Pursuing the Triple Aim: CareOregon The Triple Aim: An Introduction The Institute for Healthcare Improvement (IHI) launched the Triple Aim initiative in September 2007 to develop new models of care that will simultaneously address three goals: Improve the health of populations; Enhance the patient experience of care (including quality, access and reliability); and Reduce, or at least control, the per capita cost of care. Components of a health system fulfilling the Triple Aim will include: 1. A focus on individuals and families 2. Redesign of primary care services 3. Population health management 4. Cost control platform 5. System integration and execution In the first phase of work, September 2007 through May 2009, IHI worked with 15 sites, studying effective strategies and exchanging key findings for possible further action. Detailed information about the Triple Aim Concept Design is available at www.ihi.org/ihi/programs/strategicinitiatives/tripleaim.htm. CareOregon is one of the sites initially working with IHI on the Triple Aim. This paper summarizes CareOregon s approach and experiences so far. About CareOregon CareOregon is a not-for-profit Medicaid managed care health plan and dually-eligible Special Needs Plan serving about 110,000 low-income Oregon residents eligible through Medicaid, including 5,600 who are dually eligible through Medicaid and Medicare. Nearly half of CareOregon members are persons of color, one third speak a language other than English as their primary language, and 53 percent receive care from contracted safety net providers within CareOregon s network. CareOregon s Health Plan Provider Network includes 950 primary care providers (serving members who receive their care in community health centers, academic health centers, large health system or group practices or private practice groups); 3,000 specialists; 33 hospitals statewide; and 14 Public Health Departments. CareOregon has been pursuing a quality strategy for many years, but it is not just about quality; it is also a sound business strategy. This perspective grew out of a fiscal crisis that almost bankrupted the organization.
During the economic recession of 2001, the state of Oregon, like many other states, engaged in financial belttightening that had a significant impact on CareOregon. Benefits were cut and contracts were re-negotiated, contributing to an already challenging fiscal picture at CareOregon. The board of directors brought in new leadership and pursued a new strategy, says David Labby, PhD, MD, CareOregon s Medical Director. Part of the turnaround involved the question of how do we start thinking of ourselves as more than an organization that contracts with providers and pays claims, but rather as an organization that is responsible for the health of a population. We knew we had to do things fundamentally differently, and help our network do things fundamentally differently. A period of aggressive fiscal management including moving to an extremely high use of generic drugs through a formulary program, meticulous benefit management, and use of predictive modeling to identify atrisk patients and provide complex care management paved the way for the transition to an integrated system that manages a population of patients. Two programs that grew out of this shift are Primary Care Renewal, an initiative designed to strengthen primary care, and CareSupport, an internal case management project designed to provide stronger support for high-risk patients. Both initiatives are highly consistent with the goals of IHI s Triple Aim work. We were able to apply the Triple Aim framework to these projects because there is such great synergy between them, says Rebecca Ramsay, BSN, MPH, CareSupport Unit Manager. The Triple Aim is very congruent with our thinking, says David Labby. As a payer, we get capitated dollars from the state. Our job is to take care of the population. The healthier our population, the more successful our business model will be. So improved health care outcomes become a business strategy, which is very much in line with the Triple Aim. In 2004, CareOregon launched a comprehensive program called CareSupport and System Innovation, designed to support and fund provider organizations within its network that were willing to undertake improvement initiatives. Guided by the Institute of Medicine s six aims for improvement, and subsequently the three goals in the Triple Aim, CareOregon has invited provider organizations both outpatient clinics and hospitals to submit proposals for funding of programs that foster a culture of evidence-based practice and continuous improvement, empowering providers with the skills, knowledge and resources necessary to meet Institute of Medicine and IHI quality-related goals. System Components to Achieve the Triple Aim at CareOregon As is the case with other types of improvement work, with the Triple Aim it can be hard to fit a health care system s specific programs or initiatives neatly into each of the five categories listed above. Many initiatives touch on more than one component. For example, efforts to redesign primary care may also work to bring about cost control. Nevertheless, following is an overview of the work being done by CareOregon, organized into the five system components necessary to achieve the Triple Aim. 1. A focus on individuals and families It is a challenge every day to design for individuals and populations, says Rebecca Ramsay, because it requires balancing individualized care for unique patients with standardized, reproducible interventions and care processes for populations. CareOregon leaders say that striking this balance is an example of mass customization, a strategy that combines efficiency with flexibility. In the old way of practicing, a complicated patient might be referred to a nurse or care manager to see what they could do to support the patient, says Labby. If you gave the same case to six different people, you d get six different solutions. Mass customization, says Labby, is about building a system where the right things
