The Problem

"80% of people with T1D stop seeing an endocrinologist after turning 18 for several years." - Bill Polonsky, Founder of the Behavioral Diabetes Institute

Design Challenge

Designing an empathetic platform targeting youth with T1D who are preparing to transition to adult care, preventing long term damage to their health.

The Solution

After many rounds of user interviews and product iterations, we designed and prototyped a mobile app that connects young people with T1D with mentors who can help guide them through the transition from pediatric to adult care.

"What are we doing?"

Other teams partaking in the Diabetes Design Initiative were college students and assigned specific design challenges. Our graduate team, however, was given nothing other than the overarching theme of this quarter's projects: empathy.

In keeping with the theme and to find our design challenge, we decided to reach out directly to the T1D community. To start, we hosted an ideation workshop with Loop and Learn, a group of people with T1D who use DIY Loop to manage their diabetes. You can learn more about DIY Looping in the link above.

Our workshop included a wide range of ages and experience living with T1D. Attendees ranged between a 16 year old teenager who had recently been diagnosed, to Joanne, the founder of Loop and Learn, who has been living with T1D for several decades. During this workshop, we discussed common issues people with diabetes (PWD) go through.

We learned about just how serious T1D is, and the effort it takes to keep oneself alive. One attendee described it as "Having to tame a monster inside of you 24/7, or else it will kill you".

With a better understanding of the disease, we proceeded by brainstorming needs in the T1D community (better technology, community, mental health, education, etc), and technology platforms that could help address those needs (apps, chatbots, online forums, etc).

From here we started matching needs with technologies and brainstorming potential solutions. We wanted to get a better understanding not only of what was possible, but what had been attempted and had failed in the past.

For example, we decided to stay away from anything involving online support forums, as a previous experiment with that solution had resulted in disaster.

Now that we had some options to choose between, we began narrowing our scope a bit. As a group, we discussed the best ideas we had and listed them out.

After some discussion, we decided we wanted to address the lack of community / educational resources for newly diagnosed PWD. Our workshop attendees provided us with other contacts, including Bill Polonsky, founder and president of the Behavioral Diabetes Institute.

During our conversation with Bill Polonsky, we learned this astonishing statistic: "80% of people with T1D stop seeing an endocrinologist after turning 18 for several years." This can result in long-term, irreversible damage to the health of these young people.

Most young people don't like being told what to do by their parents, but if you have T1D, listening to them can be crucial to your survival. But not every young person understands this, and when they leave home to attend college or just live on their own, they may have difficulty managing their T1D in a new environment.

After learning about this problem from Bill, we decided we wanted our target users to be teens (16-18) getting ready to transition to adult care. We reached out to students with T1D in that age range to learn more about their experiences and needs.

These are a few takeaways we got from those interviews:

• Students with T1D have a strong desire to connect with other people their age who "just get it".
  ‍
• Several teens expressed interest in learning about managing diabetes especially as they go to college and leave the care of their parents.
  ‍
• There are some questions they'd rather ask someone they view as a peer instead of their parents.

Based on these interviews, we came up our design challenge:
‍
How might we connect young people with T1D to mentors who can help guide them through the transition from pediatric to adult care?

Ideation

With a design challenge in mind, we began ideating on possible solutions. To start things off, we created user personas for mentor and mentee users. We wanted to better understand who our users would be, and why they would use our product.

We really wanted to hone in on this mentee/mentor or "Diabuddy" relationship for a couple reasons:

• We learned during our workshop that for PWD, having a strong community and support system is extremely important. ‍
  ‍
• Two members of the Loop and Learn group (Joanne and Cassidy) really stood out to us. Their Diabuddy relationship exemplified what we wanted our users to have.
  
  Their relationship started off with Joanne (who is now in her 60s), mentoring Cassidy (30s) on how to DIY Loop and manage her T1D, but has now blossomed into a strong friendship.
  
  Cassidy told us about how the responsibility of managing a minute by minute, life or death disease like T1D with very little training could be extremely isolating until she met Joanne. Joanne was there for her through success, failure, questions, etc and Cassidy was comforted knowing that someone who understood was there for her.

Joanne and Cassidy also explained to us that although their relationship was ideal, it isn't because they just both happen to have T1D. Like any other relationship, factors like personality, interests, how they manage their diabetes, etc are also important.

Now that we had some personas down, and keeping in mind the importance of compatibility between Diabuddies, we began ideating on what features we would need to design through screen ideation and mapping user flows.

Based on our young target audience, user interviews, and brainstorming as a group, we decided to design an app and began the wireframing process.

Design Decisions

My teammates and I designed initial wireframes individually so we wouldn't get tunnel vision and influence each other's decisions. We then took the best parts from each of our designs, and put them together.

