Can Transparency Be Opaque?
A UX case study to help Original Medicare beneficiaries make the most of their healthcare coverage.
Introduction
The problem scenario
Regarding healthcare in the United States, progressive policies and data transparency are a step in the right direction but for Original Medicare beneficiaries access to information without personalization isn’t useful. Current plan resources are disjointed and display content indeterminately disallowing meaningful interpretation and interaction.
The Medicoor dashboard app enables Original Medicare beneficiaries to meaningfully engage with their healthcare plan in a personalized way. A notable design challenge was eliciting fruitful engagement without using burdensome amounts of information.
In maintaining that research and testing are paramount I employed a double-diamond approach seeking to deliver the best solution to identified pain points in an agile manner.
1. Discover
Market Research, Competetive Analysis, Heuristic Evaluations, and User Research
“How useful are advancements in technology or policy if we’re not equipped with the tooling to meaningfully engage with it?”
In 2018, patients in the United States experienced an 11% increase in out-of-pocket costs and 70% of patients claimed knowing costs before a procedure would help them anticipate charges and budget for payments. Recently, the agency that oversees the federal healthcare programs in the U.S., the Centers for Medicare and Medicaid Services, mandated that all healthcare providers including but not limited to hospitals publish a list of all their prices online. An admirable precedent, to be sure, but how might patients benefit from this alone if list prices are far different than transaction prices? How useful are advancements in technology or policy if we’re not equipped with the tooling to meaningfully engage with it?
Not All Tools Are Created Equal
My goal with competitive analysis and heuristic evaluations was to see if other payers take measures to close gaps between list prices and transaction prices and to objectively identify any disparities in refinement between private payer resources and those provided by the CMS to Medicare beneficiaries. Initial constraints prevented access to a variety of applications that exist in the marketplace including tools offered as extensions to the benefits packages many employer-sponsored payers provide their beneficiaries. Without credentials, access beyond onboarding wasn’t possible in most cases. It became evident, however, large payers like Aetna and Anthem, among others, provide sophisticated tools to their beneficiaries, so, in order to work-around said constraints to make necessary comparisons, I stormed through secondary research on the existing landscape… articles and professional critiques of the technology, I dug into payers’ progressive web app FAQ pages, burrowed into reviews on the App store, I even reached out to colleagues with employer-based coverage to garner insights on specific functionality regarding costs and probed for any scenarios they deemed meaningful.
Heuristic evaluations highlighted canonical approaches for engagement and proved to be an important measure of utility regarding interface design. The United Healthcare web portal, the Healthcare Bluebook web app, and the University of Utah Health web portal are among a few of the resources I evaluated.
I cared less about the visual design component of interfaces at this juncture and more about the distribution of information and the use of microcopy on features. I decided to focus on Medicare beneficiaries at this stage as it became increasingly clear how “subpar” CMS engagement tools were in comparison to other payer resource tools. More specifically, private payer tools resembled native applications or progressive web apps with centralized features tailored to the beneficiary whereas CMS resource features were spread over multiple static web pages. For example, you cant access the cost transparency tool anywhere on the provider resource page and vice versa. I suspected problems beyond meaningful cost interpretation.
38 Million Opportunities
Roughly 22% of the U.S population is at or nearing retirement age. Of the 72 million “baby boomers” at least 38 million of them are enrolled in the Original Medicare plan. A plan apart of the program speculated to be insolvent by 2026 which covers, arguably, one of the most healthcare dependent generations in American history. As advantage plans are offered by third-party payers (with completely separate resources from the CMS) and in light of the 20% co-insurance burden taken on by Original plan holders I decided to dial primary research efforts in on the populous qualified to lose and gain the most.
“How might we provide Original Medicare beneficiaries with accurate out-of-pocket healthcare costs?” quickly morphed into something more emblematic of the real friction in this space: disjointed resource tools of engagement. I screened for and interviewed five participants, all Original Medicare beneficiaries. I prepared for the interviews with the intent to identify their needs and goals. I considered age-related aversion to technology and overall behavior towards existing healthcare resources. As much as possible I accounted for variance in patient conscientiousness while keeping the approach simple. Now that the focus wasn't entirely on cost transparency tools, alone, I asked open-ended questions to elicit more latitude within our dialogue. Each participant provided a considerable amount of feedback most of which created new ways of looking at the space as a whole, therefore, it was time to synthesize the interview results.
