GreenPages
Helping New Immigrants Find the Right Healthcare in Their Area
Many people begin their search for medical care by asking their friends and family for recommendations. Team Green Bean (Sophie Clyde, Manuni Dhruv, Jacob Xu, and Emily Gui) began this project by asking the question, how do people who don’t have friends or family to ask find medical care? After months of conducting user research with real immigrants struggling with the US healthcare system, we created a website designed specifically to assist new working immigrants in finding a doctor in their area.
Skills: Interviews, Ethnographic Observation, Affinity Mapping, Personas, Storyboarding, Wireframing, UX Design, Figma
Introduction & Project Goals
We started this project with the problem statement, “How might we help people who do not have a network of friends or family in their area to seek non-emergency medical assistance in the United States?” We were initially interested in this problem area because we saw a need to better direct patients to the correct healthcare provider, but soon found that defining our exact user group and what we wanted to help them with would be one of the most difficult aspects of the project. As we learned more about the problem space and our user group, our problem statement continued to evolve and we had to constantly iterate on our design so that we could ensure we met their unique needs.
Gathering User Needs
We began by examining existing systems that help people find the medical care they need, conducting research on how people choose medical care around them, and mapping each stakeholder in this process and the relationships between them. As we began observing online spaces we chose to narrow our focus to helping people new to an area find healthcare when they can’t ask friends or family for help or opinions.
Once we decided on our focus, we began observing online resources and investigated existing tools that help people find healthcare in their area avoiding visiting medical facilities to avoid personal risks of contracting diseases and the privacy concerns that can arise from patient interaction. We also conducted semi-structured interviews with 6 participants to gather information about how people search for medical care, what factors they consider during their search, what information they trust, and how their prior experiences shape their current process. The data we gathered from our virtual observations and our interviews led us to multiple important findings. Some of the findings we gathered included common problems our users ran into like out-of-date information and confusing medical or insurance terminology, and why they trusted or mistrusted certain resources due to perceptions of bias or lack thereof. We used this data to create user personas, task analyses, and to form an initial set of design requirements.
After researching the problem space and gathering our user needs, we realized that we needed to narrow down the scope of the project even more, as anything we designed for such a broadly defined group would be less effective than something designed for a more specific group. We thought deeply about which groups might be particularly prone to lacking friend and family connections in an area, and after much debate refined our original problem statement to “How might we help new working immigrants, who are unfamiliar with how the US medical system works, find non-emergency medical care in their area?”
Design Process & Challenges
We began the ideation process for our design by coming up with anti-solutions for each of our design requirements, then creating metaphors for the search process (“Medical search as a…” library, adventure game, labyrinth, guided tour, etc.). We took the best ideas that came out of our metaphors and combined them to form storyboards for our two best ideas: one that framed the medical search as a dating app equivalent, and another that guided users through finding medical care in a similar way as to a cooking app or recipe guide.
Our analysis of the data gathered during the research process showed us three common themes vital to the process immigrants went through when seeking healthcare: Search/Selection, Education, and Community. As we began the ideation process, we focused our efforts on making sure that our designs would include all three pillars as integral parts of the design. However, as we moved from sketches to wireframes and finally to our our high-fidelity prototype, we found that the time limitations of our project prevented us from developing every aspect of the site we had in mind. We spent many days struggling to figure out a way to fit all our ideas in, but in the end we chose to prioritize the doctor search and selection process, as we knew that this was the core of our project and we wanted to make this process the smoothest it could be for our user group.
You can view a video walkthrough of our high-fidelity prototype here, and view the prototype itself here.
This choice meant we had to exclude many of the educational aspects that we originally wanted to include in our prototype. We instead chose to focus on incorporating education into the doctor search and selection process instead of developing the Explore Library section further. For instance, when a user enters their information to choose their preferences, we added text describing why it is important for us to have that information and how it will help the user in their doctor search process. We also had to reduce the focus on the community aspects of the site. We focused on incorporating a review system in our design instead of larger community spaces like forums or messaging systems, as we felt that user reviews were an integral part of helping users select the doctor they wanted.
User Testing & Results
Due to the time constraints of the project, we were not able to conduct a full evaluation of our high-fidelity prototype and determine how well it met all of our design requirements. We instead conducted a discount evaluation, measuring and evaluating how well the design met two of our eight original design requirements:
-
The system should reduce complexity and choice paralysis through a clear presentation of options and guidance to the next steps.
-
The system should support people with different levels of access to technology and infrastructure.
Our discount evaluation consisted of 20-minute individual moderated in-person task-based user testing with a 5-10 minute post-session survey distributed after the session to collect both quantitative and qualitative feedback. We recruited five students for the discount evaluation, two of which were part of our primary user group. Since all five participants were students at Georgia Tech, we were not able to simulate the different levels of technology familiarity or access to infrastructure resources we might find in a more standard environment. However, we asked the participants to consider the prototype from the perspective of those who lack the technology proficiency and assigned each participant a different background condition (time availability, insurance availability, and means of transportation), so that we could receive feedback that considered the needs of users with varying access to infrastructure resources.
The results of the discount evaluation showed varying levels of success in our prototype’s ability to meet our chosen design requirements, but also revealed multiple ways in which we could improve the prototype to better meet these requirements and the needs and expectations of our users in general. Users liked how doctors' profiles were displayed one by one, as well as the information provided by the featured tags. They were satisfied with the amount of guidance provided throughout the process and agreed that the site matched their expectations and prior experiences with real-world systems. But users also reported feeling overwhelmed with the amount of information displayed on doctors' profiles and had difficulty determining where to go at certain points of the prototype, such as in writing a review. Users also expected to be able to set an appointment directly through the site and became confused when unable to do so.
What's Next?
After analyzing the results of our task-based testing sessions and survey, we used our findings, initial design requirements and past research to develop a list of design implications for our system going forward, focusing on three areas where the design can improve: Information Architecture, Depth of Information, and Overall User Experience. If we were to develop this project further in the future, we’d address these aspects of our design first by making changes such as providing more options in the ‘Logistical Information’ section of Preference settings and making the option to write a review easier to find in the site.
Team Green Bean learned an incredible amount about the process of user research, design, and how to work as a team at every step of this project. Learning to keep our users at the center of our design process helped us design a truly unique and tailored solution for our focused user group that could be further developed into an incredible tool for new immigrants in the future. But another important lesson we learned is that our design will never be finished. Our discount evaluations revealed many things that we could improve about our design, and we could keep repeating the process of improving then testing our prototypes forever. It’s important for us to recognize that we can always improve our design, just as it’s important for us to recognize that sometimes we need to move on to the next project instead of chasing the moving finish line that is “perfection.”
