The United States is failing in population health. Life expectancy in the United States is falling for the first time since reliable mortality records have been available, and health equity has been on a downward trend for decades.¹

Why is the United States, among the richest countries in the world, failing to protect the health of its residents? A dysfunctional medical system that pays for sickness rather than wellness is certainly part of the problem, but so is confusion about the role of public health in the United States.

The Mission of Public Health

In 1988, an Institute of Medicine commission defined the mission of public health as “assuring the conditions in which people can be healthy.”² Yet much of public health continues to focus not on the conditions in which people can be healthy but rather on individual health.

Several forces have combined to push public health away from its historical mission.

The Cognitive Straightjacket of Randomization

First, science has been increasingly narrowly construed as the business of conducting randomized controlled trials (RCTs). The emphasis on RCTs took hold in the field of medicine as a useful antidote to expert opinion about the effectiveness and appropriateness of care. The resulting turn to evidence-based medicine has the potential—as yet only partially realized—to improve the quality of medical education and clinical care.^3-6 However, to say that RCTs can lead to an evidence base that improves clinical care is not to say that RCTs are the only avenue to improving health. This point seems to have been lost on prominent gatekeepers of science. To take one example, the JAMA family of journals prohibits the use of causal language except where an RCT has been involved. This means that, in the pages of one of the most prominent medical journals, it is not permissible to discuss cigarette smoking as a cause of cancer, colonoscopy screening as a cause of reduced all-cause mortality, global climate change as a cause of increasing heat stress or even greenhouse gas emissions as a cause of climate change. As one wry observer remarks, by this standard there is no evidence that injury in falls from airplanes can be prevented by parachutes.⁷

This policy places JAMA (and other journals with similar policies) in the company of some pretty severe science-deniers. But more problematically, the exclusive faith in RCTs generates a strong bias against the contextual determinants of health that are supposed to be at the core of the public health mission. It may be the mission of public health to ensure the conditions in which people can be healthy, but since conditions can only rarely be randomized—and then with great difficulty—much of public health is accordingly beyond the pale of science, at least as so defined.

The cult of randomization has an insidious effect on the development of public health. It is more difficult to fund, conduct, and publish research that uses statistical techniques to make subtle causal inferences about observational data than it is to randomize patients to different arms of a trial. As those difficulties accumulate, the science of health moves in the direction of the pills and procedures of medical service delivery rather than the context and conditions of health.

There is no question that randomized controlled trials make for an easier pathway to causal inference, even allowing for their documented problems of replicability and external validity.^8-11 But there are also many situations in which randomization is not feasible or ethical. In these situations, to not acknowledge the legitimacy of other forms of causal inference is flatly unscientific. Medical journals like JAMA and others in the cult of randomization should be called out for their anti-science bias.

The narrow focus on medical interventions as causes of health might be less meaningful if, in fact, medical care were a major determinant of health. It is not. Studies have shown that only one-tenth to one-quarter of the variation in health within developed countries can be attributed to variations in medical care.^12,13 But the other causes—worker protections, family income, the built environment, structural racism—are precisely the kinds of contextual factors that are difficult or impossible to randomize. The cult of randomization, therefore, leaves the great majority of the causes of health entirely outside the concern of many major funders and journals. Those things that cannot be easily randomized can be discussed only in a smaller circle of academia than can pills and procedures. Those things that can be randomized, but whose effects are visible only over a long period, such as early-childhood education or food advertising, are similarly disadvantaged. And structural factors such as poor housing affordability, exploitative working conditions, or structural racism are the least likely to be taken seriously as objects of scientific inquiry. Yet these are precisely the factors that create the conditions in which people can—or cannot without great difficulty—be healthy. The mission of public health is accordingly at a structural disadvantage in the hierarchy of what counts as serious research. It is no wonder that public health tends to wander away from its mission in favor of a narrower focus on individual health behavior change.

The Limitations of Empiricism

A second factor pushing public health away from its mission of assuring the conditions in which people can be healthy are the limits of our theoretical models. Public health can be proud of a long tradition of interdisciplinary collaboration, with economic, sociological, psychological, and other theoretical currents enriching the flow of public health research. But theoretical developments have often remained anchored in their home fields without ever creating a coherent theoretical base within public health. Within public health, our two methodological subfields—epidemiology and biostatistics—are empirical, not theoretical, fields.

Of course, there are many useful conceptual models in public health. But they tend to be descriptions of what, rather than of why or how. They speak a language of nouns without verbs. This is very different than most scientific fields, in which an overarching theoretical frame creates an established body of knowledge about how things happen.

