How Atrius Health Is Making the Shift from Volume to Value

Republican control of the White House and Congress won’t change the U.S. health care’s movement away from the traditional fee-for-service system to one that is based on value, paying providers on the basis of the quality of outcomes achieved for the money spent. Consequently, care providers will continue to be under tremendous pressure to change.

One leader in making the transition is Atrius Health. An innovative care provider organization in Eastern Massachusetts with 30 medical practices and a home health and hospice agency, it already receives almost 80% of revenue from global payments (a fixed payment for all of a patient’s medical treatment during a specified period of time) or other risk contracts. Seeing how Atrius orchestrated the transition from being a fee-for-service provider to one that predominantly relies on value-based payment can benefit others pursuing the same goal.

For years, Atrius Health had been focusing on standardizing processes, removing waste, and improving efficiency using improvement activities employing lean techniques. The original approach was dubbed “care-model improvement.” The approach worked and helped Atrius Health achieve among the highest quality scores in the Centers for Medicare & Medicaid Services’s Pioneer Accountable Care Organization program. But its leaders began to question whether even bigger breakthroughs might be possible. Instead of incrementally improving processes, the goal became replacing them and creating whole new care models. The results have been remarkable.

The first innovation project, Care in Place, focused on patients over 65 years old. When patients call the practice with concerning health symptoms and can’t come in to the practice, a visiting nurse is sent to the patient’s home within two hours. The nurse has real-time phone access to a geriatric care–trained nurse practitioner who, based on the nurse’s assessment, can order tests (including in-home imaging) and prescriptions and make treatment decisions.

Forty-four percent of the patients seen with Care in Place would have been sent to the emergency room had the service not been available. Patients achieve their goals of staying in their homes, high-quality medical care is provided, and costs are reduced by avoiding unnecessary hospital care. As of November 30, the Care in Place program has seen 213 patients and prevented 93 unnecessary ER visits and an estimated 41 subsequent hospitalizations. This has lowered the cost of care by $452,000. The program is live at 24 of 25 Atrius Health’s primary care clinics.

Here’s the approach that Atrius Health used to design and implement Care in Place.

Establish an innovation center. In order to allocate funds for innovation initiatives, Atrius Health’s leaders created a new department with its own P&L.

Create a dedicated multidisciplinary team. A team of engineers, designers, operations specialists, and clinicians was assembled and led by an experienced, well-respected physician “chief engineer” whose vision for the future of care was disruptive. The team used an outside consultant with experience in automotive product design to train and coach the leader and team members in the new capability.

All of the individuals were hand-picked to ensure highly divergent thinking, a desire to disrupt status quo, and comfort with ambiguity, among other core competencies. Interview questions were designed to test applicants for these traits. For example, a key competency, “drive to learn,” was tested with the prompt: “Tell us about one thing you taught yourself over the last six months and how you went about doing it.”

Some interview questions were used to test multiple competencies at once. An example is the question: “How would you go about washing all of the windows in the city of Boston?” This gave the interviewers a chance to assess applicants’ ability to creatively solve problems, think on their feet, handle ambiguity, think divergently, and think through complexity — all at the same time.

It must be acknowledged that assigning full-time clinicians to such roles, those that don’t entail directly delivering care to patients, is a large commitment. But having such people on the team is critical: Clinical credibility is essential to the team’s success, especially when it comes to spreading the new care model to clinicians in practices beyond the experimental site.

Probe the patients’ needs. Many organizations have included patients on improvement teams to help define value and waste in current processes from the customer perspective. However, patients are not always able to articulate all of their needs, and their needs vary. It’s important to understand unmet and latent needs.

To this end, the innovation team conducted multiple interviews with elderly patients and observed them in their homes. One of the many needs it identified was “stay in my home.” The dominant model of care delivery used in the United States is failing to meet this need, forcing patients to leave their homes for office visits and trips to the ER. This idea of keeping people in their homes became a powerful mantra for the team and drove the development of Care in Place.

Develop a complete understanding of the system. Innovation teams must find opportunities in the extremely complex environment that is today’s health care system. They must learn from every department in their own organization as well as outside places that could potentially impact the care model. Data needs to be created, analyzed, and discussed to understand the marketplace. Over the course of the Care in Place project, the team repeatedly met with over 60 different workgroups, leaders, departments, tech vendors, and so on. Insights were generated that would not have been possible without seeing all components of the system simultaneously.

