Overview

MediPal aims to decrease economic losses in the US due to issues from low health literacy by providing a health literacy AI tool for older adults.

Over 80% of adults in the US have less than proficient health literacy, (1) as defined byas one’s “capacity to obtain, process and understand basic health information and services needed to make appropriate health decisions.” (2) Low health literacy correlates to medication therapy non-adherence and low utilization of preventative care. This leads to economic losses in the healthcare industry. (3) Healthcare providers can help patients with health literacy. Despite being the most accessible health providers, even pharmacists only have a few minutes to do so. A virtual AI pharmacist would allow patients to independently increase their health literacy.

This app concept design was a solo project for which I identified the problem, conducted research, developed a solution, ran user testing, and completed all product design artifacts from sketches to prototypes.

MediPal

Role

Product Designer

Timeline

Jun 2023 - Sep 2023

(10 weeks)

Tools

Figma

Figjam

Problem Statement

Low healthcare literacy in the US costs the economy about $236 billion. (3) How might we increase health literacy among the older adult population in the United States?

Expert Interviews and Field Research

3 major barriers to health literacy

I used affinity mapping to gather insights from observations at retail pharmacies and interview responses from 2 pharmacists and 3 intern pharmacists. I realized there were 3 major barriers to health literacy: hard-to-digest information, distrust, and unaddressed fears.

White Paper Research

Characterizing the target users

A negative correlation between age and health literacy in adults

  • Health literacy tends to decrease after the age of 40. Older age is a stronger predictor of health literacy than socioeconomic status (4)

  • Age-related decline in cognition is the likeliest reason for declining health literacy, rather than loss of vision or vocabulary (4)

An increase in internet usage among older adults

  • about 30% increase in adults over 65 using social media since 2012 (5)

  • older adults use mobile devices more often than computers (6)

  • older adults need help getting started with using technology but eventually make it an integral part of their daily lives (7)

Shame is a 4th barrier to health literacy

  • 40% of patients with low health literacy admitted feeling shame (8)

  • 67% of patients with low health literacy don’t tell their spouses (8)

  • 19% of patients with low health literacy don’t tell anyone (8)

Ideation

Research gives insight to 3 themes

SWOT Analysis

A virtual AI pharmacist

Using insights from my aforementioned research, I explored other possible solutions such as feature additions to existing platforms and a duolingo-esque app for health literacy. I conducted a SWOT analysis to narrow down my solution. The 3 themes for ideation, additional white paper research, and empathy revealed that a virtual AI pharmacist offered the most strengths and opportunities with the least weaknesses and threats.

User Archetypes

Seniors are not the only users

Although my design decisions focused on being accessible to seniors, I had to realize that there were other potential users. Oftentimes older adults have family, friends, and caregivers helping them manage their healthcare. Younger adults may also struggle with health literacy. With this in mind, I created user archetypes to help brainstorm potential features for the solution.

MoSCoW Analysis

Syncing EHRs not needed

I conducted a MoSCoW analysis to prioritize the features for the minimum viable product I wanted to design. Although allowing users to sync their electronic health records would help create a more personalized experience, this feature would introduce additional issues concerning privacy and collaboration with healthcare institutions. I identified 3 key features that maximized the helpfulness of the solution without introducing those issues.

Key Features

  • chatbot with voice recognition acting as virtual pharmacist

  • pill identifier

  • over-the-counter medication recommendations

Concept Validation Testing

Incorporating language support

I conducted concept tests with 7 primary users and 4 secondary users. The primary users senior citizens and adults over 50 with visual impairment or were unfamiliar with technology. The secondary users included two caregivers, a retirement center volunteer, and a pharmacist. I originally prioritized language support low on the MoSCoW analysis. However, 7 out of 11 users (4 out of 7 primary users and 3 out of 4 secondary users) I interviewed for concept validation testing recommended incorporating it in the MVP.

Guidelines

Designing for accessibility

I cross referenced accessibility guidelines, for UI targeted towards elderly, across the latest WCAG standards as well as standards set by the Macular Society. I also used the Assistive Access mode from iOS 17 as a guide.

Iteration

Constantly improving the design

Storyboarding, accessibility audits, mentor feedback, and usability tests lead to continued improvements in the design.

1. Cutting down the FAQs and simplifying the flow

  • I reached out to the pharmacists from my interviews for help with brainstorming FAQs patients may have. Storyboarding revealed that many FAQs could funneled into the same question categories.

  • User testing further revealed overlap between question categories. 3 out 7 seniors were confused about which question category to choose in at least 2 of 5 scenarios during testing.

  • I simplified the user flow accordingly and updated the imagery on the buttons to increase learnability.

2. Improving the first few pages

  • According to feedback from 2 design mentors, the main page after choosing your language was oversimplified and unhelpful.

  • I reworked the design to increase usability while maintaining accessibility.

3. Adding micro-interactions

  • Addressing accessibility in older adults who may experience visual impairment requires using larger text styles.

  • Larger text styles take up a lot of screen retail causing the design to look clunky when navigation/scroll buttons are added for increased learnability.

  • By adding micro-interactions, such as pop-up tutorial animations for swiping, the end design has a cleaner look while still maintaining accessibility standards and high learnability.

Final Design

AI chatbot for your medication questions

Medication recommendations when you’re unwell

Easy-to-use pill identifier

How Would this Succeed as a Business?

  • Investors such as hospitals and other healthcare institutions would benefit from partnering with a start up for Medipal, because increasing health literacy correlates to increased medication therapy compliance, increase in utilization of preventative care, and subsequently decrease in financial losses.

  • A startup Medipal could generate profit by using a paid model and appealing to children of senior citizens who need help managing their healthcare.

Metrics for Success Would Include:

  • Quiz Scores between people who answer health questions with the help of Medipal vs google searches can be used to assess the helpfulness of Medipal in health literacy

  • Medication Therapy Adherence can be compared between users and non-users of Medipal to indicate how useful Medipal would be in decreasing financial loss to hospital institutions

  • Time to Complete Action would reflect how difficult it is for users to navigate the app.

If I Had More Time I Would…

  • Add more iconography to strengthen the usability

  • Ideate a corresponding app for wearable devices

What Did I Learn?

  • Balancing aesthetic design and accessible design can be very difficult. You have to master creating simple and sleek designs with high contrast.

  • Getting feedback early allows for impactful changes even before conducting usability tests.