Delivering Care in the Home for Elders with Complex Medical Conditions
“You know, that was just mind-blowing. We actually had time to hear his stories about his family and about the historic floods in Portland.” Shannon Jackson, MD, understood the sentiment, and it pleased her to hear it voiced by the colleague who had joined her team only two weeks earlier. The two doctors had met with the patient in the long-term care facility where he lived. An elderly diabetic, he had suffered multiple physical setbacks, including a stroke, but he had not lost his touch as a raconteur.
Patients and their stories animated Jackson at work. She could think of nothing more satisfying than getting to know somebody well and helping them through the most challenging transitions in their lives. That is one reason she had readily accepted the invitation to lead Elder at Home, a program created in mid-2015 by Providence Health & Services, the $12 billion multi-state health system with operations concentrated around Portland, Oregon.
Elder at Home was designed to serve a specific set of patients. They were old and frail; many were likely in the last years of their lives. They rarely left their homes because they were difficult to mobilize. When their health conditions worsened, they often stayed put until the situation reached a crisis point that landed them in the hospital or the emergency room.
The objective of Elder at Home was to bring higher-quality, proactive, personalized care directly to patients in their homes, with an eye towards improved care planning, care coordination, and medical decision-making. All patients were enrolled in the Providence Health Plan Medicare Advantage program, Providence’s insurance arm, and the company stood to win financially if the approach resulted in a decline in the total cost of care for these patients.
Jackson had worked at Providence since completing her geriatrics fellowship about a decade earlier. She had started as a primary care physician within Providence’s ElderPlace program, an offering that combined medical services with housing and social services. (ElderPlace was supported by Medicaid through the Program of All-Inclusive Care for the Elderly.) Later, to establish a work schedule that was more predictable and amenable to raising her small children, she had taken a position as a hospitalist.
The two roles together were ideal preparation for becoming medical director of Elder at Home. At ElderPlace, she had seen the challenges of delivering primary care to elders with complex medical conditions. As a hospitalist, she had become frustrated with the frequency of avoidable readmissions. She saw the same people way too often.
She also became sensitized to how easily care can go awry in a hospital setting. Doug Koekkoek, CEO of the Providence Medical Group and one of the Providence’s senior leaders supporting the initiative, elaborated:
When one of these patients arrives in the emergency room or in the hospital, they look very sick. In fact, they are very sick. But the treating physician may be seeing the patient for the first time. They have little context, and the patient may not be able to speak effectively on their own behalf. As a result, the treating physician has to explore many possibilities and run many tests just to set some priorities.
Illustrating how much better it could be, Jackson described one of the patients she had treated through the Elder at Home program:
The patient was a WWII veteran in his early 90s. He was still living in his home with the help of hired caregivers. His emphysema was worsening and he had been admitted five times in the last year for recurrent flares of his breathing trouble. He also had developed memory loss and was relying ever more heavily on others to bring groceries, cook meals, and get medications for him. His arthritis was especially problematic on a day-to-day basis and his knees made it uncomfortable to walk more than several feet at a time.
We initiated a conversation with the patient, his hired caregivers, and his son, who lived out of state. The patient decided to change the focus of his care towards comfort in the home. When he became acutely ill a few months later, we referred him not to the hospital but to hospice. Beyond simply ensuring he remained comfortable, the team was also able to communicate with his support network each step of the way.
By early 2016, about six months after the program accepted its first patient, Jackson’s team was serving more than 300 patients. The program’s design was borrowed largely from an outside contractor who had offered to run a similar program on Providence’s behalf. Koekkoek and his colleagues were impressed with the proposal but decided Providence was better off running the program themselves.
At the core of the care design is a staffing model that assigns a panel of approximately 110 patients to a team of three: an MD, an NP, or a PA, plus a nurse and a social worker. Additional staff is necessary to ensure 24-7 nursing coverage for every patient, an expensive proposition but one that makes it possible for the team to be involved any time a patient is in trouble. The model also anticipates adding chaplains, pharmacists, and nurses specialized in mental health issues, though at much higher patient-to-staff ratios.
Patients are selected for the program through a sequence of several steps. The first is a quantitative analysis that incorporates both claims data and clinical data. The program’s designers believe selecting the heavy utilizers of care in the recent past is too simplistic to be effective. Therefore, Providence is developing more sophisticated algorithms that predict which patients will need the program in the near future.
The quantitative analysis results in a list of patients. Providence narrows the list by seeking the subjective opinion of each patient’s primary care physician (PCP). For some patients, PCPs will indicate the program is unnecessary, perhaps because they believe the patient is not as sick as the numbers suggest. Or the PCP may know the patient is having little trouble making it to the clinic for appointments or that the family is providing robust support.
The remaining patients each receive a three-part invitation to the program—a letter, a referral from the PCP, and an in-person visit from a social worker. Almost all patients opt in; they welcome the additional care and support.
Jackson’s team invests heavily in getting to know each patient as soon as they are added to the program, a process the team refers to as “on-boarding.” First, the social worker engages the patient in an extensive conversation to establish the team’s understanding of the patient’s full life context, including how family members or caregivers are involved in care. Then, the MD, NP, or PA and the nurse visit for a thorough review of medical care and a discussion about the patient’s priorities and care goals.
Finally, the team strives to develop a full care plan in partnership with the patient. To do so, the team communicates with other physicians who are involved in the patient’s care, sharing observations and recommendations and asking additional questions. The team pays particularly close attention to the ongoing relationship with the patient’s PCP, aiming not to replace but to enhance his or her efforts.
Beyond the on-boarding process, Jackson’s team is still working out the specifics of the team’s daily routine. Each day typically includes one to three scheduled appointments with patients, all of which take place at the patient’s residence. Some appointments are in response to a patient’s request; others are proactive check-ins at a frequency the team sets based on need, at least once per quarter.
