The Milbank Memorial Fund is an endowed operating foundation that publishes The Milbank Quarterly, commissions projects, and convenes state health policy decision makers on issues they identify as important to population health.
We focus on a number of topic areas identified by state health policy leaders as important to population health.
The Center for Evidence-based Policy at Oregon Health & Science University is a national leader in evidence-based decision making and policy design.
Keep up with news and updates from the Milbank Memorial Fund. Get the latest from thought leaders, including Christopher F. Koller, president of the Fund.
We publish The Milbank Quarterly, as well as reports, issues briefs, and case studies on topics important to population health.
May 9, 2019
May 2019 | Hugh Alderwick, Laura M. Gottlieb | Early View, Perspective
The language of the social determinants of health is currently in vogue in US health care. Yet the same words and concepts, such as social determinants of health and population health, are often used in different ways—sometimes with very different objectives.1 This, for fans of the later Wittgenstein, is how language works: the meaning of words is in their use.2 Confusion over health care jargon is nothing new.3,4 But misunderstandings over meanings could have important implications as a growing number of health care systems design new interventions to respond to patients’ social circumstances. New collaborations among the health care, public health, and social services sectors5—each often arriving with different ways of thinking about social determinants of health and social interventions—would benefit from a shared understanding of the language being used and what it means for achieving their objectives. Clarity would also benefit policymakers exploring (and testing) new payment systems and policies to incentivize greater health care investments in social supports.6-9 This is particularly important given the potential unintended consequences of bringing language and information related to patients’ social circumstances into a market-based health care system.10
In this article, we highlight some of the key terms being used in work related to social determinants of health in US health care settings. We propose an initial set of definitions for these terms, using existing definitions where available, and make distinctions between concepts commonly conflated or confused. Our goal is to provide the scaffolding for a national discussion about these terms and surface emerging distinctions between concepts and practices related to social determinants of health in health care.
Terms and Distinctions
Social Determinants of Health and Population Health
Social determinants of health are defined by the World Health Organization as “the conditions in which people are born, grow, live, work and age,” which are “shaped by the distribution of money, power and resources.”11 They include income, education, employment, housing, neighborhood conditions, transportation systems, social connections, and other social factors. Social determinants of health operate at multiple levels.12 Underlying structural factors, such as a country’s macroeconomic policy; public policies on education, housing, social security, and other areas; and broader cultural and institutional contexts shape the distribution of resources across society and people’s social position within it. These structural factors, in turn, shape more downstream social factors, such as living and working conditions, and access to money to buy food, clothes, and other basic resources, which form the circumstances of people’s daily lives. The interactions of these factors shape health and health inequities.
Social determinants have both direct and complex effects on health. For example, poor air quality, more common in poorer neighborhoods, can quickly exacerbate asthma symptoms.13 Levels of income and education, meanwhile, shape a range of health outcomes, such as the prevalence of chronic disease and life expectancy, across the life course.14-17 The combined impact of social determinants on health is stark: the poorest 1% in the United States die around 10 to 15 years earlier than the richest.18 But social determinants of health are not just about the haves and have nots. The social gradient becomes a health gradient: people in the middle of the socioeconomic scale have worse health than those at the top, people lower down in the scale have worse health still, and so on.16-19 This means that social determinants of health affect the whole population, not just the poorest or most vulnerable patients.
Social determinants of health are not the same as population health. Population health has been defined as “the health outcomes of a group of individuals, including the distribution of such outcomes within the group.”20 Social determinants are just one group of factors that shape population health, alongside health care, genetics, behaviors, commercial influences, and more.4 Estimates differ, but most studies suggest that social, economic, environmental, and other nonmedical factors play a greater role in shaping population health than health care services.21
Population Health Improvement and Population Health Management
Health care and public health professionals may have different interpretations of the term population health, with implications for the approach taken to improving it. In one qualitative study in 2014, health care leaders of ACOs used the term population health to describe efforts to improve care for their own patients—the individuals covered under specific health plans, or served by hospitals and clinics.3 For public health leaders, however, the unit of analysis was more often the whole population living in a geographical area, such as a county or region.
