Enabling over-worked nursing staff to administer medications accurately

Pilly — Medication Schedule Tracker for Hospitals

Advisory Board, a best practices firm that uses a combination of research, technology, and consulting to improve the performance of health care organizations partnered with our graduate program and challenged students to design a solution for nurses to track patient medication in Hospitals. This case study breaks down our process, insights, and learnings.


Field study

Having only 5 days to complete the challenge, our first priority was to aggressively narrow the scope of our solution. We visited the Bloomington Hospital in the city and talked to two subject matter experts.

(L) Julia interviewing a nursing staff, (R) Tosh interviewing a nursing staff

What we learned

  • Patient records are stored electronically on a Electronic Medical Record (EMR) software and can be accessed on a ‘computer on wheels’. Any member of the caregiver team assigned for that patient has access to these records.
  • Nursing staff often works in 12 hours shifts ending or starting with a handover meeting to go over what was done, and has to be done.
  • Administrating medication to patients is one of the major task. Each round of medication is received from a Pyxis Machine and taken to the patient in their room.
  • As per common nursing medical regimen, the staff confirms Patient, Medicine, and Dosage 3 times before it is administered.
  • Nursing staff is often overworked juggling between different patients providing care, filling up tons of documents, administrative work, and meetings with doctors, physicians, and surgeons. They prioritize tasks on the go.
  • Taking notes for each medication treatment is very important on the records, so nurses prefer to type as much detail as they can.
Computer on wheels with scanner
One-time medication administration workflow

Secondary Research

Reading through several cases on AllNurses.com — a discussion forum for nursing students and new nurses, we realized that communication gap between different medicine administrators is a common problem. There are so many things to talk about that it is easy to miss out a small but important detail.

Delay in administering medication to the patient is a common problem. In itself it is not a big deal but, the time of next dose is affected by it. If not communicated to the next nurse it could lead to certain complications.


Based on all the research we did, we narrowed down the problem space to Delay in administering medication.


After several rounds of quick ideation, we focused on solving the communication gaps between nurses who work in rotating shifts.


Workflow diagrams and Ideation

Competitive Analysis

We collected all types of current medication tracking charts and applications: paper charts, digital charts, charting systems and applications for patients themselves to use.

Things we can use:

  • Common Treatment Time Category: Morning, Afternoon, Evening and Bedtime
  • Sort the information by medication
  • Differentiate Suggested Treatment Time and Actual Treatment Time

Things we can improve:

  • The “late” marker on current charts are not clear.
  • The note-taking option is not available for each specific treatment.
Examplars — Medication tracking paper charts, digital charts and tracking app for patients)




Interface Details

Rows are lead by Drug names. Our research showed that medications are administered in time ranges and not at specific times. So, we split the day in 4 parts.
Each drug name in expanded view includes Drug Type, Prescription Date Range, Dosage, Provider (Dr. who prescribed it)
Next to Drug names are anticipated or expected time of dose is shown in gray with no. of dots indicating dose. Administrated doses are shown in green and late administrated doses are in red. Notes is the most common requested feature used by medical staff to subjectively document everything including reactions, improvements, and anything usual or unusaul.
At the bottom is the Sign Bar. After every set of medicine that is administered, nurses have to “sign” to record. This determines the dose was adminstered succesfully. The Notes are usually entered after signing.

Usability Test

We conducted a usability test with an IU Masters student. We introduced the nurse scenario to him and showed him our design. The testing was positive in most aspects. He got the whole concept and the “late” mark meanings clearly.

Julia introducing the prototype to Miringu


However, the viewing all the medication details function is not clear to him. We changed the unclear arrow to a more common “expand all/collapse all” design pattern as the following:

Iteration based on the usability feedback


It was challenging trying to create software for nurses in the small timeframe. We constrained down early after conducting research on Monday, but it was difficult deciding on a design. We were able to decide on a design midway through the week, and it gave us time to iterate and discuss after testing.

A case study by Julia Zhu, Ryan Griggs, and Tosh Anand

Senior Product Designer @ Cisco • http://tosh.design previously, Dataminr