For example, through a patient-journey-mapping exercise, the Care in Place team noticed that some of the senior patients being sent to the ER did not need to go. A chart review of 400 ER visits by patients who were more than 65 years old conducted by one of the innovation center’s physicians revealed that close to 50% of ER visits did not need ER-level care. The team then spoke with the RNs doing phone triage, who explained that, in some cases, visiting the ER is the only option because patients cannot physically come to the clinic or because all appointment slots are taken.

The team also visited the local hospital for a day. Review of the data had revealed that patients over 80 years old arriving at the ER had a 70% likelihood of being admitted. ER doctors explained that many of these patients didn’t actually need hospital care, but without understanding the social environment at home, the ER doctors felt it was too risky to discharge them. Combining these insights with those from the triage RNs revealed the opportunity to reduce hospitalizations.

In addition, the Care in Place team met with a leader at a large insurance company that insures 45% of Atrius Health’s elder population. The team received verbal commitment that the insurer would be willing to negotiate new reimbursement arrangements that would defray some of the cost of Care in Place. This commitment gave the innovation team leverage when it met with Atrius Health’s own contracting and billing departments to work out details of the model.

Give time to explore. Even with fully dedicated teams that are learning very quickly, deep understanding, inspiration, and great ideas do not happen in a week. They are built through discussions, debates, and constant learning, and can take months to fully mature. Most care-delivery organizations do not have a culture that can tolerate uncertainty for such a long period. The innovation team has to be protected from the “results now” culture prevalent in health care.

The research that informed the creation of the Care in Place model took a physician and two engineers four months to complete. During this time, the team built its knowledge of the problems, needs, and constraints and continually brainstormed possible solutions, resisting the idea of converging too quickly on a single idea. This helped them gain a deeper understanding of why patients were going to the hospital and the possible alternatives.

Allocate resources for rapid prototyping and testing. Prototyping involves testing different designs very quickly and is a critical component of innovation. Prototyping care models requires quick access to care locations, medical supplies, IT systems, new medical devices and technology, patients, and time with different providers. The innovation and operations teams must work together and negotiate the use of these resources.

During the Care in Place project, the team needed to prototype home visits with nurse practitioners. Team members identified a clinic and met with its leaders to pitch the project. They emphasized the aim to reduce ER and hospital admissions, which aligned with the clinic’s key performance metrics. Simultaneously, the innovation center’s leaders made Atrius Health’s executives aware of the need to prototype in the clinic; the executives then made it clear to the clinic’s leaders that the project was an organizational priority.

The Care in Place team worked with the clinic to set up dedicated space, medical supplies, and access to their RN triage phone room for prototyping. The clinic could not take away its nurse practitioners from their normal clinic schedules for prototype visits because it would negatively impact access in the office.

Instead, the clinic’s leaders introduced the innovation team to nurse practitioners in a per diem pool — a group the clinic had set up to help deal with fluctuations in patient demand. Two per diem NPs were secured to run the prototype visits; the innovation center paid for their time out of its testing budget. These initial visits generated important learning, tested critical assumptions, and built confidence that the model could be expanded. The innovation center is now developing its own list of per diem providers as well as other processes to ensure fast access to prototyping resources for future projects.

Protect the work. Navigating the political and operational realities of the organization is a central challenge in conducting a project whose outcome at the outset is uncertain. Consequently, the executive leadership must be educated in the innovation process so it can set expectations in the organization that important research is being done that will require time to evolve.

Executive sponsors should make sure findings are shared throughout the project. Sharing them with organizational leaders and key stakeholders helps incorporate that learning into other strategic initiatives, adding value even before the creation of a new innovative care model.

As part of the innovation team’s exploration of possible future work, its members led a series of town halls across the practice. They invited attendees to discuss some of the greatest pain points and challenges that providers and staff faced. In addition, the team collected information on the patient’s perspective from various sources (e.g., prior focus groups, surveys, and Press Ganey data). This feedback was organized into themes (e.g., “I felt rushed in my visit” and ”Limited clinic hours don’t work for me”) and then prioritized by what was most important to patients. The resulting report was shared with executive leaders and used as an input in the development of Atrius Health’s overall strategy.

Align innovation with the enterprise’s strategy. Care-model innovation is very resource intensive. It must be clear to people throughout the organization that the innovation activities are aligned with the enterprise’s strategy and have its leaders’ full support. For this reason, Atrius Health found it was critical for the innovation center’s leader to be part of the senior executive team.

As Medicare and private payers shift payment models from volume- to value-based approaches, it will become more important that care-delivery systems rethink how care is delivered. Most are not equipped to do so. Developing a core capability for innovation like Atrius Health’s will be crucial for designing and implementing the new, radical care models that will be needed.