Web App to Improve Tracking of Pressure Ulcers in Long-term Care Homes

Research Shows Urgent Need for Streamlined Documentation to Prioritize Patient Care

From user interviews with 2 personal support workers, 1 registered nurse, and 1 registered nurse practitioner and secondary research, we gathered the following insights:

Deep-dive into the life of a healthcare worker

We conducted further interviews with the healthcare workers to understand how they are currently interacting with their each other and their technology (Electronic Health Records) the product the following artifacts.

PERSONAS

Personas were valuable because even though PSWs and RNs jobs are similar, they both had different needs and pain points.

SYSTEMS DIAGRAM

A systems diagram was helpful defining the interactions between the user groups and their technology.

HIERARCHICAL TASK ANALYSIS (HTA)

An HTA helped us break down the workflow of the healthcare workers and identify areas of redundancy.

Design Statement - Let's Get Specific

Based on our research, we knew that pressure ulcer prevalence was linked to the care provided by healthcare professionals, and since the care from the healthcare professionals depended on their strained work conditions, we focused our efforts on improving the workflow for the nurses and PSWs in long-term care homes.

A Two-Device System Integrated to the EMR

Bedside/Tablet Interface: This interface enables nurses and PSWs to input pressure ulcer interventions, such as shifting, applying barrier cream, or changing clothes. It ensures efficient logging of interventions, which are compared and updated with the patient's care plan.

Nursing Station: An interface connected to EMR database like PointClickCare. Provides patient info (age, medication), intervention history, and data on pain level and position tracking for informed decision-making.

Bedside/Tablet Interface V3

Nursing Station V1

User Testing with a focus on usability and cognitive load

Prototypes were evaluated using tests for performance, user satisfaction, and cognitive load. To compare with the existing system under COVID-19 constraints, a mock Electronic Medical Record (mock-EMR) was created in Microsoft Excel to compete against the Bedside/Tablet interface. These tests were performed on a virtual call or in-person.

Testing Methods

• System Usability Scale (SUS): Testers simulated nurse/PSW tasks on prototypes, including mock PCC. Post-task, a 10-question SUS questionnaire assessed usability and learnability for comparison with current and future systems.

• NASA TLX: Testers replicated tasks on prototypes, including mock PCC. Post-task, a 6-question NASA questionnaire gauged perceived workload that we could compare with the current system.

• Tapping Test: Testers tapped rhythmically with their non-dominant hand or foot while using the prototype. Areas of slow or irregular tapping indicated increased cognitive load, particularly on the bedside interface, where the test was focused.

• Sternberg Test: Testers remembered instructions, completed tasks with prototype/mock PCC, then recalled instructions. Success rate indicated impact on short-term memory, exclusive to bedside interface.

• Qualitative Feedback: After the testing session, the user would provide feedback on the interface and their experience identifying moments of vagueness, confusion, or frustration.

Round 1 Results - Bedside/Tablet Interface

Good

Our interface performed much better than the Mock-EMR. Users also found it more “intuitive”, “easy”, “user-friendly”, and “streamlined”.

Bad

Some accessibility violations were present (button size and colour contrast). Users did not know that they could select multiple items on some pages.

Round 1 Results - Nursing Station

Good

Our interface met the learnability and usability SUS requirements while also imposing minimal cognitive load.

Bad

The extended duration of Task #2 stemmed from user confusion surrounding an icon representing a late intervention. Accessing Braden scores, vital for decision-making, proved cumbersome as users had to navigate into each patient's section. Furthermore, identified accessibility issues included violations like inadequate button size and insufficient color contrast.

Addressing accessibility issues and selection confusion

Bedside/Tablet Interface

Refinement 1: Some users were confused how many options there were allowed to select on a given page. To address this, checkboxes were used on pages where you could select multiple items and radio buttons were used when only one item can be selected.

Refinement 2: Button size and colour contrast was improved to comply with WCAG 2.0 Accessibility guidelines.

Refinement 3: To reduce confusion, the alert icon was updated to a red clock to show that the patient hasn't been seen in >2 hours

Refinement 4: Button size and colour contrast were changed to comply with WCAG 2.0 and Google Material Design Accessibility guidelines.

Refinement 5: Braden scores were added to the main page because they can help make quicker, better decisions.

Validation that our refinements had a positive impact

Using the improved designs after the redesign, we tested again using the same methods to validate the improvements!

Round 2 Results - Bedside/Tablet Interface

Summary

Both iterations were successful in meeting the design requirements but V3 had slightly better results. Although, 83% of testers preferred the V4 interface saying that it was, “more mapped to my inner model”, “cleaner”, and “easier to navigate”.

Round 2 Results - Nursing Station

Summary

Nursing station V3 performed significantly better on usability and learnability scales, required the least cognitive load, and possessed the shortest task times.

The MVPs are live!

Try it out!

Things I'll carry onto my next project