Purchasing Population Health - Revisited.

More than 20 years ago, a US community “Healthopolis” was envisioned for 2020 based on a “Purchasing Population Health”[1] strategy that created appropriate financial incentives for improved population health outcomes. The proposed action steps to implement the strategy included:

1. Debate, accept, and research a focus on value issues of outcomes and cost.

2. Create outcome-based payment for integrated health delivery systems.

3. Incorporate the nonmedical determinants and sectors.

The populations in Medicare, Medicaid, and the states were suggested as places to start implementation. For this discussion, population health is defined as Kindig and Stoddart's “the health outcomes of a group of individuals, including the distribution of such outcomes within the group.”

Although much has happened since the Purchasing Population Health strategy was published, when The Commonwealth Fund released its 2017 Mirror, Mirror on the Wall comparison statistics, the United States fared poorly in outcome and process comparisons (eg, life expectancy, equity, efficiency) to other developed countries. Recent reports of declining life expectancy have added fuel to the concern. Is there any reason for optimism that we will see incentives and resources aligned for communities to have better population health outcomes? We believe there should be optimism – guarded, yes, but there are encouraging signs.

First, the need for greater value in our health care system is not debated when governments see rising health care costs eroding investments in education, job development, wage growth, and housing, among others — social determinants of health that, arguably, are more important to the health of populations than medical care. The unworkable Sustainable Growth Rate for controlling medical costs in Medicare was repealed and replaced with the bipartisan supported Medicare Access and CHIP Reauthorization Act of 2015. Of note, the legislation creates a framework for a shift from volume-based fee-for-service models to value-based alternative payment models with a goal from the Health Care Payment and Learning Action Network (LAN) of 50% of Medicare payments in these value-based models by the end of 2018. Medicare's direct move to value-based payment signals a clear direction to the market, which is not likely to be undone given its bipartisan support. However, the current attacks on the separate Affordable Care Act and the slowing or undoing of some initiatives (eg, cost accountability in the Merit-based Incentive Payments System, mandatory bundled payment models) can muddy the waters.

Second, current health care data frameworks support the inclusion of population health outcomes in payment models. The commonly accepted Triple Aim proposes a 3-part aim to simultaneously address the health of populations, the experience of care, and per capita costs. Reports such as the Institute of Medicine's Vital Signs call for a parsimonious set of measures in health and health care with an emphasis on outcomes over process measures. The “Measures by Purpose Area” in the Performance Measures white paper from the Population-based Payment (PBP) Workgroup of the Health Care Payment LAN includes health outcomes such as life expectancy at birth, quality of life, healthy social circumstances, and healthy behaviors.[2] Although incorporating global measures of quality of care or better health (similar to total cost of care) in payment is challenging, “big dot” measures of quality and health are recommended as part of performance incentives in PBP models. The National Quality Forum Measure Applications Partnership approved a geographic measure of smoking prevalence (MUC16-69) for more development – the first global outcome measure at a geographic level for use in payment models.[3] Although not focused on payment, Kottke et al[4] outline how summary measures of population health and well-being can be used by health plans and accountable care organizations.

The third step in the proposed Purchasing Population Health strategy requires much stronger multi-sectoral partnerships and resources to incorporate the nonmedical determinants and sectors. To aid state and local stakeholders with this step, County-based Rankings and Roadmaps provide measures of health and health care as well as upstream factors to guide population health initiatives. Nationally, the Robert Wood Johnson Foundation's Building a Culture of Health highlights the need for multiple sectors to embrace health and equity and includes 10 underlying principles that cross medical and nonmedical determinants. Kottke et al postulate that reframing the proposition as “well-being in all policies” might increase the willingness of non–health care sectors to engage in the effort.[5] Recent innovation models include the Center for Medicare and Medicaid Innovation's State Innovation Models and accountable communities of health (ACHs) incorporating the social determinants of health and other sectors. For example, Washington State has 9 ACHs designed to convene stakeholders from multiple health sectors to explore coordination and integration with public health, community, and social services. Although bridging between health care and the nonmedical determinants of health is early, evidence is emerging about how addressing social determinants such as food insecurity[6] is associated with decreased health care costs, which aligns the incentives for investment. In addition, all-payer global hospital budgeting systems are being expanded, which should incent upstream collaborations and investments.[7]

In summary, important steps from the Purchasing Population Health strategy are occurring, and we suggest these additional actions. First, develop pilot population health payment models to achieve accountability for health care, health, and financial outcomes, including appropriate fiscal agencies. Although providers are not solely accountable for population health measures, they can play a pivotal role in a joint accountability model. Attributed population health (ie, population medicine) and geographic (ie, total population) measures as well as measures for successful multi-sector collaborations to incorporate the social determinants of health should be developed. The models should seek to test risk-adjusted payments based on the social determinants of health such that providers caring for disadvantaged populations have adequate resources to decrease inequities.[8]

Second, establish learning collaboratives to research and test the needed processes and roles in the medical and nonmedical sectors to create the desired health outcomes. Siege and colleagues from ReThink Health have documented that multi-sector partnerships for regional transformation need more development and support for favorable stewardship, strategy, and financing, noting that health care payers and economic development organizations need to be more engaged.[9]

Third, for a purchasing population health strategy to be effective, long-term, different community investment strategies are key. Developing and implementing community benchmarks in medical and nonmedical determinants can guide investment decisions to improve outcomes and equity.[10]

Finally, there must be no significant slowing of the transition to value-based health care purchasing strategies. The signals to providers and communities must remain strong to move from volume- to value-based models that include outcomes for attributed and geographic populations.

Although the challenge of purchasing population health was daunting 20 years ago, there are developments to make the creation of Healthopolis easier. The fundamental assertion in 1997 “that population health improvement will not be achieved until appropriate financial incentives are designed for this outcome” is as true today as it was then. Given the changes in federal administrations with the shift to more local control, the imperative for experiments at the regional and state levels cannot be greater to create better health, experience, and stewardship of resources. The time is now for alignment of incentives, payment models, measures, and to support community leaders, partners, and infrastructure to make purchasing population health a reality.