These differences are, perhaps, understandable—reflecting the distinct roles, capabilities, and incentives for health care and public health in the United States. But conflicting interpretations could have unintended consequences. For example, focusing efforts to improve population health only on patients may divert attention and resources away from more upstream interventions, like improving neighborhood conditions, that are needed to improve health across whole communities. A further distinction has therefore been made between “population health improvement” and “population health management”—the latter describing approaches to improve health that focus on defined groups of patients served by health plans, hospitals, ACOs, or other health care organizations, rather than the whole community living in a particular geographical area.22 Health care systems can—and sometimes do—pursue both strategies at the same time: they can focus on improving care for their patient populations while collaborating with other services and sectors to improve health for a wider geographic population.23 But the clear risk is that health care systems predominantly pursue more downstream interventions for their sickest and most expensive patients.24
Social Determinants of Health and Social Risk Factors
Social determinants shape health for better or worse. Higher income, for example, is associated with better health, while lower income is associated with worse health. In contrast, social risk factors are specific adverse social conditions that are associated with poor health, like social isolation or housing instability. These social risks have very real impacts on health and health care. Social isolation predicts mortality at a rate similar to more clinical risk factors.25 And homeless people are at increased risk of early death, have high rates of disease and disability, and are frequent users of emergency departments and other health services.26-28 As a result, patients are increasingly being asked about selective social risk factors in clinical settings.29,30
In the emerging literature on social interventions in clinical settings, these social risks to health are referred to variously as social determinants,31 health-related social needs,32 or health-related social problems,33 and other times simply as social needs.34 Given that the term risk factors is commonly used to describe any attribute or exposure of an individual that increases their likelihood of poor health,35 we propose that an individual-level adverse social determinant of health, such as low-education level or housing instability, be referred to as a social risk factor.
Making a distinction between social determinants of health and social risk factors could be important for at least two reasons. First, it makes clear that social determinants of health are neither negative nor positive. Social factors—whether community characteristics, financial stability, social relationships, access to services, or employment status—can increase as well as constrain people’s capacity to be healthy. In other words, social determinants of health are not things a person can have or can avoid: they affect everyone. And enhancing protective factors in the social environment, such as access to education, and family and other social supports for children and young people36 can promote people’s health.
Second, defining and identifying individual-level social risk factors can help health care systems target interventions to address them. For example, the Centers for Medicare and Medicaid Services (CMS) recently developed a 10-item screening tool covering five actionable social risk factors: housing instability, food insecurity, lack of transportation access, utility insecurity, and interpersonal violence.37 Data on social risk factors could be aggregated to better understand where community-level interventions might improve health, as well as be used to target interventions to individual patients.38
Social and Behavioral Risk Factors
Social risk factors are not the same as behavioral risk factors, which include smoking, alcohol intake, other substance use, lack of exercise, and unhealthy diet. In one sense, the two are intertwined. Social factors shape individual behavior: there is a higher prevalence of unhealthy behaviors in socially disadvantaged groups.14,39-41 The relationship also moves in the other direction: obesity, for instance, is linked to discrimination in the workplace, education, and interpersonal relationships, which, in turn, can reduce socioeconomic status.42,43 Yet modifying individual behaviors fundamentally depends on addressing the upstream social factors that shape them. For example, individuals seeking to manage their weight through exercise and diet are constrained by the obesogenic environment that surrounds them—and individuals in poorer communities are even more constrained by fewer supermarkets, more fast food outlets, and fewer safe and accessible places to exercise.44 Conflating behavioral and social risks could lead to short-sighted approaches that focus too heavily on individual behaviors—a phenomenon known as lifestyle drift, which risks widening health inequities.45
To add to the confusion, the term behavioral risk factors is also often used in the United States as an umbrella term for mental health risks. Like in the case of behavioral and social risks, mental health and social risk factors can profoundly influence one another46 but are not identical. Social needs, meanwhile, are not necessarily synonymous with social risk factors—they also depend on people’s individual preferences and priorities. Distinguishing between social risks and social needs emphasizes the patient’s role in identifying and prioritizing social interventions. This concept is at the heart of efforts to implement shared-decision making in traditional medical care. The idea follows that an understanding of what matters in the context of a patient’s life, combined with clinical expertise, is needed to make high-quality decisions about the right course of treatment.47 Likewise, a screening tool may uncover multiple social risks, such as food, housing, and utilities instability, but the patient may tell her provider that her most pressing social need is to find a safe place to stay, away from a violent male partner. Failing to understand the patient’s perspective—a common failure even within the scope of traditional biomedical care48—could lead to the wrong immediate treatment.
