Streamlining Genomic Data

My Role

UX Designer in a small cross functional team. I pushed timelines (reasonably) in order to conduct adequate user research to explore and advocate for user needs. The project had nebulous requirements that I needed to refine based on design and dev resources.

Outcome

A mobile optimized, accessible, and fun section in MyChart that is adaptable to different organizations.

The Problem

Many patients struggle to understand their genomic testing results. They need a way to easily access and understand their lab results and know what to do next. 

Physicians need a structured and streamlined way to order and receive lab results in order to make informed medical decisions. Physicians also need an efficient way to document them in layman terms to send patients via MyChart.

Our Solution

Since there are 2 users in the problems we’ve identified, we needed separate project teams to approach this problem. I was the only UX Designer on both teams.

We redesigned the Genetic Profile for patients on MyChart:

  • Improved navigation flow

  • Formatted text to be digestible and accessible for all health literacy levels

  • Added fun and whimsical elements without being a distraction

We redesigned the manual genomic variant interfacing for physicians on the Epic interface:

  • Streamlined the manual workflow and increased automation in data entry

  • Increased tech support for non-SME users

The Process

Without diving into project details (per the NDA), we followed the double diamond process. This was not a linear process, but allowed for a structured plan to be put in place to keep the projects on track.

Research

Through conducting user interviews and reading through development documentation, these were some of our must haves:

The Genetic Profile for patients needs:

  • To be mobile optimized

  • To have personalization options (per each organization)

  • To be fun and whimsical

The Manual Variant Entry for physicians needs:

  • To be flexible to allow discrete variant data entry and automation

  • To be a simplified workflow (since some users aren’t genetic counselors)

  • To be a joy to use

  • To be intuitive and efficient (data entry takes up a lot of a physician’s time)

Challenges

We had robust research plans for both projects, so I struggled with balancing these two projects with other important tasks on my plate. I came up with a priority matrix to decide on which tasks to do first. This allowed me to take things one step at a time and delegate lower priority tasks to my coworkers.

During the design phase, the biggest challenge was that there were too many cooks in the kitchen. This made it difficult to come to a solution due to a wide range of opinions coming in at different (random) times. I decided to implement bi-weekly feedback sessions where different groups of leaders could see our progress and give feedback. This allowed for better allocation of time and a more structured way to receive feedback.

Genetic Profile Outcomes

Patients can now access their genetic testing results on MyChart and actually understand the data because the information is displayed effectively and put into layman’s terms.

(Read the EpicShare article here)

Manual Variant Entry Outcomes

The automated integration of discrete genomic data into Epic improved the speed of data integration by 78%.

(Read the ASCO publication here)

Outcomes

As these two projects were only recently released, outcome data is limited but positive.

My Takeaways

I learnt to balance quality and tight timelines.

We pushed our target deadline in order to produce quality work and maintain usability at a high standard.

I saw a huge improvement in my communication skills throughout the project.

I stood up for usability and negotiated timelines with the developers to ensure that UX was prioritized. 

I refined my confidence to advocate for my design decisions.

Our design kept changing throughout the project, which is normal. Though we would eventually realize that each iteration changed because of ONE piece of feedback. These people all felt like important stakeholders, so we held their feedback at a very high regard. However, none of them understood the full scope and details of the project. I realized that I had to speak up because I was the design expert of this project. If I didn’t advocate for my design decisions, no one will. Once I started doing that, our iterations were then based on design principles rather than what one person said.