The Failure of the Electronic Medical Record

We're in a time of EMR. Electronic Medical Record. There are incentives to encourage doctors to adopt an EMR, and barring that, there's the penalty for not adopting. Of course, this is if you accept medicare. If you don't accept medicare payment, the EMR movement can suck it.

I've worked with at least 4 different EMR systems, not counting the early implementation of computer accessible records at the VA and University of Washington, where I was a medical student.

I grew up a child of the 70s and 80s. We didn't start with computers, but they came along and we used them in school. In college and particularly in medical school at the end of last century (wow, do I feel old now), I did a lot of communicating with friends via online mechanisms that required me to learn to type quickly. I, of course, was aided by a fairly brilliant computer lab teacher in 7th grade who informed us that if we didn't have our hands in the right location on the keyboard, he would delete our work. That sort of thing really makes a girl learn where to put her fingers on a keyboard, not to mention how to save your document as you go. I say all of that as a way of making you understand that I can type well, and computers don't frighten me. The same could be said for many of my younger colleagues, but not for many of my older colleagues and teachers.

Having said all of that, I have grown to love and hate the EMR. On the one hand, patient data is nicely accessible without having to manually search through a room of folders. I can type in a patient name or other search parameter, and find what I need most of the time fairly efficiently without having to rely on a medical records staff person to find it for me. On the other hand, the data is sometimes packaged in a non intuitive way, and this can be frustrating, to say the least.

A quick Google search "electronic medical record options" gives pages and pages of EMR reviews, options, and news stories. Most medical associations, from the AMA to the local county medical society, have sent out bulletins, news letters and informational emails about how to choose the EMR that's right for you. Medicare has had a deluge of bulletins and updates on the requirements to meet what is called "meaningful use" in order qualify for a very hefty bonus check for physicians who chose to opt in to EMR early (up to $45,000 over 5 years paid to doctors who implement and adopt a certified EMR).

This post isn't about what makes an EMR certified. This is about my concerns as a physician both using an EMR and receiving notes from other providers doing the same. If you go back to that Google search again, you'll find that there are a surprising number of EMR vendors. They are all similar in that they sell a similar product, but they are as variable as the list of doctors you'll find in the phone book if you flip to the doctor section of the yellow pages.

Some are web based, some are server based. Some are shockingly expensive, charging an initial fee and then additional monthly fee per provider (doctor, PA or NP using the system). Some are free and supported by advertising (yes, advertising in the electronic health record). Some are closed systems, meaning the software belongs to the company and you can't change it. Some are open source (if you write code, you could very well change the look and feel of your EMR, at least to some degree). Some allow you to have multiple copies of it running at once, others log you out once you log in on another browser or computer. Some allow multiple people to edit a chart at once, and others do not. Some have integrated lab and xray ordering and reports (this is required by medicare to attest for meaningful use), and others don't.

An interesting aside here is that lab and radiology ordering requires the lab or radiology location to also integrate with your EMR, which costs money. Routinely debated is who pays for that integration: the lab? the EMR vendor? The provider's office?

Regardless of all of those choices, one thing is consistent from EMR to EMR: they're not a book. I know that sounds dumb. But the truth is that a patient's chart is a story. It's their medical story. And each visit is a chapter in that story from the doctor's point of view. A doctor's note should be clear: this is why the patient came to see me, this is what I discovered from his history and his examination, this is what I think is going on, and this is what we plan to do about it. That's what every patient visit should be: what the patient tells you, what you see, what you conclude, what you plan.

The problem is that more and more doctor's notes are generated from templated EMR notes, and filled with extraneous details that do not matter. The end of a note will be filled with codes to be submitted to medicare to prove that we deserve that incentive money (smoking cessation, med list reconciled, fall assessment done, etc). While those are good measures of good care, we shouldn't have to insert them into the chart with a code every time the patient is seen. And it certainly shouldn't gum up the data.

Also, there are multiple parties involved in integrating into an EMR: pharmacies, radiology offices, labs, etc. They all have different standards for formatting and even which computer language. They all feel like they are the most important aspect of the system.

My ultimate experience with EMR is an overwhelming sense of optimism that is constantly peppered with the disappointment of reality. Notes don't make sense. Formatting is a challenge to adjust. Routine 'updates' often cause several days' worth of errors that make the EMR so much less useful. EMR software is written by programmers, not providers. They layout and input is often intuitive to a programmer, but not a doctor. Systems that require a 'few extra clicks' often turn into minutes or hours of delay in a busy office. While a few seconds here and there doesn't seem like much, the end of the day is filled with incomplete charts, unsent prescriptions, unfiled documents and lab results, because time is of the essence. Few charts have effective or efficient voice to text translation, so almost all notes are done from templates (so many notes look identical, despite patients all being unique) or they are overwhelmingly time consuming, even for a fast typist for me (and often left incomplete due to time constraints).

What has been lost in all of this process of assured documentation of required aspects of care in order to assure reimbursement is the basic fact that a patient's medical record is his story. A patient's chart should unfold the story of her health, her illnesses, her recoveries, her needs, her concerns. There is no room for such humanity in the current EMR model. Everything has been boiled down to a series of yes or no questions and codes to assure incentive checks keep coming.

Gone are the dictated notes that outlined the patients risks, his choices, his symptoms, his treatment plan and his prognosis. We're in too big a hurry to make sure we see enough patients in a day that seems to feel shorter and shorter to stop and tell the patient's story. We have all of the patient data, and we're making sure to document it all: race, ethnicity, vital signs, medication lists, allergies, fall risk, on and on the list goes. The problem in all of this for me is that in this sea of data, there is little information. And this, ultimately, is the failure of the modern EMR.