As states launch their federally mandated health insurance exchanges, officials in Oregon say they are poised to set a high bar for just how smoothly such a system can operate.
Challenges facing the states are considerable. In simplest terms: Exchanges must serve a vast and varied population, many of whom are unfamiliar with the machinations of health care, many of whom loathe the notion of free-market insurance and many of whom simply don’t want to confront the process.
Oregon’s answer was to approach the construction of its $300 million system using a methodology known as design thinking. Oregon joined 10 other states and the federal government to participate in Enroll UX 2014, a program that used design thinking to create a reference model for state and federal health insurance exchanges.
Design thinking is generating buzz in engineering and design as a better way to solve business problems. It’s a process that attempts to combine empathy for the context of a problem, creative insights and rational analysis to arrive at a solution. UX 2014 was led by IDEO, a Palo Alto, Calif.-based innovation and design firm that lists big companies among its clients, including Wells Fargo, Samsung and AT&T.
Leaders of Cover Oregon, which operates the state exchange, say that design thinking helped them to put together a working system under extraordinary time constraints. “We will have been barely 2 years old by the time we go live, and we are coming into a completely new marketplace, trying to capture and maintain a whole new class of consumers,” said Cover Oregon CIO Aaron Karjala.
Design thinking is one of those notions that, while potentially game-changing, can also be somewhat fuzzily defined.
IDEO says it “brings together what is desirable from a human point of view with what is technologically feasible and economically viable. It also allows people who aren’t trained as designers to use creative tools to address a vast range of challenges.”
Traditionally, “developers approach these projects as ‘IT deployments,’ focusing closely on technical specifications but not fully taking into account the needs of real users,” said Natalie Foley, vice president and COO of design thinking and strategy firm Peer Insight. “The entire notion of creating a service for an end-user is often lost on development teams, resulting in the creation of user experiences that are underwhelming and subsequently fail to capture the results — financial or otherwise — that were projected.”
Not everyone is ready to tout design thinking as the solution to such problems. Some suggest that the notion already has achieved buzzword status, becoming less meaningful as it becomes more ubiquitous. Others worry that while this approach may be effective at serving user needs, it may not always take into account the business case driving a project.
Despite such concerns, Oregon designers say design thinking has served them well.
The design effort began with an IDEO-inspired separation of potential users into four basic groups: the “engineers” who want to know everything in all its glorious detail; some who want hand-holding from the outset; others who want to take a high-level view of the situation before they dive in; and those who wish to find their own path but want to be guided along the way.
With each step of the design process, engineers kept in mind the need to serve all these groups simultaneously. This was made possible thanks to a broad range of input from potential users. Rather than work in a vacuum, developers sought direction from workgroups and advisory councils representing small business, individuals and tribal entities, as well as technical personnel. In order to manage the input of literally hundreds of advisers, Cover Oregon opted to work from the 10,000-foot level.
“We made the decision not to make the process overly formal. We wouldn’t say, ‘Yes, we will put this in or not put that in,’” Karjala said. “Everyone knew they were informing a design that we would be responsible for, so we wouldn’t get caught in the loop of trying to satisfy 1,000 people in terms of seeing their feedback implemented.”
Informed by these volumes of input, Cover Oregon went into heavy testing mode. That’s a fundamental precept of design thinking: Rather than build it once and then test drive, design thinking encourages engineers to take incremental steps, creating components that can be tested and modified on the fly.
“For example, we built a plan selection page for small employers, and it was set up in a series of tiles, with as much information above the fold as we could get,” Karjala said. “We also had a series of filters across the top. When somebody came to that page, they saw every plan we could offer, which could be hundreds of plans, and we thought that would be really helpful. What we found is that it overwhelmed everybody visually.” The team went back and broke the tiles into columns, pushing filters off to the side.
(The actual prototypes were shopped out to Deloitte Digital with Oracle developers eventually building the finished product onto the Oracle Health and Human Services Framework.)
One major move was to break up the entranceway into a series of discrete dashboards. “The dashboard will have the same basic layout, and depending on who you are, that layout will display different things,” Karjala said. “So an agent registers as an agent, or someone may register as an individual or as an employer, and they each will get a different view.”
At an even more granular level, an employee who identifies as working with a particular company will see only the health plans that company has selected, and will then comparison shop from there.