happen reliably for patients on a routine basis. Rebecca Ramsay calls this creating buckets of care management solutions. The goal, they say, is to act with the individual, and learn for the population. The evolution of multidisciplinary rounds at CareOregon is a perfect example of how this works. When care management staff members evaluate patients together over time, says Ramsay, they begin to recognize themes that apply to the care of populations. This has become an important skill at CareOregon as staff assess members who might benefit from the organization s CareSupport program, which provides active case management. These patients typically are at high risk for functional health decline and/or they are within the top six to 12 percent of the health plan s membership in terms of utilization. If their risk factors are judged to be modifiable, they are entered into CareSupport. Through the course of many sessions, we learned to identify what is modifiable, what interventions might help. We put together a language and a logic that help us think about complex patients in terms of five care domains, and we use these domains to assess patients, says Rebecca Ramsay. These domains are derived from the Chronic Care Model originally developed by Ed Wagner, MD, MPH, of the MacColl Institute. Further information about the Chronic Care Model is available through the Improving Chronic Illness Care web site at http://www.improvingchroniccare.org/index.php?p=the_chronic_care_model&s=2. The five domains CareOregon uses to assess patients are as follows: 1. Medical Status: What is the prognosis? Can medical treatment improve it, and can we help facilitate that? 2. Medical Home Relationship: Is there a functional Medical Home? What can we do to facilitate this? 3. Medical Services Access: Are there system barriers to receiving needed care? 4. Self-Management Capability/Willingness: What can, or will, the member do to manage health? 5. Social Support System: Are appropriate social supports in place? These are depicted as the CareSupport model in Figure 1. Figure 1: CareOregon s CareSupport Model
The CareSupport model has not only made the assessment of patients more reliable and consistent, but it has also reduced the time and the number of clinicians necessary for the task. When we started this process two years ago, we would spend an hour talking about a complicated patient, says Ramsay. Now, however, consistent use of the five care domains as a framework for assessment has made the process more efficient and more productive. The number of health care professionals required to perform the assessments dropped from six in 2006, to two in 2008, says Ramsay, because everyone is on the same page. This saves time the amount of time required decreased from 90 minutes four times per week, to 120 minutes once a week and human and financial resources, and increases the number of patients who are assessed. Figure 2 shows the details. Figure 2: Assessments and Costs Associated with CareSupport Helping patients maximize each of the five domains often requires tailoring solutions to the individual. For this, CareOregon advocates inviting the patient to co-design interventions, a process that relies on a strong patient-clinician relationship. Motivational interviewing, a technique for eliciting from patients health-related goals that are important to them, is at the heart of creating these relationships. Customer co-design starts with the patient s personal characteristics, not their medical ones, says David Labby. Learning from the individual requires knowing the individual. And knowing the individual, says Rebecca Ramsay, starts with asking the right questions, and listening carefully to the answers. In our traditional CHF [congestive heart failure] program, trying to reduce re-hospitalizations, we found we got very little engagement from patients when we started asking medical-centric questions, she explains. When we adjusted the way we approach patients, and asked, What is important to you? we got much better engagement. You can often get at some of the same answers, but the right approach opens the door to a more meaningful dialogue with patients. 2. Redesign of primary care services CareOregon s Primary Care Redesign (PCR) initiative is currently underway at five clinic systems that represent a striking diversity of clinic types. Included are a clinic system that serves as the largest public safety net in the region; a multi-site community health clinic system serving migrant farm workers; an urban clinic serving mostly homeless patients; a university-based family practice clinic; and a hospital-based internal medicine clinic that predominantly
sees Medicare and Medicaid patients. It is challenging to define the commonality across these clinics because the populations they serve and their funding mechanisms are so different, says Rebecca Ramsay. But their leaders and providers came together under a common mission, to redesign the delivery system because they feel it is broken, that there is too much fragmentation. They believe that we can improve population health, patient experience, and decrease per capita cost, but that in order to do so we must be very deliberate. CareOregon s PCR initiative was inspired by the transformational work done by the Southcentral Foundation of the Alaska Native community in Anchorage, Alaska. (See the related story, Alaska Native Medical Center: Values-Driven System Design, at www.ihi.org/ihi/topics/patientcenteredcare/patientcenteredcaregeneral/ ImprovementStories/AlaskaNativeMedicalCenterValuesDrivenSystemDesign.htm). It is built around the creation of robust patient-centered medical homes, a model of primary care that emphasizes creating close ongoing relationships between patients and their care teams of medical and behavioral