RADNET

OVERVIEW

patients self checking in via mobile device for radiology appointment

SUMMARY
RadNet is radiology firm that operates diagnostic imaging centers (MRI scans, CT scan, breast scams; etc). Doctors and nurses use our software to manage their clients and appointments. When patients register online for an appointment, an email will be sent to them. Clicking the link will take the patient to a self check-in workflow on their mobile device.

My task was to improve the onboarding process by addressing any pain-points and reduce the workload of in-house nurses and customer representatives.

TIMEFRAME
Sep. 2022 - Dec. 2022

SKILLS UTILIZED
User Research, Customer Journey Map, Wire-framing, Prototyping, A/B Testing

TYPE
Professional

TOOLS
Sketch, InVision, Maze

TEAM
Michael Tewasart (UI Designer & UX Researcher), Randy Rogers (Product Manager), Stephen Powell (Developer)
CHALLENGE
patients are having trouble completing the self check-in workflow
PROBLEM
The self check-in progress was hastily implemented during COVID19. Patients often have to ask the receptionist for assistance or bring in a tech-savvy friend. This results in more stressed staff & patients, lowering the quality of the appointment experience.

SOLUTION
Redesign the current existing workflow. Remove unnecessary features, streamline what works, and add any features that the stakeholders request.
FINAL IMPACT
REDUCED APPOINTMENT WAIT TIMES BY 40% (~15 - 25 min). PATIENTS & STAFF MUCH ARE MUCH HAPPIER.
  • Mentored other designers on design systems, components, documentation, and visual hierarchy.
  • Streamlined communication between different departments (in different time zones) and managed expectations while balancing their feedback and customer needs.
  • Increased inclusivity of the process by making it work with blind patients.
DESIGN PROCESS
slow iterations bit by bit is the key here
NO ASSIGNED DUE DATE AT THE BEGINNING
I wanted my process to have a greater emphasis on repeated testing with multiple different user groups after each significant iteration.

USER RESEARCH

guerilla in-person interviews

OVER A PERIOD OF 2 WEEKS...
Research was conducted on 6 patients at a radiology clinic in southern California. Each patient timed when check-in on their mobile device and how long they waited before their appointment. After the appointment, they were interviewed in-person on how they felt.
TRACKING IRRITATION
A customer journey map was vital to understanding what the patient was going through.
SURVEY INFO
  • Patients greatly ranged from late 20s to mid 60s, with 100% of them being female.
  • About 60% of them have some form of tech literacy. The rest often had difficulty navigating, and would ask the receptionists / nurses for assistant.
  • While none were visibly disabled, this was something I had to take into consideration in the future.

INSIGHT

KEEp THE OUTdated, but current system. MAKE sure designs comply with wcag IF necessary.

TOO MUCH INFO EARLY ON (AND HIDDEN IN DROP-DOWNS)
Receptionists are often asked to clarify confusing aspects of the workflow. Examples include terminology, what button to click, and where to go next on the screen.

ACCESSIBILITY TO AVOID FUTURE LAWSUITS
Throughout the entire process, I had to keep in mind of the following:
  • Maintain a readability contrast ratio between foreground text and background of 4.5.
  • Using gender neutral language.
  • Working with developers to make sure the mark-ups can be read with screen readers (h1, h2, h3, title).
  • Adding alt-text for all images.
  • Links are now underlined.
  • Font size should be at least 16px for eligibility.
  • Call-To-Action buttons should be concise and understandable with a clear verb.
  • Remove any graphics, animations, and videos that can cause seizures for folks with epilepsy.

User FLOW

in the beginning, digital & physical user flows were severely segregated

USER FLOW
A visual breakdown of the self check-in workflow from opening up the email link to completing the appointment.

SKETCHES

doodles and SKETCHES for inspiraton

PENCIL AND PAPER
Sketching in the beginning helps me relax and tackle the problem with a clear mindset without worrying too much about the details.

ITERATION #1

make each drop-down it's own page with that comply with web accessibility guidelines

TOO MANY DROP-DOWNS TO HANDLE
Most frequent complaints were that it was too much cumbersome to navigate. Users were being presented with way too much information.
EACH DROP-DOWN IS NOW ITS OWN PAGE
I hypothesized that by possibly rearranging the workflow into something more linear, the completion time would be much quicker. In other words: user can only complete one page at a time. I also worked with the nurses and edited / trimmed down some of the copy.
LONGER COMPLETION TIME, BUT HAPPIER
I held a quick guerrilla test with the same patients. To my surprise, it actually took them slightly longer to complete it. However, they greatly preferred this format.

ITERATIONS #2

stakeholder additionS: adding provider info & payment system

STAKEHOLDER INPUT
When I was presenting it to my project manager & various stakeholders, a feature was requested: giving patients the ability to add new providers that aren't in the database and to confirm, add, and edit payment for the appointment on the mobile device.
COMPLETION TIME ⇧ SLIGHTLY LONGER
A 2nd guerrilla test was done on the same patients, where they were asked to complete the same workflow with the added features.

ITERATION #3

automating an alert and integrating paper-work into the mobile workflow

THE LINGERING PROBLEM...
The main bottleneck that still continued was that patients would be forced to memorized & text the center when they would arrive at. Patients can often forget this last part, prompting receptionists & nurses to assist with the last few steps.
NEW USER-FLOW ADDITION: AUTOMATING THE LAST FEW STEPS
A solution was added where the patient would automatically receive an email / text 1 hour before the appointment. Then, they would continue their workflow by signing digital forms that were once on paper. A notification would inform the receptionist when this was done.
COMPLETION & WAIT TIME SIGNIFICANTLY
2 different groups (each with 5 users) were tested on this new workflow as I believed having a new set of users would yield less biased results.

FINAL THOUGHTS

understanding what you should change & what you shouldn't for the user

UX OUTSIDE THE SCREEN
I had to take into consideration different scenarios for the user during their journey path when they begin. They could be inside or outside the building; they could have confirmed the appointment on appointment day or several days before. I certainly learned a lot of thinking outside the box.

NO NEED FOR RADICAL UI CHANGE
Sometimes sticking with the existing UI is for the best. While outdated, it made the user comfortable with navigation (as it's consistent with the web portal they signed into.

CHANGING WORKFLOW
While it may seem a bit counter-intuitive to make the user stay longer on the workflow, it allows them to complete it at a more methodical pace. Since all the paperwork has already been integrated into the workflow and notifications have been automated, nurses / receptionists aren't burdened.