User testing likewise showed a consumer need that a more traditional development process might have missed. “They need context for where they are in the process. They want to know in general how much time they have remaining in the process, and they need a place where they can leave the site and then come back to it,” Karjala said.
These unique needs — some particular to the special circumstance that is the health insurance exchange — help to explain why Oregon found design thinking such a useful approach.
While most states already have some consumer-facing presence for delivery of state services, the health exchanges demand a new and very different hybrid.
“The biggest factor is that this effort has to integrate two separate worlds: state health and state systems that have operated under a traditional paradigm, and the commercial world, the commercial suppliers with their channels of sales and support. So this is something that really hasn’t been out there before,” said Brian Patt, head of Health Care Exchanges and HHS Platforms for Infosys Public Services.
In addition, the states are addressing a relatively new audience, some 100 million uninsured or underinsured individuals who may have little or no understanding of the health insurance milieu. Systems also must be geared up to perform elaborate eligibility checks against federal systems. “There is clearly a lot of iteration that has gone back and forth with the federal side,” said Patt.
Add to this a certain lack of subject-matter expertise. States have long had offices and departments to deal with the delivery of public services, but the health insurance exchange changes everything. “In this case, there is no one person, no smart individual who can do that, because you are creating something that has never been there before. So you need to gather all the user experience and preference as part of the discovery process,” Patt said.
All of these factors make the traditional Design, Develop and Implement (DDI) approach a poor fit for creating health insurance exchanges. “DDI models work very well if you have a very defined problem domain. In this case, we know what the objectives are, but the constituents aren’t defined, the regulations are in flux and the technologies behind it, especially on the federal side, aren’t in place,” said Patt. “DDI assumes that all those parts are in place. In this space, too much is undefined for DDI to really be a sustainable model.”
In addition, the new exchanges will need to meet a high standard for service and usability.
“What’s new and different in this marketplace is this idea of ‘Travelocity for health insurance,’ the ability to view options, sort them and categorize them. That is something states have never done before,” said Phil Poley, managing director of health and public service for North America at Accenture and former chief operating officer of Medicaid in Massachusetts.
“There has been a true intent that this should be a much more customer-friendly, first-class experience, designed to attract a broader mix of consumers that would never go to a state benefits website,” he said. “If you don’t make it easy for people who are already reluctant buyers, they will simply abandon the effort.”
While Oregon fully embraced design thinking, others opted to combine traditional development with a heightened sensitivity to the user experience. Arizona, for instance, implemented key elements of the Enroll UX 2014 model, layering these ideas over the state’s existing one-stop public services portal.
“They are much more mindful of the design thinking now than they were in our initial building for them, but they didn’t go full tilt, just because of the time constraints,” said Bobbie Wilbur, co-director at nonprofit technology solutions provider Social Interest Solutions. She helped to craft some of the final language of the Affordable Care Act and is the system integrator for the Arizona exchange.
Even with the basics in place, “you still have to listen to the community-based assistors that are actually using it and ask what they are running into, what is working, where do we need to make changes? Your ears have to be enormous. You need to know what is working and what is not and be ready to move to what will work,” Wilbur said.
That kind of thing takes time, but the combination of an existing structure along with this design thinking approach allowed Arizona to keep on track. “Arizona probably gave itself six to nine months in acceleration because of the reuse,” Wilbur said.
Back to Oregon, where more than 200,000 people initially are expected to sign up through Cover Oregon.
As Cover Oregon raced to the finish line earlier this year, organizers reflected on a process that sometimes came down to the most rudimentary approaches. For instance, designers plastered a conference room wall with paper versions of their early user interface. As focus groups walked through, designers marked off the spots where they got stuck, needed help or took a wrong turn.
At the same time, higher-level cooperation between Cover Oregon and the Centers for Medicare and Medicaid Services and other federal partners ensured that security would be solid without interrupting the user experience, and that eligibility could be authenticated without interrupting the flow of registration.
“Before you can access any sensitive information, you have to have the user of the system prove who they are. You have to ask for forms of identification: Where did you live on this date, or how much is in your bank account? So we worked with our federal partners to push that further back in the process, to get users engaged in the process first before we started asking for all that additional information,” Karjala said.
“A lot of people make the assumption that the users know what they want. We could have gone ahead and just designed to what we thought the user needed,” Karjala said. Thanks to design thinking, though, the Oregon team says it feels a lot more confident of hitting that mark on the first try.