Theory is essential to science for several reasons. It creates a foundation of shared understanding that is used to advance the field—a set of facts that are no longer in dispute. All economists believe that incentives matter to behavior and that the demand curve is downward-sloping. Sociologists believe that social norms constrain and shape behavior. Anthropologists believe that cultural practices are adaptive to their environments. All of these theoretical commitments are supported by strong empirical evidence, and although none is empirically unassailable, they generally are widely endorsed and valid. Their value is not that they are always and everywhere empirically true, but rather that they are true enough, enough of the time, that they are valuable starting points for further empirical and theoretical investigation.

Theory also informs empirical models, identifying directions of causality and separating endogenous from exogenous variation. And, as thorny empirical issues become more widely understood, pressure will grow to have strong theoretical models that can help better identify empirical models. As illustrations, consider collider bias^14,15 and the stable-unit treatment value assumption.^16,17 Collider bias is an analogue of omitted variables bias, which is the familiar problem of an unmeasured variable that influences both X and Y. In collider bias, an included variable that is caused by both X and Y similarly induces a spurious correlation between them even if none would otherwise exist. A classic example is the correlation between a broken starter and a dead battery in a car. These are unrelated events, but if we focus on the car not starting, the finding of a dead battery and that of a broken starter are inversely associated. The stable-unit treatment value assumption also is called the “no interference” or “no interaction between subjects” assumption.¹⁷ A randomized controlled trial found that children whose parents were assigned to receive payment for their school attendance were indeed more likely to attend school; what it missed, however, was that their siblings, for whom no payment was received, had poorer nutritional status than siblings in the control group.¹⁸

Population health science starts with its own theoretical commitments: that the health and health equity of a population are different from and determined differently than the health of individuals. This is a point that was made long ago by Virchow and Durkheim and repeatedly since.^19-21 While individual health may be determined by health behaviors or toxic exposures, progress on population health requires understanding why those behaviors and exposures happen. Population health science is invested in the population causes of incidence and not only the individual causes of susceptibility.²¹

Attempting Policy While Fearing Politics

Finally, public health has naturally not wanted to engage in overtly political advocacy. This aversion to wading into the morass of politics is a healthy instinct. If public health becomes politicized, it will lose the trust of a large fraction of the population. Avoiding political fights is accordingly sound advice for anyone wishing to preserve the respect that our public health successes have earned us. Yet this strategy has significant disadvantages.

If the goal is to avoid inflaming political passions, it is much easier to focus on individual health behavior change than on structural factors. To the extent that the context is addressed at all, it is safest to construe the conditions in which people can be healthy in the narrowest and most literal sense: restaurant inspections and contagious disease containment. Even when systemic changes are proposed, they are oriented around individual health behavior: bicycle helmet laws, cigarette taxes, sugar-sweetened beverage taxes. Although these changes may face political headwinds, there is rarely a storm of protest at new impositions on the behavior of the most marginalized.

Of course, public health does sometimes attempt to engage the structure in ways that are unambiguously designed to improve the context, rather than just punish unhealthy behavior. Asthma home remediation, school-based health centers, and nurse home-visiting programs are all bright spots, all too small and too underfunded.

And why are such programs underfunded? Because public health must fund such programs out of budgets that are designed for much narrower purposes. Public health, unfortunately, has an impossible task: create a culture of health but without displacing any vested interests in the current culture.

It is a Faustian bargain: maintain funding by avoiding offense to any political interest, thereby abandoning attempts at wider structural change, or jeopardize funding by using public health methods and evidence to advance an agenda for creating a society in which people can be healthy. Such a society would involve substantial changes in how income and power are distributed, how cities and suburbs are built, and whose voice is heard and how. Creating the conditions in which people can be healthy, in other words, requires politics.

Public health once was more open to engaging in political fights. Delivering clean water to growing cities was one of the great public health triumphs and led to huge increases in life expectancy. But publicly financed sewers came only after a huge—and very political—struggle. ^22,23 Quarantine laws and food inspections similarly came about as a result of political struggle.²⁴ It might be nice to imagine that progress in public health occurs whenever public health simply presents its science to dispassionate decision-makers. Never has it been so.

While decades of hiding from politics has been a reasonable if sometimes uncomfortable accommodation, it is now no longer working. The allies won through the scrupulous nonpolitical approach to public health—if they were ever really with us—have now definitively abandoned us.

The definitive breakdown has occurred around an issue that is not inherently political or even controversial: the wearing of face masks during a pandemic. This and the broader response to the pandemic—closing restaurants, asking for physical distancing, bans on large gatherings, evidence-based vaccine development, and adequate safety trials for vaccines once developed—have been politically controversial.

The problem for public health is clear: notwithstanding decades of assiduously avoiding politics, politics has come to us. What’s worse, public health is now being turned to overtly political ends. For example, the Trump Administration has coopted public health guidance from the Centers for Disease Control and Prevention (CDC) to illegally deny hearings to asylum applicants.²⁵ And, as a SARS-CoV-2 vaccine is developed at warp speed in the United States by an administration that is known for its self-dealing and lying, public health is finding that the lines between politics and science have been harmfully blurred as vaccine skepticism grows.²⁶ If we continue to ignore politics, we will be crushed.

As the terms of the Faustian bargain have changed, the only way to now preserve the public health mission is to engage politically. Where once it may have been defensible to argue that dramatically narrowing the scope of public health was worth the price of preserving what the rest of the field does, that case can no longer be made. We are in no danger of alienating highly political Republicans, not because this danger is not here, but because it has already happened. Across the country, public health officials have been fired or have quit under duress for implementing the most anodyne public health actions in the face of an enormous threat to the public’s health.²⁷ If our good graces do not protect us now, they were never worth sacrificing for.

The Emergence of a Population Health Perspective

In reaction to these forces in the field of public health, population health science is emerging to bring back an emphasis on structural and contextual factors.

Population health science requires scientists from different disciplinary backgrounds to combine their knowledge and expertise to answer questions that individual disciplines alone cannot. It requires syncretic practice focused not on individual health, but on the mean and variation—the health equity—of outcomes in a population.

Each of the three barriers to engagement with the traditional mission of public health identified above is addressed in some way by population health science.

Population health science starts with one very strong claim, but a well-evidenced one: that science without randomization is not only possible, but essential. The natural sciences understand this intuitively. From Galileo’s observations of pendulums to Einstein’s observation of light bending around the sun during an eclipse, science makes bold claims that it then tries to falsify. No randomization required. The example of macroeconomics is impressive. Here is a field in which there is never any chance of randomization and in which there are many potential confounders and often very small numbers. And yet the field progresses.

The physicist Lee Smolin has written, “Science has succeeded because scientists comprise a community that is defined and maintained by adherence to a shared ethic,”^28(p.301) and he defines this ethic as having three parts:

• If an issue can be decided by [professionally competent] people of good faith applying rational argument to publicly available evidence, then it must be regarded as so decided.
• If, on the other hand, rational argument from the publicly available evidence does not succeed in bringing people of good faith to agreement on an issue, society must allow and even encourage people to draw diverse conclusions.
• Science progresses. We will know more later than we know now. As a result, what we know now is necessarily incomplete.^27(p303)

Causal inference using observational methods are not only consistent with this definition, but have been an essential means of advancing science in many fields with particular relevance to public health, including epidemiology, toxicology, economics, sociology, and psychology.^29-36

Science accordingly has nothing inherently to do with randomization, but rather with agreement on an ever-growing body of knowledge, produced sometimes by experimentation and sometimes by careful observation. When this body of knowledge is organized in a way that makes it memorable, it is theory.

Population health has several theoretical models capable of making testable predictions. Two models, the fundamental causes model³⁷ and multi-level theory,³⁸ include concrete explanations of specific causal pathways. These are promising developments, because the more concretely population health can develop ideas of how population health happens, the more rapidly the field will advance. Theoretical development is an urgent need within the field. Both models are capable of generating predictions that would not be otherwise obvious. For example, neo-materialist theory suggests that poor health arises from a failure to adequately achieve the material conditions that secure health.³⁹ One might expect that, as a society becomes richer, more and more people are lifted above the threshold of necessities, and health accordingly improves. If, as seems intuitive, there are diminishing marginal health returns to increased living standards—that an extra $10,000 a year matters more to the man under the bridge than to the woman in the McMansion—then rising living standards should also compress the distribution of health, reducing health disparities. The fundamental causes model proposes that access to society’s resources is so inflected by social position, that even if additional resources have a higher marginal health return for lower-income people than for higher-income people, health disparities nonetheless can widen when average standards of living are increasing. This claim is bold and surprising when taken piece-by-piece, yet consistent with an abundance of evidence on health disparities. The capacity to make a counterintuitive prediction that cannot be falsified is the hallmark of a strong scientific theory. It also generates important other theoretical implications: in this case, that average health and health equity are theoretically independent and driven by different processes. This theoretical finding would come as a surprise to many methodological individualists.

