Modular Inquiry / Lead Page
Applied Heuristic Principles
Visibility of System Status
Match System and the Real World
User Control and Freedom
Consistency and Standards
Error Prevention
Flexibility and Efficiency of Use
"Recognition" Rather Than "Recall"
Aesthetic and Minimalist Design
Recognize, Diagnose and Recover from Errors
Give Help and Documentation
Personas and Scenarios for Bulk Patient Importing tool.
After understanding of our users' tech knowledge from this research,
the team decided on a "no UI" option, saving valuable developer time.
PointClickCare
Intake + EHR Data Sharing
PointClickCare is a Toronto-based electronic health records system that manages the home care patient process from Intake, Start-of-Care, and Care Plans through Claims and Billing to Discharge.
The power of electronic health records (EHRs) has been broadly over-sensationalized. Dr Eric Topal, an influential thinker in the field of innovative medical technology, estimates only 4% of EHRs' systems are compatible. In other words, transferring patients remains a mostly manual process, involving many phone calls, emails, text messages(!), faxes and typically 100-150 printed pages per patient. This initative moves the product and industry closer toward a universal health record.
Objective
Reduce friction and errors within the patient intake/transfer process
Results
We digitized five key steps in the Intake process:
Lead Tracking
Duplicating the real world model we saw during our field visits (Post-It notes, white boards, etc.), we created a simplified lead form with multiple areas for basic notes, segmented controls to reduce drop-down fatigue and added quick links for next steps for the patient.
Intake UI Redesign
We introduced a crisper, modular UI to replace a static form approximately 6500 pixels long, allowing the Intake Specialist to add sections as needed. We also made it possible for companies to customize their own presets, so they could adjust sections in the UI at a global or admin level.
IVR Touch Tone Translator
Some companies use Medicare's IVR system to check eligibility. It's a touch-tone system with a staggering amount of complexity. It's layered in error potential and typically takes 30-45 minutes to complete five or so basic bits of data.
For example, the eighth number in a Medicare number is an alphabetic value, so a user needs to press "*" to indicate a letter, then "2" twice for the letter "B", and then "8" for the eighth position in the Medicare number.
Our solution is a front-end Javascript "translator" so the Intake Specialist enters the data into the webform, translates and then touch-tones the translated series into the Medicare IVR system. It reduces errors and cognitive load, and saves time. Despite improvements like this, the user experience still has significant limitations due to the US Department of Health and Human Services' infrastructure requiring the IVR.
IVR Translator Concept.pdfDocument Repository
Prior to this feature, intake documentation (100-150 pages) was stored along with the patient charting, vitals and medication information. We applied a category and tagging system, so a user's role would automatically filter out unneeded assets.
Document Repository wires (Axshare)Bulk Patient Import
Because of a dramatic restructuring in 2020 Medicare legislation, smaller senior health care organizations are liquidating at a rapid pace. This tool allows larger agencies (500+ patients) to quickly merge and acquire patients and reduce or remove any lapse in care.
Bulk Patient Import Wires (Axshare)More details ►
Goals
Productivity
Research Methodologies
- Contextual Inquiry
- Direct Observation
- Moderated Testing
- Personas
- Scenarios
- Stakeholder Interviews
- User Interviews
Key Findings
Our biggest assumption was that the Intake process was slow, confusing and prone to error. We were correct, however we found most of the work was being done outside the application. Intake specialists were using sticky notes, white boards and their own process checklists. And only when all the information was collected did they enter it into the system. We were initially confident in our application because analytics showed us the Intake UI took only about 5 minutes to complete. But after our discovery, we found the entire intake process took about two hours per patient.
Another big surprise was the intake specialists' inventiveness. They were handling a tremendous throughput of patient data; some had created their own checklists, and one user created a series of folders and actions within MS Outlook that rivaled the usability and flexibility of our application.
Comparable Products
CareCloud, Epic, GE Centricity
Deliverables
Personas, Scenarios, Wireflows, Data Hierarchy Map, Heuristic Analysis, Wireframes, High-Fidelity Designs
Team
UX/UI (myself), Junior Researcher
Challenges
The product roadmap was dense with features that had been promised to prospective clients. Many other features were driven by clients' executives' feedback instead of end users. And front-end technology understanding and UX maturity within our product team were limited as well. As a result, many features that were developed were flawed, outdated or useless to the end user.
To remedy this, I spent more time iterating and socializing smaller design iterations and rationale. It slowed our velocity, but helped the broader team understand how design hierarchy decisions are driven by strategic thinking.
I also advocated and produced more remote and in-person usability tests, user session recordings, personas and scenarios, supporting analytics, visualizing information architecture as well as showing a range of options ("T-Shirt Sizing" project estimation) to help shape the product team's view beyond a skeleton of a feature.
