Improving our system of care for older adults with complex, chronic illnesses requires wrestling with a vexing dilemma. Models of care that are readily scalable have limited effectiveness, and effective models are difficult to scale. As an example of the former, the patient-centered medical home (PCMH) has become widespread, but its impact on population health and health care costs varies and has been modest overall. As a stand-alone solution, the PCMH appears insufficient to deliver the diverse set of interventions required by chronically ill, older adults with complex needs—a growing segment of the US population that drives health care costs. Two complementary approaches provide opportunities to address this challenge: 1) strengthen the design and implementation of scalable models to make them more effective, and 2) innovate methods to improve the replication and spread of models already proven to be effective. Recent research and interest among funders and model builders working in the emerging field of complex care suggests that option two, innovating new methods of replicating effective models of care, deserves more attention.

Growing Needs And Complexity Of An Aging Population

The increasing prevalence of noncommunicable chronic diseases among older adults is contributing to a growing number of individuals with complex health and social needs. By 2050, the US population ages 65 and older is projected to reach 83.7 million, nearly double the size of this age group in 2010. In 2014, among fee-for-service (FFS) Medicare beneficiaries ages 65 and older, 4.3 million (15.4 percent) had six or more reported chronic conditions. National Medicare expenditures (Parts A and B) for this group totaled $126.1 billion, representing 51 percent of Medicare FFS spending. In addition to driving up overall health care costs, multimorbidity is associated with decreased functional status, especially among those 75 and older and in women more than men. The national toll of this modern epidemic, in terms of health care expenditures, disability, and reduced quality of life, will escalate dramatically in the years ahead if chronically ill older adults continue to receive costly care that fails to provide the kinds of services they most need. To address this challenge, new types of care models more specifically designed for this population are being paired with new approaches to identifying and engaging those at high or “rising” risk.

Effective Models Of Complex Care Management

Care management programs have received considerable attention and support as one promising approach to addressing complex care needs. Drawing on earlier work supported by the Agency for Healthcare Research and Quality, the authors of one review described complex care management as “programs in which specially trained, multidisciplinary teams coordinate closely with primary care teams to meet the needs of patients with multiple chronic conditions or advanced illness, many of whom face social or economic barriers in accessing services.” In practice, there is wide variation in the settings; target populations; interventions; duration of service; team composition; patient engagement strategies; and the information technology (IT), operational, training, and management systems used by different care management programs. To date, only a small number of programs have robust, high-quality evidence showing that, when implemented reliably in an appropriate context, they are effective in improving health outcomes or reducing the cost of care for complex, chronically ill older adults.

One such program, developed by Health Quality Partners (HQP), a nonprofit health care research and development organization, serves as an illustrative example of the opportunities and challenges of replicating and spreading an effective model of complex care management. The program uses nurse care managers and was designed according to a set of principles of “advanced preventive care” developed by HQP. The program seeks to reliably deliver a broad set of assessments and preventive interventions that proactively address the needs of chronically ill older adults. Care is delivered continuously over time and across care settings in close collaboration with primary care providers and community-based health and social services. The nurse care manager and other members of the care team require special training. A program-specific process and performance monitoring system supported by a custom-designed IT application is used for day-to-day management, continuous improvement, and to identify opportunities to enhance program design.

HQP has been implementing and refining its program of advanced preventive care since 2002. Through a variety of collaborative partnerships, the program has been delivered to more than 2,300 individuals cared for by more than 130 primary care practices in six counties in southeastern Pennsylvania. The HQP program has been involved in the Centers for Medicare and Medicaid Services Medicare Coordinated Care Demonstration and Bundled Payment for Care Improvement Model 2 initiatives and undergone independent evaluations. When fully implemented and reliably delivered to appropriate target populations, the HQP program has been associated with significant improvements in health outcomes and reductions in acute care use and cost, including: 33 percent fewer hospitalizations, a net reduction in the total Parts A and B Medicare expenditures of 22 percent, and a 25 percent reduction in all-cause mortality. Positive impacts on acute care use and net costs were also observed for five consecutive years when the program was provided to higher-risk members of a Medicare Advantage health plan in southeastern Pennsylvania.

The Challenge Of Spreading Effective Complex Care Management

Recently, the Commonwealth Fund conducted a review of complex care management models. Despite their effectiveness, none of the four models identified as having the most robust evidence of impact, including HQP’s program, has spread broadly in the United States. This may be because effective models of care for complex populations share elements that make their replication and adoption challenging for health care organizations. These include: person-focused care provided across time and space; an aggressive, proactive and preventive orientation; new or expanded roles for nonphysician disciplines that change team dynamics and organizational culture; special training; diverse and rigorous processes of care and communication within care teams and with external community service organizations; and new IT systems to support the work flows, data capture, and analytics to manage and make ongoing improvements to these models.

Without significant support, guidance, and tools to successfully assimilate these key elements, new adopters often seek to “simplify” the model, selectively choosing the subcomponents that are easiest to implement, resulting in diminished program effectiveness. Other threats to the successful replication of even previously proven models include: the unintentional loss of alignment between the target populations and programs due to shifts in case-finding methods, changes in program interventions, disruptions in the continuity of service, or shifts in the availability of health care or other services in a community. Recent experience suggests that financial incentives provided through payment reforms alone, without knowledge, will not be sufficient to spread effective models of care. More successful replication of complex care management programs will require the development of a new field of practical knowledge grounded in applied research. Designing and testing promising innovations in replication will allow us to gain such knowledge.

