Editor’s note: This is part of a periodic series of Health Affairs Blog posts discussing the Culture of Health. In 2014 the Robert Wood Johnson Foundation announced its Culture of Health initiative, which promotes health, well-being, and equity. These blog posts are being run in conjunction with the November 2016 issue of Health Affairs which explores roles for individuals, communities, commercial entities, and public policy that extend beyond the reach of medical care into sectors not traditionally associated with health.
It is now widely recognized that the health outcomes of populations often are determined more by social factors than by medical care. Much of the most innovative recent work on social determinants and population health demonstrates the value of partnerships across sectors, with health care systems partnering with community-based organizations ranging from housing authorities to nutrition support programs and beyond. These partnerships have proven themselves to be essential to populations enrolled in alternative payment models such as accountable care organizations (ACOs), and ACO leaders are beginning to recognize the need to integrate services both within and without the health care sector. Thus, health care systems are exploring how to collaborate with social service providers to keep patients away from costly care and improve health outcomes.
Given the substantial number of actors involved in these efforts, many health care systems are facing a strategic challenge: what role do they play in the creation and execution of a local population health strategy? This choice is likely to have profound implications for how the system chooses to address the social determinants of health.
Different Hub-And-Spoke Models
Two visual models may serve as useful reference points in discussions of how population health activities are orchestrated (Figure 1). Both take the form of a hub-and-spoke, where the hub allocates funding to and coordinates activities of spokes. The first model positions a health care organization at the center of the model and imagines other community-based organizations as spokes.
Health care organizations at the hub would contract or manage health promotion activities and social service delivery, by purchasing these services from community organizations, deciding to provide them itself, or some combination thereof. Health care organizations may be motivated to take on the hub role given the financial risk for outcomes in accountable care and the need to collaborate to achieve improved outcomes in social determinants. However, without those incentives, the additional time, energy, and cost required to coordinate spokes may not prove cost-effective for the hub organization.
Some early evidence suggests that this vision of the hub-and-spoke is where current policy and financing trends are headed. The Centers for Medicare and Medicaid Services recently announced demonstrations of accountable health communities, which called for health care providers to coordinate among a range of relevant community partners. Among payers, many Medicaid managed care plans have begun coordinating and reimbursing for services across the health and social services boundary and effectively playing a hub role.
There are some pragmatic rationales for placing health care at the hub. First, health care already controls huge amounts of funding which would be politically difficult if not impossible to wrestle away for other budgetary priorities, despite evidence at national and state levels to suggest that more social service spending relative to health spending is associated with better health outcomes. Second, health care has a tremendous amount of expertise in structuring and managing contractual relationships with vendors. Third, many health care organizations are re-branding themselves as organizations interested in “health” rather than “health care.” Widening their care coordination to include social services would further demonstrate a commitment to “health.”
Nevertheless, there are also legitimate concerns about placing health care at the hub. First, the health care system has a reputation for managing resources inefficiently, raising concerns of whether increasing their scope of services is prudent. Second, health care organizations frequently share a geographic community, such as a neighborhood, city, or county, and therefore could end up competing for contracts with spokes while avoiding other expensive place-based investments. Third, some community-based organizations fear that asking health care to assume a central role will ultimately allow for an inappropriate medicalization of population health strategies, as providers may prioritize medical interventions (i.e. laboratory screenings) that are well known to them as opposed to unfamiliar community-oriented interventions (i.e. sidewalks or housing).
The second model offers an alternative vision. Health care organizations would take on the role of a spoke alongside a range of other community actors, leaving another organization to take up the hub position. In most other industrialized countries, this coordinating role is played by local government, which sets budgets for, finances, and administers contracts with a range of stakeholders. However, it can be difficult to imagine every local government in the United States having the infrastructure or political standing to take on this work, particularly in the face of political pressure from the health care industry. As a result, cross-sector collaborations among a range of different organizations are being explored, with local foundations or YMCAs taking the lead in communities.
Both configurations of the hub-and-spoke model are already in practice in the US. In Baltimore, Maryland, Bon Secours Health System has taken on a hub role in coordinating population health efforts, focusing on increasing the stock of affordable housing in its neighborhood and helping coordinate the efforts of other organizations to increase access to education, job training, and community activities. In a similar way, Nationwide Children’s Hospital, in Columbus Ohio, has served as a hub organization in developing a Healthy Neighborhoods, Healthy Families network to address underlying health disparities in the Southside neighborhood adjacent to the hospital.
In contrast, in Spartanburg, South Carolina, a local family foundation, the Mary Black Foundation, has taken the lead in a large community health investment project, and has coordinated work among health care organizations, local government, non-profits, and local businesses to improve community health. Elsewhere, local public health departments are using Affordable Care Act-mandated community health needs assessments as runways to longer-term collaboration among health systems and community partners.
Given this heterogeneity in opinion and practice, we are left with an open question as to what an optimally designed multi-sector approach to health would necessarily look like. We argue that health care institutions and other stakeholders in a given community can and should intentionally discuss these, and perhaps other, models to be clear about the roles each will play in improving health. All organizations need not necessarily address social determinants of health directly, but each community should be clear about whose responsibility that piece of population health is.
Health care organizations that assume the role of a hub generally assume more responsibility for social determinants than do those that see themselves as spoke, although all are likely to have some role in this work. It remains to be seen how government will elect to support health care organizations in hub-and-spoke roles differently. Certainly, if health care organizations elect to provide social determinants of health programming, that work needs to be funded above and beyond the costs of delivering medical care. However, in the event that health care organizations choose to act as spokes, state governments in particular may wish to redirect funds from health care to social service budget lines in pursuit of population health.
Rigorous evaluation of emerging models is essential. As communities across the country develop their own population health coalitions, researchers can and should be called upon to evaluate the efficacy of a range of governance models in real time. Finally, we echo the call of other scholars who have noted the need for greater patience in evaluating the impact of efforts to improve the health of communities. No responsible policymaker wants to waste valuable time and resources pursuing programs that are ineffective — but health services researchers should be careful about importing their expectations of bio-medical interventions into the realm of organizational and social change.