segment high-need patients and match the appropriate interventions as well as from a set of key measures to support value-based payment.

MAIN THEMES AND LESSONS

The first important lessons from this initiative are that the high-need patient population is diverse, complex, expensive, and often transient, and the heterogeneity of this population suggests that a similar diversity of care models will be needed to address the range of problems these individuals experience. At the same time, there is a need to strike a balance between standardized and customized approaches to care. In that regard, segmenting patients can be a useful tool for targeting care, but there is need for more real-world testing and refinement of approaches for segmenting patients in conjunction with care models demonstrated to work with certain subpopulations of high-need patients.

Another key lesson from this initiative is that just improving medical care for high-need patients will not address all of the challenges they face, nor will it lower the cost of care. To be successful, as the examples cited in Chapter 4 and in Appendix A demonstrate, care models for high-need patients will often need to address the social risk and behavioral health factors that play an outsized role in the lives of these individuals. Going forward, care models, policies, and assessment tools need to address social services and behavioral health needs in addition to those services normally considered the purview of health care systems. The final overarching lesson is that to be actionable, policy solutions must account for existing constraints and complexities arising from the lack of integration of medical, social, and behavioral services and with the way the United States finances care models.

In addition to those overarching lessons, a number of important themes emerged from the presentations and discussions in the three workshops and deliberations among the committee members. These included:

Segmentation and Taxonomy

The high-need patient population comprises a heterogeneous group of individuals that have a diverse array of conditions, making segmentation of this population into a finite number of subpopulations an important consideration when attempting to match patients with appropriate models of care. With a patient taxonomy and menu of evidence-based care models, health systems would be better equipped to plan for and deliver targeted care based on patient characteristics, needs, and challenges and to identify gaps in their ability to deliver care for specific subgroups within their patient populations. Models of

care for high-need patients must balance the need for standardized approaches for diverse populations with the need for personalization around individual patients’ conditions, needs, and characterization. In that regard, having too many segment groups becomes too complex and impractical for broad implementation, but having too few segments makes groupings less meaningful and undermines the objective to be able to target care effectively.

The purpose and utility of segmentation must guide the development of a taxonomy for high-need patients and reflect the fact that a taxonomy will be a dynamic and interactive tool and that a single individual can move between taxonomic segments as their health—and therefore their care needs—change over time. Starting a taxonomy from a medical perspective has limitations, but it is a feasible starting point for most health systems, given the availability of data. Additionally, functional status can be “baked in” to the various medical segments in a taxonomy, with social risk factors and behavioral health considerations spanning all clinical/functional segments.

Barriers around data collection and use, particularly among smaller clinics and providers who lack a sophisticated and interoperable health information technology infrastructure, limit the use of patient segmentation. There is a need to improve our understanding of the transient nature of the high-need patient population and how health systems need to account for it when developing care delivery models.

Tailored Care and Care Coordination

In the 2 years since the germination of this collaborative project, understanding and approaches to care delivery for high-need patients have evolved, with the demonstration that multiple care models can improve care. A successful care model is designed to respond to the goals and needs of patients, and an essential tool for standardizing and centering care around patient needs and goals—as well as for assessing patient needs across disease groups—is measuring functional status. Understanding patient needs and goals also requires better measurement of patient priorities, and meeting these goals will require flexible models of care.

Care coordination is critical for high-need patients, and improving care coordination will require the development of new workforce and training efforts. Such efforts are often costly, so special consideration should be given to potential care coordination approaches that help control costs. Additionally, improving care for high-need patients requires aligning the care system with social, economic, and behavioral programs and services, a task that will be difficult because of the fragmentation that exists in these fields.

• Work with payers to develop interoperable electronic health records that can include functional and behavioral status and social needs.
• Identify the threshold for targeting programs to those elderly who are frail, since not all elderly need the intensive, coordinated care these programs provide.
• Use established metrics and quality improvement approaches to create an environment of continuous assessment and improvement for these models.
• Partner with community organizations, including schools and even prisons, as well as with patients, caregivers, and social and behavioral health service providers outside of the health care system to create patient-centered care plans.
• Assess established culture and promote changes needed to institute new and successful care models, blending medical, social, and behavioral approaches.

