Blog | Monday, February 25, 2013

Learning a new language: an insight into EMRs

OK, I'll admit it: I had no idea. I thought that the whining and griping by other doctors about EMRs was just petulance by a group of people who like to be in charge and who resist change. I thought that they were struggling because of their lack of insight into the real benefits of digital records, instead focusing on their insignificant immediate needs. I thought they were a bunch of dopes.

Yep. I am a jerk.

My transition to a new practice gave me the opportunity to dump my old EMR (with all the deficiencies I've come to hate) and get a new, more current system.* I figured that someone like me would be able to learn and master a new EMR with ease. After all, I do understand about data schema, structured and unstructured data, I know about MEDCIN, SNOMED, and HL-7 interfaces. Gosh darn it, I am a card-carrying member of the EMR elite! A new product should be a piece of cake! I'll put my credentials at the bottom of this post, in case you are interested.**

So, imagine my shock when I was confused and befuddled as I attempted to learn this new product. How could someone who could claim a bunch of product enhancements as my personal suggestions have any problem with a different system? The insight into the answer to this sheds light onto one of the basic problems with EMR systems.

Problem 1: different languages As I struggled to figure out my new system, it occurred to me that I felt a lot like a person learning a new language. Here I was: an expert in German linguistics and I was now having to learn Japanese. Both are systems of written and spoken code that accomplish the same task: communication of data from one person to another. Both do so using many of the same basic elements: subjects, objects, nouns, verbs. Both are learned by children and spoken by millions of people. But both are very, very different in many ways.

The reason for my feeling this way is that, at their core, EMR products are computer programs. They are written by engineers with physicians (many of whom have left clinical practice to work for the EMR company) consulting to help shape the product. The object of the program may be physician use, but their heart is that of an engineer. So the storage of the data, the organization of the medical information, the location of where anything can be found, is based much more on the nature of the programmer than anything else.

Problem 2: strengths vs. weaknesses The idea of an EMR is (reputedly) to simplify the task of health care providers in documenting care and retrieving the information quickly. The reality is that some things are of higher priority to one EMR manufacturer than another. Tasks that were simple in my old system (putting in labs, generating letters with structured data, getting a quick overview of a person's record) are difficult in the new system. The new system, however, does other tasks much better (auto-completion of lab data, management of referrals, interfacing with patient portal, etc.).

I am amazed at how many steps it takes to do tasks my old EMR vendor did quickly. Why did they make it so hard? It comes down to priorities, and for whatever reason (CCHIT, Meaningful Use, Moon Phase) some things get high priority, while others are consigned to the "later" pile.

Problem 3: the system The fundamental reason EMR systems are so difficult is not the nature of the programmers making it or the doctors using it; it is that EMRs are grown in the hot-house of a chaotic and arbitrary health care system.

It makes no clinical sense that there are a gazillion ICD-9 codes, but there are, and any EMR system wanting success needs to devote lots of effort to ICD-9 (and soon to ICD-10, yippee). The structure of most office notes are not to give the best clinical information in the simplest format; notes are generated for the sake of proper billing, including a 10:1 ratio of useless to useful information. Most notes are like a small gift contained in a large box of packing material, with the majority of information simply getting in the way of what is really wanted. EMR systems are well-designed to generate lots and lots of packing material.

The system I chose does the E/M office visit very well, but does so at the cost of hiding useful information and de-emphasizing what is most clinically helpful for the sake of E/M codes, or what will qualify the practice for "meaningful use" money. I don't fault the system for it, since we doctors spend far more of our time focused on E/M codes and "meaningful use" than on patient care. That is one of the big reasons I left my old practice.

The reality is that EMR systems are designed to finesse the payment system more than they are for patient care. That is because the thing we call "health care" refers to the payment system, not to actual patient care. My frustration with my current EMR system is not that it doesn't do its job well (it still is better than my old one, I think). It's that it is grown on a planet where the honor being a healer is being consumed by the curse of being a provider. Patients don't matter as much as payment in our system, so EMR systems will follow those priorities. Those who don't will not succeed.

So to those I have scorned in the past, I bow my head in shame. I got good at using a complex tool that allowed me to manage the insanity of our system. It turns out that my skill was a very narrow one.

It makes me feel like a piece of scheisse (たわごと).

*For those wondering, I was on Centricity by GE and am now using eClinicalWorks.

**My Geek Credentials: --I did my residency at Indiana University, the land where Clem McDonald, one of the pioneers of electronic records made our records electronic when personal computers were still new (I attended from 1990 to 1994). It was there I became a believer in computerized records. --In practice, I installed MedicaLogic's EMR in 1996, as one of the first users of their Windows based product, Logician. --Within 2 years I was on the user group board, and was elected president in 1998. I was a regular speaker at the conferences and known for my profuse production of clinical content (called "Encounter Forms") --In 2003, I applied for and won the HIMSS Davies Award for ambulatory care for our practice, recognizing our achievements with EMR in an ambulatory setting. --After that, I served on several committees for HIMSS, gave talks for multiple other groups (NHQA, National Governors Association), giving the keynote talks for the HIMSS series given around the country to convince docs to adopt EMR. --In 2011, I participated in a CDC Public Health Grand Rounds as a speaker from the physician perspective on the subject of Electronic Medical Records and "Meaningful Use."

After taking a year-long hiatus from blogging, Rob Lamberts, MD, ACP Member, returned with "volume 2" of his personal musings about medicine, life, armadillos and Sasquatch at More Musings (of a Distractible Kind), where this post originally appeared.