MediBridge

Inspiration

Managing medications can be difficult for older adults, especially when prescription instructions are written in a language they are less comfortable reading. Through peer interviews, we found that family members often become the bridge between older adults and their medications—translating labels, clarifying unfamiliar wording, organizing multiple prescriptions, and setting reminders.

We wanted to create a tool that could help older adults understand their prescriptions more independently while still keeping caregivers and healthcare professionals involved when needed. This inspired us to build MediBridge, an AI-powered medication assistant that connects prescription instructions, language accessibility, and medication organization.

What it does

MediBridge helps users understand and organize prescription instructions in a simple, accessible way.

Users can provide a prescription label, and MediBridge:

• Explains prescription instructions in the user's preferred language
• Converts complicated instructions into clear, patient-friendly language
• Identifies information that may be unclear and should be confirmed with a pharmacist
• Organizes confirmed instructions into a medication schedule
• Creates clearly named reminders so users can distinguish between different medications and doses

Rather than simply translating a prescription, MediBridge focuses on helping users understand what to take, how much to take, and when to take it.

How we built it

We built MediBridge around the medication-management workflow we identified through our user interviews and research.

Our process began with peer interviews to understand how older adults currently manage prescriptions and where family members are involved. We then researched medication comprehension, language barriers, and medication adherence to identify evidence-based design considerations.

We designed the system to move through a series of steps:

Prescription → Explanation/Translation → Verification → Medication Schedule → Reminders

A key part of our design is verification. If prescription information is unclear, MediBridge should not simply guess. Instead, it flags the information for clarification with a pharmacist before it is used to create a medication schedule.

Challenges we ran into

One of our biggest challenges was designing an AI system for a healthcare-related problem where incorrect information could have serious consequences. We had to think beyond simply making the AI generate an answer and instead consider when it should recognize uncertainty and direct the user to a pharmacist.

Another challenge was realizing that translation was only one part of the problem. Our interviews showed that even when older adults understood their prescriptions, managing multiple medications and remembering different schedules could still be difficult. This caused us to expand our focus from translation to the entire medication-management process.

We also had to balance independence with caregiver support. Some users may only need a translated prescription, while others may benefit from having a family member help manage their medications. Designing a system that supports both without making the older adult unnecessarily dependent was an important challenge.

Accomplishments that we're proud of

We are proud that our product was shaped by real user experiences rather than assumptions. Each of our peer interviews revealed different challenges, from translating English instructions into Telugu or Marathi to organizing multiple prescriptions and distinguishing between generic phone alarms.

These interviews directly influenced our features. For example, the need for translated instructions led to multilingual support, while difficulties distinguishing medication alarms inspired our focus on clearly named reminders.

We are also proud of designing MediBridge with safety in mind. Instead of treating AI-generated information as automatically correct, our system emphasizes verification and pharmacist involvement when instructions are unclear.

Most importantly, we turned a broad problem—"medications are difficult to manage"—into a focused workflow that addresses understanding, verification, organization, and reminders.

What we learned

Our biggest takeaway was that AI should solve the user's actual problem, not just provide an impressive technical feature.

At first, it was easy to think of our project as a prescription translation chatbot. However, our interviews showed us that translation alone does not solve the problem. Users may still struggle with multiple medications, different dosing schedules, reminders, and knowing when they need professional clarification.

We also learned how important trust is when building healthcare technology. An AI assistant should not confidently provide an answer when the underlying information is unclear. Designing for uncertainty and creating a clear path to professional verification became an important part of our product.

Finally, we learned how valuable user research is during product development. Our interviews changed the features we prioritized and helped us build something centered around actual user needs rather than what we initially assumed users would want.

What's next for MediBridge

We see MediBridge eventually becoming a broader medication-management platform rather than just a prescription assistant.

Future features could include:

• Pharmacy integration for directly connecting users with pharmacists
• Prescription storage so users can keep track of their medications in one place
• Drug interaction checks to identify potentially harmful combinations
• Duplicate medication detection, including medications with different brand and generic names (such as Advil and ibuprofen)
• Timing conflict detection between medications
• Refill reminders based on dosage and remaining medication
• Multiple caregiver accounts so family members, nurses, or other trusted caregivers can help manage medications
• Healthcare platform APIs so MediBridge can integrate with existing healthcare systems

Our long-term goal is to make medication management more understandable and independent for older adults while providing caregivers and healthcare professionals with better tools to support them.

Built With

• claude
• css
• javascript
• python