Designing a New EHR User Interface: Telling a Story on a Timeline
In January of this year, my wife Karen — who is an artist and illustrator — and I came across an illustration from a children’s book for early readers which she had illustrated back in 1980.
The book tells the adventures of the very nice Brown family (Mr. and Mrs. Brown, Sam, Jane, and Grandpa) who somehow manage to accomplish and thoroughly enjoy their daily activities despite the fact that, or perhaps because, they are incapable of abstract reasoning.
In the chapter about bedtime, Sam explains his plan for the night:
About the same time we came across this illustration, I had been thinking about how to display EHR events visually on a timeline (see my posts from 11/09/2015 and 11/25/2015 for a description of the EHR TimeBar design). I was struck by how both Sam and I wanted to use length in order to measure time.
I started doing some reading in the cognitive psychology literature. It turns out that it’s not just Sam, but all of us (including the MIT undergraduates who were the subjects of some of the experiments and whose capacity for abstract reasoning is presumably better than that of the Browns) who use length and other spatial metaphors to think about time.
When we use expressions like “a long time ago,” “a short time later,” “back in time,” “the distant past,” “spring is far away,“ and so forth, they come to us so naturally that we don’t realize we are using our concepts of space to describe time.
To what extent we have to rely on spatial metaphors when we think about time turns out to be a hotly debated topic in cognitive psychology. According to the “strong version” of one theory, the human brain just can’t intuitively grasp the concept of time. This theory holds that the human brain, like the brains of all other animals, evolved to help us perceive and interact with the physical world.
When humans subsequently acquired the capacity for abstract reasoning, we had to use our existing sensory and motor systems – the ones we use for seeing, hearing and touching things and moving around in the real world – in the service of understanding abstract concepts. In other words, according to this theory, we are mentally incapable of understanding the concept of time without thinking of it in concrete terms. For instance, by visualizing it as a physical path.
If this theory (or even a weaker version of it) is true, it has profound implications for designing EHR user interfaces. If we can present a sequence of clinical events as points in space along a visual timeline, we should be able to grasp this time-based data with little cognitive effort.
In other words, we can “co-opt” the parts of our brain — including our high-bandwidth visual processing system — that have been finely honed by millions of years of evolution to intuitively grasp the physical world in the service of grasping abstract concepts. In doing so, we spare our finite cognitive resources for patient care issues.
Consider two alternate ways of representing the sequence of clinical events (documents) in my data set for the months of March and April, 2014. Note that for clarity, in both views only the time-based (temporal) information about the documents is represented; the subject matter of the documents has, for now, been omitted.
Chronological List of Events – Numeric Format
Timeline View of Same Events – Graphical Format
We can then begin to ask which view better supports the following tasks:
Accurately shows the precise date of each event
Here, the numeric chronological list does better. In the timeline view, an interactive gesture such as a mouse hover would be required to cause the precise date to display. On the other hand, if the precise date is not important, the numeric list of precise dates can cause visual clutter.
Shows the events in the correct temporal order
Both views show the events in the correct temporal order.
Makes it easy to see how events are clustered
Here the timeline view does better. We can intuitively see how events are clustered and the time intervals between them. In contrast, with the chronological list, it takes cognitive effort to get this same information.
Makes it easy to see how many events occurred on the same day
Again, the timeline view does better. To get the same information using the chronological list requires cognitive effort.
Makes it easy to see how events relate in time to the present
In one sense, the present – that is, “today” – is always the most important point in time when taking care of a patient. The timeline view incorporates a symbol for today. In this example, it is as if we are viewing the timeline on April 30. Of course, today could just as easily be inserted into a numeric chronological list, but most EHRs do not support this option.
It’s important to remember that this comparison makes a huge presupposition. It assumes that your EHR is, in fact, capable of providing you with a single chronological list of all clinically relevant documents (notes, labs, reports, imaging, procedures, and so forth) regardless of their category.
As I discussed in my last post, many major EHRs, both ambulatory and inpatient — at least as they are usually configured — do not provide this option. Without such a view, some of the tasks above become nearly impossible cognitive challenges (see my very first post, Human-Centered versus Computer-Centered Design).
In summary, both the chronological list of events and the timeline view have advantages and disadvantages. We are so used to working with chronological lists of EHR events in numeric format that the advantages of timeline-based views remain largely unexplored.
Note: I would like to thank Stephen Few, whose article, Data Visualization for Human Perception, published in the Encyclopedia of The Interaction Design Foundation, inspired the comparative format of this post.
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. E-mail Dr. Rick.