AccuNurse was a voice-assisted clinical documentation system designed to reduce the time nurses spent writing, transcribing, and re-entering patient information. I researched IV and physical therapy workflows, designed the mobile and desktop experience, and helped turn paper-heavy documentation into a real-time voice workflow nurses could review and trust.
Hospitals already knew documentation was consuming too much nursing time. My job was to understand how that problem showed up in real workflows and design a system that could reduce duplicate documentation without disrupting care.
A voice-driven mobile application that let nurses capture structured patient information during care, verify it through spoken and visual feedback, and review the completed record at the end of the shift on a practicioner dashboard. Tie the care into billing and accounting.
Less time
Documenting
In contracts
first year
Care Facilities
Customers
Sr. Product Designer
Research, IA, UI/UX, Visual Design, Handoff
IOS-Mobile React-Web
6 months (Kickoff to Delivery)
I shadowed nurses at two different target locations with two separate work environments.
What I was seeing was that the nurses had to assemble their own working system before they could even start care. They were pulling patient information from paper, checking schedules, going to a computer, copying information down, gathering paperwork, then taking that information with them as they moved from room to room.
Once they were with a patient, the documentation process still wasn’t integrated into the care itself. They either had to stop and write things down while they were working, or rely on memory and document it afterward.
So the inefficiency wasn’t just “too much paperwork.” It was the constant switching between:
computer → paper → memory → patient care → paper again → computer again
That’s useful because it shows exactly where the friction lived.
The goal wasn’t simply to “digitize documentation.” The hospital already had a digital system. The problem was that nurses had to document twice—first on paper, then again in the mainframe.
The opportunity was to collapse those two steps into one:
say the information once → capture it directly into the appropriate digital field → eliminate the paper transcription step
And the real value wasn’t just speed. It was giving nurses back time that could be spent with patients instead of sitting at a workstation doing administrative work.
This wasn’t only about clinical documentation. The nurses were also capturing information that directly affected billing and reimbursement: time spent with the patient, supplies used, and other billable activity.
AccuNurse also had to support the financial side of care. Nurses documented time spent with each patient, supplies used, and other billable activity. By capturing that information through voice at the point of care, the system could associate it directly with the patient record and billing workflow, reducing the need for duplicate entry later.
Double documentation and physical movement between:
Patient room
Nurse station
Staff room
PC database
This created:
Redundant transcription
Latency in data
Error opportunity
Reduced patient-facing time
Documentation was fragmented:
Start with paper or memory
Go to workstation
Return to patient
Finish documentation later
This introduced:
Time loss
Cognitive load
Transcription risk
Workflow interruption
Increase therapy minutes (RUG metrics)
Decrease staffing needs
Validation of mandatory fields
Faster documentation at point of care
Accuracy of documentation
Proper integration with billing system, and supply ordering
We felt that the tough decision to leave a few things behind wouldn’t impact getting the product in users hands, fulfilling the contract, and start to get feedback.
The dashboard integration was too complex to implement before launch. It was decided to deliver post launch as overall scope of project would increase.
Communication between devices within the hospital ecosystem was a request by the client. This feature needed to be researched and a new data set and archetecture built. The featured was backlogged to post launch.
There were numerous wireframes (almost 200) outlining functionality. I am only showing two. I wired everything out for the dev team(this was before design systems). Today there are better ways to be more efficient.
Green highlights the active field, system input matches so user can see what the system is entering
2 week testing length
Every product operates within its own set of constraints, timelines, technology, business goals, functional requirements, and the realities of the environment it lives in. Good design happens within those guardrails. The role of a designer is to bring experience, the right tools, and a clear understanding of what’s being built and who it’s for.
Each case study shown here is only a snapshot of a much larger process. I tend to ask a lot of questions because understanding the people using a product, and the context they use it in, reveals insights that assumptions alone never will.
Thanks for taking the time to explore this project. If it resonates with you, I’d be glad to talk more about it.