UPMC Digital Symptom Screening Form

Overview

I collaborated with UPMC Magee Women’s Hospital to improve the patient screening process for the gynecological oncology service. I designed a more streamlined, actionable, and transparent digital screening experience that allows doctors to receive the right information at the right time.

Working in a 4-person team, I was involved in the entire design process and worked closely on research, concept development, and prototyping.

Role

Product Designer

Timeline

4 months

Featured Skills

Interaction Design

Prototyping

Participatory Design

UX Research

Team

Malavika Doshi

Anika Jain

Jason Wei

Initial Problem

Patients are routinely asked to complete symptom screening forms, but this information isn't reaching their doctor.

Only 32.7% of new patients and 57% of patients receiving chemotherapy complete these forms.

Even if patients do complete these forms, the information isn't reaching providers, which means that patients who screen positive for certain needs oftentimes do not receive timely interventions.

Our goal was to make the forms more useful for providers, while reducing burden on patients.

Solution

A smarter symptom screening experience that prioritizes the symptoms that matter most.

01

Urgent symptoms take priority

Symptoms that require immediate attention appear at the beginning of the screening, ensuring that they're identified quickly. When an urgent symptom is selected, the system immediately alerts the patient to contact their provider.

02

Logical symptom grouping

Symptoms are organized into categories that align with how providers think about symptoms.

03

Dynamic questions

Follow up questions appear as needed, based on the symptom selected by the patient. Linking symptoms with relevant follow-up supports providers in clinical decision-making.

04

Summary and resources

The system generates a summary of the most important symptoms to discuss with the provider at the next appointment. Helpful resources appear alongside reported symptoms.

Research

We interviewed 2 providers, conducted on-site observations, and audited the screening forms to identify challenges with the current system.

Why don't patients like filling out the forms?

01

They answer the same questions across multiple forms only to have them asked again during the appointment.

Why am I being asked this question again?

02

They feel as if their answers disappear into "the void". It's unclear how filling out the forms benefits their care.

Does anyone even look at these forms?

Why is this information collected but not used by providers?

01

UPMC is legally required to administer screenings, but there's no protocol attached to them.

02

Competing priorities make it easier to ask about symptoms during the appointment rather than look at forms ahead of time.

We had generated preliminary ideas based on the research, but they weren't grounded in the actual content of the forms.

As a different approach, we conducted a co-design workshop to find out how providers perceive and prioritize the screening questions.

We found that the current screening forms use a one-size-fits-all approach that doesn't match how providers think about symptoms.

01

Providers want to know about urgent symptoms first, but these get buried among less urgent ones.

02

The symptoms that matter evolve throughout the cancer journey, but patients are asked the same questions regardless.

03

Providers mentally categorize symptoms by body systems, but the current forms lack this logical organization.

The consequence? A lot of potentially valuable clinical information falls through the cracks and patients may miss out on timely interventions.

Reframing the Problem

How might we bring more clarity, ease, and efficiency to the symptom screening process so that urgent concerns from patients are captured and addressed at the right time?

Design Process

How can we better structure the information to align with what doctors want to know?

❗Providers want to know about urgent symptoms first, but these get buried among less urgent ones.

⭐ Organized symptoms based on urgency

I created an "SOS" category, appearing at the start of the screening so symptoms requiring immediate attention can be identified first.

Patients are asked the same questions regardless of what phase of the journey they're in.

⭐ Adapted the symptoms to the patient's current phase of the cancer journey.

After the SOS category, patients see typical symptoms specific to their current phase rather than generic questions asked to all patients. This makes the screening feel more relevant.

Symptoms are scattered without any organizational structure.

⭐ Grouped symptoms into categories that align with body systems.

Existing forms treat each symptom as a standalone data point, missing critical patterns.

⭐ Asked targeted follow-up questions

Connecting related symptoms helps providers evaluate symptoms holistically. Selecting "tingling sensation" triggers a follow up about joint mobility since this combination indicates neuropathy.

How can we reduce cognitive load and completion fatigue?

The current forms are overwhelming and present too much information at once.

⭐ Presented one category at a time and simplified how questions are asked.

The categories are presented one at a time, limiting the amount of information shown on the screen. Instead of answering each question one-by-one, the patient is presented with a list of symptoms and selects the ones that pertain to them.

How can we make the forms more actionable?

Patients feel as if their answers disappear into the void.

Patients and providers like to discuss urgent issues verbally.

⭐ Created a summary page with resources

The summary page serves as a reminder for the patient to bring up these symptoms to their doctor at the next appointment and connects them with relevant resources based on the selected symptoms.

Testing and Iteration

Highlighting urgent symptoms is necessary but not enough

From user testing with a UPMC oncologist, we found that when patients report emergency symptoms, they should contact the care team immediately, so I added an instant call-to-action that appears when patients try to proceed after selecting a critical symptom.

Before

❌ Patients can report emergency symptoms but receive no guidance on what to do next.

After

When patients select an emergency symptom and try to continue, a modal immediately appears directing them to call their care team.

Conclusion

Presenting to UPMC stakeholders

At the final presentation, audience members acknowledged the chaotic nature of the challenge we had tackled and were enthusiastic about the solution's potential to improve the symptom screening experience if implemented.

Of course, implementing new solutions in healthcare involves navigating complex workflows, regulatory requirements, and organizational change management. Our project doesn't solve these challenges, but it does provide a vision for how symptom screening could work better. If UPMC chooses to take on this initiative, our project offers design strategies they can build upon.

Potential Impact for UPMC

⭐ Increased form completion rates

Moving from 32.7% (new patients) and 57% (chemo patients) to higher completion rates provides more robust clinical data.

⭐ Earlier detection of critical issues

Flagging urgent symptoms sooner could prevent emergencies and complications.

⭐ Delivering more actionable information

Providers receive organized, treatment-specific symptom data with urgent concerns prioritized for efficient decision-making.

⭐ Reducing administrative burden

Consolidating multiple screening formats into one streamlined digital experience would reduce staff time spent on form management.

Let's work together! ⊹ ࣪ ˖₊˚⊹⋆

© Katherine Niu 2025

Let's work together! ⊹ ࣪ ˖₊˚⊹⋆

© Katherine Niu 2025

Create a free website with Framer, the website builder loved by startups, designers and agencies.