Home » Rick Weinhaus » Currently Reading:

EHR Design Talk with Dr. Rick 3/12/12

March 12, 2012 Rick Weinhaus 12 Comments

Humans Have Limited Working Memory

Consider a very common, high-level EHR design. The screenshots that follow are from a particular EHR, but many vendors use a similar design.


A row of clickable tabs at the top of the screen is used to designate the different categories of data that make up the patient visit. When a tab is clicked, the window for that category of data opens to full screen size. The tabs can be clicked in any order.



The screenshot above shows what I would see after having clicked on the History of the Present Illness (HPI) tab and having entered some data.



If I were then to click on the History (Hx) tab and enter some data, the new screen would look like the one above. The HPI data is no longer visible because the HPI window automatically closes when the Hx tab is clicked.

This EHR design is completely logical. It is also completely usable, if usability is defined as being able to easily navigate from one part of the record to another with a single click. In fact, it is a totally reasonable design if it weren’t for one problem — humans have absolutely terrible short-term (working) memory.

It used to be thought that humans could retain about seven unrelated elements in working memory, but recent work suggests that the actual number is more often in the range of four to five. In contrast, a modern computer has no problem retaining thousands of unrelated data elements in random access memory.

Given our severe limitation in working memory, this EHR design doesn’t work very well. Every time I click on a new tab, the previous window closes and that data is no longer visible. I have to carry that information in my head. Furthermore, the row of tabs itself contains no information. It just serves as a navigation tool.

In other words, this design is based on how a computer — not a human — thinks. It is a computer-centered, not a user-centered design (see my first post).

As a clinician, I need to devote my full cognitive resources to my patient’s health issues. I need to be able to retrieve information from any part of the record quickly and effortlessly. While completely logical, this very common EHR design just doesn’t do a good job of extending my working memory. From personal experience, I can tell you that using a system like this is enough to drive you crazy.

So what’s the alternative? The alternative is to design an EHR based on what humans are good at — using our visual system to make sense of the world. The data needs to be organized spatially, assigning each module to a fixed location on the screen the way that T-Sheets and other paper forms do (see my previous post). Instead of making the overview of patient data just a row of information-less tabs, display the actual data in a one- or two-screen view, allowing the clinician to see the information rather than forcing him to remember it.

Of course, every design requires compromises. If you decide to use a compact, fixed spatial layout for your high-level design, then you need to solve the twofold problem of what to display in the default view and how to display more information on demand.

In my next post, I will present an example of one widely used EHR design solution to this problem.

Next post:

The Problem with Scrolling

Rick Weinhaus MD practices clinical ophthalmology in the Boston area. He trained at Harvard Medical School, The Massachusetts Eye and Ear Infirmary, and the Neuroscience Unit of the Schepens Eye Research Institute. He writes on how to design simple, powerful, elegant user interfaces for electronic health records (EHRs) by applying our understanding of human perception and cognition. He welcomes your comments and thoughts on this post and on EHR usability issues.

View/Print Text Only View/Print Text Only

HIStalk Featured Sponsors


12 Responses to “EHR Design Talk with Dr. Rick 3/12/12”

Pages: [2] 1 » Show All

  1. 12
    Rick Weinhaus, MD Says:

    An EHR Developer —

    Thanks so much for your note! You raise two very interesting issues.

    1. We are so used to thinking of collecting and recording information as one task and summarizing and reviewing it as another that we tend to program a separate user interface for each. An alternative design would be to use the same pane for both tasks. For instance, a summary view of the past medical history could be assigned to a pane in a particular location on the screen. For data entry, that same pane could expand to a larger size (but not full screen in order to preserve as much screen layout as possible) and then contract to its summary size afterwards.

    A design of this type would reduce the total number of screens and views needed to present the patient data. It would probably be more difficult to write the code for this kind of dual-purpose pane, but the reward would be a simpler user interface.

    2. I have been thinking about the tab design as well and why it works for some applications and not others. I think that the difference is that with many websites and applications, the user is looking for a particular piece of information. In this case, tabs can work pretty well.

    For instance, you might visit the website of a university to see your team’s basketball schedule or you might visit it to see what a particular faculty member had recently published, but you probably wouldn’t be looking for both pieces of information at the same time.

    In contrast, when using an EHR, the clinician is always trying to explore connections between different categories of data. Having the data visible on the same screen facilitates these kinds of open-ended queries.


  2. 11
    An EHR Developer Says:

    Dr. Weinhaus,

    I always enjoy reading your articles. Two comments/thoughts (for disclosure, I do work for a major EHR vendor, though not the one you’ve mocked up in this case):

    1. In your example of documenting HPI/Hx/etc., I would say that there are two separate activities being performed by the physician. The first is simply documentation – collecting information and recording it. The second activity is using the collected information to make a clinical decision.

    To me, the UI presented is appropriate for the first activity – the collection of information. It’s certainly true that you won’t be able to hold all the pieces of information from another tab in working memory, but if your goal is to simply document the information, in general you wouldn’t need to.

    I wholeheartedly agree that this UI is completely inappropriate for the second activity – reviewing the information collected and making a decision based upon it. What would your thoughts be about a hypothetical design that used the exact same UI for collection, and presented the information in a second, summarized form for the purposes of review and decision-making?

    2. To what extent do you think the optimal logical organization of information varies from person-to-person? Clearly things such as the capacity of working memory are less variable, but to again go back to your tab-based layout – there’s a new generation that will be highly familiar with this UI, as it’s the fundamental way that a web browser organizes pages. How do you think their experiences will shape the optimal way to format information for them?

Pages: [2] 1 » Show All

Leave a Reply

Subscribe to Updates



Text Ads

Report News and Rumors

No title

Anonymous online form
Rumor line: 801.HIT.NEWS



Founding Sponsors


Platinum Sponsors




































































Gold Sponsors





















Reader Comments

  • Riyengar: Wonder what the market is for stand-alone scheduling..it seems most HIS provide a built in component....
  • FLPoggio: Re: Streamline & DSS - DSS offers products and services to government and commercial clients based on the VA’s Vis...
  • RL: Re: EHR Nomad. I wouldn’t completely give up on the idea of migrating some data into Allscripts. To the point Mr...
  • FLPoggio: Fred, Couldn't agree more. From my HISTalk piecein 2013: "The real world keeps changing. Yet many of the predictive ...
  • CC: Thank you for refusing to give in to that patient's demands for an antibiotic. Too many providers are pressured into pre...

Sponsor Quick Links