How, or if, America will ultimately achieve better integration of medical and social services is an open question.
This article was written by Jon Christianson, PhD, Medica Research Institute senior fellow, and James A. Hamilton chair in health policy and management at the School of Public Health at the University of Minnesota.
A few months ago, I was driving through a small community in southern Minnesota that, according to its residents, was the first town west of the Mississippi founded by Norwegian immigrants. (To you, this may sound like the answer to a very difficult trivia question, but it has more significance to some of us in this part of the world!) The town was in the midst of celebrating Norwegian Constitution Day. In Norway, the celebration consists mostly of children’s parades and other events that highlight and reaffirm that country’s shared cultural values. Coincidentally, it also was the last day for potential grantees to submit proposals to participate in CMS’ Accountable Health Communities (AHC) initiative. If you are familiar with the AHC model, you probably guessed the connection.
The AHC initiative provides an interesting example of how health services research can influence US health policy development over time, and perhaps even change the way health policy is viewed in a broader sense. Somewhat arbitrarily, one could argue that this particular journey began with the emergence of “population health” as both a term and a framework for assessing the performance of our health care system. Three of the leading players in the story are David Kindig, MD, PhD; Don Berwick, MD, MPP, FRCP; and Elizabeth Bradley, PhD, MBA; all are well-known health services researchers but also expert translators of their work to policy makers. Berwick took the further step of “crossing over” to direct the actual implementation of health policy as acting head of CMS.
The Concept of Population Health: Laying the Foundation
The concept of “population health” has been discussed for decades, but it was elevated to a much higher level of consciousness through an article published in 2003 by Kindig and Canadian colleague Greg Stoddart, PhD.1 In their article, “What Is Population Health?,” they offered a very brief, but also very broad, definition: “…the health outcomes of a group of individuals, including the distribution of such outcomes within the group.” (p. 380) They followed with the suggestion that population health could be thought of as a field of study: “…the field of population health includes health outcomes, patterns of health determinants, and policies and interventions that link these two.” (p. 380) The authors offered specifics in each of these areas, but anticipated that some readers might find their definition too broad to be useful. (p. 382) They were right.
Since then there have been a number of attempts to revise the definition and, in some cases, provide more specifics through the addition of modifiers (eg, population health management, population health medicine).2 Other efforts have sought to distinguish the use of “population health” to describe residents of a geographic area (“total” population health) from its application to specific subgroups of the population defined by disease state, sociodemographic characteristics, or organizational affiliation.2
In this ongoing discussion of how best to think about and measure “population health,” it’s easy to overlook other observations made by the authors. Of particular importance for our purposes, they noted that the determinants of “population health” (in addition to good quality medical practice) were likely to include such factors as “the social environment (income, education, employment, social support, culture) and the physical environment (eg, air and water quality)”1 (page 381). The implication was that improving population health—as measured by such things as mortality and quality-adjusted life years—requires not only better, more cost-effective medical care and social services, but also improved coordination of services across these domains and possibly a different balance between the 2.
The Triple Aim: Expanding the Discussion
Five years later, Berwick and colleagues Thomas Nolan, PhD, and John Whittington, MD, MBA, incorporated the concept of population health in their immensely influential article3 that proposed a framework for guiding public and private efforts to improve the healthcare system. They argued that “the United States will not achieve high-value healthcare unless improvement initiatives pursue a broader system of linked goals [consisting of] improving the individual experience of care; improving the health of populations; and reducing the per capita costs of care for populations” (p. 760).3 With that statement the Triple Aim was born. A focus on the health of populations (the “population as the unit of concern”) was critical, they argued, in pursuing the Triple Aim. They wrote, “only when the population is specified does it become, in principle, possible to know about its experiences of care, its health status, and the per capita costs of caring for it” (p. 762).3
The authors proposed that an “integrator,” would be needed to link healthcare organizations and other groups, such as social service providers, and coordinate efforts in pursuit of the Triple Aim. Specifically, among its other responsibilities, an integrator would “encourage and cooperate with government policies, agencies, and programs to discourage smoking, provide alternatives to violence and substance abuse, and address community determinants of mental health problems” (p. 164).3 In a single-payer health system, integrators could be provided with a global budget and charged with using it efficiently to “take care of the healthcare needs of a defined population” (p. 768).3 In regard to the US healthcare system, the authors offered examples of organizations currently playing the envisioned integrator role, but acknowledged the number of true “integrators” was very limited.
The Triple Aim concept has become the directional beacon for many health care organizations and public policy initiatives. It has highlighted the importance of including population health in health policy discussions and, while not its primary focus, advanced the idea that improving population health will require effective integration of medical and social services. Until the “integrator” function is better understood and developed, the authors suggest that large healthcare delivery systems might be best positioned to be integrators in pursuit of the Triple Aim.
