Orchestrating Patient Flow:
Designing a Holistic Healthcare App from Ground Zero

Executive Summary

Patient discharge delays are costly and frustrating for care teams. At Intermountain Health, a grassroots spreadsheet was being used to coordinate discharges across departments, but it was slow, inconsistent, and difficult to scale. I partnered with data scientists and hospital staff to transform that manual process into a robust digital tool tailored to the real users: nurse managers.

I led research, design, and iterative testing to uncover the key workflows and data needs behind daily discharge huddles. This included simplifying complex forms, designing quick-scan list views, and collaborating with analysts to integrate data from over 85 systems, including pharmacy, physical therapy, respiratory therapy, and surgery. The new tool completely replaced the spreadsheet, with 100% of nurse managers incorporating it into their daily workflow.

The result was a redesigned process that saved time, improved visibility, and gave clinical teams a reliable source of truth for discharge coordination. This project highlighted the importance of uncovering hidden users, designing for urgency, and aligning technology with real-world care.

Background

Hospital teams at Intermountain Health were managing patient discharge planning using a shared Google spreadsheet. It started as a local workaround but quickly spread beyond a single unit. Over time, it was adopted by teams across hospital floors and eventually used in more than a dozen hospitals throughout Intermountain’s system. The spreadsheet became a central tool for flagging discharge barriers, coordinating handoffs, and tracking patient progress in real time.

Although widely used, the spreadsheet was difficult to navigate and maintain. Another problem was that no real data from patient or hospital records was shown on the spreadsheet. The clinical staff updated it manually, which resulted in inconsistent formats and outdated information. It was never designed to support such a large scale or complexity. Even so, teams relied on it daily to coordinate discharges.

Recognizing the impact and limitations of this grassroots solution, Intermountain invested in transforming it into a formal application. The goal was to reduce discharge delays, improve coordination across departments, and develop a more reliable and scalable tool that clinical teams could trust.

My Role

I was brought in to lead the design and research for the new patient flow tool. I began by analyzing the original spreadsheet, interviewing the data analysts and clinical leaders who created it, and mapping out the logic behind each section. I also participated in working sessions and interviews with frontline nurses and care teams to understand how they were utilizing the tool in their day-to-day work.

As the project evolved, I led iterative design work based on ongoing research and usability testing. I partnered closely with data scientists to understand the numerous sources feeding into the system, including over 85 data streams from across the hospital, such as pharmacy, physical therapy, respiratory therapy, and surgical services. I also worked with product and engineering teams to translate user needs into functional screens and components that could support the workflows of multiple roles.

When testing revealed that nurse managers were the tool’s most frequent and essential users, I redesigned the interface to accommodate their daily routines, including morning huddles. I simplified the views, broke up complex forms, and created a quick-access list view to help them scan information quickly. These designs served as the foundation for the new application, which was subsequently built and adopted by hospital teams.

Research & Insights

Understanding how the spreadsheet was being used required close collaboration with data analysts and clinical staff. I began by reviewing the spreadsheet structure itself, working with the original data team to clarify the meaning of each column and how values were being entered. This analysis helped surface gaps and assumptions, such as whether specific fields were consistently updated or being used at all.

To gain a deeper understanding, I met with care coordinators, nurses, and administrators to observe how they utilized the spreadsheet during their daily routines. These conversations revealed that the original intent of the spreadsheet no longer matched how teams had adapted it over time. What stood out most was that nurse managers had become the primary users. Every morning, they relied on the spreadsheet to lead huddle meetings, quickly reviewing patients who were nearing discharge and identifying potential barriers. These meetings moved quickly, often covering ten to twenty patients in just a few minutes.

The spreadsheet made that process difficult. Information was hard to scan. Updates were buried in long rows and scattered columns. There was no consistent way to flag progress or communicate status between shifts.

After initial prototypes were created, I returned to nurse managers and data scientists to gather more specific requirements. We worked together to identify which types of data needed to be both visible and editable within the application. These included status indicators for therapy progress, notes from respiratory and physical therapy, surgical clearance, medication timing, and other departmental updates. Each of these sources had its own formatting and update logic, which required close coordination with the data science team to ensure accurate modeling and display.

Through this extended research, one insight became clear. Nurse managers needed a tool designed specifically for their daily discharge huddles. They needed to see the right data at the right time, presented in a way that was easy to scan, update, and share with others.

Persona Development

Who We Were Really Designing For

To better understand and design for the real users of this application, I created a persona based on interviews and observation of nurse managers. This persona helped clarify their specific goals, constraints, and information needs during the patient discharge process.

Nurse managers used the original spreadsheet every morning during a brief, high-pressure meeting called the discharge huddle. They reviewed 10 to 20 patients in under an hour, needing to scan information quickly and make updates on the fly. They were interrupted often, so the tool had to support fast re-entry and cognitive continuity.

The persona reinforced that this app was not for abstract data analysis or slow administrative review. It was for real-time decision-making in a clinical environment. That insight led directly to the development of the list view, short-form interactions, and barrier communication screens.

Design Process

Based on early research, I designed three initial screen types to reflect the main use cases identified by analysts and care coordinators. These included an analytics dashboard for administrators, a care coordination view that displayed patients by hospital floor, and a detailed input form for adding and updating patient status. Each screen was informed by the spreadsheet structure and aimed to streamline the existing workflows.

After usability testing with nurse managers, it became clear that the initial designs were too complex for daily use. The form view was overwhelming, and the care coordination screen did not support the fast-paced nature of morning huddles. Nurse managers had only about two minutes to review and update each patient, often moving through an entire floor in under an hour. The interface had to match the speed and focus of that routine.

In response, I redesigned the interface around their real needs. I created a condensed list view that displayed just the most essential discharge-related information. This quick-scan view enabled nurse managers to stay on track during meetings without having to navigate through nested menus or scroll through long, scrollable forms.

I also broke up the original input form into shorter, role-specific sections to make updates faster and easier. This modular approach reduced cognitive load, allowing staff to quickly enter information without being overwhelmed by unnecessary fields.

To support cross-department coordination, I designed a barrier view that highlighted which services were currently delaying discharge. Whether it was pharmacy, physical therapy, respiratory therapy, or surgical clearance, these blockers could be seen at a glance and addressed more efficiently during the huddle or afterward.

Outcome & Impact

This tool was fully adopted by nurse managers across a dozen hospitals, replacing the spreadsheet entirely.

The redesigned tool replaced the original spreadsheet in over a dozen hospitals within Intermountain Health. Nurse managers, who had previously relied on a clunky and inconsistent manual process, began using the new application on a daily basis. Adoption among nurse managers reached one hundred percent, and the spreadsheet was fully retired in every hospital where the tool was introduced.

The tool provided faster access to essential discharge information, with streamlined forms and quick-scan list views that matched the speed and rhythm of daily huddles. Teams no longer had to dig through complex rows or hunt for updates. Instead, they could see discharge blockers, communicate across departments, and make patient-centered decisions quickly and confidently.

Although I was laid off before final metrics could be gathered, the widespread adoption and consistent daily use by nurse managers reflected the tool’s success. It became a reliable foundation for discharge planning, saving time, reducing communication delays, and improving care coordination across clinical teams.

This project demonstrated the value of close collaboration between design, data science, and clinical users. It also reinforced the importance of identifying and focusing on real users, especially when organizational assumptions or legacy processes obscure those users.