Ignacio Ignacio Alvarez

      As a UX Designer, I had the same thought as you. We cannot base a research on one particular individual who has a good experience with his particular EHR and his particular use of it.

      Although it is not significant for a statistical point of view, it is interesting to understand it. As we read this post, we can understand that it took him many years of manipulation and updates for him to master it. Therefore, he probably uses routine tricks that, if understood by EHR vendors, could help building the next generation of EHR.

      I also think that by declaring that it will « revolutionize medicine » you put data and administration has top priorities of the healthcare system leaving the patient beside. Do you even know what your patient thinks of you using the EHR during the consultation? How can a patient trust a doctor that was on his computer for most of the consultation time? Revolution, maybe, but is really a good one? Are we taking the right step?

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

    I don’t know why we need to turn something as mundane as using a computer into crusades and renaissances….
    RhythmDoc is right. Medicine is becoming computerized… slowly (and wastefully, as usual in health care). So what?

  • wiseword

    I would LOVE to use a rotary phone again. I have two (unconnected, of course) for old times’ sake. They worked, unlike digital junk.