Bento
Bento is a toolkit that helps newly diagnosed IBD patients to learn and keep track of their diets
Team: Jiasi Tan, Elva Fu, Audrey Zheng, Tzu-Chao Zhou
Contribution: Design Research, UX/UI Design, Visual Design, Interaction Design
Context: CMU School of Design, Learner Experience Design Course Project, Client: UPMC
Skills and Tools: Interviewing, Think-Aloud, Storyboarding and Speed dating, Wizard of Oz, Experience Prototyping, Sketch, Principle
For full documentation of our design and research process, please visit our Medium publication:
The Challenge
UPMC aims to reduce the readmission rates of patients by closing existing learning gaps often caused by a lack of resources and understanding that result in inactivity by those involved in the care process. This project called for us to the design of learning experiences that focus on educating patients or physicians, caregivers to reduce readmission rates at local medical facilities.
Our Solution
Bento is a toolkit that helps IBD patients to learn their new diets on an interactive website during inpatient, and keep track of their diets with a personalized food journal after discharged.
Phase 1
In-patient using the website
01 - Patients learn the high level diet principle
02 - Patients learn what they can eat and why
03 - Patients learn real-life special cases
04 - Help the patient learn how to plan balanced meals and supplement their nutrition intake
Phase 2
Post Discharge, using a personalized food journal
Minimum effort required increase patients’ motivation using the food diary. Patients will have handy reference to the safe/trigger food list. The planned well balanced meal will also loaded to the diary.






The toolkit also comes with the food cards that serves as a handy resource, as well as patients encoding memory.
Process
01
Investigating problem
To investigate the problem, in the initial research phase, we used some blockbusting activity and applied different technicals to better analyze the problem from different perspective.
Interviewing
Transition Care Nurses
After the initial problem investigation phase, we visited UPMC and interviewed two transition care Nurses (TCs). UPMC health plan (the insurance company) hire TCs to specifically deal with readmission issue. Their job is to make sure the patient is safe for discharge.
After visiting the hospital and talking to the two TCs, our team gained more in-depth knowledge of readmission. Which helped us on narrowing down the project scope and define the problem.
Our Focus Area
After the interviews, our team decided to focus on IBD (Inflammatory bowel disease). IBD is a disease that could only be maintained but not cured. Having this disease means that patient need to adopt a major lifestyle shift (especially eating) for the rest of his life. The reasons why we have decided to focus on IBD include:
Secondary Research
Understanding IBD
By conducting further secondary research, we got to know more about the facts of IBD regarding diagnosis, treatment, stress, symptoms, diet and support.
We then summarized all our findings and identified the major problems that IBD patients have.
02
Identifying Gaps
After we have investigated the major problems in the current states, we described the characteristics of preferred states where we imagine problems as eradicated. Our team then determined the types of gaps that emerged, which directed our focus in defining our teaching goals.
In the book, <Design for How People Learn>, Julie Dirksen provides a structure for determining learning gaps that also apply broadly to human-centered design. She presents the common gaps that essentially bridge current and preferred states:
• Knowledge gaps: audiences don’t have enough information to perform the task
• Skill gaps: audience lack sufficient practice relative to a specific task
• Motivation or attitude gaps: audience have pertinent knowledge and skill but aren’t interested in performing the task
03
Defining Learning Goals
Based on our framing of current and preferred states, our team defined what we aim to teach our audience in an effort to address the learning gaps that emerged. We believe that by targeting our teaching goals we will create learning experiences that lead our audience to preferred states:
Help the patients further understand this disease and be more reflective about connections between their diet and the symptoms
Help the patients transfer their learning to everyday situations in order to manage their symptoms better
Help the patients learn about how to manage their symptoms through lifestyle changes and help them realize a positive outcome is attainable
04
Ideation
We started by brainstorming many different ideas and scenarios. Our team further discussed and selected several promising ideas and developed these ideas into storyboards. We then did a speed-dating session to test these ideas within our class and also had a panel discussion with staff from UPMC.
Selected Initial Concepts
01 <IBD Chat Bot>
#1 Tim is a busy man. Luckily, IBD chatbot helps notify Tim. This helps with habit formation and reminder (short term + long term goals) - regular check in, medication plan reminder.
#2 Tim can message the chatbot. He doesn’t feel embarrassed talking to the machine. The patient can release stress through conversation with a friendly UI.
#3 IBD chatbot helps with basic Q&A (knowledge) and targeted feedback - what Tim did today, how Tim feels now, and gets feedback from Chatbot (advice and confirmation).
#4 After conversing with Chatbot, Tim remembers to take his medicine at precisely 10:30am.
This concept is more targeting at reducing embarrassment and help with depression, and get personalized and immediate feedback whenever and wherever. Most IBD patients are diagnosed before 30 which means that there are still many years to go. A personalized learning goal from short term to long term is important to provide. However, IBD situations are very different from each other. Personalized treatments are required. Chatbot will be a very goo tool to provide a personalized goal.
02 <Make your Food Choice>
#1 The patient is diagnosed with IBD. Besides giving out the flyer, the patient could play the game to learn about the disease.
#2 Furthermore, the patient could also know the basic principle of their diet plan. But since every patient's case varies, personalized case management will be required.
#3 There will be a version for the caregiver or family member to learn basic knowledge about the disease so they could be more supportive or even supervised
#4 Patients could document their eating and share to the support network and get feedback,
This concept is more about motivation and attraction. We came up with this idea because we figured sometimes it’s not about ‘not knowing what to do’, it’s more about ‘not wanting to do the right thing’. We want to design an intervention that helps patients shift attention.
