What Is Population Health?

Population health is a rela­tively new term that has not yet been precisely defined. Is it a concept of health or a field of study of health determinants?

We propose that the defini­tion be "the health outcomes of a group of individuals, in­cluding the distribution of such outcomes within the group," and we argue that the field of population health includes health outcomes, patterns of health determinants, and poli­cies and interventions that link these two.

We present a rationale for this definition and note its dif­ferentiation from public health, health promotion, and social epidemiology. We invite cri­tiques and discussion that may lead to some consensus on this emerging concept. (Am J Public Health. 2003; 93: 380-383)

ALTHOUGH THE TERM

"population health" has been much more commonly used in Canada than in the United States, a precise definition has not been agreed upon even in Canada, where the concept it denotes has gained some prominence. Proba­bly the most influential contribu­tion to the development of the population health approach is Evans, Barer, and Marmor's Why Are Some People Healthy and Oth­ers Not? The Determinants of Health of Populations,¹ which grew out of the work of the Pop­ulation Health Program of the Canadian Institute for Advanced Research. No concise definition of the term appears in this vol­ume, although its authors state the concept's "linking thread [to be] the common focus on trying to understand the determinants of health of populations."^1(p29)

The idea that population health is a field of study or a re­search approach focused on de­terminants seems to have evolved from this work. Early discussions at the Canadian Insti­tute for Advanced Research also considered the definition and measurement of health and the processes of health policymaking, but the dominant emphasis evolved to the determinants themselves, particularly the non­medical determinants. John Frank, the scientific director of the re­cently created Canadian Institute of Population and Public Health, has similarly called populadon health "a newer research strategy for understanding the health of populations."² T. K. Young's re­cent book Population Health also tends in this direction; he states

that in Canada and the United Kingdom in the 1990s, the term has taken on the connotation of a "conceptual framework for thinking about why some popula­tions are healthier than others as wed as the policy development, research agenda, and resource allocation that flow from this framework."•^1ft*^,

However, Young also indicates that in the past, the term has been used as a "less cumbersome substitute for the health of popu­lations," which is of course its lit­eral meaning. Evans and Stod­dart, while supporting an emphasis on "understanding of the determinants of population health," have also stated, how­ever, that "different concepts [of health] are neither right or wrong, they simply have different purposes and applications..,. [W]hatever the level of definition of health being employed, how­ever, it is important to distinguish this from the question of the de­terminants of that definition of health."^I(p28) The Health Promo­tion and Programs Branch of Health Canada has recently stated that "the overall goal of a population health approach is to maintain and improve the health of the entire population and to reduce inequalities in health be­tween population groups."^{4ipl 1} They indicate that one guiding principle of a population health approach is "an increased focus on health outcomes (as opposed to inputs, processes, and prod­ucts) and on determining the de­gree of change that can actually be attributed to our work."'¹'¹¹

Dunn and Hayes, quoting the definition of the Canadian Fed­eral/Provincial/Territorial Advi­sory Committee on Population Health, write that "population health refers to the health of a population as measured by health status indicators and as influenced by social, economic and physical environments, per­sonal health practices, individual capacity and coping skills, human biology, early childhood development, and health ser­vices. As an approach, popula­tion health focuses on interre­lated conditions and factors that influence the health of popula­tions over the life course, identi­fies systematic variations in their patterns of occurrence, and ap­plies the resulting knowledge to develop and implement policies and actions to improve the health and well being of those populations."³'*⁵ ' Kindig has suggested a similarly broad defi­nition: population health is "the aggregate health outcome of health adjusted life expectancy (quantity and quality) of a group of individuals, in an economic framework that balances the relative marginal returns from the multiple determinants of health.~"^;?4" This definition pro­poses a specific unit of measure of population health and also in­cludes consideration of the rela­tive cost-effectiveness of re­source allocation to multiple determinants.

Recently, even in the United States, the term is being more widely used, but often without clarification of its meaning and definition. While this develop­ment might be seen as a useful movement in a new and positive direction, increased use without

80 | Mcaefs tor Population Health | Peer Reviewed | Kindig ana Stoddaa

American Journal of Public Heal:- Marc- 2003. ,ci 93. No. 3

precision of meaning could threaten to render the term more confusing than helpful, as may al­ready be the case with "commu­nity health" or "quality of med­ical care." For this reason, we propose a definition that may have a more precise meaning for policymakers and academics alike; our purpose is to stimulate active critiques and debate that may lead to further clarification and uniformity of use.

