An award-winning Master thesis project, HealthTalks is an Android app that allows patients to record a small summary of their appointments dictated by their physicians. The app then transcribes that summary, provides definitions for each medical term used, and highlights the prescribed medication.
Team Lead, UX/UI, Researcher, Developer
Android Studio, Adobe XD, LaTeX, R Studio
Oct 2016 - Dec 2018
After extensive secondary research, our understanding of the doctor-patient relationship and a patient’s personal health information management habits was quite thorough but didn't fully cover the Portuguese population, specially in our region. So we decided to survey those patients, focusing on three major issues: 1) note-taking behaviours regarding personal health information; 2) relationship between patients and doctors and both parties’ behaviours during a medical appointment; and 3) their interest in the functionalities we had planned for HealthTalks.
Our goal was to evaluate the characteristics of the patients in our region and their acceptance of our solution, HealthTalks.
Given our scope, we tried to create questions that were unambiguous, easy to understand, and avoided biases.
We tested our prototype questionnaire with 10 people of different education levels to ensure it was universally understood.
The pilot study heavily influenced the structure, layout and phrasing of the final questionnaire, which was answered by more than 1200 people over a period of 3 months. The answers gave us a clearer picture of the issues Portuguese patients are facing, what they currently do to solve them, and invaluable feedback for the development of our app.
The general consensus among UX researchers is that five users should be enough to find the vast majority of errors in usability testing. However, since we had a very heterogeneous group of possible end users for HealthTalks, that number could be insufficient to represent all users. Hence, we deliberately tested the app exclusively with seniors, a target group that was expected to benefit greatly from this solution, while at the same time assumed to be one of the less receptive to it. A consequence of that choice was that, if our 5 seniors were indeed less technologically versed, then some problems that only more proficient users would find remained unidentified.
We wanted to weigh the results with a quantitative measure of the testers' perspectives on technology to identify possible biases. To that effect, we employed the Technology Readiness Index 2.0, a tool created by A. Parasuraman and Charles L. Colby for that purpose.
Our app had many concepts for which there are no standard icons, most of them related to the medical field. As such, it was our goal to discover what icons people most associated with the terms we used to classify and organise data.
We used the moderating method “concurrent think aloud”, which meant that the testers were incentivised to express their thoughts and reasoning while completing 7 increasingly complex tasks. We also kept track of time spent, number of clicks, and task success.
HealthTalks was a complex and far-reaching thesis of which I am extremely proud. I was given a lot of freedom in my approach, and I appreciate the gentle but firm supervision I had while navigating so many of the themes that I love. It made me learn a lot about the scientific method, which helped to mold my way of thinking and approaching problems. It was also an excellent experience in writing and presenting, which I have made use of ever since.