2. Define
Affinity Diagrams, Empathy Maps, and Personas
Data Prognosis
My goal during the definition stage was to bring about clarity and focus by organizing and interpreting the data. Affinity diagramming formed clusters of meaning around all these new insights and recruited me away from overthinking matters beyond the scope of the user. I broke the user research feedback down into five specific categories:
- Personalization
- Plan Infrastructure
- EHR Integration
- Value-based reimbursement model
- Cost transparency
The subcategory insights, although distinct, suggested interrelations between pain points. As dialogue became more dynamic, participants began shedding light on a host of their problems by highlighting their needs. The feeling of isolation and dependence emphasized the need for personalization and autonomy. “All those years of contributions, I want solutions!” called attention to the plans infrastructure. “How is it possible to feel without means with a program this large?” howled another participant. Other remarks suggested passive use of claims data considering its potential to yield dynamic interactivity in real time. Also, the desire for provider engagement called upon reimbursement model implications which inevitably poured into cost transparency and the beneficiaries obscured positioning in all matters.
I continued to organize insights and make observations through empathy mapping. This brought me closer to the users' experience in a real way. I became more sensitive to the fact that what users think, what they say, what they do, and how they feel can all be different. After spending some time with the mappings I noticed overlapping characteristics of concern and similar motives so I converged on a primary persona.
Meet Keisha our target user! She needs a personalized healthcare tool she can engage with in a meaningful way. Current plan resources display content indeterminately disallowing meaningful interpretation and interaction. The CMS stand-alone web resources make the attempt to provide dynamic interactivity from claims data on search results for participating providers, among other things, but users feel bogged down with info-dense fine print and technical language. Additionally, the current cost transparency tool provides indefinite numbers regarding expenses associated with various procedures and treatments so the beneficiary is left in the dark concerning additional costs. A strong motivation for the ideation stage was turning this apparent apathy into useful engagement.
3. Develop
Sketches, Guerilla Usability Testing, Wireframes
Problem Diagnosis
I began solution ideation with a new, more insightful, question: How might we present useful information to Original Medicare beneficiaries in meaningful ways? This involved a unique design challenge: centralizing features of engagement without inducing overwhelm. I developed momentum by determining two important scenarios (based on user feedback) that benefit our target users as they look to engage with their plan. These red routes paved the way for early usability feedback.
Red Route #1: Find a Physician and schedule a consultation
Our key persona wants to find providers based on sophisticated reviews.
Red Route #2: Search for the cost of a procedure or treatment
Our key persona needs accurate out-of-pocket costs for procedures or treatment.
Before guerilla usability testing I took a moment to reflect on the evolution of the project. This idea that sometimes the actual problem exists further upstream, less recognizable, stood with me. Were initial curiosities focused on solutions or problems? Data-transparency (in this context) is a solution that succeeds basic affordances that elicit satisfactory results when engaged with. The real problem is beneficiary orientation towards information and the lack of regard for what constitutes useful engagement. At the end of the day, beneficiaries want something simple and I kept that in mind as I reached out for early usability feedback.
My goal with the guerilla usability tests was to gather insights from verbal feedback. I asked four random participants to think aloud as they navigated the screens. I was concerned with their perception of the flow for each scenario. Were the scenarios useful? Is there confusion? How might they describe the utility of the flow? The responses were fairly consistent. The participants recognized certain elements of each “search feature” and found little trouble navigating through the red routes at low-fidelity. It’s important to remember these features aren't particularly novel, the utility is in the coordination. A couple of participants voiced interest in how these scenarios were to be integrated with other pertinent information which spurred ideas surrounding information architecture.
I constructed a site map to visualize the architecture of content. To validate the subject matter and re-align with user needs I created user stories.
The user stories fundamentally represent explicit insights gathered from user interviews and reflect the goals of our key persona. They also gave me the opportunity to think about prioritization of screens and levels of importance. To bridge the gap between sketches and wireframes I created user flows for the identified critical routes. This process helped visualize user touch points and facilitated additional coordination.
