Opportunity Knocks: Population Health in State Innovation Models

John Auerbach, Debbie I. Chang, James A. Hester, Sanne Magnan*

August 21, 2013

*Participants in the activities of the IOM Roundtable on Population Health Improvement

The views expressed in this discussion paper are those of the authors and not

necessarily of the authors’ organizations or of the Institute of Medicine. The paper is

intended to help inform and stimulate discussion. It has not been subjected to the

review procedures of the Institute of Medicine and is not a report of the Institute of

Medicine or of the National Research Council.

Copyright 2013 by the National Academy of Sciences. All rights reserved.

1

Opportunity Knocks:

Population Health in State Innovation Models

John Auerbach, M.B.A., Northeastern University; Debbie I. Chang, M.P.H., Nemours; James A.

Hester, Ph.D., M.S.; Sanne Magnan, M.D., Ph.D., Institute for Clinical Systems Improvement1

These are historic times for health care and health. The Affordable Care Act (ACA) has

unleashed novel initiatives such as the Health Care Innovation Awards and State Innovation

Models (SIMs), and we applaud Congress and the Center for Medicare and Medicaid Innovation

(CMMI) for their foresight in creating such learning opportunities. Our country’s National

Quality Strategy focuses our efforts on the Triple Aim2 of improving population health,

improving the experience of care and lowering per capita costs of care. And this is matched with

growing on-the-ground experience striving to achieve the laudable goals of the Triple Aim,

spurred in part by CMMI’s funding in this area.

Nevertheless, our current health care payment system rewards medical care for

individuals, neglecting rewards for changing the factors that make people healthy, e.g., the places

outside the doctor’s office where people live, learn, play, and work. One clear need is to develop

models that reward making the population healthy. We believe that the SIMs developed by states

provide a unique opportunity to test new alignments, payments, and incentives that focus our

current delivery system on achieving health for all. Unless we start now to develop such tools

and models and accelerate their use, an orientation toward population health will always be

underrepresented and underresourced. Furthermore, we are unlikely to achieve the goal of health

care reform until we address the underlying drivers of increased prevalence of chronic disease

such as tobacco use and obesity.

The major models currently being tested are focused primarily on the aims of controlling

total costs of care delivery and improving the patient experience and do not significantly reward

improvements in population health. They include measures of population health that focus on

clinical preventive services but do not track “upstream” or higher-level determinants of health,

such as school days missed, patient-reported health statuses, or health outcomes for a community

as defined by a geographic region. Although clinical care contributes to population health, we

have learned that other factors, such as healthy behaviors and the local built environment, are

much more important. Another issue with the current payment models is that the time horizon for

improvements is determined by the annual cycle of changes in medical spending, which

precludes interventions with longer-term impacts and a wider range of benefits, such as

interventions in the realms of employment and education. The unfortunate reality is that we have

a relatively poor understanding of how to pay for population health in a sustainable way.

This creates the opportunity, no, the imperative, that the states receiving Centers for

Medicare & Medicaid Services (CMS) funding to test and implement SIMs dedicate a portion of

their resources to pilots and experiments that are focused on the third aim of improving

population health. With input from and ownership by the community and providers, we believe

these pilots should be structured with goals and actions at the community level and integrate

1 Participants in the activities of the IOM Roundtable on Population Health Improvement.

2 A concept developed by the Institute for Healthcare Improvement in 2006 (see Berwick et al., 2008) and forming

the basis for the Three-Part Aim described in CMS and other Department of Health and Human Services programs.

2

clinical services, public health programs, and community-based initiatives targeting the upstream

determinants of health. They should include the implementation of a core set of metrics for

tracking changes in population health for both program improvement and accountability (see, for

example, the recommendations in the 2013 IOM report Toward Quality Measures for Population

Health and the Leading Health Indicators). They should also include aligned payment models

for key stakeholders that reward and incentivize demonstrated improvements in the health of the

community.

Where could these models start? An optimal approach would involve a portfolio of

measures paired with financial incentives that are balanced in the following dimensions:

substantively balanced to meet the prioritized needs of the community;

designed to capture and link both clinical and community-wide measures for process

and outcome; and

intended to produce both short- and long-term impacts.

For example, a balanced portfolio might include both practice- and community-wide

measures and intentionally seek ones with relatively quick positive and measurable health

benefits and/or cost-saving outcomes, such as effective prevention interventions (e.g., influenza

vaccinations, alcohol screening/brief counseling), asthma intervention measures (which decrease

emergency room visits and hospitalizations), and behaviors responsive to city- or state-wide

interventions (e.g., tobacco use levels). Mental health measures could be included (e.g., Patient

Health Questionaire-9 for depression, which can be used for screening and follow-up).

Alternatively, there might be complementary metrics for which significant benefits may be seen

over a longer period of time, such as the prevalence of risk factors (e.g., obesity) and illness (e.g.,

diabetes, HIV), and/or summary measures of population health (e.g., Centers for Disease Control

and Prevention healthy days or health-adjusted life-years [see, for example, IOM, 2011]).

In addition, a balanced approach might incentivize proven or promising process measure

goals best achieved by large provider systems such as accountable care organizations. Examples

of process measures that could be constructed to focus on population health outcomes include (1)

evidence of meaningful partnerships between health care organizations and state and local public

health departments, (2) systematic use of community-based health workers in underserved

communities and among racial and ethnic minority populations to assist in care transitions and

reduce environmental risk factors, and (3) active participation in community-wide efforts to

improve conditions affecting health.

We encourage states and CMMI to take advantage of this historic opportunity to fully

realize the Triple Aim. Both should push the limits of innovation, recognizing that population

health initiatives are unlikely to achieve the understandably aggressive cost-saving goals being

pursued in health care over a 6-month, 12-month, or even 3-year period. Intentionality is the

operative word for states and CMMI. States will need to be intentional about including

population health community partners and agencies and focusing their grants in this area, even

though the perceived financial payoff seems uncertain. The complexities of improving the health

care system alone often result in population health being treated as a last-step add-on or being

addressed only superficially or not at all. Similarly, CMMI will need to be intentional and

realistic about population health outcomes and the costs-saving goals that can be achieved.

CMMI will need to consider different criteria for success in evaluating interventions that include

population health, including allowing for longer time frames for achieving results. Finally, let us

3

not let the risk of failure and the difficulty of achieving cost savings and improved outcomes in

the short term discourage states and CMMI from taking advantage of this opportunity to focus on

population health.

We believe that there is no better time for innovation in population health than now.

CMMI opportunities, supplemented by complementary initiatives supported by other sponsors,

have the potential to accelerate the growth of innovative approaches. Harnessing the lessons

learned through state collaborative networks offers the potential for further acceleration and

advancement in the field, but first we need to build strong proposals, together with continued

support and flexibility from CMMI. Current investment by states and CMMI in this area have the

potential to reap even greater rewards in the future as CMMI focuses on dissemination and

spread.

We urge states and CMMI to open the door and step through quickly as opportunity knocks

for population health.