We presented these two selected ideas to UPMC by investigating problems, identifying learning goals and gaps, defining teaching goals and grounding design approaches.
Revised Concepts
Our team constructed a scenario and storyboard that describe the premise of the design concept. We used them to guide the making of our prototype.
01 - <IBD Website and Physical Journal for learning and tracking diets>
#1 Tim is a busy man. Tim though his symptoms were caused by a stomach flu, but his doctor just diagnosed him with IBD.
#2 To help him manage IBD, Tim’s doctor directs him to a website targeted at newly diagnosed IBD patients.
#3 Tim’s doctor explains that both the website and the food journal are tools to manage and learn more about Tim’s specific food intolerances.
#4 After going home, Tim continues to track his safe vs. flare foods. Tim feels happier knowing which foods to avoid.
02 - <IBD App for learning and tracking diets>
#1 Tim is a busy man. Tim though his symptoms were caused by a stomach flu, but his doctor just diagnosed him with IBD.
#3 Tim’s doctor explains that the IBD app is a tool to manage and learn more about Tim’s specific food intolerances.
#2 To help him manage IBD, Tim’s doctor helps him download an app made for newly diagnosed IBD patients.
#4 After going home, Tim continues to track his safe vs. flare foods. Tim feels happier knowing which foods to avoid.
Learning Experience Concept
Our team used our teaching goals and learning theories to inform our design approach, which is represented in the conceptual model below. Functioning as an experience journey, the model illustrates the sequence, hierarchy, dependency, and relationships among all facets.
05
Grounding Design Approaches
Through the research and design process, we leveraged the following learning theories as structural bases for our learning concepts and to inform the design of their facets.
Structured Flow of Goals
Shows how short-term and mid-term goals are necessary to achieve to reach long-term goals
<Design for How People Work> — Julie Dirksen
Our long-term goal is to help the patient achieve a state where they can full control over their diet, through intervals of learning experience (short-term goal) and keeping a diary (intermediate goal).
Learning Flow Model
Describes the benefit of helping learners acclimate and assimilate to new information
<About Learning> — Bernice McCarthy
We used the learning flow model as the guiding principle for designing the modules of the learning experience to provide continued support and gradual challenge.
The 4MAT System
A simple and effective way of moving through learning
<About Learning> — Bernice McCarthy
The learning modules overall followed the 4MAT system, taking the learner through the principles (why) - the instances (what) - the scenario (how does it work) and the comprehensive (what if?)
06
Prototyping and Testing
Low-Fi Paper Prototyping






Mid-Fi Prototype
Hi-Fi Prototyping
Interaction Testing - Wizard of Oz
07
Final Design
Part 1
In-patient learning activities
On-Boarding
Select the doctor’s recommendations on diets
Activity One - <Learn Diet Principle>
Description: Based on the diet recommendations the patient entered, the system will present a series of food cards, when hovering over, the patient can see if the food fits the high-level principle and how possible it is to induce a flare; the patient can favorite the safer food that he/she likes.
Learning Goal: To reinforce the high-level principles with some examples in order for the patient to contextualize the principles given; build expectancy that there are foods the patient can still enjoy that are relatively safe.
Tap food cards to see if the food is safe to eat
Activity Two - <Arrange Foods by Nutrition Content>
Description: This activity presents more instances of food that more or less of the nutrition component the patient is supposed to pay attention to (e.g. high protein, low fiber); the patient will first arrange the food based on nutrition content, and then adjust their answer based on system feedback.
Learning Goal: To help the patient apply the diet principles to more instances of food in order to practice the principles and make it transferable to more real-life situations.
Tap food cards to see nutrition numbers on the back side
For this card flipping activity, we also designed a physical version, to facilitate discussions between nurses and patients about their diets.
Activity Three - <Dining scenarios - Choose a Drink>
Description: The patient will be presented with more complicated dining choices that require the patient to utilize the principles they have already learned and applied comprehensively. The system will provide prompt feedback and reasons for why certain choices are beneficial or not.
Learning Goal: To help the patient apply the diet principles to more complicated real-life situations; to learn about.
Provides context and Immediate feedback
<Dining scenarios - Sushi Or Not?>
Activity Four - <Plan Balanced Meal>
Description: The setting is when the patient is in remission when the patient is supposed to try eating more varieties of food and supplement the nutrition lost during flare periods. The patient will be asked to plan balanced meals with sufficient calorie and they can favorite those combinations.
Learning Goal: To help the patient learn how to plan balanced meals and supplement their nutrition intake during remission.
Feedback of Trigger Food
Part 2
Post Discharge, using a personalized food journal
Reflection
The final presentation of this project was on May, 7th, 2019 at UPMC Downtown Pittsburgh, U.S. Steel Building. The presentation we very well and we got lots of useful feedbacks. It was great that we can talk directly with staff at UPMC who have been working on the readmission issue and physicians who have expert knowledge of IBD.
Next Step
We want to test the effectiveness of our designed learning experience. Due to our time constraints and recourse limitations, we were not able to directly communicate with enough IBD patients. In the future, we plan to talk to more IBD patients to conduct usability testing, as well as to test the effectiveness of our designed learning experience.
What I learned
Empathy is not sympathy. To better help the patients, we better empathy more. Solving a real life issue can be very complicated, as it involved not only expert knowledge, but also complicated emotions. As designers, we should listen not only with our ears but with all our senses.
Keep being curious. We need to constantly be curious in order to learn new things. If we’re not familiar with the topic, then do research, searching online until feeling comfortable and knowledgable with the topic.
For full documentation of our design and research process, please visit our Medium publication: