When It Comes To The EMR, Little Things Matter A Lot


                                                         By Fred Pelzman, MD

Click. Click. Click.

All day long, we interact with the electronic medical record (EMR), and in so many ways it does an incredible job of helping us take care of our patients.

Everybody having the access to the same chart, remembering the days when we would have a paper chart in one office and someone else would have their own chart in a subspecialty clinic, and never the two shall meet, and then separate charts for the inpatient service that only sometimes found their way to the outpatient world, and specialty practitioners who used a completely different system and never even sent us their consultation letters, all made for endless chaos and often dangerous errors in medical care.

A shared electronic medical record is a beautiful thing. But those of us who have used this, as well as other systems, for many years, have long known that these were not designed by the physicians, nurses, technicians, pharmacists, and others interacting with patients in either inpatient or outpatient settings.

Sure, the general outlines are there, and much of what we need is available to take care of our patients. Things like sending in prescription refills, tracking results, communicating with patients, communicating with each other, and ordering tests and referrals are all dramatically better than in the old days.

But systems and processes have been built up over years and years of use, design and redesign, and far too often little glitches and mistakes have been built in and left there, felt by those who do the programming to not be that much of an inconvenience.

Take, for instance, the simple process of ordering a new prescription. Let's say my patient and I have decided that they should take lisinopril 10 mg, once daily, for their high blood pressure. In the order entry box, I type lisinopril 10, a few choices pop up, and I select the one I want. Then I have to toggle through information required to complete the "sig" section of the order, so that it can be filled by the pharmacy as directed and taken by the patient as we intended.

When I type "1" for the number of pills to take, then "tablet" for the formulation it comes in, the computer then informs me that it's unable to proceed because I have not indicated the route that I want the patient to take it by, even though the system hasn't given me the opportunity to get to that field yet -- that's one more click. The computer yells at me that I made a mistake, that I must be a fool for trying to get away with not specifying the route of administration.

I understand that they need to leave the option for multiple routes of taking a medication, such as oral, transdermal, via PEG tube, or rectally, but the fact that it tells me I've made an error when it hasn't given me that choice yet seems contrary to what makes common sense for all of us. These types of tiny mistakes are embedded and persistent in the system, despite our best efforts to get them removed.

Another example of one more wasted click occurs when a prescription refill comes in from a patient or from a pharmacy. Even though that prescription is already in our system the way it was prescribed by me last time, the system makes me reenter multiple pieces of data about this prescription -- such as its indication -- when it already knew why the patient was taking that medication. Somehow the system isn't smart enough to remember that the patient took it for their hypertension last time we filled it.

Over the years, the system developers who work with us have been incredibly responsive, working with our teams to eliminate massive numbers of these little problems, and even helping us detect and overcome some of the big ones. For example, when we first turned on these systems, the original set of allergy cross indications and warnings about medication interactions was far too large and inclusive -- nearly every medication led to a prompt that we needed to respond to. A large team of people worked on eliminating these, paring it back to just the ones we just really needed.

Right now, the team that works with us to help us fix these minor glitches and increase optimization in the electronic medical record has been on standby, since they are instituting an upgrade to their systems. So for now, the ability to make changes and respond to our wishes to make things better is on hold.

For those of us on the front lines taking care of patients and dealing with this stuff every day, we can only hope that the people who produce these systems listen to us and help us find a way to make the process of getting these issues fixed more streamlined and easier. Right now, requests for removing flow errors and increasing optimization need to go through multiple layers of bureaucracy, being approved by multiple committees and going up the chain of command before things can change.

We recognize that fixing something in one place may cause a problem in another, and everyone needs to be thoughtful when we institute changes that affect workflows for multiple different people in multiple different clinical settings. But at the end of the day, because we're all spending so much of our time doing this, we need to find a way to make a kinder and gentler electronic medical record that works for everybody.


Articles in this issue:

Journal of Medicine Sign Up

Get the Journal of Medicine delivered to your inbox.

Thank you for subscribing.

No membership required*


    • Editor-in Chief:
    • Theodore Massey
    • Editor:
    • Robert Sokonow
    • Editorial Staff:
    • Musaba Dekau
      Lin Takahashi
      Thomas Levine
      Cynthia Casteneda Avina
      Ronald Harvinger
      Lisa Andonis

Leave a Comment

Please keep in mind that all comments are moderated. Please do not use a spam keyword or a domain as your name, or else it will be deleted. Let's have a personal and meaningful conversation instead. Thanks for your comments!

*This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.