ABOUT THE PROJECT
I designed HealthTrack for my Master's thesis at Maryland Institute College of Art. It is a mobile application that guides patients through documenting their symptoms and organizes the information so that it is easily accessible.
Project Manager, User Researcher, UX/UI Designer
Figma, Google Forms, Google Sheets
8 weeks from October to December 2021
How might we help patients reduce their chance of diagnostic error?
A mobile app that guides patients through documenting their symptoms, recognizes trends in their health, and organizes the information so that it is easily accessible.
WHAT IS DIAGNOSTIC ERROR?
"Diagnostic error can be defined as a diagnosis that is missed, wrong or delayed, as detected by some subsequent definitive test or finding." - Johns Hopkins Medicine
WHAT ARE THE CAUSES?
There are reasons within the healthcare system, the physicians, and the patients. For this project, I focused on the patients. In my research I found that...
they may be unfamiliar with medical terms or they might not provide timely and accurate information around their symptoms and medical history.
I began the research phase by first uncovering assumptions to validate during my user research process and whittling them down to these core three:
People with ongoing health issues have difficulty organizing their health information.
People have challenges with accurately relaying their medical symptoms to their doctors.
People are comfortable with entering their medical information online to third party applications.
To find users, I sent out a screener and received 11 responses. 3 people said that they had experienced difficulty getting diagnosed, and they had recorded their symptoms. I chose to interview these 3 users and 2 others who had said that they do record symptoms, but had not experienced difficulty with diagnosis.
2 20 - 29
1 30 - 39
1 50 - 59
1 60 - 69
Interview question topics
Why they records symptoms
How they record symptoms
How they keep track of records
Issues in getting diagnosed
Trust in using 3rd party apps
People with a history of chronic illness have experienced more trouble with getting diagnosed than those who don’t have chronic illnesses.
People with a history of chronic illness have found recording symptoms helps in identifying causes, patterns, and treatment efficacy.
People make their best judgements on what symptoms are important to record whether that be physically or mentally.
What people record does not always match up with what the physician asks them in their visit.
People have difficulty managing and organizing their health records which can be in multiple places.
People are open to trusting third party applications with their medical information.
To synthesize the users I interviewed, I developed this persona, which defined the user goals, challenges, and needs.
Since I wasn't able to consult with subject matter experts in diagnostic error, I read many academic papers on the subject, which lead me to 3 design principles.
Allowing users to record their symptoms and make notes however they need to as there are many different symptoms.
Not influencing the user in what to do with their health. There are many applications out there that attempt to diagnose patients, put ideas in their minds, and tell them what they should do, however these tools can be harmful. They induce panic, and they do not consider patients' family history or regional trends.
The solution should teach users how to explain what they are experiencing in words that physicians use. It should bridge the gap in communication between what the patient is experiencing and the physicians' diagnostic procedure.
COMPETITIVE ANALYSIS & HEURISTIC EVALUATIONS
My next step was to research competitors in the space. I found that healthcare companies were promoting pdf printouts of tables and mobile applications where users can record their symptoms. I found it interesting that of the users I interviewed who recorded symptoms, none used either of those methods. Two users said that they used the Notes app on their phones, and one user said that they used a pocket calendar book, binder, and spreadsheets.
I performed heuristic evaluations of the symptom tracker mobile applications on the market and found that they were either overwhelming and confusing to use (CareClinic, Symple), or not specific enough for the user (Tally). The applications that were overloaded with features, could be unusable to people when they are sick and their minds are not as sharp, making it difficult for them to record what they are experiencing. On the other end of the spectrum, blank documents lacked all features.
I began the design process by making the decision to design a mobile application. I felt that this was the best way for the tool to be easily accessible whenever users need to enter information as well as portable with them as they went to doctor's visits. I then created basic user flows of what the logical steps should be for the users’ main tasks: entering known and unknown conditions.
Wireframing in low-fidelity helped me quickly flush out each step in the user flow as well as get user feedback early on and iterate quickly. At this stage I was able to define the main features of the application.
To test my designs, I recruited participants for usability testing. I was able to test with 4 users. Here are the most common usability issues that came up and the changes I made.
Passcode log in to Face ID
First question upon logging symptoms
Insights placement in symptom logging flow
Record symptoms for known and unknown symptoms
Users may have health illnesses that they have been diagnosed for, so they may add these logs from their user profiles and enter records for them. They may also experience symptoms that do not belong to their health conditions and need to enter those records.
Points of education and opportunities to learn more
This will allow users to not have to go to other resources to learn about their symptoms and have the chance of experiencing the negative side effects of researching their symptoms online. Here, the users can learn from a trusted resource and get relevant information.
Logs with questions that are specific to the health issue
The questions that are prompted to the user will depend on the symptom that they have selected, to ensure that the patient is entering all of the information that the physician may ask of them in a doctor's visit.
The tool will give the user metrics if it notices trends in the users’ symptoms.
The user can view their symptoms in a calendar since this is the most common and fundamental way users organize their symptoms. It also gives them the ability to add symptoms to past dates.
REFLECTION & NEXT STEPS
As I reflect on this project, the biggest challenges I faced were gathering user research as it was difficult to find users who could pass my screener. I also wish I had the opportunity to speak with physicians, but those I could get in contact with were not able to in the time that I needed. I think this is a very interesting and serious topic that hasn’t been properly tackled, so I would love the opportunity to take this project further. From this project, I learned to work with what I have and be scrappy. Which meant spending a lot of time reading through dense research papers.
As for next steps, I would like to further explore more ways I can create meaningful insights for users as users were heavily interested in that feature. I would also like to conduct contextual research and have users try using the application when they are experiencing symptoms and see what they experience when they bring this into their doctor’s visit. It would be highly interesting to test this with a group of patients within a hospital and see if the hospital experiences reduced diagnostic error as a result.