How I Optimized Patient Search through Design Thinking

My Role

Senior UX Designer

Project Overview

Healthcare providers do not have a lot of information to help them find the right patients and determine if their medical insurance covers services.

Patients are expected to have insurance cards available when setting appointments and arriving for procedures, but most do not. This leaves providers offices to use the patient's name, date of birth, social security number or a combination of the three to figure out if a patient can be seen.

Helping healthcare providers search for patients is critical to delivering improved care.

Process

Empathize
Define
Ideate
Prototype
Test

Empathize through Research

I guided an external research firm who was brought in to supplement the innovation team during the discovery phase of the project. I helped to identify organizations that have shown interest in providing feedback on our products.

Then we facilitated a series of focus groups and user interviews. During this process, I found out that most providers, regardless of size, face similar challenges with limited patient information.

However, I found that larger hospitals and health systems didn't see the value in accessing previous searches, because they don't see the same patients regularly. While smaller provider offices did see reoccurring patients and desired this capability. This key piece of feedback helped me as I worked on the information architecture.

Qualtitative Interview Participants

Total: 45 people representing 40 organizations
Organizations: 60% Small, 18% Medium, 13% Large, 9% X-Large

User Persona Development

In hospitals and large health systems, I focused on the people working the front-desk to inform the basics of the search. I created a persona to showcase how, unlike smaller practices, her reasons for searching for patients were centered around checking coverage and benefits.

Define

Problem Statement

How might we make it easier to locate the right patients and view insurance coverage information quickly, accurately and efficiently?

Goals & Objectives

• Less data entry - Allow providers to find patients without requiring a member ID
• Speed - Enable providers to access previous searches
• Efficiency – Providers should be able to determine a patient's coverage at a glance
• Accuracy – Utilize AI to determine if coverage information is up-to-date
• Flexibility – Providers should be able to check coverage from anywhere

Ideate

Inspiration and Competitive Analysis

I drew inspiration from a number of places, but the three apps that had the largest impact on my design were Google Health, USAA and Apple Music.

Information Architecture Questions

One of the tasks that I facilitated with users was a card sorting exercise using miro to help guide the information architecture decisions. What providers wanted to see varied based on the reason for searching for the patient in the first place.

I discovered that some providers were searching for patients to see if they had active coverage, while others were searching for patients to check the status an authorization or a claim. Additionally, the providers used different pieces of information to search depending on the reason.

Prototype

Instead of jumping into high fidelity mockups, I decided to work with wireframes as long as I could. I used wireframes to layout the information architecture, for a round of usability testing and even to interact with developers to solidify the interaction design. I used figma for both the original wireframes and prototypes, but have recreated a version of the application in App Plus for this usecase.

Mockup

View Prototype

Test

I collaborated with an external research firm to perform two rounds of testing with quick design iterations between sessions based on feedback. I sat in on all of the moderated sessions and took direction from the research team as to what to quickly change in the wireframe or prototype for the next session.

Round 1: Wireframes & Moderated Task-Based Testing

I created a wireframe in figma and the research team provided the wireframe's link to the providers and facilitated the task-based testing. I worked with the research team to create a test plan and script and to make modifications to the wireframe as needed. I also served as an observer for most of the sessions, so I was able to quickly incorporate user feedback.

Round 2: Prototype & Moderated Testing

I designed a high fidelity prototype of a mobile search application for the next round of testing. For this round, I served as an observer for the majority of the sessions, but did have to step in and facilitate a few sessions.

From this testing, I found that most providers would only use the mobile app for quick tasks like checking the status of claims, coverage, and authorizations, but they did not feel comfortable submitting claims or authorizations. This allowed me to focus my efforts on designing quick summaries of information.

Outcome

• By the end of the second round of testing, most users were able to locate patients in 30 seconds or less regardless of the types of information they had available.
• Providers were able to find 87% of patients in the first search attempt.
• On a satisfaction scale of 1-5 with 5 being the highest, providers rated their overall search experience as a 4.5
• 99% of providers surveyed would recommend the patient search application to a collegue.

Lessons Learned

This was a fun project because I worked with a cross-functional team and the developers were involved from the beginning, so we were able to move at lightning speed. However, everything from the design handoff to the development environment was new. We were using Figma to collaborate with the developers for the first time and the developers were working in AWS and leveraging AI/ Machine learning. With all the new things we were doing, we ran into a lot of struggles and a lot of rework.

With that in mind, I took away the following:

• Working in an innovation environment with very tight deadlines can be an incubator for amazing ideas, but settling on one that can be delivered through production can be hard to do.
• As a UX designer, you have to be flexible when working with new technology. There is a compromise between user desires and what can be immediately delivered in a consumable way