DEFINITION AND CONCEPT

As indicated above, the pri­mary tension or confusion at present seems to be between defining population health as a field of study of health determi­nants or as a concept of health. The Group Health Community Foundation has recently stated that "some observers see popula­tion health as a new term that highlights the influential role of social and economic forces in combination with biological and environmental factors, that shape the health of entire populations ... others interpret population health primarily as a goal—a goal of achieving measurable im­provements in the health of a de­fined population."⁷'^p'*

We think that there are 3 gen­eral possibilities: population health (a) is only concerned with the independent variables (the multiple determinants), (b) is only concerned with the depen­dent variables (health outcomes), or (c) is concerned with both the definition and measurement of health outcomes and the roles of determinants. While none of the three is normauvely correct or incorrect we believe that the lat­ter is more appropriate, primarily because the concept and mea­surement of health and health

outcomes focuses attention and research effort on the impact of each determinant and their inter­actions on some appropriate out­come. It also allows one to con­sider health inequality and inequity and the distribution of health across subpopulations, as well as the ethical and value con­siderations underpinning these issues.

While the original Evans and Stoddart "field model" did not discuss a population health con­cept in these terms, the idea is implicit in the evolution of the dependent variable from "health care" to "health and function" to "well being."'to*^33-531 The Insti­tute of Medicine has given seri­ous attention to measuring population health, thereby en­couraging some kind of sum­mary measure that includes mor­tality and health-related quality of life.⁹

Given these considerations, we propose that population health as a concept of health be defined as "the health outcomes of a group of individuals, including the dis­tribution of such outcomes within the group." These popula­tions are often geographic re­gions, such as nations or commu­nities, but they can also be other groups, such as employees, eth­nic groups, disabled persons, or prisoners. Such populations are of relevance to policymakers. In addition, many determinants of health, such as medical care sys­tems, the social environment, and the physical environment, have their biological impact on indi­viduals in part at a population level.

Defining population health this way requires some measure(s) of health outcomes of populations, including their distribution throughout the population. We chose the broader term "health

outcomes" rather than the more narrow term "health status"; we believe the latter refers to health at a point in time rather than over a period of years. We do not believe that there is any one definitive measure, but we argue that the development and valida­tion of such measures for differ­ent purposes is a critical task for the field of population health research.

Our definition does imply the necessity, of one or more broad summary measures capable of being a dependent variable for the spectrum of all determinants (generally including length of life and health-related quality and function of those life years), along with a family of other sub-measures for different policy and research purposes. For example, the Health Utilities Index is being used in the Canadian National Population Health Survey,¹⁰ Years of Healthy Life have been used in Healthy People 2000," and the EuroQuol has been re­cently added to the Medical Ex­penditure Panel Survey.¹²

We support the idea that a hallmark of the field ofpopulation health is significant attention to the multiple determinants of such health outcomes, however measured. These determinants include medical care, public health interventions, aspects of the social environment (income, education, employment, social support, culture) and of the phys­ical environment (urban design, clean air and water), genetics, and individual behavior. We note with caution that such a list of categories can lead to a view that they operate independently: pop­ulation health research is funda­mentally concerned about the in­teractions between them, and we prefer to refer to "patterns" of determinants

Population health researchers tend to use a set of methods and approaches that have the follow­ing important characteristics: ex­amination of systematic differ­ences in outcomes across populations, complexity of inter­actions among determinants, bio­logical pathways linking determi­nants to population health outcomes, and the influence of different determinants over time and throughout the life cycle.^13-15

In our view, a population health perspective also requires attention to the resource alloca­tion issues involved in linking de­terminants to outcomes. Part of the study of population health in­volves the estimation of the cross-sectoral cost-effectiveness of different types and combina­tions of investments for produc­ing health.¹ Because improve­ment in population health requires the attention and actions of multiple actors (legislators, managers, providers, and individ­uals), the field of population health needs to pay careful atten­tion to the knowledge transfer and academic-practice partner­ships that are required for posi­tive change to occur.¹⁷*¹⁸ Figure 1 shows how we view the field of population health. The field in­vestigates each of the compo­nents shown in the figure, but particularly their interactions.

CRITIQUES

We expect and welcome cri­tiques of the definition presented here. As noted above, one cri­tique will be that the tasks of defining and measuring concepts of health are large enough to constitute a subject of their own, rather than being combined with the study of determinants of health. We have already given our rationale for including them

March 2003. Vol 93. No. 3 I American Journal of Public Health ₃ i n r> ., i ..-v. ,■ „

i _u ruu.n. ncou.1 nir.^garj s.oddart | Peer Reviewed | Models for Population Health j 381

in population health as a field of study, but we would add that the need for accountability argues strongly for the inclusion of out­come and distributional consider­ations if a population health ap­proach is to be useful in guiding policymaking regarding resource allocation across determinants and sectors. Without such a framework, advocacy and finan­cial incentives for individual de­terminants can proceed indepen­dently of their impact, as some would argue is now the case for some medical care expenditures in the United States.