The engagement of population health science with politics starts with a clear-eyed understanding that politics matters. To be clear, population health science should be used to inform political judgments, never to support political ends. Public health accordingly has two tasks: it must resist the use of population health science for the objectives of politics, and it must simultaneously engage the tools of politics to pursue the population health objectives identified by science.

One example of using politics to pursue population health science is offered in a recent paper in this journal showing a high and positive correlation, which persists when adjusting for potential confounders, between a state becoming more liberal over time and increasing life expectancy.⁴⁰ The magnitude is large: The five states with the greatest movement toward more liberal policies from 1970 to 2014 saw life expectancy increase by about one year associated with this policy shift. The five states with greatest movement toward conservative policies saw life expectancy decrease by about one year associated with this policy shift. This difference is on a par with the life expectancy between the Netherlands and Puerto Rico. There is no way to embrace science and to deny the import of results such as these. Public policy matters to population health outcomes, and for that reason politicians must be held accountable to population health science to the best of our abilities.

Another strand of population health science seeks to expand our abilities to influence political outcomes—or more precisely, outcomes of policy debates. There is increasing recognition that how we talk about our research is as important as what we say. A large body of research in political science suggests that voters are typically not swayed by facts about policy issues.⁴¹ Instead, recent research on message framing suggests promising strategies to make population health interventions more widely acceptable. For example, political conservatives become more open to government actions to prevent obesity when the rationale is presented in terms of military readiness or consumer sovereignty.^42,43 Evoking themes of fairness can increase support for universal health insurance.⁴⁴ Stories have been shown to be effective in shifting public opinion about policy options to improve the context in which health happens.⁴⁵

Population Health Science on the Ground

Population health takes two distinct but interacting forms: population health management and population health research.

Population health management is pursued in the United States by medical groups, accountable care organizations, managed care organizations, and insurance companies. It involves the tracking of outcomes of the entire membership, whether or not individuals present for treatment.⁴⁶

The emphasis in population health management is on proactively ensuring genuine access to needed care, on management of chronic conditions, and on disease and injury prevention. Because most medical care in the United States is compensated on a fee-for-service basis, population health management remains a relatively small part of what US medicine does, but it is rapidly growing. Most of the top medical schools now have departments of population health. One of the priorities of the field is to develop shared definitions and systems of benchmarks that can be used to assess progress on population health and communicate among actors.⁴⁶ While many OECD countries have some kind of shared health records that are accessible to—and used by—all health care providers, the United States lags severely in implementing interoperable electronic health records.

On the academic side, research in population health has increased dramatically in the United States, partly in response to the evolving policy context of medical care delivery and partly because of growing recognition of the importance of upstream factors to health outcomes. It has long been recognized in public health that medical care delivery accounts for a very small portion of the determinants of health.^12,13,47,48 Increasingly, though, it is seen as inadequate to simply ascribe differences in health to differences in behavior. Instead, research is revealing the role of the political, social, and economic context, operating through forces such as structural racism, environmental injustice, and labor exploitation in the patterns of behavior that are observed within a population. Given this understanding, public health then becomes more a matter of changing this context by changing policy than of exhorting individuals to change their behavior.

While this perspective has been within public health at least since Rudolf Virchow, it has gained increasing salience and theoretical coherence in the past several years.

The Interdisciplinary Association for Population Health Science (IAPHS) was founded in 2015 to bring scholars of population health together and to build the science of how some populations are able to achieve high levels of health, with strong health equity. Also in 2015, the CDC initiated a Population Health Training in Place Program to build capacity in policy analysis and population health improvement.

These institutional changes are the early signs in the United States of a growing commitment in population health science to bring scientific tools to the historic mission of public health to assure the conditions in which people can be healthy.

Conclusion

Creating the conditions for health is difficult work: far more difficult, for example, than admonishing people to act more healthfully. It is more difficult for medical delivery systems to take responsibility for keeping people healthy than to treat them when ill. It is more difficult to engage in politics with scientific integrity than to avoid political controversy altogether. And it is more difficult to think critically about the theoretical basis of what causes health in populations than to conduct randomized trials of clinical interventions. But with declining health and eroding health equity in the United States, it is clear that the conditions in which health can happen have not been fostered. It has become urgent to do the harder work to assure population health. Fortunately, the perspective of population health science has engendered several institutional changes that support this hard work. Whether health and health equity in the United States improves or not will depend on how well these institutions are supported.

