Caring for Elders at Sutter Health
When Jeff Burnich, MD, was a practicing physician, he never believed it was possible to take great care of elderly patients during the last 12–18 months of their lives. The patients, he said, were “incredibly complex to manage.” He recalled:
I’d spend a lot of time on the phone with social workers, dietitians, and nurses. And I never felt I knew enough about what was really going in these patients’ homes. Sometimes I felt I had no choice but to send them to the hospital. It never made me feel good, but it was the only way to keep up with the other appointments on my schedule that day.
Burnich had a chance to improve the situation in 2008 when he accepted a role heading up Sutter Health’s network of several thousand physicians. In partnership with several colleagues, he built the company’s Advanced Illness Management (AIM) program. By deploying additional clinicians, particularly nurses and social workers, the program’s intent is to improve care through home visits, 24-7 telecare, care coordination, structured team communications, extra effort to educate and engage patients, and enhanced clarity about care goals.
Starting with a small pilot in Sacramento in 2009, by 2016 AIM had grown to a team approaching 200 clinicians practicing in 19 counties and serving nearly 3,000 patients. The results were powerful. Numerous metrics showed that the care was better. Simultaneously, the program slashed the total cost of care by nearly 50 percent, saving the system tens of millions of dollars annually. AIM’s senior team was actively engaged in Washington, DC, advocating for payment reforms that could help spread similar programs elsewhere in the country.
Hospice benefits augment the resources available to patients during the last six months of life, but the preceding period of several months to a year can also be fraught. Furthermore, many patients equate “hospice” with “giving up” and decline hospice care even when eligible for it. As such, the last 18 months or so of life are a period in which risks are great, expenses are high, and opportunities for improving the health-care system are plentiful.
During this time frame, patients are commonly afflicted with more than one chronic illness and struggle with one or more poorly controlled symptoms. They experience an acceleration in functional decline and commonly find they need hospital or emergency care more frequently. They may struggle with nutrition. And while many may acknowledge that they are in only fair or poor health, few have a clear understanding of their health condition. One reason for this is they may have received crucial information while dealing with a medical crisis that limits their capacity to absorb information and grasp what it might mean for their life at home.
Gaps in care can be particularly dangerous. For example, after a hospitalization, a patient may be eligible for a period of home health care. After that eligibility expires, however, there is no model for continuing care other than intermittent visits to physicians. If the physician is unavailable for any reason, many patients believe their only option is to dial 911. Avoidable hospitalizations frequently follow.
Settling into his new role at Sutter in 2008, Burnich saw that Northern California offered some advantages over his prior home in Ohio. The physicians in California were familiar with managed care and with working under payment models in which providers were at risk. They even had some experience in bringing together competing entities to better integrate care.
Additionally, in 1996, through its home care affiliate, Sutter Care at Home, Sutter Health had launched an initiative aimed at improving care for elders. Although the program improved care, it also had shortfalls. It was not accessible to all patients and not integrated with Sutter’s other services.
The expanded and redesigned AIM program was envisioned from inception as a collaboration of Sutter’s home care, clinic, and hospital entities. Thirty-six leaders from all three groups convened for a comprehensive discussion about options at a two-day off-site meeting in 2008. The consensus was that while the three groups were performing well in isolation, coordination between them was a critical failure. Patients commonly fell through the cracks in the system and ended up in the emergency room. Even when patients were hospitalized repeatedly with the same symptoms, hospital staff and home care staff were unlikely to collaborate on a long-term care plan.
The meeting generated a wealth of ideas and suggestions for the design of the AIM program. Over a period of several months, the design team studied the literature and other programs before finalizing the design for a pilot.
Burnich and his colleagues saw Sacramento as an ideal location. Sutter had a well-developed outpatient care-management program there already. Furthermore, Sacramento looked financially advantageous because Sutter already was at risk for a larger proportion of patients there than in other areas. So Sutter had care managers and a home care team that could be redeployed for the pilot, reducing the need to hire new staff.
Sutter launched the pilot in 2009, focusing on a single group of 40 physicians and adding staff to the clinic, hospital, and home care teams where necessary. Two years later, Sutter formally evaluated the pilot’s progress. The evaluation so impressed Sutter’s board of directors that they encouraged expanding the program. Sutter subsequently applied for and received a grant from the Center for Medicare & Medicaid Innovation in 2012 that accelerated AIM’s growth.
Prior to launching the pilot, Burnich and his colleagues had established a number of design principles for the AIM program. At the top of the list, care would be aligned with patients’ goals. Every patient would have an advance care plan, and those plans would be revised as patients’ health status changed.
In the view of the program’s leadership, conversations about care goals were important but not sufficient. It was also crucial to identify and understand what was important in an individual’s life—what they wanted to be able to do the next day or the next week—in order to establish goals that would engage and motivate the patient in managing their own care.
If a patient’s motivation was a crucial factor, knowledge was equally critical. The team believed every AIM patient should have an adequate understanding of their health condition. At the very least, patients needed to be aware of when escalations in symptoms suggested that contacting a clinician was prudent or, even better, should have some guidance on how to get the most out of each physician visit.
