Aging Policy From a Multilayered Geographic and Life Course Perspective

Contemporary challenges to healthy aging in communities across the United States are shaped by the complex, multilayered nature of the federal, state, and local contexts in which aging occurs. We introduce a framework for thinking in this multilevel way, as well as acknowledging the differences experienced by more and less socially advantaged groups as they age through the life course in sometimes distinct health contexts. Our framework draws on two prominent models—the ecological model (Bronfenbrenner, 1979; Smedley & Syme, 2000) and the life course perspective (Elder et al., 2003)—but focuses specifically on key elements of space and time relevant for the within- and between-country disparities we discuss. We highlight the promise and challenge of this framework for improving research and policy aimed at supporting healthy aging and reducing disparities in health and longevity.

Healthy Aging Across the United States: Increasingly Unequal

While the United States spends more on health care than any other wealthy nation, by a considerable margin, the population health returns are disappointing (House, 2015). The U.S. health and longevity crisis—marked by the faltering standing in morbidity and life expectancy relative to peer nations—has occurred despite massive spending on health care and has galvanized attention from researchers and the media in recent years. It is often illustrated by contrasting post-1980 trends in U.S. life expectancy versus those in other high-income countries. For example, Figure 1 shows trends in female life expectancy across 22 high-income countries, where the United States is the enlarged dot. The U.S. trend is unequivocally alarming. Moreover, the poor performance of the country as a whole obscures profound disparities between U.S. states. Overlaying states with the highest (Hawaii) and lowest (West Virginia) female life expectancy in 2018, Figure 1 makes it clear that some states are actually performing on par with high-income countries, while others are extremely far behind. Within-state geographic units also show substantial heterogeneity, with gains in life expectancy at age 65 greatest in large metropolitan areas and coastal regions, while nonmetropolitan areas and interior areas of the United States lag behind other parts of the country and all Organization for Economic Cooperation and Development (OECD) comparison countries (Vierboom & Preston, 2020).

Figure 1.

Life expectancy in high-income countries, 1979–2017. The United States is the square; Hawaii (HI) and West Virginia (WV) are the extremes within the United States.

Such striking disparities across states exist for many outcomes, including life expectancy at age 50 years and morbidity, disability, and cognition among midlife and older adults (Montez et al., 2017; Wilmoth et al., 2011). These pronounced disparities in midlife and older adult health and longevity across U.S. states underscore the necessity of incorporating a nuanced geographic perspective in explaining the U.S. crisis. For example, a recent study (Montez et al., 2019) documenting educational disparities in adult mortality at the U.S. state level from 1986 to 2011 found that: (1) disparities widened in only 24 states; and (2) the rising mortality of low-educated adults (Case & Deaton, 2015; Montez et al., 2011) that has received much attention has not occurred in all states.

Another crucial factor to consider is that the disparities across U.S. states were not always so pronounced. For instance, in 1980, life expectancy at birth was just 1.6 years longer in New York than Mississippi; by 2014, it was 5.5 years longer (Montez, 2017). This post-1980 divergence is alarming in part because it reflects large gains in some states alongside stagnation or absolute declines in others (Fenelon, 2013). Those states making sizable gains tend to be the same ones that actively implemented policy changes that aggressively invest in residents across the life span. States like New York increased spending on public education, raised the minimum wage, enacted their own Earned Income Tax Credit, levied high tobacco taxes, and expanded Medicaid, for example. While states like New York have been raising the ceiling on life expectancy, states like Mississippi have kept the floor intact, with considerably less investment in broad-based social, economic, and health policies. This backdrop must be taken into account when explaining current levels of and disparities in U.S. adult health and longevity. It also underscores that the crisis was not created in recent years; rather, it has been percolating for decades.

Looking Beyond Nations and States

Further complicating research that attempts to understand poor population health in the United States is the reality that people are embedded in multiple contexts, such as counties, cities, neighborhoods, and either urban or rural areas. Each layer may have an independent influence on health and longevity, or the influence of one layer may depend on another. For instance, state vaccination policies may affect population health, and so too may living in a rural area; however, those policies may have weaker consequences for adults residing in rural areas because of their lower population density and, consequently, lower risk of infection. Additionally, contexts like local labor or housing markets may vary drastically in conditions that shape opportunities for health and aging, even if state policies are applied without regard to these differences. Investigating how the conditions and policies prevailing in state and local contexts might interact is essential for understanding how and why they may have contributed to the U.S. health and longevity crisis. It is essential for developing appropriately targeted strategies and interventions.

Other findings make a nuanced multilevel approach necessary. Local areas may be more similar within than between states, which seems to be the case for life expectancy (Arcaya et al., 2012), so ignoring state boundaries when studying neighborhood effects may be limiting. The relative importance of state and local contexts also could change across the life course; a recent study showed that U.S. state and local contexts jointly predict adult disability risk, but the relative importance of each geographic level changes across the life course (Montez et al., 2017).

When and for Whom Does Place Create Disparities?