Connecting Social Services and Population Health: Establishing the Evidence Base
While Kindig, Berwick, and colleagues noted that social services, living environment, and other non-medical factors can have an important impact on population health, Bradley and her colleagues advanced the evidence base for this notion. In a 2011 article,4 and later in their book The American Health Care Paradox: Why Spending More Is Getting Us Less, the researchers examined macro-level spending data from the United States and other industrialized (primarily European) countries. What they found was striking. The ratio of spending on social services to spending on health services was the lowest in the United States and Mexico (the only countries in their data where it was less than 1) but greater than 2 to 1 in countries like Sweden, Belgium, Denmark, and Austria (page 829).4 (The residents of the aforementioned southern Minnesota community might be disappointed to learn that the ratio for Norway was approximately 1.75 to 1, less than Sweden but at least higher than the United States!)
More importantly, Bradley et al,4 found that “the ratio of social to health expenditures, adjusted for GDP per capita, was significantly associated with greater life expectancy, lower infant mortality, and fewer potential years of life lost.” The authors acknowledged that while this favorable association did not occur in all the population health measures they examined (p. 826), it was still thought-provoking. Bradley and colleagues were careful to point out that their statistical findings did not imply causation and that the number of countries was small. Nevertheless, they modestly suggested that in addition to efforts to improve health status by increasing—or more appropriately targeting—health expenditures, “additional attention to social services is also needed” (p.830). In other work, they reported results from a literature review suggesting that specific US social programs have been found to improve program-related health measures.5
The researchers took a step further, performing essentially the same analysis using the different states in the United States.6 The results were quite similar: states with higher ratios of social to health spending (where social spending was calculated as the sum of social service and public health spending, and health spending was the sum of Medicare and Medicaid spending) had better population health outcomes. They suggested that their findings “may inform efforts among policy makers, clinicians, and researchers to leverage social services and health spending more effectively to improve population health” (p.761).6 But how might this be done in practice?
Accountable Health Communities: Learning by Doing
In January, 2016, CMS’ Center for Medicare and Medicaid Innovation announced its intent to test the effectiveness of different variations of what it termed the Accountable Health Communities model.7 According to CMS, the AHC model “is based on emerging evidence that addressing health-related social needs through enhanced clinical-community linkages can improve health outcomes and reduce costs.”8 Further, it “addresses a critical gap between clinical care and community services in the current healthcare delivery system by testing whether systematically identifying and addressing the health-related social needs of beneficiaries impacts total healthcare costs, improves health, and quality of care.”8 The AHC initiative is a direct outgrowth of Goal 5 of the CMS Quality Strategy (2015), which states that “successful efforts to improve social determinants of health and access to appropriate healthcare rely on deploying evidence-based interventions through strong partnerships between local healthcare providers, public health professionals, community and social services agencies, and individuals.”
The AHC initiative is ambitious and somewhat complicated. CMS plans to contract with up to 44 “bridge organizations” (a direct counterpart to Berwick et al’s “integrators”) for 5 years. These organizations can choose to pursue 1 of 3 tracks that have various levels of requirements: increasing beneficiary awareness of available community services; providing navigation services to assist beneficiaries in accessing services; and encouraging alignment of community partner organizations to assure services are available for beneficiaries. Track 3 would include all activities, while track 2 would contain only the first 2 and track 1 only screening and referral of beneficiaries to social services. Bridge organizations are prohibited from using their AHC program funds to directly purchase services.8
An extensive evaluation of the AHC effort is planned, which will involve randomization and/or matched comparison groups. Participants likely will vary not only in their “tracks,” but also in the types of beneficiaries they choose as a focus for social services (eg, those with diabetes), the details of the programs involved and the types of services they offer, and the characteristics of the “bridging” organizations.
The Work of Many Hands
While the AHC evaluation is underway, ideas about of how best to integrate social and health care services will continue to emerge. Looking forward, Bradley and Taylor observed in their Health Affairs blog that “integrating the work of health and social services will be long-term work requiring many hands.”9 Some have expressed hope that Medicare accountable care organizations will find it advantageous to play the integrator role, facilitating access to appropriate social services for their assigned beneficiaries, or even for residents of the larger communities in which they provide services.10,11 Others have expressed concerns that this presages a “takeover” of the social services enterprise by more powerful health care organizations.
There are examples of voluntary, community-level attempts to better integrate some types of social services with healthcare delivery (Accountable Care Communities and Accountable Health Communities)12;13 Related to this, the National Quality Forum has created an Action Guide (2014) to assist entities in coordinating efforts to improve population health.14
Proposals regarding how best to integrate medical and social services, and rebalance medical and social expenditures, have arisen from all points on the political spectrum. For instance, an affiliate of the conservative Manhattan Institute has proposed that Medicaid programs be given greater flexibility at the state level to allocate funds in ways that best improve the health of beneficiaries, including spending Medicaid dollars for social services where population health gains warrant it.15 Others have suggested more careful study of how some European countries coordinate medical and social services.16 (Perhaps it’s time to brush up on my Norwegian language skills.)
How, or if, America will ultimately achieve better integration of medical and social services— and whether significant improvements in population health occur as a consequence—is an open question. Even so, over the past 16 years the cumulative efforts of health services researchers have significantly contributed to a broader discussion of how to efficiently spend resources in the pursuit of better health outcomes. But, of course, more research is needed.