A substantial fraction of the team’s daily work is unscheduled, fluid, and dedicated to dealing with urgent issues. Patients are encouraged to utilize the 24-7 availability of the team’s nurses, and they do. Each workday begins with a 30-minute standup meeting, during which all overnight calls are reviewed and decisions are made about what the follow-up will be and who will take responsibility for it. The team also makes adaptations from normal routines in order to cover appointments for staff members who may be absorbed in a patient’s urgent needs or who simply are away on vacation. The team meets on two other occasions each week, with a broader and longer-term focus on the evolution of the program as a whole.
Providence has slated the program for growth to at least 1,500 patients. It is not entirely clear how long it will take to get there. Growth implies a heavy workload for Jackson and her nascent team. One limiting factor is the fact that on-boarding a new patient is the team’s most labor-intensive activity. Furthermore, growth requires recruiting, which diverts time away from the program’s operational demands.
Recruiting is tricky because the roles within the program are evolving. Although the Providence team planned extensively—even estimating the time required for each program task—the staffing model is still best described as an educated guess. There remains much to be determined about who does what at work each day. Well aware of the uncertainty, Jackson and her business partner, program director Tanya McGee, have gone well out of their way to learn from a similar program in Minnesota.
The team’s social workers illustrate the recruiting challenge. Until the program’s growth rate declines, the social workers will be almost entirely focused on on-boarding new patients. On-boarding is a long way from traditional social work. For potential new hires, it is difficult to understand the job before experiencing it; as such, some new hires are surprised and perhaps a bit disappointed. Meanwhile, the social workers witness some patients who could benefit from their classical skills because the patients may be going through tough transitions or navigating complicated family dynamics. The social workers are then caught between the lure of the work they have been well trained for and the pressing demands of the growing program.
The team’s nurses also find themselves in an unexpected position. When applying for the job, many are focused either on direct care or care coordination, but the reality is that all of the program’s nurses are doing both kinds of work every day. Furthermore, the care coordination role is quite distinct from other care coordination jobs at Providence. It is more intensive and more personal.
The team’s growth and evolution present a learning curve for all. Jackson herself is discovering she needs to spend most of her clinical time with the patients who are unstable or transitioning to a new care setting. These patients typically need a thorough review of their care plan and goals of care.
Jackson is also being tested in her first leadership and managerial role. She is forthright about her inexperience. As such, she leans heavily on McGee, her longtime partner and ally, and she actively engages mentors. Her current view is that the program’s most critical success factors are recruiting people she can trust and building a team culture. Another new role for Jackson is that of overseeing physician for physician assistants. It has required her to become familiar with Oregon’s scope-of-practice laws as she guides the team towards standard roles and routines.
The fact that Elder at Home is a small and new program within the large Providence organization also presents challenges. In particular, Jackson’s team members must work hard to build relationships with other Providence groups with whom they coordinate frequently. This starts simply by explaining the purpose of the program and differentiating it from similar or similar-sounding programs, like ElderPlace. In Jackson’s view, maintaining relationships with PCPs is particularly critical. She is navigating a mix of envy (“How come you get to spend so much time with these patients?”) and relief (“I’m so grateful you can help me with these patients for whom I never have adequate time!”).
Another challenge is information technology. Effective mobile technology is crucial for Jackson’s team because they do so many visits in the home. Thus, the team needs laptops or tablets that are fully integrated with Providence’s EMR systems. There have been a few hiccups along the way.
Koekkoek attributes Providence’s success in getting the program off the ground not just to the fact that Providence offers a health plan but to his close relationship with one of the leaders of that health plan. He also believes Oregon’s unusually aggressive move towards population health principles in the 1990s helped set the stage.
For Koekkoek and his colleagues, satisfied patients plus a reduction in inappropriate utilization of care will define the program’s success. So far, the program seems to be on track. McGee estimates that through the first several months of the program, 70 percent of calls to the program’s 24-7 nursing line prevented an avoidable emergency room visit or hospitalization.
Jackson is monitoring other metrics as well. Some are task oriented, including the number of care plans completed, the number of new patients added, and the number of cognitive evaluations. Some are outcome oriented, including the number of falls.
Overall, Jackson feels she has adequate time and leeway to get the program up and running. An analyst distant from the program will be gathering data to assess the overall savings, but the analysis will not be completed until roughly the one-year point in the program’s evolution.
Koekkoek is frank about the financial risk that the program represents—several million dollars per year in nonreimbursable care. He is optimistic nevertheless. Indeed, he would not be surprised to see the program ultimately deliver savings measured as a multiple—2x or even 3x—of the amount invested in Jackson’s team. Over the long term, such results could dramatically strengthen Providence’s competitiveness in the Medicare Advantage market.
Jackson knew the head-and-neck surgeons were extremely technically skilled. She also knew they did not know the patient’s situation nearly as well as she did.
The patient had been on tube feeding at his residential nursing facility for more than five months after his cancer surgery. Now, what initially appeared to be an infection was in fact a recurrence of the cancer. The patient was so weakened that he could not get out of bed on his own.
Jackson was able to participate in the conference with the surgical team, the patient, and his children, who called in by phone. After a review of the surgical and chemotherapeutic options, Jackson brought up another possibility—moving the patient from the nursing facility back to his home with hospice care. The patient warmed to this option and even asked if his dog could come home with him. He spoke about the difficulty of the last five months, his dread of news about his cancer, and his fear of dying. He also said that his biggest fear was having to go through more surgery that wouldn’t restore his ability to eat.
The patient was happiest with the idea of returning home, where he would be more comfortable and could reunite with his dog. The Elder at Home team helped him and his family make the transition. Jackson was grateful for the opportunity to be by his side at a critical moment in his life.