This distinction is illustrated by three recent studies of social screening interventions in US health care settings. In one study that focused on food insecurity screening in pediatric primary care, only half of the patients screened as food insecure asked for referrals to food-related interventions, while only half who asked for food-related interventions screened as food insecure.49 In another study, 29% of patients screened positive for food insecurity but only 17% wanted help to address it—and only 4% selected it as their top priority.50 A third study focused on multiple social risk factors found that nearly all screened patients (97% to 99%) reported one or more social risks to health. But once clinics added questions asking whether patients wanted help to address these risks, the numbers decreased—15% of patients with documented social risks requested help in one clinic and 21% of patients requested help in another.51 While clinical teams should consider the range of factors that might prevent patients from seeking social supports from their health care providers when they need them—such as stigma, fear of discrimination, previous trauma, lack of trust, or legal issues—surfacing patients’ perceptions of social need could help target resources toward interventions that could most benefit patients.
Social Needs–Informed Care and Social Needs–Targeted Care
Based on information about patients’ social risk factors and needs, health care systems across the United States are increasingly developing interventions to address them.52 Some make referrals to social service agencies providing food, housing, or income assistance. Others have established medical-legal partnerships or food pantries to improve access to basic social resources. And many use data about patients’ social needs to improve or target traditional medical care.
There is no single term to define these activities in clinical settings. In the United Kingdom, identifying and responding to patients’ social needs in general practice is often referred to as social prescribing.53 This language has some merits. Using established medical terms to describe social interventions might encourage clinicians to see the process of identifying social needs and linking patients with basic resources as part of their regular clinical toolkit. But it also has big drawbacks—for example, the risk of medicalizing socioeconomic hardship and alienating the community-based organizations that health care systems depend on to deliver social services. A bigger challenge, perhaps, is that the term prescription implies a solution that is known and easy to find—like a drug quickly located behind the pharmacy counter. What is the prescription given to a person without stable housing when shelters are full and affordable housing scarce?
Two terms have been proposed to categorize practices in health care settings used to respond to patients’ social needs.54 The first is social needs–informed care: activities that involve modifications to traditional medical care to account for patients’ social circumstances. Some examples of social needs–informed care are common in clinical practice—like providing transportation to hospital appointments, or translators for patients with language barriers. Other examples of social needs–informed care may become more widespread as social screening tools offer clinical teams new data on patients’ social risks. For example, data on patients’ access to food could be used to inform decisions about medications (such as antibiotics that should be taken with meals) and dosing (such as distribution of basal versus meal-time insulin).55 The second is social needs–targeted care: activities in clinical settings that seek to address patients’ social needs directly—for example, helping patients access income assistance if they lack financial resources, or linking them with transportation, food, or housing supports.
Both types of patient care are important, but social needs–informed care alone is insufficient if the aim is to improve the underlying social risks that adversely impact patients’ health, rather than just providing better medical care. And both types of care are insufficient if the aim is to contribute to improvements in population health for the whole community—not just better care and social supports for patients in a single clinic or health plan. This is a bigger task than health care systems can accomplish alone. Improving population health requires that health care organizations collaborate with other services and sectors, such as local government, housing agencies, schools, and community-based organizations. But it also depends on wider policy interventions at the state and federal level—for example, government decisions about the level and distribution of public spending on housing, income support, transportation, and other social services—that fundamentally influence the structural social and economic conditions shaping health for individuals within those communities.
The idea that health is shaped by social determinants is not new. Nor is the idea that health care professionals can—and often do—play an important role in decreasing the health impacts of adverse social circumstances.56,57 But these ideas are now entering the mainstream of US health care. Various federal and state policy initiatives, such as the federal demonstration on accountable health communities,28 and Medicaid waivers in Oregon, California, and elsewhere,9 have strengthened the role of health care systems in identifying and addressing patients’ social needs. Most of these efforts (and their evaluations) are still nascent, which means that we still know relatively little about how health care payers and providers can best intervene to improve patients’ social circumstances in different contexts.46 As health care initiatives related to the social determinants of health grow in number and scale, greater clarity over the language used and concepts underlying them is likely to aid both their implementation and evaluation. While we have focused on developing a social determinants of health lexicon for health care systems in the United States, several of the terms and distinctions described will be relevant in other countries.
Acknowledgments: Support for Hugh Alderwick’s research that contributed to this article was provided by the Commonwealth Fund. Alderwick was a Harkness Fellow in Health Care Policy and Practice in 2017-2018. The views presented here are those of the authors and should not be attributed to the Commonwealth Fund or its directors, officers, or staff.
Download the Article
Read on Wiley Online Library
Published in 2019