clinicians, care managers, and medical and team assistants. The teams either deliver or coordinate all of a patient s care, with a particular focus on providing preventive care, screenings and chronic care management. Primary care clinicians who practice in medical homes often refer out to specialists less often, and when they do, there is greater coordination and collaboration, leading to greater continuity for patients. Expanded access and after-hours coverage is also a hallmark of a medical home, so that patients do not turn to the emergency room for non-emergencies simply because it is the only alternative outside of regular office hours. David Labby believes that over the long term, the PCR initiative will also help to address the shortage of primary care physicians that is being felt in Oregon and throughout the US. Creating care teams allows all health professionals to focus on what they are best qualified to do working at the top of their license and capability which should make primary care more attractive and help bring more medical students into primary care, increasing the availability of primary care doctors. 3. Population health management The medical home model also emphasizes a population perspective. David Labby says that this is a fundamental shift in thinking for most physicians. In most current practices, physicians move from room to room seeing patients, providing care, making referrals, trying to stay on schedule. As they say at Southcentral, Your job is not the schedule. Your job is the health of the population. A population approach requires everyone to adjust his or her thinking. Case managers are usually very specific, creating individual solutions for each patient, says Rebecca Ramsay. While the system components necessary to achieve the Triple Aim do emphasize a focus on individuals and families, that focus must be balanced with care solutions that have population impact. Labby offers an example of how this population perspective works. We decided to look at a population of high emergency department utilizers, he says. We want to redirect them. We identified best practices with this population, asked our case managers to listen to patients reasons for using the ED and to do PDSA [Plan-Do-Study-Act] cycles. We learned that a good number of them didn t have a robust primary care home. Now we are able to set them up with that. This not only provides patients with better continuity of care, but is also an important cost control measure. CareOregon s Primary Care Renewal initiative also focuses on this new way of thinking about care. Computerized patient registries are used to support this focus on the population, helping medical home teams get the big picture of how specific categories of the patients on their panel are doing, and helping them proactively reach out to patients who may not be coming in for necessary care. 4. Cost control platform The basic challenge of cost control, explains Labby, is to create savings that can be reinvested in the system. There is no new money to put into the system, he says. We need to identify what we can do that both decreases cost and increases health.
A new payment methodology is being created as part of the PCR initiative to support the new medical home model. The goal is to pay for work that matters. In other words, says Labby, We want to focus on the things that a primary care practice can impact. If they don t do inpatient care, they probably can t change how long someone is in the hospital, whether two days or three days. But they might be able to keep a patient out of the hospital to begin with, by providing care management, say, for congestive heart failure, which we know decreases readmissions. CareOregon s leaders are also exploring ways to pay for health outcomes that matter. We are actively talking with clinics about what we collectively value, says Labby. We want payment to be done in partnership with clinics so that payment funds continuous improvement toward the population health goals that matter to the community. 5. System integration and execution We are not an integrated system, but we are trying to create a virtual integrated system, says David Labby. In fact, CareOregon prefers the term integrated care home to medical home, says David Ford, CareOregon CEO, because it is not just medical care that patients need. There are social, behavioral and life-quality aspects of care that must also be coordinated. Comprehensive, integrated care is the vision that has driven a lot of the board s direction in leading our efforts around policy and clinical evolution. There is a larger vision here of creating health for Oregonians. Especially because CareOregon providers serve as a safety net for so many patients, the organization is particularly focused on the bigger picture of what it takes to help patients get and stay healthy. To provide world-class kids care, for example, says Ford, how do we influence their diet and activity? What are we doing about family violence? These issues should not be outside our sphere, and our job is to create a community care model that links supports together around patients. Measures related to the Triple Aim Metrics in both the Primary Care Renewal and the CareSupport initiatives reflect the goals of IHI s Triple Aim. Specific measures from these programs can be sorted into the Triple Aim goals: The Experience of Care Satisfaction, both patient and provider Access (number of days to third next available appointment) 30-day hospital readmission rates Population Health Receipt of evidence-based care Improved patient functional status Cost Control Hospital and emergency room admissions per thousand Per member per month costs Hospital admissions for ambulatory care sensitive conditions Sample Results So Far The Experience of Care A recent CareOregon patient satisfaction survey shows that 89 percent of patients surveyed are satisfied with the care they receive (Figure 3).
