Should EMRs Force Workflow Changes?

Today, I was lurking in the EMR and HIPAA Facebook chat where some readers and publisher John Lynn were talking of things EMR-related.  During that chat, one exchange really brought home to me how far we have to go in even agreeing on how the ideal EMR should work.

During the discussion, one chatroom member said that the biggest problem with EMRs is still that they force doctors to break their workflow.  Another stalwart chatroom member, the insightful @NateOsit, retorted that EMRs should break workflow patterns, as this would promote healthy change.

Well, there you have a conundrum,  if you look closely enough. While people seldom speak of the issue this directly, we’re still arguing over whether EMRs should fit doctors like a glove or change their habits for the (allegedly) better.

This isn’t just an academic question, or I wouldn’t bore you with it. I think the EMR industry will be far more wobbly if the core assumption about its place in life hasn’t been addressed.

At present, I doubt EMR vendors are framing their UI design discussions in these terms. (From the looks of some EMRs, I wonder if they think about doctors at all!) But ultimately, they’re going to have to decide whether they’re going to lead (create workflow patterns that follow, say, a care pathway) or do their best to provide a flexible, doctor-friendly interface.

I’d argue that EMRs should give doctors as many options as possible when it comes to using their system.  Perhaps the system should shape their workflow, but only if the users vote, themselves, that such restrictions are necessary.

But the truth is that when a hospital spends a gazillion bucks on a system, they’re not doing it to win hearts and minds, no matter how much they may protest otherwise.  And when a practice buys a system, they’re usually doing it to meet the demands of the industry, not give their colleagues their heart’s desire.

So let’s admit it.  Though I don’t argue that they’ll ultimately be put to great uses in some cases, ultimately, EMRs are about dollars and bureaucratic face-saving.  So, today’s workflow will just have to take a back seat.

About the author

Anne Zieger

Anne Zieger

Anne Zieger is a healthcare journalist who has written about the industry for 30 years. Her work has appeared in all of the leading healthcare industry publications, and she's served as editor in chief of several healthcare B2B sites.


  • Anne,
    Thanks for bringing up a really important conversation. EMR workflow is so important. I think the real answer to EMR workflow is a balance. Far too many EMR software require you to completely change how you do things. Far too many doctors don’t want to change one iota of what they do. Bring these two together and you have the current EMR workflow conundrum.

    I think the point that Nate made is important, but I would have phrased it differently. I’d suggest that a doctor will go through workflow changes when implementing an EHR, because there are certain things you can do with an EHR that you couldn’t do in the paper world. For example, many EHR software has patient statuses that change as the patient is seen by the various staff in the clinic. These statuses provide a different workflow for knowing where the patients are at in the process. This can often replace the paper chart on the door workflow.

    I think your last sentence is a little off base too. Although, I would agree that many people are treating an EHR as free government money and meeting a government requirement. Those that follow this route are going to be very disappointed by their EHR.

  • Ah, the age old question – should the system fit the user or should the user fit the system? How many years has this question been asked? The answer is, most usually, both!

    Time to whip out the old car analogy! Do you get to decide which side of the car the steering wheel is on? Of course not – that is defined by a standard! And that standard is set to apply to the majority of drivers in a locality, thus why in some countries the steering wheel is one one side of the car, versus the other. Cars are adapted to where they are used, but en masse and not individually, for a feature as specific as the steering wheel.

    So, to extend the analogy – things that can be standard in an EHR, like a steering wheel, should be the same to encourage instant understanding, standard training, common use, safety, cost efficiency, etc. And that “standard” should be the one that fits the masses.

    For things in an EHR which are NOT like the steering wheel, for example, the paint color of the car, those should be customizable. Those extension points, those customizability points, can also be seen as STANDARD FEATURES (or strategic advantages of one system over another) that every EHR (car) should have.

    Customization points can be “standard”. But there does not need to be a definition for the ways in which the functionality within that customization point be assembled, rather, only what it needs to achieve.

    So I think the trivial answer (non-specific but useful in framing the discussion) to this question is: people should be forced to do certain things the same way, only when it makes sense (a set of criteria can be developed to determine this), and be given complete flexibility through standard customizability points for *everything* else (where possible).

    Since everyone’s interpretation is relative to their situation, that’s why we have such a tough time pinning specifics on the situation created by the answer above.

    My 2c.

  • A physician’s workflow is not linear. An application that allows rapid task switching with minimal cognitive dissonance can adapt to many different workflows. The number of clicks are less important than the amount of cognitive interruption they cause. At any given time in the ED I may have 5 immediate tasks in my mind. By the third click that makes no sense, I have forgotten my train of thought, possibly with dangerous consequences.

  • I just left a response in LinkedIn, where I found this article, but I thought I would share here as well!

    My concern with assuming that the work flow has to change, it means you have to assume that all EMR vendors know how to practice medicine and manage the clinic better than those actually doing it.

    Before you get tangled up with the EMR or any other IT solution, I would recommend to get together as a clinic and talk about what you do well and what could be better in terms of workflow – i.e. communication, staff training (yes, the clinic has to know how to use EMRs), your brand message and awareness of it in your community, patient experience and retention as appropriate for continuing care, patient education.

  • Bob,
    The idea that every doctor has to change doesn’t necessarily mean that the vendor knows how to practice medicine better than the clinic. It does mean that every EHR vendor has technical capabilities built into their EHR that could benefit a clinic that didn’t have those capabilities before.

    Let me give another simple example. Our workflow on lab pricing in a clinic was to look at a paper which had the list of pricing on it. Of course, this was a pain as prices changes and it didn’t include every lab. With our EHR, when you ordered the lab it told you the price of that lab in the EHR. That’s a change of workflow that was a change that everyone loved.

    There are dozens of examples like this and it happens in every clinic and every EHR.

    I do agree that the clinic should know their workflow and look at how their current workflow will be impacted by an EHR. This is essential and most clinics don’t spend enough time considering this.

  • Some existing workflows are efficient and viable. However, any observant person who spends enough time in a doctor’s office can spot piles of things that could stand a change in workflow. Think about it for a moment; most offices have more paper pushers then clinicians, few use e-prescribe, few would let a computer offer an opinion on diagnosis, few use electronic orders to labs, let alone take in results electronically, few have a doctor put enough (if any) info into a system to 1. have good records and 2. generate proper coding. Each one of these has high value and if not being done, a change in workflow is in order. If they are all being done, there is a lot less room for change.

  • A note to Brian, the ER doc. I’ve watched ER docs go from patient to patient with a pile of charts, then go back to the nurse’s station and spend the next half hour dictating into a phone everything they think they remembered. It causes huge delays in labs and treatments (and doesn’t allow them to be done in intelligent order) and is obscenely wasteful and inaccurate. It gets even worse when a critical patient comes in and everyone runs to take action.

  • A practice needs to know the workflows/processes currently used as part of preparation for implementation of an EMR system. Half a year late, but the discussion is still fresh. Jon’s question “should the system fit the user or should the user fit the system?” leads to a follow up.
    Any organization is comprised of a group of workflows/processes that have developed by design or by need. Some of these processes will not match the capabilty of any EMR system. An early step in adoption of a control system like an EMR includes making a list of individual processes and documentation of the details. (consider mapping each on paper). An example is collecting information on a new patient. 1. Patient arrives at desk 2. Receptionist finds new patient form and gives to patient. etc. An EMR system might allow the use of tablet computer to document the information directly to the system.
    This change can be planned before implementation of the EMR system, but only if the ‘new patient process’ is identified and discussed as part of the planning process.

Click here to post a comment