Payers

• Actively support the adoption of care models or specific elements of models that research has shown to be effective at improving care for high-need patients.
• Work with policy makers to continue progress toward a value-based system, using alternative payment models, including those that work within a fee-for-service structure, to support more effective care for high-need patients.
• Expect that return on investment for most models of care for high-need patients will take time and that a return in 2 to 3 years is unlikely.
• Develop financing models to provide social and behavioral health services that will both improve care and lower the total cost of care for high-need patients, recognizing that even cost-neutral programs are worth supporting if the outcome is positive for patients.
• Support recognition, training, and education for patients and caregivers as part of care teams.
• Lead efforts to identify and share information about high-need patients and the potential for different models to positively affect the care of those populations.

Providers

• Meet patients in their communities or connect patients to community and other social resources and accept that much of the care they need will be delivered by family and unpaid caregivers or professionals outside of the health care system.
• Learn to work collaboratively in teams, and understand that many successful care models work best when everyone works at the top of their licenses.
• Engage with patients, care partners, and their caregivers in the design and delivery of care.

• Fully adopt the proven practices of health literacy to improve patients’ and caregivers’ ability to follow care plans developed with their input.
• Identify and work to change cultural norms that may hinder adoption of successful care models.
• Identify and engage patients’ care partners as integrated team participants.

Patients and Their Care Partners

• Seek out formal training and education experiences to enhance care, understand complex medical situations, limit injuries and other errors, and identify problems earlier.
• Explore with your care team the potential benefits of home-based care, including improved financial, social, and psychosocial outcomes.
• Request formal recognition as part of the care team.
• Participate in active communication with providers regarding quality of care, needs, and services.
• Work with a care coordinator or care coordination team to amplify self-advocacy efforts and fully utilize care models.
• Contribute to the development of quality measures to assist in better decision making around care and care delivery.

Research community

• With the involvement of patients, caregivers, and other key stakeholders, continue research on approaches for identifying and segmenting high-need patients in practice settings and matching those individuals with successful care models.
• Gather better data for care models that work, including the effective integration of social and behavioral health services.
• Develop and test a parsimonious set of metrics for measuring outcomes and return on investment for models of care.
• Identify the best methods of care coordination, workforce training, and education for caregivers.
• Study effective culture change implementation techniques to promote spread and scale of successful care models.

Policy Makers

• Increase and expand efforts to engage patient and caregiver involvement in discussions around policy options for improving care and reducing costs for high-need patients.
• Harmonize and coordinate Medicare and Medicaid programs to increase access to needed services and to reduce the burden on patients and caregivers.

• Continue payment policy reforms and alignment initiatives to incentivize pay-for-performance instead of fee-for-service.
• Incentivize adoption and use of interoperable electronic health records that include functional, behavioral health, and social factors.
• Create state- and community-level data-sharing tools which include integrated claims databases that link and share information across payers, service sectors, and provider networks, such as the Predictive Risk Intelligence System (PRISM) that Washington State developed to support care management for high-risk Medicaid patients.
• Explore the expansion of programs to mitigate financial strain of caregiving, like Medicaid’s Cash & Counseling.
• Modify existing regulations, such as 42 CFR Part II and data-sharing rules in the Health Insurance Portability and Accountability Act of 1996 (HIPAA), to improve data flow among and within agencies and providers.

CONCLUSION

Common to the presentations and discussions among participants was the notion that improving the care management of high-need patients will require engagement and coordination of a broad range of stakeholders at multiple levels. While each stakeholder sector individually may impact a patient’s life, a community, or even a regional health delivery system, one of the most expensive and challenging populations for the current health care system will remain underserved until there is a unified effort—rather than small, incremental steps—to improve care for the nation’s high-need patients and to reduce the cost of delivering that care. It is important that different stakeholder groups convene to discuss opportunities for actions and improvement, using the potential activities identified here to guide discussion and action. The taxonomy to guide care team and care model design needs further discussion, refinement, testing, and validation, as do the implementation tactics and practices to determine elements of successful care models. Policies to accelerate the spread and scale of proven models, new workforce development initiatives, suitable quality measures, and expanded data infrastructure are all at the forefront of the national health care goals of balancing quality and associated costs. Sustained attention to these areas, too, is needed.

This page intentionally left blank.