The wireframing of early solutions addressed the balancing act of achieving desired informedness without burdensome amounts of text. I wanted to alleviate cognitive burnout and instigate meaningful engagement. That started with critical screens and extended to the prescription page and the plan screen layout. This was a great opportunity to revisit thoughts on the technical feasibility of core features. Simple affordances amplified by integrations and big data analysis actuate interaction design.
4. Deliver
Interactive prototype, Usability Testing
Value-based Design
Knowing who you're designing for is as important as what you're designing. Research suggests that seniors perceive information differently than younger people. Their cognitive structure is sequential whereas, let's say, millennials’ are parallel. In other words, when interfacing with technology seniors require a step-by-step approach at a slower pace while the cognitive structures of millennials enable the processing of multiple bits of information at once. I coupled this knowledge with specific principles in order to guide the visual design process.
Empowerment and accessibility, integrity, transparency, functionality, and appeal were among said principles guiding style decisions. Simple compositions, thoughtful affordances, the omission of any unnecessary copy or technical language, a minimalistic approach. Our key persona needs to feel confident her healthcare resources provide value. In order to do that, I created a dashboard app that focuses on high priority features, all coordinated in one place.
Ironically, four out of my five user research participants use tablets and three of them mentioned how they use them in the waiting area before doctor visits. This inspired the decision to create the Medicoor dashboard app with tablets in mind.
Current CMS resources cultivate discord, the Medicoor dashboard app seeks to unite Original Medicare beneficiaries with their plan. To validate this statement I prepared a usability test plan and script and began the process of recruiting participants.
Care Coordination
The high-fidelity prototype was primed for testing. Is it intuitive? Is it easy to use? Are there an adequate amount of plan details displayed? I developed a series of questions including two tasks to find out.
Task #1: Search for and schedule a consultation with a physician
Task #2: Search for the costs of a procedure or treatment
For each task, I read aloud a scenario keeping sure not to include words that lead the participant in any specific direction. My goal was to find out if the design elements, specifically, the copy and affordance measures advanced the participant towards the end goal of each critical flow in an easy way. Other sections of the app represent key features of the Original Medicare Plan and I approached testing the usefulness of each section by asking the participants to describe what they saw to elicit open-ended dialogue and qualitative insights.
Usability insights:
- Users want to engage with their Primary Care Physician before scheduling a consultation with a specialist
- Users were confused on where to begin searching for the costs of a procedure
- Users found the search bar for Task#1 hard to find
- Users couldn't see the dates to schedule a consultation as they scrolled
I implemented solutions to all of these insights within the final design. The user is now able to see the date of choice when deciding on a time slot for consulting with a provider. Also, a critical usability issue reared itself as users found the search bar for task #1 at the last moment. I moved things around on the Provider screen in order to flag attention more quickly. I remain committed to instructive and simple copy so making revisions to the “out-of-pocket” cost at-a-glance button extinguished the confusion of where to begin searching for the costs of a procedure. Providing alternatives to account for differing approaches to seeking specialty care truly represents the effort to empower each beneficiary with the control they deserve in taking healthcare matters into their own hands. Therefore, implementing the messaging feature to the office of the Primary care physician provides the beneficiary with the flexibility they deserve.
Impact and Takeaway
I learned a lot working through this project. I learned to be mindful of biases and to more efficiently consider alternative ways of thinking. My success criteria evolved as did my understanding of the problem space. I intended to capitalize on quality feedback in order to iterate solutions and remain lean in the process. Medicoor seeks to coordinate the experience of healthcare for Original Medicare beneficiaries so that they feel more comfortable… more confident, investing their time in their own wellness. Success means users find the app useful and easy to use meaning they feel more empowered to proactively interact with their plan which, by the way, indirectly effects overall population health. The goal is to inspire and, eventually, incentivize a new patient conscientiousness one meaningfully simple step at a time.
As our work is never absolute I do believe there are improvements to consider moving forward. With more time and resources I’d look to extend the depth of user research in order to uncover additional friction. I’d engage and interview providers to understand the incentive models they operate under in order to ideate on behalf of the patient-provider relationship as a whole. The future of healthcare is promising and I believe progress begins with patient education and empowerment.