Thoughts on Potential Next Steps Toward a Balanced Portfolio of Measures in the SIMs and Beyond

Although the specific parameters of the ideal balanced portfolio are not currently known, the SIMs and Innovation Awards offer opportunities to contribute to the knowledge base. However, even though these CMS initiatives are an important resource, they will not be sufficient by themselves to develop the tools and models needed to pay for improvements in population health, as many of the “population health” initiatives being proposed are focused targeted high utilizers (super-utilizers) to achieve medical cost savings in the short term as CMS requires. Commercial payers and large self-insured employers have greater flexibility to test models and have advanced payment reform in the past through the development of innovations such as the Alternative Quality Contract designed by Blue Cross of Massachusetts. Private foundations also have a role, as exemplified by the social impact bond for asthma that was sponsored by The California Endowment (Social Finance, 2013) and the exploration of the role of Community Development Financial Institutions sponsored jointly by the Federal Reserve, the Kresge Foundation, and the Robert Wood Johnson Foundation (see, for example, Erickson, 2013). Ideally, potential sponsors will look to the SIM states as fertile ground for testing and spreading new approaches to creating a sustainable balanced portfolio for improving population health. Ultimately, public- and private-sector collaboratives deploying innovative approaches with a focus on shared learning and harnessing and spreading what works would help to move the field even further.

Suggested citation: Auerbach, J., D. I. Chang, J. Hester, and S. Magnan. 2013. Opportunity knocks: Population

health in State Innovation Models. Discussion Paper, Institute of Medicine, Washington, DC.

http://nam.edu/wp-content/uploads/2015/06/OpportunityKnocks.

REFERENCES

Erickson, D. 2013. Linking community development and health. Testimony Prepared for the Robert Wood Johnson

Foundation Commission to Build a Healthier America, June 2013. Available at http://www.rwjf.org/

content/dam/farm/reports/reports/2013/rwjf406408 (accessed August 16, 2013).

Berwick, D.M., T.W. Nolan, and J. Whittington. 2008. The triple aim: care, health, and cost. Health Aff (Millwood)

27(3):759-69.

IOM (Institute of Medicine). 2013. Toward quality measures for population health and the Leading Health

Indicators. Washington, DC: The National Academies Press.

IOM. 2011. For the public’s health: The role of measurement in action and accountability. Washington, DC: The

National Academies Press.

4

Social Finance. 2013, March 25. The California Endowment awards grant to social finance and collective health.

News release, March 25. Available at http://payforsuccess.org/sites/default/files/fresno_asthma_

demonstration_project_press_release.pdf (accessed August 16, 2013).

American Journal of Public Health | March 2003, Vol 93, No. 3380 | Models for Population Health | Peer Reviewed | Kindig and Stoddart

MODELS FOR POPULATION HEALTH

What Is Population Health?| David Kindig, MD, PhD, and Greg Stoddart, PhDPopulation health is a rela-

tively new term that has not yetbeen precisely defined. Is it aconcept of health or a field ofstudy of health determinants?

We propose that the defini-tion be “the health outcomesof a group of individuals, in-cluding the distribution of suchoutcomes within the group,”and we argue that the field ofpopulation health includeshealth outcomes, patterns ofhealth determinants, and poli-cies and interventions that linkthese two.

We present a rationale forthis definition and note its dif-ferentiation from public health,health promotion, and socialepidemiology. We invite cri-tiques and discussion thatmay lead to some consensuson this emerging concept. (AmJ Public Health. 2003;93:380–383)

ALTHOUGH THE TERM“population health” has beenmuch more commonly used inCanada than in the United States,a precise definition has not beenagreed upon even in Canada,where the concept it denotes hasgained some prominence. Proba-bly the most influential contribu-tion to the development of thepopulation health approach isEvans, Barer, and Marmor’s WhyAre Some People Healthy and Oth-ers Not? The Determinants ofHealth of Populations,1 whichgrew out of the work of the Pop-ulation Health Program of theCanadian Institute for AdvancedResearch. No concise definitionof the term appears in this vol-ume, although its authors statethe concept’s “linking thread [tobe] the common focus on tryingto understand the determinantsof health of populations.”1(p29)

The idea that populationhealth is a field of study or a re-search approach focused on de-terminants seems to haveevolved from this work. Earlydiscussions at the Canadian Insti-tute for Advanced Research alsoconsidered the definition andmeasurement of health and theprocesses of health policymaking,but the dominant emphasisevolved to the determinantsthemselves, particularly the non-medical determinants. John Frank,the scientific director of the re-cently created Canadian Instituteof Population and Public Health,has similarly called populationhealth “a newer research strategyfor understanding the health ofpopulations.”2 T.K. Young’s re-cent book Population Health alsotends in this direction; he states

that in Canada and the UnitedKingdom in the 1990s, the termhas taken on the connotation ofa “conceptual framework forthinking about why some popula-tions are healthier than others aswell as the policy development,research agenda, and resourceallocation that flow from thisframework.”3(p4)

However, Young also indicatesthat in the past, the term hasbeen used as a “less cumbersomesubstitute for the health of popu-lations,” which is of course its lit-eral meaning. Evans and Stod-dart, while supporting anemphasis on “understanding ofthe determinants of populationhealth,” have also stated, how-ever, that “different concepts [ofhealth] are neither right orwrong, they simply have differentpurposes and applications. . . .[W]hatever the level of definitionof health being employed, how-ever, it is important to distinguishthis from the question of the de-terminants of that definition ofhealth.”1(p28) The Health Promo-tion and Programs Branch ofHealth Canada has recentlystated that “the overall goal of apopulation health approach is tomaintain and improve the healthof the entire population and toreduce inequalities in health be-tween population groups.”4(p1)

They indicate that one guidingprinciple of a population healthapproach is “an increased focuson health outcomes (as opposedto inputs, processes, and prod-ucts) and on determining the de-gree of change that can actuallybe attributed to our work.”(p11)

Dunn and Hayes, quoting thedefinition of the Canadian Fed-

eral/Provincial/Territorial Advi-sory Committee on PopulationHealth, write that “populationhealth refers to the health of apopulation as measured byhealth status indicators and asinfluenced by social, economic,and physical environments, per-sonal health practices, individualcapacity and coping skills,human biology, early childhooddevelopment, and health ser-vices. As an approach, popula-tion health focuses on interre-lated conditions and factors thatinfluence the health of popula-tions over the life course, identi-fies systematic variations in theirpatterns of occurrence, and ap-plies the resulting knowledge todevelop and implement policiesand actions to improve thehealth and well being of thosepopulations.”5(p57) Kindig hassuggested a similarly broad defi-nition: population health is “theaggregate health outcome ofhealth adjusted life expectancy(quantity and quality) of a groupof individuals, in an economicframework that balances therelative marginal returns fromthe multiple determinants ofhealth.”6(p47) This definition pro-poses a specific unit of measureof population health and also in-cludes consideration of the rela-tive cost-effectiveness of re-source allocation to multipledeterminants.

Recently, even in the UnitedStates, the term is being morewidely used, but often withoutclarification of its meaning anddefinition. While this develop-ment might be seen as a usefulmovement in a new and positivedirection, increased use without

March 2003, Vol 93, No. 3 | American Journal of Public Health Kindig and Stoddart | Peer Reviewed | Models for Population Health | 381

MODELS FOR POPULATION HEALTH

precision of meaning couldthreaten to render the term moreconfusing than helpful, as may al-ready be the case with “commu-nity health” or “quality of med-ical care.” For this reason, wepropose a definition that mayhave a more precise meaning forpolicymakers and academicsalike; our purpose is to stimulateactive critiques and debate thatmay lead to further clarificationand uniformity of use.

DEFINITION ANDCONCEPT

As indicated above, the pri-mary tension or confusion atpresent seems to be betweendefining population health as afield of study of health determi-nants or as a concept of health.The Group Health CommunityFoundation has recently statedthat “some observers see popula-tion health as a new term thathighlights the influential role ofsocial and economic forces incombination with biological andenvironmental factors, that shapethe health of entire populations. . . others interpret populationhealth primarily as a goal—a goalof achieving measurable im-provements in the health of a de-fined population.”7(p7)

We think that there are 3 gen-eral possibilities: populationhealth (a) is only concerned withthe independent variables (themultiple determinants), (b) isonly concerned with the depen-dent variables (health outcomes),or (c) is concerned with both thedefinition and measurement ofhealth outcomes and the roles ofdeterminants. While none of thethree is normatively correct orincorrect, we believe that the lat-ter is more appropriate, primarilybecause the concept and mea-surement of health and health

outcomes focuses attention andresearch effort on the impact ofeach determinant and their inter-actions on some appropriate out-come. It also allows one to con-sider health inequality andinequity and the distribution ofhealth across subpopulations, aswell as the ethical and value con-siderations underpinning theseissues.8

While the original Evans andStoddart “field model” did notdiscuss a population health con-cept in these terms, the idea isimplicit in the evolution of thedependent variable from “healthcare” to “health and function” to“well being.”1(pp33–53) The Insti-tute of Medicine has given seri-ous attention to measuringpopulation health, thereby en-couraging some kind of sum-mary measure that includes mor-tality and health-related qualityof life.9

Given these considerations, wepropose that population health asa concept of health be defined as“the health outcomes of a groupof individuals, including the dis-tribution of such outcomeswithin the group.” These popula-tions are often geographic re-gions, such as nations or commu-nities, but they can also be othergroups, such as employees, eth-nic groups, disabled persons, orprisoners. Such populations areof relevance to policymakers. Inaddition, many determinants ofhealth, such as medical care sys-tems, the social environment, andthe physical environment, havetheir biological impact on indi-viduals in part at a populationlevel.

Defining population health thisway requires some measure(s) ofhealth outcomes of populations,including their distributionthroughout the population. Wechose the broader term “health

outcomes” rather than the morenarrow term “health status”; webelieve the latter refers to healthat a point in time rather thanover a period of years. We donot believe that there is any onedefinitive measure, but we arguethat the development and valida-tion of such measures for differ-ent purposes is a critical task forthe field of population healthresearch.

Our definition does imply thenecessity of one or more broadsummary measures capable ofbeing a dependent variable forthe spectrum of all determinants(generally including length of lifeand health-related quality andfunction of those life years),along with a family of other sub-measures for different policy andresearch purposes. For example,the Health Utilities Index is beingused in the Canadian NationalPopulation Health Survey,10

Years of Healthy Life have beenused in Healthy People 2000,11

and the EuroQuol has been re-cently added to the Medical Ex-penditure Panel Survey.12

We support the idea that ahallmark of the field of populationhealth is significant attention tothe multiple determinants ofsuch health outcomes, howevermeasured. These determinantsinclude medical care, publichealth interventions, aspects ofthe social environment (income,education, employment, socialsupport, culture) and of the phys-ical environment (urban design,clean air and water), genetics,and individual behavior. We notewith caution that such a list ofcategories can lead to a view thatthey operate independently; pop-ulation health research is funda-mentally concerned about the in-teractions between them, and weprefer to refer to “patterns” ofdeterminants.

Population health researcherstend to use a set of methods andapproaches that have the follow-ing important characteristics: ex-amination of systematic differ-ences in outcomes acrosspopulations, complexity of inter-actions among determinants, bio-logical pathways linking determi-nants to population healthoutcomes, and the influence ofdifferent determinants over timeand throughout the life cycle.13–15

In our view, a populationhealth perspective also requiresattention to the resource alloca-tion issues involved in linking de-terminants to outcomes. Part ofthe study of population health in-volves the estimation of thecross-sectoral cost-effectivenessof different types and combina-tions of investments for produc-ing health.16 Because improve-ment in population healthrequires the attention and actionsof multiple actors (legislators,managers, providers, and individ-uals), the field of populationhealth needs to pay careful atten-tion to the knowledge transferand academic-practice partner-ships that are required for posi-tive change to occur.17,18 Figure 1shows how we view the field ofpopulation health. The field in-vestigates each of the compo-nents shown in the figure, butparticularly their interactions.

CRITIQUES

We expect and welcome cri-tiques of the definition presentedhere. As noted above, one cri-tique will be that the tasks ofdefining and measuring conceptsof health are large enough toconstitute a subject of their own,rather than being combined withthe study of determinants ofhealth. We have already givenour rationale for including them

American Journal of Public Health | March 2003, Vol 93, No. 3382 | Models for Population Health | Peer Reviewed | Kindig and Stoddart

MODELS FOR POPULATION HEALTH

FIGURE 1—A schematic definition of the field of populationhealth.

in population health as a field ofstudy, but we would add that theneed for accountability arguesstrongly for the inclusion of out-come and distributional consider-ations if a population health ap-proach is to be useful in guidingpolicymaking regarding resourceallocation across determinantsand sectors. Without such aframework, advocacy and finan-cial incentives for individual de-terminants can proceed indepen-dently of their impact, as somewould argue is now the case forsome medical care expendituresin the United States.

A second critique is that sucha definition and concept is sobroad that it includes everythingand is therefore not useful toguide either research or policy.We understand this concern butdo not agree with it. We believethat a guiding synthesis is essen-tial for considering both the rela-tive impacts of the pattern of de-terminants and their interactions.Integration of knowledge abouthealth and its multiple determi-nants seldom occurs. Policy man-agers typically have responsibilityfor a single sector; advocacygroups typically have an interestin only one disease or determi-nant. No one in the public or pri-vate sectors currently has respon-sibility for overall health

improvement. We suggest thatthe importance of a populationhealth perspective is that it forcesreview of health outcomes in apopulation across determinants.For population health research,specific investigations into asingle determinant, outcomemeasure, or policy interventionare relevant, and may even becritical in some cases, but theymust be recognized as only apart and not the whole.

Those in public health orhealth promotion may legiti-mately feel that populationhealth is simply a renaming ofwhat has been their work or leg-acy. Hamilton and Bhatti have at-tempted to show the complemen-tarity and overlap betweenpopulation health and health pro-motion,19 building on the Cana-dian Achieving Health for AllFramework for Health Promo-tion20 and the World Health Or-ganization Ottawa Charter onHealth Promotion.21 Frank hasindicated that historic conceptsof public health were similarlybroad, until the biomedical para-digm became dominant. Thosewho define public health as the“health of the public” would notdisagree with the definition ofpopulation health proposed here;in the words of Frank, the “shiftin thinking entailed in population

health should be a small one forpublic health workers . . . in factit is not so much a shift as a re-turn to our historical roots en-compassing all the primary deter-minants of health in humanpopulations.”22(p163)

However, much of publichealth activity, in the UnitedStates at least, does not havesuch a broad mandate even inthe “assurance” functions, sincemajor determinants such as med-ical care, education, and incomeremain outside of public healthauthority and responsibility, andcurrent resources do not evenallow adequate attention to tradi-tional and emerging publichealth functions. Similarly, webelieve that the emerging promi-nence of social epidemiology is avery important development forpopulation health but does nothave the breadth, or imply all ofthe multiple interactions andpathways, of the definition pro-posed here for population health.

CONCLUSION

We believe that the time hascome for a clarification of themeaning and scope of the term“population health.” We have of-fered a clarification of the termthat combines the definition andmeasurement of health outcomesand their distribution, the pat-terns of determinants that influ-ence such outcomes, and thepolicies that influence the opti-mal balance of determinants. Wewelcome discussion and debateregarding these suggestions as away of moving toward some con-sensus on this important andemergent concept.

About the AuthorsDavid Kindig is with the Department ofPopulation Health Sciences, University ofWisconsin—Madison School of Medicine,

Madison. Greg Stoddart is with the De-partment of Clinical Epidemiology andBiostatistics, McMaster University HealthScience Centre, Hamilton, Ontario.

Requests for reprints should be sent toDavid A. Kindig, MD, PhD, Suite 760,610 Walnut St, Madison, WI 53726-2397 (e-mail: dakindig@facstaff.wisc.edu).

This article was accepted November15, 2002.

ContributorsBoth authors contributed equally to theplanning and writing of this article.

AcknowledgmentsD. Kindig was funded in part by theRobert Wood Johnson Health and Soci-ety Scholars Planning Grant to the De-partment of Population Health Sciencesat the University of Wisconsin MadisonSchool of Medicine.

We acknowledge the helpful com-ments of John Mullahy, Stephanie Rob-ert, Donn D’Alessio, John Lavis, andmany members of the Robert WoodJohnson Foundation Health and SocietyScholars Planning Group.

References1. Evans R, Barer M, Marmor T. WhyAre Some People Healthy and OthersNot? The Determinants of Health of Pop-ulations. New York, NY: Aldine deGruyter; 1994.

2. Frank J. Canada Institute of Popu-lation and Public Health. Message from:Dr John Frank, scientific director, Insti-tute of Population and Public Health.Available at: http://www.cihr-irsc.gc.ca/institutes/ipph/about_ipph/ipph_message_director_e.shtml. Accessed De-cember 23, 2002.

3. Young TK. Population Health: Con-cepts and Methods. New York, NY: Ox-ford University Press; 1998.

4. Health Canada. Taking Action onPopulation Health. Ottawa, Ontario:Health Canada; 1998.

5. Dunn JR, Hayes MV. Toward a lex-icon of population health. Can J PublicHealth. 1999;90(suppl 1):S7–S10.

6. Kindig DA. Purchasing PopulationHealth: Paying for Results. Ann Arbor:University of Michigan Press; 1997.

7. Kreuter M, Lezin N. ImprovingEveryone’s Quality of Life: A Primer onPopulation Health. Atlanta, Ga: GroupHealth Community Foundation; 2001.

8. Williams A. Conceptual and empiri-cal issues in the efficiency-equity trade-off in the provision of health care or, ifwe are going to get a fair innings, some-one will need to keep the score! In:

March 2003, Vol 93, No. 3 | American Journal of Public Health Kickbusch | Peer Reviewed | Models for Population Health | 383

MODELS FOR POPULATION HEALTH

Maynard AJ, ed. Being Reasonable Aboutthe Economics of Health. Cheltenham, En-gland: Edward Elgar; 1997:322–349.

9. Summarizing Population Health Di-rections for the Development and Applica-tion of Population Metrics. Washington,DC: Institute of Medicine, Division ofHealth Care Services; 1998.

10. Wall R, Foster R. Beyond life ex-pectancy. Health Policy Res Bull. 2002;1:32–33.

11. Erickson P, Wilson R, Shannon I.Years of Healthy Life. Bethesda, Md: Na-tional Center for Health Statistics; 1995.Statistics note no. 7.

12. Gold MR, Muenning P. Measure-dependent variation in burden of dis-ease estimates. Med Care. 2002;40:260–266.

13. Berkman L, Kawachi I. Social Epi-demiology. New York, NY: Oxford Uni-versity Press; 2000.

14. Keating DP, Hertzman C. Develop-mental Health and the Wealth of Nations:Social, Biological, and Educational Dy-namics. New York, NY: Guilford Press;1999.

15. Adler N, Marmot M, McEwen B,Stewart J. Socioeconomic status andhealth in industrial nations: social, psy-

chological, and biological pathways. AnnN Y Acad Sci. 1999;896.

16. Drummond M, Stoddart G. Assess-ment of health producing measuresacross different sectors. Health Policy.1995;33:219–231.

17. Lavis JN, Ross SE, Hurley JE, et al.Examining the role of health servicesresearch in public policy making. Mil-bank Q. 2002;80:125–154.

18. Lomas J. Using “linkage and ex-change” to move research into policy ata Canadian foundation. Health Aff.2000;19(3):236–240.

19. Hamilton N, Bhatti T. Population

Health Promotion: An Integrated Model ofPopulation Health and Health Promotion.Ottawa, Ontario: Health Promotion De-velopment Division; February 1996.

20. Epp J. Achieving Health for All: AFramework for Health Promotion. Ottawa,Ontario: Health and Welfare Canada;1986.

21. World Health Organization(WHO). Ottawa Charter on Health Pro-motion. Copenhagen, Denmark: WHORegional Office for Europe; 1986.

22. Frank JW. Why “populationhealth”? Can J Public Health. 1995;86:162–164.

The Contribution of the World Health Organization to a New Public Health and Health Promotion

| Ilona Kickbusch, PhDThe author traces the de-velopment of the concept ofhealth promotion from 1980spolicies of the World Health Or-ganization. Two approachesthat signify the modernizationof public health are outlined indetail: the European Health forAll targets and the settings ap-proach. Both aim to reorienthealth policy priorities from arisk factor approach to strate-gies that address the deter-minants of health and em-power people to participate inimproving the health of theircommunities.

These approaches combineclassic public health dictumswith “new” strategies, somesetting explicit goals to inte-grate public health with gen-eral welfare policy. Health forAll, health promotion, and pop-ulation health have contributedto this reorientation in thinkingand strategy, but the focus ofhealth policy remains expen-diture rather than investment.(Am J Public Health. 2003;93:383–388)

IN 1986, AT AN INTERNATIONALconference held in Ottawa, On-tario, Canada, under the leader-ship of the World Health Organi-zation (WHO) (and with a strongpersonal commitment from thenDirector General HalfdanMahler), a broad new under-standing of health promotionwas adopted. The Ottawa Char-ter for Health Promotion hassince exerted significant influ-ence—both directly and indi-rectly—on the public healthdebate, on health policy formula-tion, and on health promotionpractices in many countries.1,2

The work on this document wasspearheaded by the WHO Euro-pean Regional Office and wasdeveloped over a period of 5years of intensive research anddebate. It was based on the“Health for All” philosophy,3 theAlma Ata Declaration,4 and theLalonde health field concept.5

The Ottawa charter initiated aredefinition and repositioning of

institutions, epistemic communi-ties, and actors at the “health” endof the disease–health continuum,a perspective that had been la-beled the “salutogenic approach”by Aaron Antonovsky.6 In over-coming an individualistic under-standing of lifestyles and in high-lighting social environments andpolicy, the orientation of healthpromotion began to shift from fo-cusing on the modification of indi-vidual risk factors or risk behav-iors to addressing the “contextand meaning” of health actionsand the determinants that keeppeople healthy. The CanadianLalonde report is often cited ashaving been the starting point ofthis new development. Recentlythe director of the Pan AmericanHealth Organization, Sir GeorgeAlleyne, reflected on this issue,stating that “it is perhaps not acci-dental that the impetus for thefocus on health promotion for themany should have risen in Can-ada which is often credited with

maintaining a more egalitarian ap-proach in all health matters.”7

In its Health for All strategy,WHO positioned health at thecenter of development policy anddefined the goal of health policyas “providing all people with theopportunity to lead a socially andeconomically productive life.”3 Itproposed a revolutionary shift inperspective from input to out-comes: governments were to beheld accountable for the health oftheir populations, not just for thehealth services they provided.Lester Breslow, the father of theAlameda County study and oneof the world’s leading epidemiol-ogists, had argued in 1985 that“the stage is set for a new publichealth revolution.”8 The Ottawacharter echoed this challenge aswell as the link to public healthhistory in its subtitle, “The MoveTowards a New Public Health.”

Fourteen years later, in a com-mentary published in the Journalof the American Medical Associa-

ContributorsAll authors contributed equally to theresearch and writing of this article.

References1. International Energy Agency. Keyworld energy statistics. 2012. Availableat: http://www.iea.org/publications/freepublications/publication/kwes.pdf.Accessed August 1, 2012.

2. Colborn T, Kwiatkowski C, SchultzK, Bachran M. Natural gas operationsfrom a public health perspective. HumEcological Risk Assess. 2011;17(5):1039---1056.

3. DiGiulio DC, Wilkin RT, Miller C.Investigation of Ground Water Contami-nation near Pavillion, Wyoming. Draft.Ada, OK: Office of Research and Devel-opment, National Risk Management Re-search Laboratory; 2011.

4. Energy Policy Act of 2005, Pub. L.109-58, 119 Stat 594, § 322.1(B)(ii).(2005). Available at: http://www.gpo.gov/fdsys/pkg/PLAW-109publ58/pdf/PLAW-109publ58.pdf. Accessed July 7, 2012.

5. New York State Department ofEnvironmental Conservation. RevisedDraft Supplemental Generic EnvironmentalImpact Statement on the Oil, Gas andSolution Mining Regulatory Program: WellPermit Issuance for Horizontal Drilling andHigh-Volume Hydraulic Fracturing to De-velop the Marcellus Shale and Other Low-Permeability Gas Reservoirs. Albany, NY:New York State Department of Environ-mental Conservation; 2011: Section6.11.1.1, p. 810.

6. Howarth RW, Santoro R, IngraffeaA. Methane and the greenhouse-gas foot-print of natural gas from shale formations.Clim Change. 2011;106(4):679---690.

7. Howarth RW, Ingraffea A, EngelderT. Should fracking stop? Nature.2011;477(7364):271---275.

8. Tollefson J. Methane leaks erodegreen credentials of natural gas. Nature.2013;493(7430):12.

9. Steinzor N, Subra W, Sumi L. Gaspatch roulette: how shale gas develop-ment risks public health in Pennsylvania.Earthworks. Available at: http://www.earthworksaction.org/library/detail/gas_patch_roulette_full_report#.UOwyQe0hTKg.Accessed December 20, 2012.

10. Witter R, Stinson K, Sackett H, et al.Potential exposure related human healtheffects of oil and gas development: a litera-ture review (2003---2008). 2008. Avail-able at: http://docs.nrdc.org/health/hea_08091702.asp. Accessed August 3, 2012.

11. Bamberger M, Oswald RE. Impactsof gas drilling on human and animalhealth. New Solut. 2012;22(1):51---77.

12. Meyer JL, Frumhoff PC, HamburgSP, de la Rosa C. Above the din but in thefray: environmental scientists as effectiveadvocates. Front Ecol Environ. 2010;8(6):299---305.

13. Reed S. With controls, Britain allowshydraulic fracturing to explore for gas.New York Times. December 14, 2012:B3. Available at: http://www.nytimes.com/2012/12/14/business/energy-environment/britain-approves-fracking-for-shale-gas-exploration.html. AccessedDecember 20, 2012.

14. Centers for Disease Control andPrevention. CDC/ATSDR hydraulic frac-turing statement. May 3, 2012. Availableat: http://www.cdc.gov/media/releases/2012/s0503_hydraulic_fracturing.html.Accessed August 5, 2012.

15. Finkel ML, Law A. The rush to drill fornatural gas: a public health cautionary tale.Am J Public Health. 2011;101(5):784---785.

Achieving Population Health in Accountable Care OrganizationsAlthough “population

health” is one of the Insti-

tute for Healthcare Improve-

ment’s Triple Aim goals, its

relationship to accountable

care organizations (ACOs)

remains ill-defined and lacks

clarity as to how the clinical

delivery system intersects

with thepublichealthsystem.

Although defining popu-

lation healthas“panel”man-

agement seems to be the

default definition, we called

for a broader “community

health” definition that could

improve relationships be-

tween clinical delivery and

public health systems and

health outcomes for com-

munities.

We discussed this broader

definition and offered rec-

ommendations for linking

ACOs with the public health

system toward improving

health for patients and their

communities. (Am J Public

Health.2013;103:1163–1167.

doi:10.2105/AJPH.2013.

301254)

Karen Hacker, MD, MPH, and Deborah Klein Walker, EdD

WITH THE PASSAGE OF THE

Affordable Care Act (ACA),1 theUnited States has turned its atten-tion to improving the quality ofhealth care while simultaneouslydecreasing cost. As we move to-ward alternative and global pay-ment arrangements, the need tounderstand the epidemiology ofthe patient population will becomeimperative. Keeping this popula-tion healthy will require enhanc-ing our capacity to assess, monitor,and prioritize lifestyle risk factorsthat unduly impact individual pa-tient health outcomes. This is es-pecially true, given that only 10%of health outcomes are a result ofthe medical care system, whereasfrom 50% to 60% are becauseof health behaviors.2,3 To changehealth behaviors, it will be neces-sary to engage in activities thatreach beyond the clinical settingand incorporate community andpublic health systems.4

The Institute for HealthcareImprovement (IHI), a leading

not-for-profit organization dedi-cated to using quality improve-ment strategies to achieve safeand effective health care, has de-veloped the Triple Aim initiative5

as a rubric for health care trans-formation. The three linked goalsof the Triple Aim include improv-ing the experience of care, improv-ing the health of populations, andreducing per capita costs of healthcare.6 However, although two ofthe three aims---experience of careand cost reduction---are self-explan-atory, there is little consensus abouthow to define population health.Words like “panel management,”“population medicine,” and “pop-ulation health” are being usedinterchangeably. Berwick et al.6

describe the care of a populationof patients as the responsibilityof the health care system and usebroad-based community healthindicators as evidence of improve-ment. Other recent publicationshave attempted to describe popu-lation health from the hospital,7---10

primary care,11 and communityhealth center perspectives.12 The“clinical view” identifies the pop-ulation as those “enrolled” in thecare of a specific provider, provideror hospital system, insurer, orhealth care delivery network(i.e., panel population).7 Alter-natively, from the public healthperspective,8 population is definedby the geography of a community(i.e., community population) or themembership in a category of per-sons that share specific attributes(e.g., populations of elderly, minor-ity population). In either case, thecontext of a community and theexisting social determinants ofhealth, ranging from poverty tohousing, are known to have sub-stantial impact on individualhealth outcomes. Thus, ensuringthe health of a population is highlydependent on addressing thesesocial determinants and requirescollaborative relationships withcommunity institutions outsidethe health care setting.13,14

COMMENTARIES

July 2013, Vol 103, No. 7 | American Journal of Public Health Hacker and Walker | Peer Reviewed | Commentaries | 1163

Two key concepts that willgreatly influence the definitionand actualization of populationhealth in the post-ACA erainclude the accountable careorganization (ACO)15 and thepatient-centered medical home(PCMH).16 The ACO representsan integrated strategy at the de-livery system level to respond topayment reform.15 These inte-grated systems of care are poisedto manage a population of pa-tients under a global paymentmodel. The PCMH is focusedon transforming primary care tobetter deliver “patient-centered”care and to address the wholepatient, including their health andsocial needs.17,18 Both models willneed to identify, monitor, andmanage their “population” of pa-tients. However, their ability toextend their definition of popu-lation health to encompass theentire community will depend onresources, market share, and thestrength and capacity of collabo-rating community and public healthorganizations. As integrated deliv-ery systems are asked to do morethan focus on their own patients,they will require additional re-sources. These may come froma realignment of existing programs(community benefits), a return oninvestment from effective preven-tive care, or collaborative rela-tionships with existing communityand public health organizations.

