Managing tasks for inpatient care teams

Today, doctors, nurses and other members of the inpatient care team often keep track of their to-dos using scribbled notes on paper. While efficient in the moment, these notes are hard to decipher after the fact, leading to duplicated work, dropped tasks and information lost in translation. HospitalHelper is a task-management app specially designed for hospital staff in mind, helping the team record their tasks efficiently and pass that information on quickly at the end of their shift.


My Contributions

  • User interviews
  • Research analysis
  • Product Strategy
  • Persona creation
  • Wireframes
  • Copywriting
  • Design—Lead
From left to right: Add Patient screen from Start Shift flow, End Shift screen, and Edit Checklist screen.

Project Goals

The HospitalHelper team had already created an alpha version of the product, but product strategy was missing. The team had spoken to their target audience and implemented all user-requested features exactly as requested, leading to a product with many abilities, but lacking an overarching narrative, structure or opinion.

My first task was to reduce the product’s complexity by first identifying the specific problems our users were facing in their current task management practices, and then optimize the app to solve those core problems. This case study will focus on the solution to one particularly important user need.

Example of the handwritten notes currently used by inpatient staff to manage their tasks and patients.

Discovering the Dealbreaker

I conducted a series of exploratory interviews with doctors, nurses, and other medical professionals to understand pain points, relationships and context. These conversations unveiled many pain points. We coded and prioritized user needs and decided on our feature new set. It meant cutting out several features the app currently supported, but allowed the paths through the app to be more structured and focused. The new version of the app allowed users to:

  • Add and manage patients during their shift
  • Record tasks and notes for each patient
  • At the end of their shift, produce hand-off note for the next shift of doctors and nurses.

The third feature solved a major pain point for our users and was a major differentiator. However, it required the first two features in order to function well, and those features were up-hill battles. Through our research, we knew that caretakers typically recorded their notes on paper and while they didn’t enjoy typing up their scribbled notes at the end of the shift, paper’s speed and ease in the moment made it hard to break the habit. We realized the instant gratification of paper outweighed long-term benefits of using a more efficient solution. This posed a problem: if we couldn’t convince users to drop the paper and manage all their patients and tasks in the app during their shift, we couldn't offer feature 3 and produce the hand-off note.

From left to right: the Open Tasks UI that details the caregiver's current to-dos, the Add Task screen that allowed the caregiver to quickly add a task and its details, and the slide-out Patient List, with a Bloomberg Terminal-style density.

Better than Paper: Optimizing for Speed and Ease

Speed and ease in task entry became a make-or-break feature: if it wasn't perceived as faster and more convenient than paper, our product couldn't inspire habit change.

As the revised, less-featureful version of the app began circulating and our beta test pool grew, we gained access to data. Digging through, I noticed patterns in the naming of tasks. More often than not, users were trying to do one of four things: follow up on something, check something, order something or contact someone.

Taking advantage of this predictability, we developed an autocorrect-inspired UI. When the user brought up the Add Task screen, the four common verbs (follow up, check, order, contact) are offered as tappable suggestions under a free-entry field. If one of the verbs is tapped, it flew up to the free entry area and a new set of logical suggestions populated.

Inspired by user data, I designed an auto-suggest style UI to speed up task entry.

Even in its initial hard-coded state, this small tweak was able to dramatically cut down on the amount of typing required; in tests, it was shown to speed up task input by as much as two seconds per task. In the healthcare world where every second counts, this was a massive and exciting improvement.

I organized my designs in swim lanes to track iterations and ensure others could navigate my Sketch files.

RIP, HospitalHelper

Although HospitalHelper was gaining traction and had a very positive userbase, the parent company of the app pivoted and shut down the project to better allocate resources to other aspects of healthcare. It was a sad moment to not see this exciting product go further, it was a great experience and learning opportunity.

Desperate Parents

  • Frequent night terrors over a long period of time
  • Normal life disrupted
  • Actively seeking solutions
  • Highly motivated and uses product consistently

Dealing Parents

  • Less than four night terrors a week
  • Interested in solution, but not actively seeking solutions out
  • Need support to use consistently

Read Next

Detecting eyeball defects. 2015 — Android app — UX Lead