After several iterations of this process, including user testing and feedback meetings with our mentors and Loop and Learn members, we arrived at our final high fidelity prototype.

We focused on 3 main features of the mentee user flow:

• Compatibility Questions Survey:

  

  We decided to move away from multiple questions on the same page to giving each question its own page. By doing this, we reduced visual overload, and included a progress bar so the user could estimate how much longer the process will take.
  

  To reduce effort for the users, we also tried to make as many questions multiple choice as possible. Questions included a range of T1D management questions to hobbies/interests.
  

  

  
  
  

• Finding Potential Diabuddies:

  

  This was arguably the most important screen for users, as it's where they can find potential mentors to be Diabuddies with and connect with. After comparing our initial wireframes, we decided to use carousels to showcase potential Diabuddies of certain categories. By doing this we reduce visual overload while showcasing a similar amount of information.
  

  The top carousel contains mentors who had the most similar answers in the compatibility survey questions as the user. After user interviews and iterating, we decided to add more specific categories with their own carousels as well. For example, people who use the same CGM pump, are pre-bolus, etc are added beneath the top carousel.
  

  

  
  
  

• Messaging a Diabuddy:

  

  This is where users can keep track of their messages with their Diabuddies. In early wireframes, we wanted Diabuddy relationships to be "exclusive". We realized this was detrimental to the mentees who might want to ask different mentors different questions. In future wireframes, mentees could message as many mentors as they want.
  

  

After we had the Mid-Fi wireframes completed, we also had to consider a logo. We wanted the app to look as professional as possible, as we would be presenting it to an audience of professionals in the diabetes medical device community.

• Logo:

  

  In 2006, the United Nations declared the blue circle as the symbol of diabetes worldwide.
  
  By intertwining two of these symbols, we represent a Diabuddy relationship as a venn diagram; Two different people supporting each other through their shared disease.
  

User Feedback and How We Responded

Our team was constantly in contact and doing user research/testing with members of the Loop and Learn community, as well as people from our target audience. We wanted to make sure we were always on the right track with our users in mind.

Here's some of the feedback we received and how we addressed it:

• Mobile App's feel impersonal:

  Some Loop and Learn members have had bad experiences with mobile apps in the past, and were wary that they lacked a personal touch. Some users that we interviewed also mentioned they wished the survey questions contained more vernacular a person with diabetes would be familiar with.
  

  

  In order to address both parties' concerns, we took the time to research more specific diabetes related questions, and added them to our survey.
  

• Seems like a dating app:

  People we interviewed also mentioned that the early aesthetic / wording of the app made it seem like a dating app. This was because our initial wireframes displayed mentors as "matches", and only allowed you to "match" with one person at a time.
  

  In our final prototype, we made sure to change any language referencing "matches" to "potential Diabuddies". Users could also reach out to as many potential Diabuddies as they wanted, increasing choice and erasing "exclusive relationship" connotations.
  

• Safety features for minors

  Because many of our target users are minors, our user interviews with them often included parents as well. Many parents brought up worries over their children's safety and privacy, and we sought to address their concerns.
  

We decided to include a third type of account: the "Parent/Guardian". A minor who signs up for the app must connect their account to their parent's, and the parent has the ability to see who their child is messaging and monitor their activity. Parents also have the ability to message mentors that their child has reached out to.

Most teens obviously don’t want their parents to see exactly what their text messages include, so parents have a toggle to view these messages and the parent and child can discuss if this needs to be turned on. When a user turns 18 they have the option of disconnecting their account from their parent's.

Also as an added level of security, mentors are unable to message mentees unless the mentee reaches out first.

We also included disclaimers during the onboarding process so users can have a safe, transparent experience.

Results and Reflection

After several rounds of user interviews and product iteration, we showed that there is both a need for and path towards an app like Diabuddies. Several of the potential users we interviewed expressed interest in having a mentor who can help guide them through college, or just be there when they hit a rough patch.

We finished up our 10 weeks by presenting our prototype and research findings to an audience of experts in the T1D community. Our hope is that we generated interest from these experts, and their companies can consider allocating resources to the continued design and engineering of Diabuddies.

I personally had a great time working on this project. It was very fulfilling to work on a project that I knew would help people. Members of my own family have diabetes, and I learned a lot about the disease and the struggle they go/went through.

Our next steps would be to iterate our prototype further based on our last round of user testing. We also hope to collaborate with medical device companies we presented to like DexCom and Tandem, and place our design in the hands of talented programmers T1D community leaders who can help develop the app.

Special thank yous to Joanne and Cassidy for always answering our diabetes related questions, Lars and Heidi for your guidance, and Maya, Kian, and Hannah for being great teammates.