In this article, we discuss twomajor points regarding ACOsand their approach to populationhealth. First, ACOs should becommitted to serving the healthof the people in the communitiesfrom which their population isdrawn, and not just the popula-tion of patients enrolled in theircare to achieve the populationhealth goal. Second, to achievethis expanded definition of pop-ulation health, ACOs will need to

engage in collaborative effortswith community agencies andthe public health system. We de-scribe a “community” definitionof population health to be used inlieu of the “panel” definitionand then outline the resourcesneeded and strategies for collab-oration. Finally, we offer recom-mendations to assist ACOs inrealizing their population healthgoal.

DEFINING POPULATIONHEALTH

Population health connotesa high-level assessment of a groupof people.9 This epidemiologicalframework is often in direct op-position to the manner in whichthe health care system has caredfor patients in a fee-for-servicemodel: one individual at a time.Currently, population health isbeing seen in two distinct ways: (1)from a public health perspective,populations are defined by geog-raphy of a community (e.g., city,county, regional, state, or nationallevels); and (2) from the perspec-tive of the delivery system (indi-vidual providers, groups of pro-viders, insurers, and health deliverysystems), population health con-notes a “panel” of patients servedby the organization.

In the post-ACA world, as pay-ment models shift from fee-for-service to global payment, ACOswill necessarily reorient froma disease focus to a wellness focusto improve quality and containcosts. Although they will have anethical and contractual obligationfor the patients for which theycare, their engagement in thelarger community may be highlydependent on which members ofthe community population actu-ally end up being part of a par-ticular ACO or PCMH panel. Thelarger the overlap between an

ACO panel and the communitypopulation, the more the overallhealth of the community willcontribute to the ACOs’ ability tokeep their patients healthy. Sim-ilarly, the larger the overlapbetween community populationand ACO panel, the more ACOhealth outcomes will drive com-munity health indicators. Table 1displays how an ACO might ad-dress a variety of characteristics,depending on the chosen definitionof population health (none, panel ofpatients in the delivery system, allmembers of a community).

Resources

As provider organizations areasked to embrace the broadercommunity definition of popula-tion health, resources will beneeded to support this role. Theseresources include access to data,funding, and collaborativerelationships.Data. With the emergence of

the electronic medical records,ACOs should become more facileat viewing their population asa whole and identifying trendsacross their panel’s health (age,gender, race, chronic conditions).The data needed for this endeavorare largely collected at the visitlevel by registration and clinicalstaff. With adequate health infor-mation technology, systems cannow examine issues such as riskfor future disease, comorbidities,and quality metrics across a de-fined population. Using these data,the ACO can also determine thezip codes and communities wherea majority of their patients resideand compare their health indica-tors to the community health in-dicators for the same geography.

Data on community healthindicators (e.g., preventive ser-vices use, infectious disease rates,lead paint exposure, occupationalhealth issues, cancer rates, births,

and deaths from vital statistics) aremore accessible than ever before.The National Prevention Strategy19

and the Healthy People 2020 goalsfor the nation20 include healthindicators for population healthat the community level. Much ofcommunity health informationresides with state and county or cityhealth departments, some of whichhave online interactive data toolsthat are available to the public(MassCHIP-Massachusetts21). Newtools, such as the County HealthRankings22 and the CommunityHealth Status Indicators,23 arepublicly available and allow usersto obtain county-level health data.In some jurisdictions, providerorganizations are identifyingways to share de-identified datawith community health leadersto jointly identify priority pre-vention strategies.24

Funding. ACOs will also needto identify financial resources toachieve population health goals.The current fee-for-service struc-ture does not support populationhealth efforts, and although dem-onstration grants may help, theycannot sustain ongoing work.Today, nonprofit hospitals are re-quired to provide some supportfor community programs throughthe recently revised communitybenefit in the ACA.25 Realigninghospital community benefit pro-grams with population healthefforts can help support theexpanded role.

Simultaneously, ACOs need toassess which preventive strategieswill yield the best return on in-vestment (ROI) for their patients.Evidence-based services thatdemonstrate ROI and improvedhealth outcomes can help in thisendeavor. Nationally, two sets ofevidenced-based prevention ser-vices have been identified: clinicalpreventive services, such as mam-mography, immunizations, and

COMMENTARIES

1164 | Commentaries | Peer Reviewed | Hacker and Walker American Journal of Public Health | July 2013, Vol 103, No. 7

smoking cessation26; and commu-nity preventive services, such asfluoridation, lead testing, andcommunity screening.27 Many ofthe clinical preventive measuresare considered quality measuresby major accrediting systems(e.g., Healthcare EffectivenessData and Information Set or theNational Committee for QualityAssurance) and are also includedin health coverage under the ACA.Assuming an ROI is realized,dollars saved can shift to supportcommunity and public healthinitiatives. Additionally, the fed-eral public health trust fund pro-vides a new revenue stream tosupport prevention strategies di-rectly tied to health improvementand cost containment.28 Thiswas recently replicated in Mas-sachusetts with the passage ofChapter 224.29

Collaboration. Many of theseevidence-based prevention prac-tices fall within the purview ofcommunity agencies and thepublic health system outside ofACO responsibility. For example,smoking bans promulgated bypublic health authorities have af-fected smoking rates and second-hand smoke exposure and haveled to lower risk of hospitalizationfor cardiac and pulmonary condi-tions.30 Therefore, ACOs thatstrive to improve population healthwithin geography will need todevelop partnerships to supportprevention activities while integrat-ing complementary efforts intoclinical settings. In particular, theACO’s relationship with the localpublic health authority or authori-ties is essential. Although the publichealth authority is not the onlyorganization with which an ACOwill need to collaborate, it is theonly agency that has legal au-thority and mandates to protect,promote, and assure the healthfor every individual in the