References

1. Zimmerman FJ, Anderson NW. Trends in health equity in the United States by race/ethnicity, sex, and income, 1993-2016. Jama Network Open. 2019;2(6):e196386.
2. Institute of Medicine Committee for the Study of the Future of Public Health. The Future of Public Health. Washington, DC: National Academy Press; 1988.
3. Sackett DL, Rosenberg WM, Gray JM, Haynes RB, Richardson WS. Evidence based medicine: what it is and what it isn’t. BMJ. 1996;312(7023):71-72.
4. Fritsche L, Greenhalgh T, Falck-Ytter Y, Neumayer H-H, Kunz R. Do short courses in evidence based medicine improve knowledge and skills? Validation of Berlin questionnaire and before and after study of courses in evidence based medicine. BMJ. 2002;325(7376):1338-1341.
5. Bates DW, Kuperman GJ, Wang S, et al. Ten commandments for effective clinical decision support: making the practice of evidence-based medicine a reality. J Am Med Inform Assoc. 2003;10(6):523-530.
6. Timmermans S, Mauck A. The promises and pitfalls of evidence-based medicine. Health Aff. 2005;24(1):18-28.
7. Smith G, Pell JP. Parachute use to prevent death and major trauma related to gravitational challenge: systematic review of randomised controlled trials. BMJ. 2003;327(7429):1459.
8. Ioannidis JP. Contradicted and initially stronger effects in highly cited clinical research. JAMA. 2005;294(2):218.
9. Davis C, Naci H, Gurpinar E, Poplavska E, Pinto A, Aggarwal A. Availability of evidence of benefits on overall survival and quality of life of cancer drugs approved by European Medicines Agency: retrospective cohort study of drug approvals 2009-13. BMJ. 2017;359:j4530.
10. Prasad V, Kim C, Burotto M, Vandross A. The strength of association between surrogate end points and survival in oncology: a systematic review of trial-level meta-analyses. JAMA Intern Med. 2015;175(8):1389-1398.
11. Laffers L, Mellace G. Identification of the average treatment effect when SUTVA is violated. Discussion Papers on Business and Economics. Odense, Denmark: University of Southern Denmark; 2020.
12. Hood CM, Gennuso KP, Swain GR, Catlin BB. County health rankings: relationships between determinant factors and health outcomes. Am J Prev Med. 2016;50(2):129-135.
13. McGinnis JM, Williams-Russo P, Knickman JR. The case for more active policy attention to health promotion. Health Aff. 2002;21(2):78-93.
14. Morgan SL, Winship C. Counterfactuals and Causal Inference: Methods and Principles for Social Research. Cambridge: Cambridge University Press; 2007.
15. Elwert F, Winship C. Endogenous selection bias: the problem of conditioning on a collider variable. Annu Rev Sociol. 2014;40(1):31-53.
16. Boesche T. Reassessing quasi-experiments: policy evaluation, induction, and SUTVA. Br J Philos Sci. February 1, 2019. https://doi.org/10.1093/bjps/axz006
17. Schwartz S, Gatto NM, Campbell UB. Extending the sufficient component cause model to describe the Stable Unit Treatment Value Assumption (SUTVA). Epidemiol Perspect Innov. 2012;9(1):3.
18. Gitter SR, Manley J, Barham B. The Coffee Crisis, Early Childhood Development, and Conditional Cash Transfers. IDB Working Paper Series; 2011.
19. Syme SL. Foreward. In: Berkman LF, Kawachi I, eds. Social Epidemiology. New York: Oxford University Press; 2000.
20. Evans RG, Barer ML, Marmor TR. Why Are Some People Healthy and Others Not? The Determinants of Health of Populations. New York: Routledge; 1994.
21. Rose G. Sick individuals and sick populations. Int J Epidemiol. 1985;30(3):427-432.
22. Bell S. A Democratic History of Sewerage. The Pelican Map Blog. 2013.
23. Halliday S. A Government for London. In: The Great Stink of London: Sir Joseph Bazalgette and the Cleansing of the Victorian Capital. Gloucestershire: Sutton Publishers; 1999.
24. Beauchamp DE. Community. In: Beauchamp D, Steinbock B, eds. New Ethics for the Public’s Health. New York: Oxford University Press; 1999:57-67.
25. Letter to DHS Secretary Wolf and Attorney General Barr Signed by Leaders Of Public Health Schools, Medical Schools, Hospitals, and Other U.S. Institutions [press release]. New York: Columbia University Mailman School of Public Health; 2020.
