By Daniel M. Fox, PhD and David A. Kindig, MD, PhD
The Accountable Care Organizations (ACOs) described in the draft rules for the Medicare Shared Savings Program, published in the Federal Register on April 7, 2011, have a “three part aim”: achieving “better care for individuals…better health for populations…and lower growth in expenditures.” For the first time a federal program that finances health care has set improving population health as a prominent goal.
There is, however, a contradiction in how the draft rules address population health. The rules recognize that health has multiple determinants, but they make stronger demands for engaging in clinical prevention than for addressing these multiple determinants. This post gives examples of this contradiction and suggests ways to address it. We invite readers to comment and add their suggestions; and to note that the Centers for Medicare and Medicaid Services has established a deadline of June 6th for submitting comments.
Despite these constraints, the draft rules do ask ACOs to address broad determinants of population health. They must, for example, educate Medicare beneficiaries “about the upstream causes of ill health like poor nutrition, physical inactivity, substance abuse [and] economic disparities.” They must establish “community health teams” that “implement interdisciplinary…care plans that integrate clinical and community preventive and health promotion services for patients, including children” (19535).
However, the draft provisions relating to population health lack teeth. ACOs are required to “promote evidence-based medicine” in health care, but not to encourage the use of evidence-based health research in services to populations (19546). They are urged to consult state and local health departments about evaluating the “population health needs” of their patients in the context of “community health needs” but not directed to other, often richer, sources of data about population health. They must “promote patient engagement” and offer care that “incorporates the values of…transparency, individualization, recognition, respect, dignity and choice (19547) but they are not required to seek patient input on community health improvement. ACOs must “partner with community stakeholders,” but they would “be deemed to have satisfied this requirement” if they merely “have a stakeholder organization serving on their governing body” (19551).
Nor do the measures the rules propose for assessing what ACOs do and how they report about “better health for populations” (19578-19591) adequately address broad health determinants. Of 39 proposed measures, 29 assess process, mainly in clinical preventive services. Only 6 of them measure patient behaviors that are risk factors for chronic disease. A single measure—of preventing falls-- acknowledges the effects of patients’ physical environment on their health.
The final rules should set a reasonably high bar with respect to population health improvement without excessive fiscal or administrative burden, such as requiring ACOs to:
- Acquire and apply data about local population health status from an array of sources such as the County Health Rankings
- Negotiate memoranda of understanding about data sharing, referrals and coordination with community organizations that provide services that address other determinants of health than care
- Convene groups of patients and their caregivers to discuss the determinants of health in their communities and how to address them
- Educate collaborating organizations about evidence-based interventions that address multiple determinants of health and
- Provide incentives to devote a part of shared savings to local population health improvement activities
We applaud these draft provisions and recognize the health care system’s critical role in population health improvement -- but stronger provisions are needed. The eventual real-world impact of these rules will largely depend on the extent to which they require action across all three pillars of the “three part aim” (which is also known as the Triple Aim model articulated by the Institute for Health Improvement).
Daniel M. Fox, PhD is President Emeritus of the Milbank Memorial Fund. David A. Kindig, MD, PhD, is Emeritus Professor of Population Health Sciences and Emeritus Vice-Chancellor for Health Sciences at the University of Wisconsin School of Medicine and Public Health.
*****
For further reading:
I'd agree with the merit of most of the contentions Dan and David raise, but I think that many communities can't even get started. In my reading of the regulations, the anti-trust provisions have effectively wiped out the possibility of ACOs working across whole communities except for rural and single-hospital areas. The illusion of competition being important is striking down efforts to integrate across time and providers. I call it an illusion because I do not see the price competition for care plans for Alzheimers, strokes, and frailty generally.
I am not sure I am right, having no experience yet with anti-trust, but I do think that one strategy might be to engineer a more reasonable assessment of the merits and nature of competition as a price control. In the current rules, what matters is the combined market force of hospitals in each of a score of service lines. However, the very patients who most need integrated care are also people for whom the government already picks up most of their costs. Thus, we already have other ways to set prices for the elderly and frail and probably also for maternal-infant care. Medicare (and the VA) covers more than 80% of us as we come to the end of life, and now more than half of all babies are born in Medicaid (and another sizable group in DoD). One might argue for the FTC and DOJ to split off some populations in the Bridges to Health population segmentation model (http://www.milbank.org/quarterly/8502feat.html) and acknowledge that a community-wide approach might not run afoul of the antitrust laws because we have other governmental ways to set prices.
Some such approach is important because the best work available on improving care across settings relies upon community cooperations in setting standards, building information exchange, and learning to support self-care as a public health endeavor.
Posted by: Joanne Lynn | 05/01/2011 at 10:38 PM
The development of ACOs could be the occasion of important community organization activity especially in regard to the health of individuals within a given geographical/political jurisdiction(s). Promoting public health by identifying and modifying the particular determinants of good or bad outcomes would "take a village"! Multiple agencies each with its own agenda and funding streams as well as overlapping general purpose governmental structures each with many special purpose agencies leaves the health of the population as a whole as no one's responsibility. Since ACO development proposes a new "structure" with many interested parties (would you believe some of whom may not have "the common good" or "common goods" like the health of the population at the center of their concern) it might be the time to call to the attention of the community that these new entities both in and of themselves but as potentiating essential building blocks in community development and well being. The health of the public and its determinants should be all agencies' agendas. However what is needed is somebody(ies)to bring the community together.
Perhaps this could be the stuff of a Pulitzer Prize were a local newspaper to take this on as a cause.
Posted by: Charles Fahey | 05/03/2011 at 09:34 AM