TABLE

1—Characteristics

ofVariousApproachesto

PopulationHealthin

AccountableCareOrganizations

Approach

Focuson

IndividualPatientsinPrimaryCareSettings

PanelPopulation=PopulationHealth

CommunityaPopulation=PopulationHealth

Medicalhome

Mayormaynothavemedicalhome

Medicalhomeimplemented

Medicalhomeimplemented

Carecoordination

Focuseson

coordinationwithinprimarycaresetting

Focuseson

coordinationwithindeliverysystem

andpotentially

somecommunity

resources

Focuseson

coordinationwithindeliverysystem

andall

community

resources

Clinicalpreventionservices

Implementallclinicalpreventionservicesinprimarycare

Implementallclinicalpreventionservicesinprimarycare

Implementallclinicalpreventionservicesinprimarycare

Community

preventionservices

Noimplementationofcommunity

preventionservices

Limitedimplementationofcommunity

preventionservices

Fullimplementationofcommunity

preventionservices

Healthindicatorsmonitored

Measuresforprovidersettings,butno

alignmentwith

deliveryorcommunity

orpublichealthsystems

Measuresforpatientsinthedeliverysystem,butno

alignmentwithcommunity

orpublichealthsystems

Measuresfordeliverysystem

includemeasuresatthe

community

populationlevel

Needsassessment

Noattentiontocommunity

needsassessment—focus

onlyon

primarycaresettings

Mayhavesomejointneedsassessmentbutfocuseson

decisionswithinthedeliverysystem

Jointneedsassessmentrelatedtocommunity

population

outcomesandjointselectionoftargetareasforaction

Relationshiptopublichealthsystem

Norelationship

Coordinatingstructuremayexistwithpublichealth

Governance

andcoordinatingstructuresinplacewithpublic

healthagenciestoimprovecommunity

populationhealth

Relationshiptocommunity

agencies

Norelationship

Coordinatingstructuremayexistwithsomeagenciestopromote

healthforpatientsindeliverysystem

Formalcoordinatingrelationships

withcommunity

agencies

tosharecommunity

populationhealthgoals

Useofcommunity

healthworkers

Usewithinprimarycaresystem

withlittle

linktocommunity

Usetocoordinateacrossdeliverysystem

andsomecommunity

resources

Useinclinicalandcommunity

settingstoimprovecommunity

populationhealthforallindividualsinthecommunity.

FinancingforpopulationhealthinitiativesNone

withinafee-for-servicesystem

Limitedfinancing

withinfee-for-servicesystem;community

benefitssupportslimitedactivitieswithcommunity;special

grantsanddemonstrations

butno

dedicatedsource

Increasedfinancing

forpublichealthentitiesthroughstate

orfederalstreamsorPreventionTrusts;globalfee

systemsfordeliverysystemscommit5%

tocommunity

populationhealthoutcomes

Governance

topromotepopulationhealth

None

inplaceinprimarycaresetting

Limitedgovernance

structuresindeliverysystem;might

participateon

community

coalition

orininformalpartnerships

Formalgovernance

structuresinplacewithcommunity

and

publichealthagency;deliverysystem

hasadesignatesenior

lead

forpopulationhealthanddashboardmeasureson

populationhealth

a Community

canalso

equalgeographicarea.

COMMENTARIES

July 2013, Vol 103, No. 7 | American Journal of Public Health Hacker and Walker | Peer Reviewed | Commentaries | 1165

community.31 Despite the logic ofthis partnership, integrating pub-lic health and the delivery systemhas proven difficult.32,33 Today,the ACA poses an unprecedentedopportunity to refocus these efforts.While ACOs are contemplatingthe best strategies for populationhealth improvement, public healthauthorities are also recognizingtheir changing roles34,35 andtheir need to effectively alignwith providers.36 As health in-surance expands, public healthclinical services are likely to de-crease, and core functions includingsurveillance, regulation, and qualityassurance will be more importantthan ever before. States such asMassachusetts, Minnesota, Wash-ington, and Vermont have alreadyevolved from delivering direct ser-vices to providing “wrap around”services (e.g., outreach, care coor-dination) and maintaining the corepublic health functions. Underglobal payment models, ACOs willdepend on public health authoritiesto address regulatory and policyissues that have wide-reachinghealth impact.37

Figure 1 presents three possiblerelationships between healthdelivery and public health sys-tems. When a community isserved by one health system andone public health authority, in-tegration efforts may be moreeasily achieved. However, in

other cases, the delivery systemwill need to work with a numberof public health authorities orthe public health authority willneed to work with numerousdelivery systems.

Strategies to Overcome

Obstacles

To achieve alignment betweenprovider organizations and com-munity and public health agencies,strategies are needed to overcomemultiple obstacles. For example,in highly competitive environ-ments with multiple providers,a strategy of cooperation be-tween clinical delivery systemsand community and public healthagencies is required to jointlyimprove population health. TheInstitute of Medicine report,Improving Health in the Commu-nity38 presented a method formultiple stakeholders in a com-munity coming together to “shareaccountability” for populationhealth outcomes. Weak publichealth infrastructure is anotherobstacle, and in these cases, thedelivery system may need toshore up core public healthfunctions (assurance, assessment,policy).31 In communities withstrong public health systems,public health can address healthfrom a policy and regulatoryperspective while the health caresystem provides individual

clinical prevention and treat-ment.37,39 ACOs may lack theappropriate skills and resourcesto achieve population healthgoals, posing another challenge.A strategy that identifies andconnects an ACO to communityand public health resources canenhance population healthefforts. For example, manycommunity and public healthagencies have extensive experi-ence and programs serving vul-nerable populations and canassist ACOs in their outreachefforts. Overall, ACOs and publichealth systems can play comple-mentary roles in improvingpopulation health goals as seenin the following examples.

1. An urban ACO serving a largecity works with a local publichealth authority to identifygeographic pockets of patientswith diabetes. The ACO fo-cuses on improved diabetesmanagement in the clinicalsetting while linking to com-munity resources for patientsrequesting exercise and phys-ical activity options. Publichealth can lead a campaign toimprove access to fresh fruitsand vegetables and change poli-cies related to menu labeling.

2. An ACO serving a number ofsuburban communities identifieshigh use of the emergency room

from alcohol-related issues inyoung adults as a focus for im-provement. Working with thepublic health authority, localschools, and substance abuseagencies, the collaboration cre-ates a safe rides program anddevelops policies to monitor un-derage liquor sales.

3. An ACO serving a large ruralpopulation has trouble provid-ing enough access for immuni-zations to elders. Community-wide access to immunizations isprovided by working with thepublic health authority and lo-cal pharmacies. Communicationstrategies that link pharmaciesand public health to the ACOare developed, along with animmunization registryfor public health population-level surveillance.

Recommendations

It will take time for newlyemerging ACOs to developmeaningful collaborative relation-ships with public health entities.We recommend the followingsteps for ACOs:d Determine in which geographiccommunities patients resideand what the overlap is betweenthe ACO panel and the commu-nity population.

d Compare the health of thepopulation served by the ACOwith that of the community.

d Decide what level of overlapin any geographic area meritscollaboration. The more marketshare an ACO has in the area,the more investment in collabo-ration might be made, and themore impact that investment willhave on health outcomes.

d Engage in collaboration withpublic health and key commu-nity agencies, including con-ducting a joint needs assessment.

d Collaboratively select healthoutcomes for focus.

IntegratedDelivery System

IntegratedDelivery System

IntegratedDelivery System

PublicHealthSystem

PublicHealthSystem

PublicHealth System

a b c

FIGURE 1—Relationships between integrated delivery system and public health system.

COMMENTARIES

1166 | Commentaries | Peer Reviewed | Hacker and Walker American Journal of Public Health | July 2013, Vol 103, No. 7

d Set up a formal agreement withthe public health authorities toshare data and monitor progresstoward goals in clinical andcommunity settings.

d Identify population health indi-cators to be included on theACO dashboard.

d Use a portion of global paymentfee to support community publichealth activities.