Although the disparities in a range of later-life population health outcomes across places are well documented, it is unclear when in the life course they emerge and why. For instance, are the disparities in older adult cognition across states mostly due to state-level differences in childhood conditions, such as states’ compulsory schooling laws and vaccination strategies? Or do they primarily reflect conditions in adulthood, such as states’ tobacco control strategies and minimum wages? Might they reflect conditions across the life course? In other words, is the lower life expectancy in West Virginia relative to Hawaii in Figure 1 due to an amalgamation of adverse childhood conditions in West Virginia, adverse adult conditions in West Virginia, or a lifetime of adversities among those who live their full lives in West Virginia? What role does geographic selection (i.e., immigration and interstate migration) play in these patterns? And what do the answers to these questions imply about potential policy strategies to shrink disparities and improve population health? Recent work underscores the necessity of integrating both the life course and geographic perspectives to address these and related critical questions (Montez et al., 2017).

Additionally, the evidence increasingly suggests that the importance of state contexts for health varies by socioeconomic status (SES; Montez et al., 2017; 2019), gender (Montez et al., 2016), and race/ethnicity, with marginalized groups generally most vulnerable to the conditions that prevail in their state of residence. Historically, gains in U.S. adult health and longevity occurred broadly across sociodemographic subgroups and geographic areas, although some subgroups and areas gained more than others. During the 1960s and 1970s, high school graduates made greater gains in life expectancy than did nongraduates (Crimmins & Saito, 2001), and certain U.S. states and counties made greater gains than others (Ezzati et al., 2008; Fenelon, 2013). However, there has been a fundamental change in recent decades. Since the mid-1980s, some sociodemographic subgroups and geographic areas continued to make impressive gains, while others experienced small gains or even absolute declines (Ezzati et al., 2008; Fenelon, 2013; Kindig & Cheng, 2013; Montez et al., 2019). In other words, the U.S. adult health and longevity experience has transitioned from one in which “some benefit more than others” to “some benefit and some lose.” The trends are particularly worrisome for middle-age adults, women, adults with low levels of schooling, and adults residing in certain states and rural areas in the South or Midwest (Case & Deaton, 2015; Fenelon, 2013; Ho, 2013; Sasson, 2016). Despite extensive research attention and media coverage of these trends and disparities, we have only a limited understanding of their causes, and have identified relatively few strategies to reverse them. One hypothesis that has emerged for the disparate trends focuses on systemic racism, which has been shown to be meaningfully shaped by state-level policy (Smith et al., 2020). Improving public safety vis-à-vis law enforcement is a possible strategy toward reducing the harmful effects of systemic racism (Alang et al., 2017). A second hypothesis focuses on ageism, which could be combated by moving toward a more universal design in environments and policies that addresses needs for persons all across the life course.

Importantly, it is becoming increasingly clear that the growing socioeconomic status (SES) and geographic disparities are interrelated (Chetty et al., 2016; Montez et al., 2017; 2019). Simply put, the importance of SES for adult health and longevity varies across geography, the importance of geography depends on one’s SES, and the relative importance of geography and SES (and other social factors) changes across the life course. A clearer understanding of how social and geographic exposures jointly shape health and mortality across the life course holds great promise for explaining the U.S. crisis.

A clearer understanding of how social and geographic exposures jointly shape health and mortality across the life course holds great promise for explaining the U.S. crisis.

What is Needed?

The research needs we have identified are challenging to address. They will necessitate coordination across data sources collected at different levels and the development of methods to model the multidimensional nature of contextual data. For example, the sheer number of possibly relevant state-level contextual characteristics and the fact that they tend to cluster (e.g., states with low tobacco taxes tend to have low minimum wages and be in the South) present formidable challenges, conceptually and methodologically. Modeling these exposures across the life course adds further complexity. One group that has begun this work is the Network on Life Course Health Dynamics and Disparities in 21st Century America (https://gero.usc.edu/nlchdd/), funded by the National Institute on Aging. The work of many core members of the Network is included in the present chapter and the remainder of the issue. In its initial five years, the Network laid crucial groundwork by advancing understanding in five specific areas of focus: (1) trends in the health of American women; (2) widening SES disparities in health; (3) changes in racial/ethnic disparities in health; (4) the inferiority of U.S. population health as compared to other high-income countries; and (5) developing a better initial understanding of the internal geographic disparities in U.S. population health. In its new cycle of funding, the Network is focusing explicitly on understanding the complex, multilayered nature of the federal, state, and local contexts in which aging occurs across the life course, as well as inequalities in the way these multilayered contexts exert their influence. Already, Network members have provided key publications and a collection of curated state-level data resources (https://gero.usc.edu/nlchdd/resources/), enabling a broader community of researchers to contribute to research and policy to reduce geographic disparities in healthy aging. The Network will also encourage users to engage policymakers to think about “health in all policies” (Schoeni et al., 2008) in ways that reflect geographic and life course complexity in healthy aging.

Funding

We gratefully acknowledge grant R24 AG045061 from the National Institute on Aging, which has supported the research network that has stimulated our work in this area.

Conflicts of Interest

None reported.