In addition, the team felt that AIM clinicians ought to have genuine and deep insight into how patients lived. What help did patients have in their homes? How were they managing their medications? Were their homes safe, given any limitations on their mobility?
Another design principle was that care had to be continuous. In the last 18 months or so of life, care needs are no longer episodic. AIM had to be readily available to patients at all times.
Furthermore, any AIM clinician involved in caring for a patient ought to be able to jump in and provide care at any time. This implied that every clinician needed access to the same notes, and those notes needed to be structured in a simple and standard way.
A final principle was that AIM was not to supplant or displace existing clinicians, clinics, or hospitals. It was to be the glue that held everything together.
The AIM program in place today evolved from these principles as well as experience. Physicians refer patients into the program. Then begins a period of intensive interaction in the patient’s home. Over a period of 4–6 weeks and 4–6 visits, a nurse and a social worker establish a foundation for continuing care. The patient’s case is thoroughly documented in a structured note with five sections: personal goals, symptoms, medication, follow-up visits, and patient engagement and self-management. Patients always receive a symptom chart coded green, yellow, and red indicating when all is fine, attention is necessary, or urgent attention is required.
As the cost of ongoing home visits is prohibitive, patients shift to tele-support after this initial period. While program clinicians are available to patients 24-7, the program’s nurses initiate most phone calls to check in with patients and offer support. The team keeps physicians up to date and ensures that patients see physicians regularly, especially when symptoms escalate.
During a hospitalization, AIM staff located in the hospitals advocate for patients, consult with the hospital team, and support patients’ symptom management when necessary. After hospitalization, patients may revert to a period of home care. Depending on the severity of the situation, this may range from only one visit to several visits before reverting to telephone support.
The AIM program is staffed primarily by nurses and social workers with palliative care experience, plus a smaller number of nurse practitioners. A palliative care certified physician oversees the team. Those staff providing in-home visits typically are assigned full-time to the program. Although those in hospitals and clinics typically support the program only part-time while sustaining other care management duties, AIM is the top priority. AIM patients are not assigned to a one-on-one relationship with a clinician. The design team considered the benefits of such an approach but recognized that only a team could be available to patients at all times.
The AIM leadership team highlights that the ability for AIM clinicians to shift from a task orientation to a continuity of care orientation as the critical success factor. Rather than checking tasks off the list, AIM clinicians needed to engage in problem-solving with patients and their families over multiple interactions, working closely and communicating fully with others on the team.
AIM’s central management team, including Burnich, Monique Reese, DNP, & Lori Bishop, CHPN, and others, guides the program as a whole with the help of regional groups of stakeholders, including physicians, palliative care experts, hospital case managers, and other leaders. Each local team also has a manager and other clinical support specialists. While the program is guided by the same processes and principles in every location, local teams have flexibility to determine the frequency and intensity of service above the minimum requirements depending on the patient panel and their needs. The average home-visit clinician in the program serves 15–18 patients; nurses conducting telephonic visits carry a caseload of 60–80 patients.
With AIM providing the heavy lifting for care and resource management for these complex patients, frontline primary care physicians have a lighter role and can focus on working at the top of their license. In fact, in Burnich’s view nothing is really required of them other than patient referral based on criteria that are brief and simple enough to fit on a credit card–sized card. But Burnich also understands that physicians could easily obstruct progress. The program depends on their support.
To gain trust, Burnich ensures that physicians are always connected and informed. They have full access to the program’s notes about each patient and they receive routine updates and requests for follow-up visits from AIM staff. They can also ask the patient to come in for a visit whenever they feel a visit is prudent.
Sustaining physician support requires demonstrating that the program improves patients’ lives and makes physicians’ workdays easier. While a small number of physicians prefer to hold on to responsibility for patients that could be referred to AIM (and of those, many do impressive work in caring for these patients), Burnich feels the vast majority of physicians in the Sutter network buy in to the program. He is particularly pleased to see AIM valued by some specialists, particularly oncologists. In 2015, over 1,800 physicians referred patients to the program.
The program has as many referrals as it can handle.
To assess AIM’s effectiveness, the leadership team tracks dozens of metrics. According to their own analyses, hospitalizations are down by more than 55 percent for AIM patients, ER visits are down roughly 20 percent, and ICU days are down more than 60 percent. In the last 30 days of life, only 3 percent of patients are seen in the emergency room and only 7 percent are seen in the ICU. Only 12 percent of enrollees die in the hospital, compared with about 60 percent prior to AIM.
Roughly 90 percent of patients have completed an advance care plan. Acceptance of hospice has risen from roughly 25 percent to over 40 percent. Median length of stay in hospice is 22 days, up from 8 days, and patients are known to live longer when they accept hospice. Alongside these improvements in care is a stunning reduction in total system costs—slightly greater than 50 percent.
Payment reform is the key to catalyzing the spread of similar programs. Burnich remains noncommittal about specifics, such as whether the payer or the provider takes on risk. Though Sutter is both familiar and comfortable with at-risk models, such as capitation and in-house health plans, the company also is engaged in a small per-member-per-month contract with UnitedHealthcare.
Burnich and the Sutter team have no interest in keeping competitive secrets. They are eager to share what they’ve learned to support the growth of similar programs elsewhere in the country.