Sprint One: Looking Back, Looking Forward
And thus, Summer begins.
The past few weeks have been quite the whirlwind. We were all thrilled to have walked during our commencement ceremony, with many of us fortunate enough to share these moments with loved ones. Although our ‘graduation’ marked the end of ‘traditional’ classes for most of us, we certainly have our work cut out for us regarding re-imagining a better physician experience!
Retrospective
In order to plan out the logistics for our summer semester, we first started with a retrospective session to catalyze constructive conversations around our workflow processes, learning goals, and more. So, to look forward … we first had to look back. What were our greatest strengths? Challenges? Areas of improvement? The column-style sticky note activity lasted more than three hours, where we spoke objectively about our experiences over the spring semester. We decided on best practices moving forward — one of which includes making retrospective sessions a regular ritual.
EHR Crash Course
Following our retrospective session, we hit the ground running with a 4-hour EHR training session,where we got a chance to explore the intricacies of a popular EHR software. It was immediately clear that even the ‘best’ EHRs violated almost every aspect of Nielsen’s Usability Heuristics, most notably “aesthetic and minimalist design.” Due to the sheer amount of information displayed on each screen, we noticed the persistent use of hover states to further identify certain items or actions. While this is a standard way to reveal information when needed, hover boxes can oftentimes create visual noise and might cover up pertinent information.
Seeking Guidance from Subject Matter Experts
A theme that ran through much of our retrospective session was a desire to stimulate more expansive thinking. As such, we decided to speak with some external experts — both within the field of healthcare, as well as within the field of human-computer interaction. For instance, we met withNik Martelaro, an Assistant Professor at the Human-Computer Interaction Institute (HCII) here at CMU, who carries extensive research experience surrounding safety-critical systems, particularly within the automotive space. As our primary goal of this sprint was to gain more clarity into what we may actually build over the next few months, we found it helpful to get fresh set eyes on our work and push the boundaries of explorations and expand our problem-solving approach. Some insightful questions Nik left us with include:
- How might we reframe the overall physician-patient interaction?
- What might the patient be able to do, that physicians are currently burdened with?
- Is there a better framework within which to understand the overall expectations of physicians’ responsibilities?
Literature Review — Phase II
While our initial phase of literature review focused on the problem context, history, and human factors considerations, among other topics — we decided to kick off our first design sprint with a second round of literature review aimed at six design opportunities. While most of our initial ideas surrounded reducing workload, a design activity we conducted with Optum stakeholders — coupled with the feedback we got by engaging external subject matter experts — allowed us to expand our thinking and ask ourselves questions such as:
- Is there any way to increase the appreciation that physicians feel?
- Would patients be willing and able to engage in the curation of their own health data to optimize the patient visit and reduce some of the workload placed on physicians?
- What digital interventions can we employ to force physicians to prioritize breaks?
- How might interfaces intelligently adapt to the physician and task, suggesting context-awareness?
As such, our “Lit Review 2.0” centered around the following topics:
- Alternate Modalities
- Data Visualization
- Information/Alert Organization
- Time/Task Management
- Gratitude/Appreciation
- Patient Involvement
- Adaptive User Interfaces
In order to visualize the relationships between these opportunity areas, we took all of the concepts generated and mapped them based on which aspect of cognitive load they addressed: workload or emotional load. This exercise helped us begin to understand the impact our solutions could have, and how some concepts might be able to address both the emotional and workload burden.
Market Scan
In addition to our second round of literature review — much of which referenced new frameworks and/or precedent products — we performed a market scan aimed at identifying key competitors in each opportunity area. Some vendor landscapes were busier than others — but overall, the activity allowed us to evaluate the strengths, weaknesses and gaps of what’s currently on the market. We found some pretty interesting products, and were even introduced to some relevant prototypes/demos shown at the most recent CHI conference.
Research through Design
We read … a lot. The amount of literature surrounding physician burnout, cognitive load, the impact of displays on cognitive processing, among other subjects, gave us a bit of information overload. In order to avoid analysis paralysis, we decided that the best way to begin to process our ideas was just to begin wireframing. Through this process alone, we quickly realized that one person’s interpretation of a solution area might be very different from another’s. It’s one thing to say that we plan to work on information organization, but when it comes to visualizing what this concept might look like, we found ourselves looking up more and more details. For example, when sketching inbox filters, we had to ask ourselves the following questions:
- What would x, y, and z be categorized under?
- How urgent would this patient’s request actually be?
- How do physicians even address one another?
So while initial design considerations stemmed from our spring research, we can already tell that the process of designing itself will engender the need for much more domain-dependent research, especially within a field as complex as healthcare.
