Helping at-risk patients find providers who understand their nuanced care needs

Under the direction of an industry sponsor, we worked to re-design a doctor search tool. My team focussed on helping patients find doctors that truly understand their unique medical and personal values to create long lasting relationships, reduce doctor shopping, increase satisfaction, and reduce doctor avoidance.

Under the direction of our sponsor: a major health insurance provider, our goal was to design and visualize a unique doctor search experience, to present our work to our clients, and to capture and communicate our design through documentation.

My Role

All team members participated in ideation, secondary research, and concept development. Together with Donglu, I worked on the prototype and visual design of the website. For our final presentation, I created an animated walkthrough of the user interface.


Group project for ideation studio with Tobin Valenstein, Kristen Olsen and Donglu Song

Role/Project Aspects

Research, ideation, design, concept presentation


Oct 2014 - Nov 2014 (6 weeks)

Secondary Research

Secondary research is a great way to very quickly understand a new industry domain. We poured over relevant white papers, journals and articles looking at the Affordable Healthcare Act, recommendations for how to find and choose a doctor, how the United States healthcare system works, and why people switch doctors. We also looked at current tools used to find a doctor and what kinds of information they offer.

We identified two major areas that degrade the doctor-patient relationship: when someone is part of a socially stigmatized group, and when a person's chronic condition or conditions are not being addressed comprehensively.

Secondary Research: Rapidly understanding a new domain


We then looked at different venues used to inform doctor searches such as websites like ZocDoc, WebMD, and RateMD. We also looked at other mediums like Seattle Magazine's Best Doctor list, and board certifications. We compared sources by expert vs consumer opinion, and qualitative vs quantitative data.

It became clear current tools focus on impersonal data and do not facilitate easy decision making. We saw a design opportunity to find a solution that focuses on qualitative data

2x2: Finding areas of opportunity

Affinity Diagram: Search criteria

Having narrowed down on an area of focus, we set out to brainstorm ideas for a solution. I sketched out a concept for the UI before the group collaborated over some paper prototypes to be able to discuss finer points of the interface.

As a team, we struggled to decide on the flow of the patient questionnaire. We created a paper prototype so we could have something more tangible to play with. It also allowed us to do some quick and dirty user tests with classsmates

Sketching the layout before diving into Photoshop

Simplified user-flow diagram

Questionnaire Experience

Our design initially provides users with basic search results and suggests that they use the personalized search filters to find better matches.

Patient preferences are then optionally saved to their profile to automatically provide personalized preferences in the future

Following are some of the initial designs of our questionnaire. These were then modified to the aesthetic you saw at the top of this page.

The first step prompts the user to select which kind of doctor they want (GP, podiatrist, psychiatrist, etc) and the clinical interests they want their doctor to have. Clinical interests can be broad categories like cardiac health or can be specific like astigmatism.

The second step is to determine the style of care the user wants: do they want lengthy conversations or are they more interested in brief and to the point consultations, are they medication adverse, are they surgery adverse?

The third step determines if the user has any language needs, they may be speaking english as a second language and would feel more comfortable with a medical provider who speaks their language. We included a fluency scale since not everyones language requirements or skills are at the same level.

The fourth and last step is the most personal. It ask for the user to identify and core parts of their personal lives that they want to be reflected in their medical care. There are five categories: religion, ethnicity/culture, personal identity, and medical identity. We made a distinction between personal and medical identity since not everyone feels that their conditions represent or define them.

An early concept of high-fidelity UI


Our team presented our solution to the clients at their headquarters. In addition to the design, we were commended for tackling a really hard problem and thinking through how the solution would impact the workflow for not only the patients, but also the care providers and insurance company.