CONCLUSIONS

To fully meet the goals of theTriple Aim, including improvingthe health of a population, ACOsmust define “population health.”We recommend that they em-brace the broad community defi-nition of population health andtake steps to work collaborativelywith community and public healthagencies. Future financing andvalue-based purchasing shouldreward collaborations that resultin population health improve-ments at the community level. Ashealth care moves toward alter-native and global payment ar-rangements, the need to under-stand the epidemiology of thepatient population is imperative.Keeping the population healthywill require enhancing capacity toassess and to monitor and priori-tize lifestyle risk factors and socialdeterminants of health that undulyaffect health outcomes. j

About the AuthorsKaren Hacker is with the Institute forCommunity Health, Cambridge, MA;Cambridge Health Alliance, Cambridge;and Harvard Medical School, Boston,MA. Deborah Klein Walker is with AbtAssociates, Cambridge.Correspondence should be sent to Karen

Hacker, MD MPH, Institute for CommunityHealth, 163 Gore Street, Cambridge, MA02141 (e-mail: Khacker@challiance.org).Reprints can be ordered at http://www.ajph.org by clicking the “Reprints” link.

This commentary was accepted January18, 2013.

ContributorsK. Hacker led the writing process andwas involved in all aspects of the articlefrom conceptualization to writing andediting. D. K. Walker was involved inall aspects of the article fromconceptualization to writing andediting.

References1. A more secure future: what the newhealth law means for you and your family.The White House. Available at: http://www.whitehouse.gov/healthreform/healthcare-overview. Accessed June18, 2012.

2. McGinnis JM, Foege WH. Actualcauses of death in the United States.JAMA. 1993;270(18):2207---2212.

3. McGinnis JM, Williams-Russo P,Knickman JR. The case for more activepolicy attention to health promotion.Health Aff (Millwood). 2002;21(2):78---93.

4. Walker DK. Time to embrace publichealth approaches to national and globalchallenges. Am J Public Health. 2008;98(11):1934---1936.

5. Institute for Healthcare Improve-ment Triple Aim Initiative. Available at:http://www.ihi.org/offerings/Initiatives/TripleAim/Pages/default.aspx. AccessedNovember 4, 2012.

6. Berwick DM, Nolan TW, WhittingtonJ. The triple aim: care, health, and cost.Health Aff (Millwood). 2008;27(3):759---769.

7. Health Research and EducationalTrust, American Hospital Association.Managing Population Health: The Roleof the Hospital. Chicago, IL: AmericanHospital Association; 2012.

8. Institute of Medicine. Primary Careand Public Health: Exploring Integrationto Improve Population Health. Washington,DC: Institute of Medicine; 2012.

9. Kindig D, Stoddart G. What is pop-ulation health? Am J Public Health. 2003;93(3):380---383.

10. Lewis S. Creating incentives to im-prove population health. Prev Chronic Dis.2010;7(5)A93.

11. Institute of Medicine. Primary Careand Public Health: Promoting Integration toImprove Population Health. Washington,DC: Institute of Medicine; 2012.

12. Cantor J, Cohen L, Mikkelsen L,Panares R, Srikantharajah J, Valdovinos E.Community-Centered Health Homes. Oak-land, CA: Prevention Institute; 2011.

13. Evans RG, Stoddart GL. Producinghealth, consuming health care. Soc SciMed. 1990;31(12):1347---1363.

14. Marmot M, Wilkinson RG. SocialDeterminants of Health. 2nd ed. Oxford,UK: Oxford University Press; 2006.

15. Fisher ES, Shortell SM. Accountablecare organizations: accountable for what,to whom, and how. JAMA. 2010;304(15):1715---1716.

16. Barr MS. The patient-centeredmedical home: aligning payment to ac-celerate construction. Med Care Res Rev.2010;67(4):492---499.

17. Andrews E, Toubman S. Patient-centered medical home: improving healthcare by shifting the focus to patients. ConnMed. 2009;73(8):479---480.

18. Cox JV, Kirschner N. Patient-cen-tered medical home: renewing primarycare. J Oncol Pract. 2008;4(6):285---286.

19. National Prevention Council.National Prevention Strategy. Washington,DC: US Department of Health andHuman Services; 2011.

20. US Department of Health andHuman Services. Healthy People 2020.Washington, DC: US Department ofHealth and Human Services; 2012.

21. MassCHIP. Massachusetts Depart-ment of Public Health. Available at:http://www.mass.gov/eohhs/researcher/community-health/masschip. Accessed November 24,2012.

22. County health rankings & roadmaps. County Health Rankings. Availableat: http://www.countyhealthrankings.org/#app. Accessed November 4, 2012.

23. Community health status indicators.US Department of Health and HumanServices. Available at: http://wwwn.cdc.gov/CommunityHealth/homepage.aspx?j=1.Accessed November 25, 2012.

24. Diamond CC, Mostashari F, Shirky C.Collecting and sharing data for pop-ulation health: a new paradigm.Health Aff (Millwood). 2009;28(2):454---466.

25. Internal Revenue Service. InternalRevenue Bulletin: 2011-30. Notice 2011-52. Washington, DC: Internal RevenueService; 2011.

26. United States Preventive ServicesTask Force. Recommendations for the USPreventive Services Task Force. Washing-ton, DC: Agency for Healthcare Researchand Quality, Department of Health andHuman Services; 2011.

27. Task Force on Community Preven-tive Services. Guide to Community Pre-ventive Services: What Works to PromoteHealth? New York, NY: Oxford UniversityPress; 2005.

28. Haberkorn J. Health policy brief: theprevention and public health fund. 2012.Available at: http://www.healthaffairs.org/healthpolicybriefs/brief.php?

brief_id=63. Accessed February 27,2012.

29. Massachusetts Public HealthAssociation. Massachusetts Preventionand Wellness Trust Fund. Available at:http://www.mphaweb.org/documents/PrevandWellnessTrustFund-MPHAFactSheetupdatedOct12.pdf.Accessed November 30, 2012.

30. Meyers DG, Neuberger JS, He J.Cardiovascular effect of bans on smokingin public places: a systematic review andmeta-analysis. J Am Coll Cardiol. 2009;54(14):1249---1255.

31. Institute of Medicine. The Future ofPublic Health. Washington, DC: Instituteof Medicine; 1988.

32. Brandt AM, Gardner M. Antagonismand accommodation: interpreting the re-lationship between public health andmedicine in the United States during the20th century. Am J Public Health.2000;90(5):707---715.

33. Stine NW, Chokshi DA. Opportunityin austerity---a common agenda for medi-cine and public health. N Engl J Med.2012;366(5):395---397.

34. Jarris PE, Leider JP, Resnick B,Sellers K, Young JL. Budgetary decisionmaking during times of scarcity. J PublicHealth Manag Pract. 2012;18(4):390---392.

35. Young J, Resnick B, Leider JP. Per-ceived and anticipated impacts of theAffordable Care Act on state public healthpractice. Paper presented at: AmericanPublic Health Association Annual Con-ference; October 27---31, 2012; SanFrancisco, CA. Available at: https://apha.confex.com/apha/140am/webprogram/Paper260623.html. Accessed February27, 2013.

36. Benjamin GC. Transforming thepublic health system: What are we learn-ing? Available at: http://iom.edu/Global/Perspectives/2012/TransformingPublicHealth.aspx. Accessed November 30, 2012.

37. Frieden TR. A framework for publichealth action: the health impact pyramid.Am J Public Health. 2010;100(4):590---595.

38. Institute of Medicine. ImprovingHealth in the Community: A Role forPerformance Monitoring. Washington,DC: Institute of Medicine; 1997.

39. Institute of Medicine. The Future ofthe Public’s Health in the 21st Century.Washington, DC: Institute of Medicine;2003.

COMMENTARIES

July 2013, Vol 103, No. 7 | American Journal of Public Health Hacker and Walker | Peer Reviewed | Commentaries | 1167