5 ways to make your EMR more user-friendly
One of the biggest objections to the adoption of an <a href="/directory/electronic-medical-record-emr" target="_blank" class="directory-item-link">EMR is its usability (or lack thereof), which is no surprise considering the ease of its predecessor: paper. Thankfully, there are a few ways to make your system not only more bearable, but significantly easier to use.
“There are several guidelines that have been published, [and each] cover particular OS, whether it be Mac, Unix, or Windows,” said Bob Hunchberger, a clinical informaticist for a 500-bed hospital. “If your application will be deployed in the PC world, it’s important that you adhere to the standards that are implemented in the Windows world. Why? Because Microsoft has ‘trained’ its users for more than a decade what behaviors to expect from applications that run in that environment.”
Hunchberger suggests five practical ways to make your EMR more user friendly.
1. Use appropriate controls consistently. Placement is everything, said Hunchberger. He suggests controls that initiate actions be near the top left, while those that complete actions near the bottom left. “Placement and behavior of the ‘OK,’ ‘Cancel,’ and ‘Apply’ buttons is important,” he said. “Users look for them in the bottom right. The three buttons have a consistent behavior of accepting the user’s response and closing the dialog, canceling the response and leaving the dialog, and accepting many responses without leaving the dialog. Placement is crucial in a top-left-bottom-right reading/scanning society.” Additionally, Hunchberger pointed out controls that appear in different locations or placements are distracting. “From dialog to dialog and page to page, the behavior should be the same; users expect it,” he said. “They become confused and frustrated when they invoke an action, and it doesn’t do what they’ve come to expect from hours of experience with Windows Office products or Windows programs at home.”
2. Standardize task sequences. Hunchberger said the system should be consistent with how users complete tasks, whether it’s left to right or top to bottom in dialogs and pages. “Controls that initiate tasks should be located at the top left, and the workflow should follow a logical sequence from top to bottom,” he said. “Don’t have the user start at the bottom, then move to the top, then back to the bottom – if the task is too complex, use multiple pages or a wizard, but be consistent in how the user completes the task.” Completing the task, he continued, should involve the same “button grammar” and placement, so the user quickly learns the process and isn’t distracted or interrupted. “The user shouldn’t have to say to themselves, ‘On this page, I need to start at the bottom and work up,’” he said. “Or, ‘On this page, the final action button is on the left.’”
Continued on the next page.
Showing 6 Comments
Dwight Arthur say: Where are the EMR visionaries
The failure of EMR is not just because of poorly built products, it's because the industry as a whole lacks vision. Throughout most of the medical profession, the ideal EMR is imagined as something that's as easy to use as a pencil. This is an impoverished vision that results in medical professionals spending an ever-increasing proportion of their time tending to the needs of the EMR and less of their time tending to the needs of the patient.
Consider this: using technology that's available today, Google's "autonomous vehicle" project has computer-controlled cars that have logged 200,000 accident-free miles on public roads. If computers can achieve that degree of autonomy, how can we justify the millions of hours that physicians and nurses spend copying information from one device (eg blood pressure monitor) to another device (eg EMR screen)?
I mean no disrespect to this article, which is one of the better treatments of EMR usability issues. But by focusing on issues like optimizing the top-down left-right navigation of a data entry screen, aren't we condemning medical professionals to a future as glorified data-entry operators? The doctor should be asking the patient, "when did the pain start" and "does it hurt when I press here" and the computer should handle the collection of data from other monitors and databases and the transcription and summarization of spoken conversations.
The healthcare industry is critical to the economy and to the quality of life of every one of us, and yet the technological progress of this industry appears to be guided by the most pedestrian of visions.
bob hunchberger say: no disrepect taken
I became a nurse for personal reasons and when I encountered the EMR I was amazed at how poorly designed they were. I spent over 20 yrs in the IT industry as both an owner of a consulting practice as well as a software engineer. It is as if the developers of EMRs have been living under rocks. There are published standards of usability that have been in existence since the late 80's. Of course they have evolved over time but it seems that the builders/designers of EMRs are unaware of this body of work that has preceded them. This is very disappointing. There are standards developed for the Apple, Windows and Unix platforms. There are bodies of work dedicated to human factors and usability ranging from a Windows platform to a mobile platform.