26. Hoffman J. Mistrust of a Coronavirus vaccine could imperil widespread immunity. New York Times. July 18, 2020. https://www.nytimes.com/2020/07/18/health/coronavirus-anti-vaccine.html. Accessed November 17, 2020.
27. Smith MR, Weber L. Public health officials are quitting or getting fired in throes of pandemic. Kaiser Health News. August 11, 2020. https://khn.org/news/public-health-officials-are-quitting-or-getting-fired-amid-pandemic/. Accessed November 17, 2020.
28. Smolin L. The Trouble with Physics. New York: Houghton Mifflin; 2007.
29. Hill AB. The environment and disease: association or causation? Proc R Soc Med. 1965;58:295-300.
30. Rubin DB. Estimating causal effects of treatments in randomized and nonrandomized studies. Journal of Educational Psychology. 1974;66(5):688-701.
31. Sobel ME. Causal inference in the social sciences. J Am Stat Assoc. 2000;95(450):647-651.
32. Galea S, Riddle M, Kaplan GA. Causal thinking and complex system approaches in epidemiology. Int J Epidemiol. 2010;39(1):97-106.
33. Glass TA, Goodman SN, Hernán MA, Samet JM. Causal inference in public health. Annu Rev Public Health. 2013;34:61-75.
34. Greenland S. Causation and Causal Inference. In: Lovric M, ed. International Encyclopedia of Statistical Science. Berlin: Springer; 2011:208-216.
35. Hernán MA. Does water kill? A call for less casual causal inferences. Ann Epidemiol. 2016;26(10):674-680.
36. Brady HE. Causation and Explanation in Social Science. In: Goodin RE, ed. The Oxford Handbook of Political Science. Oxford, UK: Oxford University Press; 2011.
37. Phelan JC, Link BG, Tehranifar P. Social conditions as fundamental causes of health inequalities theory, evidence, and policy implications. J Health Soc Behav. 2010;51(1 suppl):S28-S40.
38. Zimmerman FJ. Habit, custom, and power: a multi-level theory of population health. Social Science & Medicine. 2013;80:47-56.
39. Lynch JW, Smith GD, Kaplan GA, House JS. Income inequality and mortality: importance to health of individual income, psychosocial environment, or material conditions. BMJ. 2000;320(7243):1200.
40. Montez JK, Beckfield J, Cooney JK, et al. US state policies, politics, and life expectancy. Milbank Q. 2020;98(3):668-699.
41. Esberg J, Mummolo J, Westwood SJ. Political Misperceptions, Group Identity, and Policy Preferences: Evidence from Criminal Justice Experiments. Princeton, NJ: Princeton University; 2020.
42. Gollust SE, Niederdeppe J, Barry CL. Framing the Consequences of Childhood Obesity to Increase Public Support for Obesity Prevention Policy. American Journal of Public Health. 2013;103(11):e96-e102.
43. Ortiz SE, Zimmerman FJ, Adler GJ. Increasing public support for food-industry related, obesity prevention policies: the role of a taste-engineering frame and contextualized values. Social Science & Medicine. 2016;156:142-153.
44. Lynch J, Gollust SE. Playing fair: fairness beliefs and health policy preferences in the United States. J HealthPolit Policy Law. 2010;35(6):849-887.
45. Niederdeppe J, Bu QL, Borah P, Kindig DA, Robert SA. Message design strategies to raise public awareness of social determinants of health and population health disparities. Milbank Q. 2008;86(3):481-513.
46. Jacobson DM, Teutsch S. An Environmental Scan of Integrated Approaches for Defining and Measuring Total Population Health. Paper presented at: National Quality Forum, Washington, DC; 2012.
47. Kaplan RM, Milstein A. Contributions of health care to longevity: a review of 4 estimation methods. Ann Fam Med. 2019;17(3):267-272.
48. Group CCoSDoHSDoHFT. A Review of Frameworks on the Determinants of Health. Canadian Council on Social Determinants of Health; 2015.

 Acknowledgments: This Perspective was presented as my Presidential Address at the 2020 meeting of the Interdisciplinary Association for Population Health Sciences (IAPHS). The views expressed are my own and do not necessarily reflect either those of UCLA or IAPHS.

Read on Wiley Online Library