Electronic Health Records -

Prescription Ordering Prototype

UX Challenge

I was hired as a User Interface Designer for a contract project through Partners HealthCare and the Brigham and Women’s Hospital Center for Patient Safety Research. The project set out to reimagine the Computerized Prescriber Order Entry (CPOE) process, and help to reduce medical errors by optimizing prescriber workflows and enhancing patient comprehension. The hypothesis of the project was that surfacing a medication’s indication, or reason for use, to a higher level of visibility on both the prescription’s label and in the medication ordering process would allow patients and doctors to feel more confident about consuming and ordering the drug. This is a change from the current reality in some of the leading Electronic Health Record (ERH) products.

Design Process & Solution

This project consisted of a multi-year effort which is documented on the Agency for Healthcare Research and Quality website. During the first year, a group of UX Researchers, Physicians, and Pharmacists held 6 different discussion panels involving a variety of key stakeholders in the medical community. The goal was to discuss the critical issues facing the industry when it comes to prescribing medications.

I started my work during Year 2 of the project, which focused on leveraging human-centered design activities to design a working prototype that proposed a new CPOE workflow. Following is a summary of my process for designing the prototype.

Initial Research:

Before starting my prototype, I reviewed the panel discussion recordings from Year 1 of the project, and completed literature reviews to better understand the domain. I also worked closely with a User Researcher and team of medical practitioners to brainstorm key requirements. Using these inputs, I documented ideas on how to satisfy the requirements. I also did a competitive review of how other EHR products on the market (e.g. EPIC and Cerner) approached prescriber order entry and how these workflows could be improved.

Early Mockups:

I did a lot of early mockup ideations, sharing different approaches with the project team. Using the expert resources in this group, I continued to ideate and add new features. At various checkpoints throughout the project, we held review sessions with groups of medical practitioners around the greater Boston area. This included some participatory design sessions to help inform the early drafts of the prototype, as well as more structured task-based testing using later iterations of the design. These screenshots illustrate how the design evolved throughout the user-centered design process.

Interactive Prototype:

At the completion of the project, I built a fully-functioning prototype in Axure that satisfied a variety of prescribing uses cases with different levels of complexity. A team of researchers were able to use my prototype for formal usability testing after the completion of my contract. The findings of this usability testing in Year 3 were published in The Journal of the American Medical Association. See: Comparison of a Prototype for Indications-Based Prescribing With 2 Commercial Prescribing Systems. The study found a statistically significant reduction in both user clicks and user error when completing tasks on my prototype, versus two commercial prescribing systems currently on the market. Below are a few screenshots from my prototype.