Figure 3: Patient Satisfaction Results for CareOregon Population Health CareOregon measured the functional health status of patients in the CareSupport program using a tool called the Health Utilities Index (HUI3). This measures health-related quality of life through eight attributes: vision, hearing, speech, ambulation, dexterity, emotion, cognition, and pain and discomfort. Figure 4 shows that patients in CareSupport for four months achieve an improved health status. Figure 4: Changes in Functional Health Status Mean HUI-3 Utility Score Comparison of overall HUI3 scores at baseline and 4 months for CareSupport managed (CM) and non-caresupport managed (Non-CM) 0.6 0.5 0.4 0.3 0.2 0.1 0 0.19 0.24 CM (N=165) 0.25 0.25 Non-CM (N=37) Baseline 4 Month Follow-up
Cost Control CareSupport has resulted in significant cost savings (see Figure 5). CareOregon reports that from 2004 to 2005, CareSupport participants reduced the hospitalization rate by 43 percent, compared with growth of nearly 11 percent for members without CareSupport. Short-term (i.e., less than one month) CareSupport participants reduced hospitalizations by 17 percent over the same period. CareOregon officials estimate that fewer hospitalizations saved between $5,000 and $20,000 per high-risk member during that period. They further estimate that overall return on investment in CareSupport over this period was $5.9 million. Figure 5: CareSupport Member Costs Analysis: University of North Carolina Figure 6 shows a breakdown of cost savings by service category. CareOregon case managed 1,991 unique patients during the period of January 1, 2007 through May 31, 2008. Figure 6 demonstrates the cost per capita (excluding cost for mental health services) for 1,991 case-managed patients between January 1, 2007 and May 31, 2008. A significant reduction in the costs associated with hospitalization in this cohort is primarily responsible for a total cost reduction of $5,054.88 per member per year. In 2007, CareOregon calculates a $7,309,966 savings for the 1,445 complex members served by the CareSupport program. Taking into account the yearly cost of the program, this amounts to a 1:4.22 return on investment.
Figure 6: CareSupport Cost Savings by Category Cost for CareSupport Enrolled Members (N=1991) 1/1/2007-5/31/2008 Pre-CS = claims paid for 12 months prior to enrollment in CareSupport Post-CS = claims paid for 12 months following enrollment in CareSupport. 1200 1000 1044 Pre Case Management Post Case Management Total Expenses 800 600 400 697 271 275 316 260 200 0 26 26 10 14 42 41 109 119 65 63 1 1 54 49 Service Category Observations Working toward achieving the Triple Aim is both challenging and exhilarating. Following are some observations from the leaders at CareOregon who are pursuing this challenge: If you ask what we are trying to do, it s simple: we are trying to reduce mortality and increase longevity. But if you think about what are the drivers of longevity, it is so much more complicated than just medical care. It forces you to look at things in a different way. Our job is not just to live within our budget, our job is to have healthy Oregonians. David Labby For complex patients, we talk about what is within our solution space. What can we impact? We won t have much affect on people s values or history, but we might be able to affect their skills, or knowledge, or health activation. So that s where we focus. Rebecca Ramsay In the old model of running a clinic, your economic unit was visits. In the new model your unit is health outcomes. It s a paradigm shift. David Labby Within the system there is a starvation for how to move things forward. Once you begin, that s the transforming engine that keeps people engaged. It is not just a project. It is a transformation. And we need internally to develop the methodology to be serial innovators. David Ford