The project

While working at AWS Healthcare and Life Sciences, I was working on a new healthcare product addressing patient affordability for a particular treatment type. I joined the team as the first UX person, and there was already a working prototype.

Patient context

When patients are recommended this particular treatment, it frequently comes after months or even years of getting a proper diagnosis. There's some hope in the treatment improving quality of life, but high treatment costs can dampen that hope. The treatment suppliers offer cost support to patients, and our product was creating a new integrated channel for cost support sign up.

Testing the prototype with patient participants

The existing prototype effectively asked users to step through supplier legalese in order to sign up for cost support. When I tested this prototype with folks qualified as a patient persona, they naturally were off-put with legalese as the core content.

"So invasive"

"I'd ask for another treatment"

It turns out, legal heavily influenced the design to be this way and so the product team ended up with that prototype.

Having a discussion with legal

I shared the results and anecdotes with our legal team, and expressed that abandonment rate is likely to be very high, distrust in the supplier will be high, and that could also reflect poorly on AWS, who is creating this sign up experience. The goal of the product, to help patients easily sign up for cost support, will not be met with this current design.

I proposed an alternative design that provided clear information about the benefits patients get with cost support in clear language, while still providing a way for patients to view supplier terms and conditions.

Sharing results with the supplier

I also shared my findings with our pilot supplier. The findings and a vision for a clearer patient experience helped influence them to change their privacy and data sharing copy.

Measuring trust in the patient experience

In the next round of patient research, we examined trust levels in "new patients" and "existing patients". We found that existing patients' self-ratings for trust increased after going through our prototype, while new patient ratings didn't change. Participants didn't express concerns as in the previous study.