A second critique is that such a definition and concept is so broad that it includes everything and is therefore not useful to guide either research or policy. We understand this concern but do not agree with it We believe that a guiding synthesis is essen­tial for considering both the rela­tive impacts of the pattern of de­terminants and their interactions. Integration of knowledge about health and its multiple determi­nants seldom occurs. Policy man­agers typically have responsibility For a single sector; advocacy groups typically have an interest in only one disease or determi­nant. No one in the public or pri­vate sectors currently has respon­sibility for overall health

improvement We suggest that the importance of a population health perspective is that it forces review of health outcomes in a population across determinants. For population health research, specific investigations into a single determinant outcome measure, or policy intervention are relevant, and may even be critical in some cases, but they must be recognized as only a part and not the whole.

Those in public health or health promotion may legiti­mately feel that population health is simply a renaming of what has been their work or leg­acy. Hamilton and Bhatti have at­tempted to show the complemen­tarity and overlap between population health and health pro­motion,¹⁹ building on the Cana­dian Achieving Health for All Framework for Health Promo­tion²⁰ and the World Health Or­ganization Ottawa Charter on Health Promotion.²¹ Frank has indicated that historic concepts of public health were similarly broad, until the biomedical para­digm became dominant Those who define public health as the "health of the public" would not disagree with the definition of population health proposed here; in the words of Frank, the "shift in thinking entailed in population

health should be a small one for public health workers... in fact it is not so much a shift as a re­turn to our historical roots en­compassing all the primary deter­minants of health in human popuIauons."^2:W

However, much of public health activity, in the United States at least does not have such a broad mandate even in the "assurance" functions, since major determinants such as med­ical care, education, and income remain outside of public health authority and responsibility, and current resources do not even allow adequate attention to tradi­tional and emerging public health functions. Similarly, we believe that the emerging promi­nence of social epidemiology is a very important development for population health but does not have the breadth, or imply all of the multiple interactions and pathways, of the definition pro­posed here for population health.

CONCLUSION

We believe that the time has come for a clarification of the meaning and scope of the term "population health." We have of­fered a clarification of the term that combines the definition and measurement of health outcomes and their distribution, the pat­terns of determinants that influ­ence such outcomes, and the policies that influence the opti­mal balance of determinants. We welcome discussion and debate regarding these suggestions as a way of moving toward some con­sensus on this important and emergent concept. ■

About the Authors

David Kindig is with the Department of Population Health Sciences. University of Wisconsin-Madison School of Medicine.

Madison. Greg Stoddart is with the De­partment of Clinical Epidemiology and Bioslatistics, McMaster University Health Science Centre. Hamilton. Ontario.

Requests for reprints should be sent to David A. Kindig. MD. PhD. Suite 760. 610 Walnut St. Madison. WI53726-2397 (e-mail: dakindig@facstaff.iasc. edit).

This article was accepted November 15, 2002.

Contributors

Both authors contributed equally to the planning and writing of this artide.

Acknowledgments

D. Kindig was funded in part by the Robert Wood Johnson Health and Soci­ety Scholars Planning Grant to the De­partment of Population Health Sciences at the University of Wisconsin Madison School of Medicine.

We acknowledge the helpful com­ments of John Mullahy. Stephanie Rob­ert, Donn D'Alessio, John Lavis, and many members of the Robert Wood Johnson Foundation Health and Society Scholars Planning Group.

References

1. Evans R. Barer M. Marmor T. Why Are Some People Healthy and Others Not? The Determinants of Health of Pop­ulations. New York, NY: Aldine de Gruyter; 1994.

2. Frank J. Canada Institute of Popu­lation and Public Health. Message from: Dr John Frank, scientific director. Insti­tute of Population and Public Health. Available at: http://www.cihr-irscgc.ca/ instilutes/ipph/about_ipph/ipph_ roessage_direaor_e.shtml. Accessed De­cember 23. 2002.

3. Young TK. Population Health: Con­cepts and Methods. Sew York. NY: Ox­ford University Press: 1998.

4. Health Canada Taking Action on

Population Health. Ottawa. Ontario: Health Canada: 199S.

5 Dunn JR. Hayes MV. Toward a lex­icon of population health. Can J Public Health. 1999:90(suppl 1):S7-S10.

6 Kindig DA. Purchasing Population

Health: Paying for Results. Ann Arbor: University of Michigan Press. 1997.

7, Kreuter M. Lez:.-. N. Improving
Everyone's Quality of Life A Primer on
Population Health. Atlanta. Ga Group
Health Community Foundation: 2001

8. Williams A. Conceptual and empiri-
cal issues in the eff.cency-equity trade-
off in the provision of health care or. if
we are going to get a fair innings, some-
one will need to keep the score.' (n:

362 | Models for Population Health | Peer Reviewed | Kindig and Stoddart

American Journal of Public Health j March 2003. Vol 93. No. 3