CASE STUDY
Building AVA Care’s Care Recipient Experience
Over 82 millions Americans are 65 or older, and half of those take prescription medications every day. Not only is the rising population of seniors driving the demand for prescription medications, but it’s also driving the demand for caregivers.
Givatar’s mission is to ease the demand for caregivers using AI technology. Their product, AVA Care, is an AI assistant in the form of a tablet application. AVA helps care recipients stay on top of their health plan, and keeps family members up to date on their health progress.
I joined the team as a freelance product designer and manager to create an end-to-end tablet interface that works in tandem with the existing desktop site. The desktop site is for caregivers of patients or loved ones who plan to use the tablet. Caregivers input the medical history and medication schedule that will be used to create a more customized tablet experience.
TIMEFRAME6 months + ad hoc
TOOLSFigma
Maze
MY ROLEUX Designer
UI Designer
Product Manager
THE PROBLEM
The existing desktop website only assists caregivers with patient onboarding
The initial tablet design is too clinical and cold
THE SOLUTION
Design an end-to-end tablet experience for patients to input health updates and share information with their care team
Create a notification system to help patients develop a daily check-in habit
RESEARCH
How do care recipients & caregivers manage health plans?
Understanding how care recipients and caregivers track and manage health plans
After de-briefing with the company founders, I came into the research phase expecting seniors cared for by unpaid family members to be the target user. To validate this, I interviewed six individuals, including three seniors with health conditions and three family caregivers.
With seniors, we discussed how they communicated with their caregivers and tracked their own health progress.
With caregivers, we walked through the desktop site to understand if the requested health inputs were relevant to the health updates they wanted.
“I manage my own doctors visits and medications schedule and definitely prefer it that way”
— DR, 85 years old, Care Recipient
“While I check-in on my mom often, she also relies on live-in care and my brother to help with different aspects of her health plan”
— CC, 58 years old, Caregiver
“My daughter and I talk weekly over the phone about my health and anything else that comes up”
— TR, 75 years old, Care Recipient
User assumptions proven wrong
Care recipients vary greatly in level of health support
I was surprised to find a lot of contradictions to my initial assumptions. A senior’s care network was often divided by many individuals including family, paid care, and doctors who provide varying levels of support.
On one end, seniors with low to moderate health issues preferred to stay as independent as possible, managing their own doctors appointments and medications schedule on a regular basis. They prefer to check-in with their family members over the phone or on their health portals.
On the other end, seniors with severe health issues relied heavily on live-in paid caregivers. Family members primarily check-in with the caregivers for health updates.
4 out of 6 participants have regular check-in phone calls with their family member or care recipient
4 out of 6 participants use their health care provider’s desktop portal to track a family members health or their own health
5 out of 6 participants believe that care recipients are still very independent and like managing most aspects of their day-to-day health themselves.
4 out of 6 caregivers check-in on the health of a care recipient along with many other people (siblings, paid care, etc.)
Who benefits the most from AVA Care?
Developing personas that encompass a range of needs
I envisioned two personas that represented the users most likely to maintain a check-in habit on AVA Care. Persona One is an adult with a temporary injury that needs frequent monitoring for a short period of time. Persona Two is a paid caregiver who provides regular care for an adult with a serious health condition.
Persona 1
Independent, Low to Moderate Health Problems
Persona 2
Dependent Patient, More Severe Health Problems
STRATEGY
Dual focus
How might we design for multiple user types?
Two ‘how might we’ statements were identified to help further focus the project and better prioritize our target users throughout the strategy phase.
How might we help caregivers create a better system to deliver consistent health updates to family members?
How might we help care recipients build a daily wellness check-in habit and log their health progress?
Key Features
Defining the MVP
After reviewing the research results with the team, we spent some time ideating on features. It was important to enhance the existing feature set by brainstorming on new ideas. After gathering the list, I started prioritizing what features were necessary to launch the product.