My personal opinion is we need to return to basics first. By that, lets build a system that at the very least adheres to the behavior we see in our day to day consumer software products. Lets build in personas that enable the EMR to adapt to the expertise level and/or clinical role of the user. Not one size fits all.
In my perfect world, someday, we will have EMR software that functions the way practitioners work and it will be intuitive like that found in an Apple app but that day is a long way off.
dch say: Nicely said
Hope there are more people out there who think like you.
Dovetails especially with my last bullet point below.
dch say: Good.
Pretty common sense. I'm glad someone in the geekosphere knows this stuff. Whoever designed the EHR system I use didn't ... at all. (I have zero control over purchasing.) On the positive side, the generous flaws in my system give me much fodder for thought.
Other thoughts - some of my wish list:
* High availability information - assemble/push into dashboards.
* User customizability - let me decide what I want to see to fit how I do workflow.
* Searchability - Basic - e.g., search for a medication name or diagnosis anywhere within a patient record to see if anybody else has considered it. Or, pull up all CXR or EKG reports onto one page for my review.
* Searchability - Wolfram Alpha style. Let me ask my own questions and conduct my own instant QI projects. "What % of diabetics had a HgbA1C done in the past 6 months?"
* Do something with data beyond static reports. Make it instantly analyzable. OK ... so the sodium or the Hgb is low today ... what does that mean in the context of the past two years? Or, the QTc is slightly long ... what's been the QTc trend over the past 5 years? Or, display a longitudinal medication history for the past 2 years, superimposing ANC values upon it to assess for correlates.
* High availability references - e.g., when I pull up Mrs. Smith, my dashboard alerts me to a serious drug/drug interaction now known to exist in her regimen. Or, the low ANC is recognized by my system as neutropenia, and a reference link for neutropenia etiologies is inserted next to the low lab value.
* Data entry - the big, quiet bugaboo - How do I generate my patient notes quickly and uniquely (no cookie cutter notes), without obtruding upon the human encounter between me and the patient, and without adding more to an already high overhead? Whatever company actually fixes this will leapfrog to the front of the industry.
bob hunchberger say: I agree with you
Another one of my frustrations with EMR vendors is the fact that they do not understand the concept of self service. In the private sector there are numerous products that enable end users to mine data to get the answers they need to solve business problems or improve processes. I don't see that in the EMR world. End users have to beg IT to build reports that take weeks and months to develop. Then when it is finally developed the dynamics of the world have moved on and the report needs to be repaired at the cost of more weeks and months. How can we improve healthcare processes if we don't have access to data in real time to drill in and drill out, analyze the results and make corrections in processes that enhance patient lives or reduce costs?
It is as if the healthcare industry is still living in the 80's vs. the 2000's. The concept of self service has been around since the 90s. Corporate IT woke up one day and said we do not want to be in the business of writing reports for people. There is no value-add. The reports constantly change and the customer is never happy. Lets give them them the tools to do this themselves. Someday this awakening will happen in the healthcare world. I hope sooner than later.
A good example of a self service product, not in healthcare, but in customer management is Salesforce.com.
This brings up another topic. We need to learn from other verticals. Many solutions to our problems have been solved in industries outside of our own. We need to start thinking outside the box, ah, industry, otherwise we will be doomed to reinventing the wheel and wasting valuable healthcare dollars that could be directed elsewhere.
axeo say: Usability
Keep us in mind for organizations that need to “own, their own” EMR.
A wave of M&A is expected with a number of products leaving the market due to consolidation. Risks are high for off-the-shelf systems as a result.
Also, usability, or lack thereof, is still a big issue.
www.axeoEMR.com