Case study: Heuristic eval & redesign of Intergy EHR (SaaS)

Identifying usability issues in Intergy — An Electronic Health Records (EHR) software used every day by thousands of healthcare providers.

Background

In order for a healthcare provider to maximize their time and energy on their patients, they need a system that is intuitive, easy to navigate, and simple — without sacrificing the complexity of valuable chart information.

I decided to interview multiple healthcare providers, heuristically evaluate Intergy EHR’s charting system for physical therapy, and suggest design improvements.

I participated in this case study individually, all work here was done by me.

Healthcare providers | photo credit

Side notes

• The interfaces used by healthcare providers are called EHR or EMR. (Electronic Health/Medical Records).
• These are typically cloud-based Software as a Service (SaaS) platforms.

Ethnographic Study

• Nielsen Norman describes lab studies vs. field studies as: the user comes to us versus we go to the user to conduct research.
• In this case, I had the unique opportunity to observe and interact with ‘the user’ in their natural work environment. i.e. healthcare providers using the interface (Intergy EHR) in the outpatient physical therapy setting (my current job).

Pain Points

After speaking with numerous healthcare providers about their experience with utilizing Intergy EHR for patient documentation, the following 3 pain points stood out the most.

Issue 1: The present documentation aspect of a “Daily Treatment Note” for a patient is similar to a Word Document. This causes issues when providers only want to edit some details of the document, however are given the opportunity to edit the entire document instead. This inevitably leads to errors and mistakes.

Actual Physical Therapy patient encounter note

Issues 2 & 3 are self explanatory. 😁

Heuristic Evaluation

The above chart was created to represent the major usability issues within Intergy EHR. I went through each of the 10 usability heuristics for interface design and rated them off of 3 criteria: generally satisfactory, unsatisfactory, or somewhere in between.

This in conjunction with findings from interviews and field study helped to then determine the next course of action to improve the UX of the system.

Key Takeaways

⚡ ️User control & freedom — too much control with editing a patient’s daily treatment note.

❌ Error prevention — the current format inevitably leads to documentation mistakes such as editing areas that should stay the same or accidentally deleting important charting information.

👉 Recognition rather than recall — it takes 7 clicks to print a patient schedule. One has to memorize this flow as it is imbedded deep within the software.

👔 Aesthetic & minimalistic design — the present design could be more clearly structured and clean, there is an overall lack of visual hierarchy and the amount of information displayed is overwhelming.

Sketching solutions

After much reflection and thought, I decided that the best way to present a plethora of data in a organized and accessible manner was to utilize a dashboard interface.

I showed this sketch to my coworkers and they agreed that this solution was viable. They stated at their previous healthcare jobs the software interface was more akin to a dashboard as well.

Designing in Figma — Challenges arise

Once I had the basic structure of the dashboard, one obstacle I faced was how to break apart the information in both a meaningful and aesthetic presentation.

There is a lot of information to be presented, how should it be displayed without looking too busy? How big should each info card be?

My solution to this was to divide the workspace (minus spacing) by 3 then begin filling in the information into each atom. Eventually I realized some of the atoms needed to be 2x in length or width as the information was lacking hierarchy.

First iteration & feedback

Below was the first iteration of my dashboard solution. It’s not beautiful, but the information is all there.

After reviewing it with coworkers and gaining feedback,

I had a list of areas of improvement:

1. The text feels too crammed inside of the cards.
2. The gray boxes within the cards feels bland.
3. The hierarchy of information feels a little off.

Second iteration

Points improved

1. Typography was adjusted to give text breathing room as well as visual hierarchy.
2. White space utilized more appropriately within cards so they didn’t feel so cramped.
3. Headings more distinguished from body text, easier to differentiate at first glance.

Additional Solution

It is impossible to display ALL of the important information in the dashboard, so the user has the option to individually expand each card to reveal more details:

Expanding individual cards to reveal more information

Addressing all pain points

✅ Notification of patient check-in

In the present system, if a provider is editing a patient note, and their next scheduled appointment checks in, there is no system in place to notify the provider. The only way to know if a patient checked in is to back out to the main schedule and see the checkmark next to their name (see above image for example).

Solution

Breadcrumbs: dismissible notification appears at top of screen that does not obstruct work flow.

Dashboard with breadcrumbs at top of screen to notify provider of patient arrival.

💻 Printing patient schedule

In the present system, It takes 7 clicks to print a provider’s daily patient schedule. Below is an actual screenshot of the word document I created that I give to new-hires when I train them on our system, because no one can figure it out on their own.

Actual word document I created to teach new providers how to print their schedule.

Solution

Adding a button in the available space above the patient schedule is not only intuitive and easily accessible, but takes the process from 7 clicks-to-complete to 2 clicks.

Provider main screen with button to print daily schedule.

Final thoughts / Lessons learned

• With common applications, a UX Designer can make educated guesses on usability issues, however more technical platforms (such as Saas) are much harder gain a read on without speaking directly with the user.
• Heuristic evaluations are a relatively cheap and effective method of quickly assessing major usability issues within an interface.
• While dashboards are an efficient way to present key information, there is certainly a balance between displaying enough information to be useful and simultaneously not overwhelming the user with too much info.
• You likely will not get to the “pretty” solution without first having an ugly solution. (“does the solution solve the problem?” is of greater importance than “does this look pretty?”)

Talk to me

I’d love to hear your thoughts and opinions!

Connect with me on LinkedIn

Andrew Cox - Physical Therapist Assistant - Florida Medical Clinic | LinkedIn