Must Haves
The core set of features I would need to designDashboard
Voice Assistant
Wellness Checks
Wellness Journal
Medications Log
Nice to Haves
Additional features that are build in addition to the core feature set, timeline permittingLocal Time/Date
Share Button
Wellness Goals
Calendar
Can Come Later
Fun ideas that can come later after the MVP is launched and testedGames w/ AVA
App Messaging
Guided Fitness
Video Calling
Doc Integration
911 Calling
Cust. Avatar
Lifestyle Tips
Sitemap Revisions
Mapping out the tablet
The existing desktop sitemap is for caregivers to input health details for their care recipient. It was important to revise the tablet sitemap for care recipients who intended to check-in with AVA and track their health progress.
The revised sitemap includes the new features meant to assist users with their health tracking process.
Desktop Sitemap (Old)
Tablet Sitemap (New)
How will users complete key tasks?
Envisioning three key task flows
I outlined the key tasks flows that a user would be most likely to start. This includes answering questions prompted by AVA and creating notifications to keep track of the health plan.
WELLNESS CONVERSATION: Checking in with AVA
WELLNESS JOURNAL: Adding a new goal
TRACKING MEDICATIONS: Scheduling Medication Notifications
DESIGN
Ideating on feature layout
Low Fidelity Sketches
I sketched out two to three variations per core screen. Because maintaining user engagement was important to the clients, I explored layouts that were more playful and interactive.
Defining the brand
Style Guide
The color palette was selected to bring a more personable and engaging energy to the brand. The team wanted to stay away from a clinical design, but still deliver a clean and professional image.
Building the layout
Mid-Fidelity Screens
Three core tasks flows representing the most essential features were mapped it. I incorporated a combination of the low fidelity elements to create the most harmonious layout that best aligned with the style guide.
WELLNESS CONVERSATION: Checking in with AVA
WELLNESS JOURNAL: Adding a new goal
TRACKING MEDICATIONS: Scheduling Medication Notifications
V1 Designs
High-Fidelity Screens
Testing
Self-guided testing
Validating the design with usability testing
I recruited seven self-guided usability test participants to understand if users could complete the three primary tasks with little error. Using heat mapping reports I identified clear pain points (in red) throughout each task.
Subjective test conditions
Analyzing the test results
I had trouble finding participants willing to participate in the test that also represented caregivers. I had to open the test to a more varied groups of users that might not typically use this product, which may have impacted the results.
Additionally, since most participants skewed older and less tech savvy, many had trouble navigating the test platform (Maze) and understanding the test format. This resulted is more mis-clicks than if I was available to answer questions during the test.
4 out of 7 users were able to complete all tasks without any errors
5 out of 7 users would be very likely to use a tool like this to monitor their health
6 out of 7 users were able to complete all tasks in under 5 minutes
Optimizing for Visibility
Frequency to Severity Mapping
Using a frequency to severity map I was able to organize participant feedback and prioritize iterations I would make. The more frequent and severe the feedback, the more likely I was to implement the iteration.
Making Iterations
Dark Mode Screens
As a tablet with an always-on screen it was important to reduce the brightness from the screen while remaining easy to read.
Visible Features Placement
Important features like the chat microphone needed to be placed in more prominent areas to encourage user engagement
Share Button
A share report button was added as quick action to encourage users to update their care team
Filter Hierarchy
Create a clear hierarchy between different filter types to help define filter order
Relevant Color Coding
Revise colors from ailment’s bubble chart to correlated ailments by color and better visualize health trends
Final Prototype
Key Learnings
Keep an open mind with your target audience until you complete interviews
Your assumptions can skew your beliefs about the target audience. Recruiting from a diverse pool of participants will only improve the quality of your work.
Set clear boundaries around your project timeline and client scope
Clear communication about project expectations and your timeline are key. My client had many great ideas about the different site features. It was necessary to prioritize only the most important features in order to deliver a high quality product.
Future Considerations
The design phase of the project was completed in a 6 week sprint, however I am supporting the development phase, which will include
Revising the always-on home screen to cycle through multiple screen options
Handing off the Figma file to the developer
Testing the prototype on the actual tablet with touch-to-text capabilities