Replication Consultancy: Example Of A Testable Innovation

Despite an extensive body of knowledge about factors relevant to the dissemination of innovations in health care, little research has explored the role innovators of new care models could directly play in facilitating the replication and dissemination of their models. As part of research undertaken for a doctoral dissertation, the lead author conducted key-informant interviews of a sample of “early adopter” health systems and physician-led organizations, including one like HQP (which was unnamed), to assess their interest in engaging in various types of partnerships with innovators in complex care management to help their organizations adopt a program.

All sixteen senior leaders interviewed universally agreed that their organizations would be allocating more resources to better manage chronically ill older adults in the future. There was widespread interest in implementing HQP’s version of complex care management by partnering with program innovators, but no single specific model for partnering was viewed favorably by all participants. Some participants spontaneously expressed a desire for an alternative approach that was not originally included among those tested in the interviews.

Based on recommendations made by research participants, HQP designed a new model of collaborative partnership: replication consultancy. This approach consists of a long-term consultancy engagement whereby originators of the complex care management program train and advise staff of the adopting organization to implement the program and the systems needed to support it. The intensity and scope of the consultancy support is progressively reduced as measurable evidence of effective adoption and assimilation of key model elements and the operational integrity of supporting systems is validated. The ongoing evaluation and validation of adoption is achieved through qualitative, observational assessments as well as quantitative measures of core program processes. Many of the quantitative measures use statistical process control analyses to assess variation over time and between nurse care managers. The end goal is for the adopting organization to be able to operate independently of the original program developers, while continuing to implement the model with high fidelity and reproducible effectiveness.

Replication consultancy is one example of an innovation that may have promise in replicating and spreading HQP’s advanced preventive care program. And as HQP’s program shares key attributes with other complex care management programs serving other vulnerable populations, replication consultancy may be an innovation worth testing for a wide array of such programs. As the field of “complex care” or “advanced preventive care” for populations with complex health and social needs matures, innovations in replicating these models of care will need to develop in parallel to ensure that effective programs come into wider use.

What It Will Take

Through the lens of HQP’s 17-year experience undertaking applied research and development in this field, several insights emerge about what it will take to successfully replicate effective models of care for older adults with complex needs. Innovative partnerships that further the goals of organizations committed to improving the health of complex care patients are essential to advancing this work. Martin’s Point Health Care (MPHC), a physician-led health care organization based in Portland, Maine, with divisions offering health care services and health insurance, has already signed on to a replication consultancy partnership with HQP for guidance providing HQP’s advanced preventive care program to higher-risk Medicare Advantage members. HQP’s program has also been adapted to support the emerging population health management needs of hospital-based health systems. Doylestown Health, a community health system in Pennsylvania, is using HQP’s advanced preventive care program and design support to develop population health management capabilities to succeed under performance-based, gain share, or risk contracts with payers.

In addition to innovative partnerships, disciplined systems thinking applied to the design, implementation, and evaluation of initiatives is essential. Prematurely cutting corners while the concept for a new model of care is still on the drawing board is a common practice, but one often ill-informed by old mental models and untested assumptions. Ineffective interventions continue to be widely deployed merely because they are easy to implement or relatively inexpensive, although they offer no real benefit and have a negative return on investment. It is often better to invest first in the design and testing of a robust “proof of concept” model and make subsequent modifications to design and implementation in multiple rounds of testing to improve the model’s performance on measures of efficiency, cost, usability, resilience, and replicability.

Such an iterative approach, building progressively on prior experience and knowledge, is one that requires a constancy to purpose and a long-term, organizational learning view of this work. Just as AIDS did not become a manageable chronic disease or cancer curable through a single drug trial, so too the field of complex care will not yield miracles overnight. Partial successes should be carefully reviewed for lessons learned and built upon whenever possible. New models of complex care management can greatly alleviate the burden of human suffering and disease, and these models warrant at least the same level of sustained commitment that major disease-specific initiatives are afforded.

Institutions such as the National Center for Complex Health and Social Needs, which is funded by the Robert Wood Johnson Foundation, AARP, and the Atlantic Philanthropies and hosted by the Camden Coalition of Healthcare Providers, could serve as catalysts on a national level to spawn new partnerships, along with a more committed and disciplined approach to diffusing effective complex care management. Because new care models for vulnerable populations with complex needs share key attributes, strengths, and challenges, joining forces to share knowledge and develop this field together is crucial.

There is nothing magical about what’s needed to learn how to spread effective programs of complex care: well-designed innovations in replication, tested in prospective trials with robust evaluation to learn from and improve the intervention, with a goal of validating effectiveness as warranted. If replication efforts of sufficient size and duration are well implemented and undertaken with a clear aim, such initiatives could rapidly advance the field of complex care and bring us closer to the day when effective models of complex care management are available to the millions of Americans who need them.

Authors’ Note

Generous support for this post was provided by AARP, a founding sponsor of the National Center for Complex Health and Social Needs and its scholar-in-residence program to advance the evidence base for complex care.